Google
This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project
to make the world's books discoverable online.
It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject
to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books
are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover.
Marks, notations and other maiginalia present in the original volume will appear in this file - a reminder of this book's long journey from the
publisher to a library and finally to you.
Usage guidelines
Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing tliis resource, we liave taken steps to
prevent abuse by commercial parties, including placing technical restrictions on automated querying.
We also ask that you:
+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for
personal, non-commercial purposes.
+ Refrain fivm automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the
use of public domain materials for these purposes and may be able to help.
+ Maintain attributionTht GoogXt "watermark" you see on each file is essential for in forming people about this project and helping them find
additional materials through Google Book Search. Please do not remove it.
+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner
anywhere in the world. Copyright infringement liabili^ can be quite severe.
About Google Book Search
Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web
at|http: //books .google .com/I
SULPHURIC ACID AND ALKALI
WORKS BY SAME AUTHOR.
The Manufacture of Sulphuric Acid and Alkali*
A Theoretical and Practical Treatise.
Vol. I. in Two Parts (1200 pp.), not sold separately. Sul-
phuric Acid. Third and much Enlarged Edition.
Just published - - £2 12 6
Vol. II. — Salt CakCy Hydrochloric Acid^ and Lcblanc Soda.
(900 pp.) Second Edition, revised and enlarged £2 2 0
Vol. III. — The Ammonia-Soda and various other processes
of Alkali-makinff, &c. (800 pp.) Second Edition
revised and enlarged - - - £2 2 0
Coal-Tar and Ammonia* Third and Enlarged Edition.
(900 pp.) - - - - -£220
Handbook of Technical GaS'Analysis* By
Clement Winkler, Ph.D., and translated by George
Lunge, Ph.D. (200 pp.) Second English Edition lOs. 6d.
GURNEY & JACKSON, Paternoster Row, London, E.C.
(Successors to Mr. Van Voorst.)
THEORETICAL AND PRACTICAL TREATISE
ON THX
MANUFACTURE
SULPHURIC ACID AND ALKALI,
WITH THB
COLLATERAL BRANCHES.
GEORGE LUNGE, Ph.D.,
rROFUSOR OF TICHHICjtL OIIMIBTKT AT THB FKDRRAL POLmCDKIC ICHOOL, miCH.
(FOaHBRLV UANAOBB OF THR TYHB ALKALI-WOBEB, SOUTH BHIBLDB.)
THIRD EDITION, REVISED AND ENLARGED.
VOLUME I.— Part ]
SULPHURIC ACID.
LONDON:
GURNEY AND JACKSON, PATERNOSTER ROW.
(Sdccbbbobs to Mr, VAN VOORST.)
CW«ywN.%c\c^(^,03
f
■■il
- X i ' . ' ? V
^
/
LKAJ^A^0^
PBIVTSD BY TAYLOB AND FRAVCIS,
BBD LION COURT, FLEET STREET.
PREFACE.
The Second Edition of Vol. I. of 'Sulphuric Acid and Alkali/
embracing the Manufacture of Sulphuric Acid, appeared in 1891^
but a copious Appendix at the end of Vol. III. brought the subject
down to 1896. It might perhaps appear somewhat early to publish
another edition of such a voluminous treatise *^ but the first glance
at this book will prove that there was sufficient cause for such
a proceeding. Not merely the enormous development of the
manufacture of sulphuric anhydride and even of sulphuric acid
itself by the contact-process required an altogether new treatment
of the subject, but also the vitriol-chamber process with all its
ramifications, which in all probability will yield the principal
portion of ordinary sulphuric acid for many years to come, has
experienced so many changes and improvements that its descrip-
tion ought to be once more brought up to date.
A mere compilation of the voluminous literature of the subject,
of patent specifications, and so forth, even combined with frequent
visits to acid-works, would not have achieved my purpose — to
furnish chemical manufacturers with a trustworthy guide for actual
practice, as well as exhaustive scientific and technological informa-
tion for all students of this branch of industry. But fortunately
I have been able to enlist for this edition, like its predecessors,
the co-operation of a large number of experienced practical acid-
* It is not intended to publish a third edition of Vols. II. and III. for some
time to come.
a2
IV PREFACE.
maniifactttrers who are far-seeing and large-hearted enough to
place their knowledge at my disposal for this book^ and therefore
for everybody in a similar position, without giving way to the
somewhat narrow-minded apprehension that by enlightening their
neighbours they might injure their own interests. This senti-
ment has certainly not been universal. The great acid-trusts in
several countries do not give any information, even on such
apparently trivial matters as statistics. On the other hand,
the United States Census Bulletin, No. 210, compiled by Messrs.
Munroe and Chatard, and issued in June 1902, contains much
that is welcome to all practical men, over and above the com-
pletest and most carefully worked out statistics ever published
in that line.
If a jealous restriction of experience within the precincts of
one's own business circle were really practised in all cases, it would
be a very bad thing, not merely for the cause of technology and
therewith of technical progress in general, but even for the largest
and most potent manufacturers, who, after all^ are not all-powerful
or omniscient. But that such a spirit is fortunately far from
universal, is best proved by the possibility of compiling a book
like the present.
For the lead-chamber process I have received (irrespective of a
number of smaller notes) valuable remarks or long communications
from the following gentlemen or firms, to whom my best thanks
are herewith publicly tendered * : —
P. Benker, 129 Rue Martre, Clichy, Paris; Dr. Brauning,
Manager of the Government Works at Oker; G. E. Davis,
32 Blackfriars Street, Manchester; The Royal Saxon Govern-
ment Works, Muldenerhiitte, Freiberg; Herman Prasch, Cleve-
land, Ohio; W. C. Heraeus, Hanau; L. Kessler, Clermont-
* This refers merely to the third edition ; the names of those who rendered
similar services to me iu the previous editions are not repeated here.
PREFACE. V
Ferrand ; Dr. G. Krell, Hiisten - Bruchhausen (Westphalia) ;
Metallnrgische Oesellschaft^ Frankfurt-a.-M.; H. H.Niedenfuhr^
3 Friedrich-Wilhelmstrasse^ Berlin N.; Chemische Fabrik Rhe-
nania^ Stolberg (through the late Dr. B. Hasenclever); Dr. A.
Zanner (Brussels).
So far as the contact-process is concerned, I have been enabled
through the special communications of nearly all those who have
been mainly instrumental in developing it^ especially of the large
firms concerned^ to completely elucidate, for the first time, the
history of this branch of industry which has now acquired such
great proportions^ and to describe its present state so far as this is
compatible with business interests. A host of hitherto unknown
facts is contained in the communications of Dr. Arndt, Dr. Jacob,
Dr. Krauss (Hochst), Dr. Messel, Monsieur Meunier - DoUfus,
Dr. Babe, Dr. Schroeder, Verein chemischer Fabriken in Mannheim,
and Professor Clemens Winkler. The Badische Anilin- und
Sodafabrik had already allowed Dr. Knietsch to publish a great
number of interesting facts in his celebrated lecture delivered
before the Berlin Chemical Society, and that firm has favoured me
also with new valuable information for this book.
All contributions to this book by other persons are acknowledged
as such in their proper places.
The mass of material thus placed at my disposal would have
swelled this book to an altogether inconvenient size^ if I had not
made some room by cutting out part of the former contents.
When preparing this third edition, I adopted the same plan as in
the second edition — shortening many descriptions and discussions
and entirely omitting others, also making innumerable alterations
in details and rewriting whole chapters. But all the more
important omissions of this kind are pointed out in the text
with reference to the former editions, so that the reader of this
third edition is always made aware of the existence of older
n PREFACE.
materials. This seems the proper course for a work of this kind^
whose object^ among others^ is to be of service to inventors as a
book of reference, to show them what has been previously done in
the field, and thus, on the one hand, to suggest to them ideas for
further improvements, and, on the other hand, to spare them
disappointments in patent matters. Some, indeed, will miss in
this edition various items contained in its predecessors, whilst
others may think that I ought to have restricted myself to a
description of such materials and apparatus as are still in use, and
of such views as are now prevalent. It will never be possible to
supply each reader with everything, and only everything, he
happens to require ; I had to content myself with taking a middle
course.
In this edition I have not scrupled to replace even valuable and
voluminous illustrations by new ones, where I was able to obtain
the most recent descriptions. No doubt some readers will miss,
for example, the splendid drawings of Schaffner^s shelf-burner,
from which hundreds of such burners have been erected, but I
thought it a waste of space to give these again, along with the new
drawings by Niedenfiihr. In the same way I have sacrificed my
own drawings of the Glover tower, of the vitriol-chambers, and
many others. Those who wish to consult them will find the
former editions referred to in the respective places. A copious
index will facilitate the use of this book.
It is now twenty-four years since the first edition of this volume
saw the light. The two former editions have earned for the
author every commendation he could desire, and more than he
honestly could have claimed for himself. It will not, I hope, be
thought egotistic if I say that hardly anybody is better able to
appreciate the weak sides of this work than its author. Errors,
omissions, matters of questionable value will be found by the
expert reader here and there. This third edition is still very
PREFACE. Vn
far removed from the ideal of such a treatise as it exists in the
mind of the author. But nobody can do more than is in his
power, and such readers as know (or think they know) this or
that thing better, and who may be inclined to sharp criticism,
had best ask themselves the question whether they would come
up to the ideal if they had to write such a book, and whether
they cannot derive some valuable information even from this im-
perfect performance, which probably represents the last opportunity
the author will have of treating the subject.
Zurich, August tfth, 1903.
CONTENTS.
Page
Fb£fac£ iii
GoHTisirn ix
IlTTROOUOlION 8
Definition of Snlphurio Acid and Alkali Manufacture, 3 ; Sulphuric
Acid, 4; Sulphate, Leblanc process, Hydrochloric Acid, '5;
Eecovery of Sulphur, 6.
CHAPTEK I.
HiSTOBlCAL AND GeNEBAL NoTES ON THE MaNUFACTUBE OF SuLPHUBIC
Acid 7
History of the Manufacture of Sulphuric Acid, 7.
General Principles of the Manufacture of Sulphuric Acid, 11;
Division of the Subject, 13.
CHAPTER II.
The Baw Matbeials of the Svlphubic-Acid Manufaotube (ineluding
NiT&ic Acid) 14
1. Natural Sulphur (Brimstone) 14
Historical, 14; properties of sulphur, 15; allotropio conditions, 15;
thermic properties, 16 ; combinations, solubility, 16 ; refined
sulphur, 16.
Natural occurrence of sulphur, 17 ; Sicilian sulphur industry, 17 ;
new processes for extracting sulphur, 18 ; analyses of Sicilian
sulphur, 19 ; prices, exportation, 20 ; sulphur in Northern
X CONTENTS.
Page
Italy, 20 ; Spain, Germany, Austria, Russia, 21 ; Africa,
Iceland, Japan, 22 ; New Zealand, United States, 23 ; other
parts of America, 24.
Sulphur manufactured from other materials, 25 ; from spent oxide
of iron, 23 ; from pyrites, 26 ; from sulphur dioxide, sul-
phuretted hydrogen, 29.
World's production of sulphur, 30.
Technical analysis of sulphur, 31 ; commercial grades, 3 1 ; detection
of arsenic, 31 ; selenium, 32 ; direct estimation by solubility in
carbon disulphide, 32 ; degree of fineness, Chancel's sulphuri-
meter, 33 ; refined sulphur, flowers of sulphur, 34.
2. Pyntes 34
Properties of pyrites, 34 ; of mareasitie, 35 ; magnetic pyrites (pyr-
rhotite), 36 ; copper-pyrites, 36.
First application of pyrites, 37 ; introduction of cupreous pyrites, 39.
Sources of pyrites, 40 ; Great Britain, 40 ; Ireland, 41 ; coal-brasses,
41 ; importations of foreign pyrites into Great Britain, 42 ;
pyrites in Germany, 42 ; in Austria, 45 ; Switzerland, Poland,
Belgium, 46 ; France, 47 ; Italy, 50 ; Sweden, Norway, 5J ;
Spain and Portugal, 53; United States, 57; Canada, New-
foundland, Australia, 59.
World's production of pyrites, 59 ; average composition of the
world's pyrites, 60 ; prices, 61 ; proportional value of poor
and rich pyrites, 61.
Analysis of pyrites, 62 ; decomposing the ore in the wet way, 62 ;
precipitating by barium chloride, 64 ; previous removal of the
iron, 66 ; Lunge's method, 67 ; other methods, 69 ; estimation
of sulphates by titration, 70 ; estimation of available sulphur,
72 ; expeditious pyrites assays, 74 ; testing burnt-ore, 75 ;
magnetic pyrites, 76 ; marcasite, 76 ; estimation of other con-
stituents of pyrites, 77 ; arsenic, 77.
3. Zinc-blende 79
Occurrence, 79 ; composition, 80 ; analysb, 81.
4. Other Metallic Sulphides 81
Noxious vapours, 82 ; total acids contained in these, 83 ; from
copper-pyrites, 83 ; materials used at Oker, 83 ; galena, 85 ;
coarse metal, 85 ; lead-matte, 85.
CONTENTS. XI
Page
5. By-jproducU of other Manufaetureg 85
Spent oxide of gas-works, 85 ; alkali waste, 88 ; sulpharetted
hydrogen £rom the manufacture of sulphate of ammonia, 89 ;
other sources of sulphur dioxide, 89.
6. NiiraU of Soda 89
Properties, 89 ; solubility, 90 ; occurrence in Chili, 91 ; in other
places, 93 ; potassium nitrate, 93 ; consumption, 93 ; treatment
of nitre-bags, 94 ; composition of commercial nitrate of soda,
94 ; occurrence of potassium and of perchlorate, 95 ; analysis
of nitrate of soda, 96 ; '' refraction," 96 ; nitrometer method, 96.
7. NUrie Acid '. 98
Boiling-points, 98 ; specific gravities, 99 ; correction for temperature,
101 ; influence of lower nitrogen oxides, 102 ; oxidizing pro-
perties, 103.
Manufacture of nitric acid, 103 ; French cylinders, 104 ; Griesheim
cylinders, 107; French stiUs, 111; Valentiner's stills, 111;
Guttmann's still, 114; other stills, 115; Prentice's continuous
still, 115; Uebel's process, 116.
Cement for nitric acid, 120 ; asbestos cement, 120.
Gases and vapours evolved during the working of the retorts, 121 ;
Volney*8 observations, 121.
Condensation of nitric acid, 123 ; old receiver apparatus, 123 ;
cooling pipes and worms, 124 ; condensing exit-gases by coke-
towers, 124 ; by plate-towers, 125 ; Bohrmann-Lunge*8 system,
125; recovering nitric acid from lower oxides, 125; Nieden-
fiihr's apparatus for this purpose, 127 ; refining or bleaching
nitric acid, 127 ; Griesheim condensing-process, 129 ; Gutt-
mann's system, 131 ; various systems, 132 ; Valentiner's
vacuum process, 134.
Concentration of nitric acid, 139.
Nitre-cake, 139.
Cost of manufacturing nitric acid, 140.
Utilization of nitric acid from nitrating processes (nitrobenzene,
nitroglycerine, &c.), 141.
Various processes for the manufacture of nitric acid, 142.
Protection against accidents by nitric acid, 145.
Transportation of nitric acid, 145.
Pumping of nitric acid, Pulsometers, 146.
Analysis of nitric add, 147 ; impurities in commercial nitric add, 147.
XII CONTENTS.
CHAPTER III.
Page
The Pbopbbtibs ajtd Analtbib of the Tbohnicallt £hplot£d Oxides
and aoidb of sulphub 149
Stdphur Dioxide (Sulphurous Add) 149
Properties of sulphur dioxide, 149 ; reactions, 151.
Sulphurous aoid, 152 ; soluhility, 152 ; production from sulphuric
acid and charcoal, 153 ; salts, 154.
Injurious action of SOj on health and vegetation, 154 ; noxious
vapours (acid-smoke), 154.
Detection and estimation of sulphurous acid and sulphur dioxide,
159 ; qualitative reactions, 159 ; quantitative estimation, 160.
Applications of sulphurous acid (sulphur dioxide), 161.
Sulphuric Anhydride (Sulphur Triaxide) 162
Pyrondphuric Acid 164
Nordhausen or fuming oil of vitriol, 165 ; melting- and boiling-
points, 166 ; specific gravities, 166 ; specific heats, 169 ; heats
of solution, 170 ; various properties, 170.
Sulphuric Acid 170
Occurrence, 170; monohydrated sulphuric acid, 171; dissociation,
vapours, 172 ; formation, 172 ; sulphuric acid containing
water, 173 ; distilled acid (critical concentration), 173 ;
absorption of SO, at that concentration, 175 ; rectified oil
of vitriol, 176 ; hydrometers, 176 ; Baume's hydrometer,
176 ; Twaddell's hydrometer, ] 78 ; specific gravities of dilute
sulphuric acid, 179 ; table by Lunge, Isler, and Naef, 180 ;
maximum density, 186; correction for temperature, 186;
hydrates of sulphuric acid, 187 ; influence of impurities on spe-
cific gravities of sulphuric acid, 187 ; of sulphurous acid, 188 ;
of nitrogen acids, 189 ; of arsenic, 189 ; of lead sulphate, 190 ;
unreliability of hydrometric testing for high specific gravities,
190 ; Anthon's table for mixtures of sulphuric acid and water,
190 ; melting-points of sulphuric acid, 191 ; boiling sulphuric
acid, 193 ; boiling-points, 193 ; tension of aqueous vapour in
sulphuric acid, 195 ; specific heat of sulphuric acid, 197.
Chemical behaviour of sulphuric acid, 198 ; rise of temperature on
mixing sulphuric acid with water, 198 ; effects of the affinity of
sulphuric aoid for water^ 200 ; decomposition of sulphuric acid,
1
CONTENTS. XUl
Page
201 ; add propertiefi, 201 ; salts (stdphates), 202 ; action on
metals, 202 ; on platinum, 203 ; on cast-iron, 203 ; ou wrought-
iron, 204 ; on lead, 206 ; sine, tin, 211.
Behaviour of sulphurous and sulphuric add towards the oxides of
nitrogen, 212 ; nitrous oxide, 212 ; nitric oxide, 21 2 ; nitrous acid,
213 ; chamher-crystals (nitrososulphuric add), 215 ; nitrogen
peroxide, 221 ; tendon of nitrous add in presence of dilute
sulphuric acid, 227 (tahles, 228) ; behaviour of nitrososulphuric
acid towards reducing-agents : sulphur dioxide, 227, coke,
231.
Analysis of sulphuric add, 2S2 ; qualitative detection, 232 ; quanti-
tative estimation, 233 ; volumetrical estimation, 233 ; litmus,
phenolphthalein, methyl-orange, 234 ; standard alkaline solutiou,
236; standard acid, 236 ; analysis of fuming oil of vitriol, 238.
Taking the sample, 238 ; weighing in bulbs, 239 ; in glass tap-tubes,
240 ; bulb-tap pipette, 241 ; sampling solid anhydride, 242 ;
Frdberg method, 242 ; table for comparing free and combined
SO,, 243; producing mixtures with a given strength of SO,, 244;
influence of SO, on testing, 244 ; testing-methods based on
suppressing the filming, 24t5 ; definition of percentage of SO,,
247.
Estimation of the impurities of sulphuric acid, 248 ; qualitative
detection, 248 ; arsenic, 248 ; volatile impurities, 249 ; nitrogen I
adds, 249 ; selenium, 252 ; quantitative estimation, 252 ; of ^
nitrogen acids, 253 ; total nitrogen acids, 253 ; nitrometer method,
253 ; reduction of gas-volumes fpr normal conditions of pressure ]
and temperature, 258 ; gas-volumeter, 259; estimation of nitrous j
acid (nitrososulphuric acid) by the permanganate method, 262;
table for testing nitrous vitriol, 264 ; colorimetric estimation of
small quantities of nitrous acid, 266 ; of nitric acid, 267.
CHAPTER IV.
The Pboductioh of Sulphub Dioxide 268
A. Brimstone-burners 268
Old burners, 268; improved burners, 272; Stahl's burners with
boiling-down pans, 274 ; coutinuous burners, 277 ; Blair's, 277 ;
H. Glover's, 279 ; de Hemptinne's, 280 ; burners for wood-pulp
works, 283 ; residue from sulphur-burners, 285 ; cooling the gas
from brimstone-burners, 286.
XIV CONTENTS.
Pag©
B. The Production of Sulphur Dioxide from Pi/rites 286
1. Breaking ihspyrites 286
Breaking by hand, 286 ; Bieving, 287 ; mechanical stone-breakers,
287.
2. PtfriteS'humers for lumps 29J
Necessity of keeping lumps and smalls apart, 291 ; burning in heaps,
292 ; in kilns or burners, 292 ; Freiberg kilns, 293 ; burners
with grates and ash-pits, 295; English grate-burners, 296; gas-
flue, 300 ; grate-bars, 300 ; ash-pit, 303 ; tilting-box for dis-
charging the burnt ore, 303 ; arrangements for preventing any
blowing out during discharging, 303 ; Rhenania burners with
such arrangement, 305 ; doors of burners, 307 ; brickwork, 307 ;
size of burners, 308 ; grate-surface, 308 ; sets of burners, 309 ;
concentrating-pans placed on top of burners, 310 ; potting the
nitre in burners, 311 ; special kinds of pyrites-burners, 311 ;
burners for copper-matte, 312.
Working of the pyrites-burners for lumps, 313; starting the burners,
313; objects aimed at, 314; sulphur remaining in cinders, 315;
appearance of cinders, 315 ; testing them for sulphur, 316 ;
composition of cinders, 316 ; residue from treating metallurgical
products, 318 ; admission of air, 319 ; subliming of sulphur,
319 ; slags (scars), 319 ; behaviour of different ores, 320 ; regu-
lation of draught, 321 ; work of burner-men, 323 ; regulating
the heat, 324 ; influence of dampness, 326 ; general remarks,
326.
3. Burning pyrites-smalls 327
Burning smalls together with lumps, 327; balls made with clay,
328: (a) burning smalls in coal-fired furnaces, 329; (b) by
the heat of burners for lumps, 330; Olivier & Ferret's
furnace, 330 ; Hasenclever & Helbig's tower-furnace, 331 :
(c) burning smalls without external heat, 331 ; balling pyrites-
dust by grinding with water and drying (pugging), 331;
Gerstenhofer furnace, 332 ; Maletra's shelf-burner, 334 ;
Aussig burner, 337 ; working shelf-burners, 339 : combination
with lump-burners, 342 ; other descriptions of shelf-burners,
343.
Mechanical dust-burners, 343 ; MacDougall's furnace, 343 ; modifi-
cations, 346 ; Frasch's burner, 348 ; Herreshoff burner, 349 ;
CONTENTS. XV
Page
other mechanical dust-burners, 351; Spence's dust-bumer» 352;
other burners, 356; Bruckner's, 367; Walter's burner for
" peas," 367.
4. Furnaces for roasting zine-hlende 368
Nuisance caused by the open roasting of blende, 368 ; development
of processes for utilizing the sulphur. 368 ; Hasendever and
Helbig's furnace, 369 ; Liebig's furnace, 360 ; Khenania furnace,
361 ; other blende-furnaces, 363 ; mechanical furnaces, 364 ;
treatment of complex ores containing blende, 366.
5. Burners for the spent oxide of gas-worJes 366
6. Burners for sulphuretted hydrogen 367
7. Processes for absorbing sulphur dioicide contained in aeid-smohe,
fire-gases^ and the like 370
Dealing with ordinary coal-smoke, 370 ; chimneys, 371 ; dilution with
air, 371 ; smoke from glass-works, 372 ; condensing by water,
372 ; by sulphuric acid, 373 ; absorption by caustic lime, 373 ;
by limestone, 374; by magnesium or aluminium hydrate,
376 ; by zinc carbonate or oxide, 376 ; by ferric oxide and coal,
376 ; by metallic iron, copper, zinc, 377 ; by copper and air,
377 ; by hydrogen sulphide, 377 ; by calcium or barium sul-
phide, 378 ; by vegetable charcoal, 379 ; by passing over red-
hot coals, 379.
8. Preparation of sulphur dioonde in the pure state 379
Old methods, 379 ; from sulphuric acid and coal, 379 ; from sulphuric
acid and sulphur, 380 ; from sulphuric acid and ferrous sulphide,
380 ; by heating ferroas sulphate with sulphur, 380 ; by burning
sulphur in air, 381 ; process of Schroeder and Haenisch for the
preparation of liquid sulphur dioxide, 381 ; vessels for sending
out liquid sulphur dioxide, 386 ; uses for liquid sulphur
dioxide, 387.
9. Draught^pipes and flues 388
Cooling the gases by cast-iron pipes, 388 ; utilizing the heat of the
gases, 389 ; Albert's process, 389.
Flue-dust, 390; removal, 392; dust-chambers, 393; Freiberg
cooling-flues, 394 ; centrifugal dust-catcher, 396.
Purification of sulphur dioxide from SO, for the manufacture of
wood-pulp, 397.
XVI CONTENTS.
Page
10. ITie burner^as 397
CompoBition of the burner-gas from burning brimstone, 397.
Composition of the bomer-gas from burning pyrites, 399.
Sulphur dioxide for manufacturing calcium bisulphite (in the manu-
facture of wood-pulp &c.), 401.
Composition of the gas from blende-furnaces, 401.
Sulphuric anhydride in burner-gas, 402.
Actual percentage of snlphur dioxide in burner-gas, 407 ; in gases
from poor ores, 408.
Comparison of brimstone and pyrites as material for sulphuric-acid
making, 409.
The quantitative estimation of sulphur dioxide in burner-gas, 411 ;
Reich's method, 411 ; Lunge's test for total acids in the burner-
gas, 416 ; estimation of oxygen in burner- and chamber-gases,
417 ; by pyrogallol, 417 ; by phosphorus, 418 ; estimation of
the oxides and acids of nitrogen in gaseous mixtures, 419 ;
Lunge and Naef's method, 419 ; estimation of nitrous oxide,
419.
CHAPTER V.
CovsTBrcnoN of the Lbad Chambkbs 421
Necessity of large spaces, 421 ; suitability of lead, 421 ; proposals
for other materials, 422.
The erection of lead chambers above ground, 422 ; foundation, 423 ;
pillars, 423 ; of wood, 423 ; of brickwork, 424 ; of stone, 425 ;
of iron, 425 ; sleepers and joists, 426 ; floor, 427 ; frame of
wood, 428 ; of angle-iron, 430 ; lead for chambers, 430 ; thick-
ness, 430 ; quality, 431 ; joining lead by solder, 433 ; by
rabbets, 433 ; by burning, 433 ; erecting a lead chamber, 437 ;
straps, 438 ; chamber-top, 441 ; chamber-bottom, 445 ; details
of chamber construction on the English plan, 448 ; chamber
buildings, 450.
Renewal of the chambers, 450 ; repairs, 450 ; pulling-down, 452 ;
wear and tear of chambers, 453.
Shape of lead chambers, 454; Rhenania chambers, 454; H. A.
Smith's proposal, 455 ; Delplace's chambers, 455 ; Th. Meyer's
tangential chambers, 455.
Combination of chambers to sets, 458; Benker's chambers, 459
size of chambers, 460 ; connecting-tubes, 464.
CONTENTS. XVll
Page
Total cubical contenta of chambers, 465 ; forced or high-pressure
work, 468 ; chambers worked by G. E. Davis, 469 ; chamber-
space for poor ores, 470 ; iuflueuce of consumption of nitre,
470 ; of temperature, 470 ; various terms of expressing chamber-
space, 471.
Proposals for diminishing cTiamher- space, 471. Use of pure oxygen,
471 ; employment of a very large quantity of nitre, 472 ; better
mixture of gases and increase of condensing^pace, 472 ; Ward's
glass sheets, 472 ; chambers with glass partitions, 474 ; with
brick partitions, 475 ; filled with coke, 475 ; apparatus of
Verstraet, 476 ; of Lardani and Susini, 476 ; of Richters, 476 ;
of Pratt, 476 ; Burgemeister's cooling-shafts, 477.
Mixing- and cooling-towers, 477 ; Thyss towers, 477 ; Sorel's plan,
478 ; Lunge's plate-towers, 478 ; theory, 478 ; construction, 480 ;
functions, 484 ; practical application of plate-towers, 487 ; ex-
periences made in various places, 488 ; prescriptions for erecting
plate-towers, 492 ; NiedenfUhr's plan of acid-making by plate-
towers only, 493 ; by a combination of plate- towers and cham-
bers, 496 ; practical results obtained by adding plate-towers to
chambers, 497 ; differences of temperature and pressure in that
case, 498.
Other apparatus on the principle of plate-towers, 498 ; Hacker and
Gilehnst's pipe-towers, 499 ; Benker's towers, 499 ; Guttmann's
balls, 600 ; cone-towers, 500.
Question of completely abolishing the lead chambers, 501.
Chamber fittings^ 501. Drawing off the acid by taps, 501 ; siphons,
502 ; acid-valves, 506 ; acid-dishes (drips), 506 ; dippers for
taking samples of acid, 507 ; man-holes, 507 ; thermometers,
508 ; pressure-gauges, 508 ; acid-gauges, 509 ; sights, 510.
Apparatus for introducing nitric acid into the chambers, 511. Com-
parison of "potting" and of the use of liquid nitric acid, 511 ;
introduction of nitric acid as vapour ("potting''), 513; in-
troduction of vapours from ordinary retorts, 517 ; introduction
of liquid nitric acid, 517 ; Mariotte bottles, 519 ; tambours,
519 ; cascades, 521 ; introduction through the Glover towers,
522; by means of spray-producers (injectors), 523; storing
nitric acid, 527 ; introduction of nitre as aqueous solution of
nitrate of soda, 527.
Feeding the chambers with nitrous gases obtained as a by*product,
529 ; Dunlop's process, 529.
VOL. I. b
XYlll CONTENTS*
Page
Supply of water as steam or spray ^ 530. Steam, 530 ; steam-boilers,
530 ; use of exhaust-steam, 531 ; registering steam-gauges,
531 ; steam-pipes, 531 ; branches and taps, 532 ; pressure-
gauge behind tap, 533 ; automatic steam-regulators, 533 ; place
of steam-jets, 533 ; regulating steam -jets from one place, 534 :
total quantity of steam required, 535 ; emplo3*ment of water in
the place of spray, 536 ; Sprengel's system, 536 ; Griesheim
system, 537 ; objections to water-sprays and refutation thereof,
538 ; Benker's spray-producers, 540 ; Korting's spray-producers,
541 ; modification of these with glass nozzle, 541 ; filtering
water for this purpose, 542; Benker*s chambers for high-
pressure style with water-sprays, 543.
Arrangements for producing the draught in viirioUchamhers, 544.
Calculation of draught from the composition and temperature of
the gases, 544 : effects of excessive draught, 547 ; draught by
outlet-pipe from the last chamber, 549 ; by steam-injectors,
549 ; by chimneys, 550 ; by the Gay-Lussac tower, 551 ;
working more than one chamber by the same chimney, 552 ;
introducing air into the first chamber, 553 ; draught produced
by Glover tower, 553; regulating the draught by dampers,
653 ; by a diaphragm in the *' sight," 554 ; by automatic
arrangements, 555 ; Strype's regulator, 556 ; mechanical pro-
duction of draught, 559 ; fan-blasts, 559 ; anemometers, 562 ;
Fletcher's, 562 ; tables for, 567 ; Swan's, 570 ; various, 571 :
Seger's, 572.
Calculation of the volume of chamber-gases according to temperature
and moisture, 573.
CHAPTEK VI.
The Recovery of the Nitrogen Compounds 575
Reasons for this, 575 ; reactions of nitrogen oxides towards sulphuric
acid, 575 ; invention of Gay-Lussac's tower, 576 ; of Glovers
tower, 576 ; principle of Gay-Lussac tower, 577 ; towers in
series, 578 ; width, 579 ; dimensions, 580 ; foundations, 581 ;
framework, 581 ; leadwork, 582 : stone-towers, 583 ; packing,
584 ; coke, 584 ; stoneware packing, 586 ; plate-towers, 586 ;
repacking towers, 587 ; dealing with dangerous gases, 588 ;
complete design of a Gay-Lussac tower, 589 ; other apparatus
on the same principle, 590.
1
CONTENTS. XIX
Page
Distribution of the feeding-acid, 592; acid-wbeels, 593; overflow
apparatus, 595 ; regulation of supply, 598 ; Mariotte's yessel,
598 ; balancing-apparatus, 599 ; box-apparatus, 601 ; " Semper
idem " apparatus, 602 ; regulation by pulsometers, 603 ; cen-
tralized system (Griesheim), 603 ; pumping-apparatus, 605 ;
air-pumps, 606 ; acid-eggs, 607 ; automatic, 611 ; prevention of
splashing, 612 ; combination of acid-eggs, 614 ; Laurent's
pulsomefcer, 614 ; Kestner's, 618 ; tanks or reservoirs for acid,
619.
Working the Gay-Lussac tower, 620 ; drying and cooling the gas,
620 ; cooling the acid, 620 ; quantity of acid for feeding the
tower, 623 ; quality of nitrous vitriol, 624 ; presence of nitric
acid in nitrous vitriol, 625 ; of nitrogen peroxide, 626 ; process
of Brivet, Lasne, & Benker, 626 ; Benker's introduction of gas
containing SO^ before the tower, 627 ; regulation of the draught,
627 ; faulty working of the tower, 628 ; loss of nitre from this
cause, 628 ; exit-gases, ruddy vapours, 629 ; acids in exit,
629 ; proposals for absorbing these, 630.
Various plans for recovering the nitre in otber ways, 631.
Denitration of the nitrous vitrol, 631 ; necessity for, 631 ; shelf-
apparatus, 632 ; Gay-Lussac's denitrificateur, 632 ; steam-boxes,
634 ; steam-columns, 636.
Glover tower, 639 ; history, 639 ; objection raised against it and
refutation thereof, 639 ; functions of the Glover tower, 643 ;
construction of a Glover tower, 644 ; materials for lining, 645 ;
foundation, 646 ; frame, 646 ; leadwork, 647 ; lip, 648 ; top,
649 ; inlet- and outlet-pipes, 649 ; lining, 651 ; arch, 652 ;
packing, 652 ; avoiding excessive height of towers, 657 ; distri-
bution of acid, 657 ; cubic contents, 658 ; description of complete
Glover towers, 658 ; circular towers, 659 ; Liity's towers, 660 ;
Niedenfuhr's tower, 661 ; placing Glover and Gay-Lussac towers
together, 662 ; working of the Glover tower, 663 ; acid used for
feeding, 663 ; mixing the nitrous vitriol and chamber-acid out-
side the tower, 664 ; temperature of acid and gas, 665 ; Glover
towers connected with shelf-burners, &c., 666 ; with brimstone-
burners, 666 ; working Glover towers for various functions, 667 ;
extent of denitration, 667 ; concentrating action, 668 ; forma-
tion of fresh acid, 668 ; time occupied by each transfer of
oxygen, 670 ; proportion of Glover-tower space to chamber-
space with reference to the formation of acid, 671 ; proposal for
XX CONTENTS.
Page
pushing the work in the Glover tower heyond the usual rate,
672 ; drawbacks of the Glover tower, 673 ; impure acid, 673.
Denitration bj other means, 674.
CHAPTER VII.
The Chambek-Procbss 675
Starting the chambers, 675 : use of acid for luting the chambers, 675 ;
introduction of nitre and steam, 676 ; stopping chambers for
repairs, 677.
Supply of air, 677 : objects, 677 ; indications of draught, 678 ;
percentage of oxygen in exit-gases, 679 ; action of excessive
draught, 680 ; of insufficient draught, 681 ; regulation of draught,
681 ; increasing draught from Glover tower by a gas-siphon,
682 ; admitting air behind burners, 682 ; influence of atmo-
spheric conditions on draught, 683; quantitative data on
draught, 683 ; influence of difierence of level, 684 ; uselessness
of mathematical formulae, 686.
Supply of water (steam), 686 : regulation, 686 ; streugth of
chamber-acids and drips, 687 ; effects of a wrong supply of
water, 693. Supply of nitre, 695 : colour of chambers, 698 ;
testing chamber-acids for nitre, 699 ; effects of too much or too
little nitre, 701 ; modified processes for introducing nitre, 703 ;
total quantity of nitre circulating in chambers, 704. Irregular
ivorking, loss of nitre, 705 : formation of nitrous oxide, 707 ;
mechanical losses, 707.
Tem2)erature, 710 : at various works, 711 ; observations, 713 ; by
Lunge and Naef, 714 ; by Sorel, 720 ; abnormal cases, 720.
DepHi of acid, 721 : general remarks, 722.
DistribtUion of gases and rate of formation in the various parts of the
vitriol-chambers, 722 : views of H. A. Smith, 723 ; observations
made on acid-trays by Mactear, 723 ; by Naef, 724 ; by Mactear
on rate of formation of acid, 725 : by Lunge & Naef at Uetikon,
727 ; absence of " free " nitrogen peroxide in normally working
chambers, 727 ; appearance of free nitrogen peroxide in the
presence of an excess of nitre, 729 ; their observations on the
distribution of gases and the progress of the process in the
chambers, 730 ; observations on the revival of the process when
the gases pass into the next chamber, 731 ; on the mixture of
CONTENTS. XXI
Page
the gases, 732 ; Abraham's ideas as to the path which the gases
take within the chambers, 734.
Carbon dioxide in chamber-gases, 735.
Duration of passage through the chambers, 735.
Testing the ehamber-exitSj 736 : low-level escapes, 736 ; high-level
escapes, 736 ; limits imposed by Alkali Act, 736 ; continuous
testing, 737 ; Mactear's plan, 737 ; simpler plans, 73S ; Strype's
apparatus, 739 ; prescriptions of British Alkali Makers' Asso-
ciation, 743 ; improvement by Carpenter & Linder, 746 ;
shape of absorbing-vessels, 740 ; estimation of nitric oxide,
748 ; estimating sulphur burnt from oxygen in exit-gases, 749.
Theory of the formation of sulphuric acid in the lead-chambers, 750 :
views of Clement & Desormes, 750 ; of Davy, 751 ; of Gmelin,
752 ; of Berzelius, 752 ; of . Feligot, 753 ; of Weber, 754 ; of
Winkler, 757 ; Lunge's former views, 758 ; Lunge & Naefs
investigations, 759 ; Raschig's theory, 761 ; Lunge's new theory,
762 ; SoreFs views, 769 ; influence of change of temperature,
771 ; the chamber-process considered as a catalytic one, 773 ;
attempts at establishing theories on the action of masses,
774; laws of chemical dynamics, 775; velocity of reactions,
777 ; impossibility of subjecting the chamber-process to mathe-
matical treatment, 779.
CHAPTER YIII.
The PrBiFiCATioN of Sulphuric Acid 780
Impurities found in chamber-acid, 780; injurious action of these,
781 ; methods of purification, 781 ; clarifying muddy acid, 782.
Purification of Sulphuric Acid from Arsefiiic, 782 : arsenic in brim-
stone-acid, 782 ; arsenic in pyrites, 782 ; arsenic in pyrites-
acid, 783 ; cases where the arsenic does no harm, 784 ; cases
where arsenic is not allowable, 785 ; acids absolutely free from
arsenic, 785.
Removal of arsenic by distillation, 786 ; as arsenic trichloride, 787 ;
precipitation of arsenic as sulphide, 788 ; by barium sulphide,
788 ; by other sulphides, 789 ; by thiosulphates, 789 ; by
gaseous hydrogen sulphide, 790 ; Freiberg process, 790 (genera-
tion of sulphuretted hydrogen, 792 ; precipitation of the arsenic
in the tower, 793 ; tests, 793 ; filtering and washing of the
XXll CONTENTS.
Page
arsenic sulphide, 796) ; other methods, 799 ; evolution of sulphu-
retted hydrogen in other ways, 800 ; damage caused hy going
too far in that treatment, 801 ; cost of treatment, 801 ; effect
of treatment, 801 ; acid for storage-batteries, 802.
Purification fro^n Nitrogen Compounds^ 802: by sulphur dioxide,
802 ; by brimstone, 803 ; by organic substances, 803 ; by
ammonium sulphate, 803.
Purification hy special methods. Electrolysis, ^c, 804.
Coloured Add^ 805 : removal of selenium, 806,
Chemically-pure Sulphuric Add, 806 : distillation, 807.
CHAPTER IX.
The Concentration of Sulphuric Acid 810
Necessity for concentration, 810 ; action taking place during boiling-
down, 810; material of concentrating-vessels, 811; lead pans, 811.
1. Lead pans heaied from the fo/>, 812 : construction, 813; Hasen-
clever pans, 816 ; pans fired by producer-gas, 817.
2. Lead pans with bottom heat, 817 : way of firing the pans, 819 ;
drawings of pans, 820 ; various systems, 822.
3. Lead pans fired hy waste heat, 823 : on pyrit-es-burners, 823 ;
drawings of such, 824 ; objections, 827 ; pans on brimstone-
burners, 828.
4. Concenirathig-pans heated hy steam, 828 : Duisburg pans, 828 ;
precautions, 829 ; multitubular pans, 831 ; remarks on steam-
pans, 832.
5. Concentration to 144° Tw. in platinum vessels, 833.
The last co^icentration of Sulphuric Acid, 834 : various plans, 834 ;
reasons for higher concentration, 835.
Manufacture of ordinary Rectified Oil of Vitriol, 836 : first use ot
glass retorts, 836 ; of platinum vessels, 837 ; present state, 838.
Concentration in glass retorts, 838; ordinary English plan, 838;
improvements, 842 ; Gridley-Chance system, 844 ; similar plans,
846 ; cost of concentrating in glass, 847.
Concentration in porcelain dishes or heahers, 848; Negrier's apparatus,
848 ; Benker's apparatus, 850 ; Webb's beaker-apparatus, 851 ;
similar apparatus, 854 ; various other apparatus, 855.
Conctntration in platinum stills, 856 : older stills, 856 ; Prentice's
stills, 856 ; Delplace's stills, 858 ; Desmoutis' stills, 863 ;
firing platinum stills, 865 ; by Liegel's gas-producers, ^6iy.
CONTENTS. XX41I
Page
CombinadoD of platinum dishes with lead hoods (Faure & Kessler's
system), 866.
Other forms of platinum stills, 875,
Loss of platinum in the concentration of sulphuric acid, 875 ; prices
of platinum stills, 876.
Gold-lined platinum stills, 877.
Crusts formed in platinum stills, 879 ; explosion in clearing out a
stiU, 881.
Concentration in iron, 881 : suitable strength of acid, 881 ; various
shapes of pans or retorts, 882 ; pans used by Quinan at Pinole,
884 ; combination of cast-iron pans and platinum covers, 888 ;
Clayton Aniline Company's pan, 889 ; Krell's apparatus, 891 ;
other plans, 892.
Concentration in wrought-iron, 896.
Action of sulphuric acid on cast-iron, 896.
Recovery of strong Sulphuric Add from waste acids (acid-tar,
sludge-acid), 896.
Cast-iron pans protected hy enamel or otherwise, 899.
Concentration of Sidphuric Acid hy hot gases, 901 : apparatus of
Gossage, 901 ; various, 902 ; Kessler^s furnace, 903 ; Kessler's
new apparatus, 908 ; Zanner's pans, 909 ; apparatus of Quinan,
912.
Concentration in vacuum-retorts, 913 : apparatus of de Hemptinne,
913; of Krell, 914 ; German form, 914 ; various apparatus, 915 ;
vacuum-retorts for acid-tar, 916,
Concentration hy electricity, 916.
General remarks, 917.
Monchydrated Sulphuric Acid, 917 : Lunge's process, 918.
Packages for Sidphuric Acid^ 920 : carboys, 920 ; filling and
emptying carboys, 922 ; tank-waggons, 925 ; acid-boats, 926 ;
sea-transport of acid, 927.
CHAPTER X.
ARBAM6EME5T OP A SXTLPHITRIC-ACID WoBKS ON THE CuiJiBEB PlAN ;
Yields and Costs 928
General remarks, 928 ; arrangement of chamber-plant, 029 ; designs
of modern chamber-plant, 931 ; estimate of cost for same, 933 ;
designs for a combination of chambers and towers, 934 ; cost of
chamber-plant in France, 935.
XXIV CONTENTS.
Page
Yields and costs, 936 ; terms for expressing the consumption of nitre
and the yield of acid, 937 ; (i) consumption of nitre, 938 ;
(ii) yield of sulphuric acid, 940 ; consumption of coals, 942 ;
results at French factories, 942 ; at Hamburg, 944.
Cause of losses in the manufacture of sulphuric acid, 945.
Statements of costs for the manufacture of sulphuric acid: (1) from
brimstone, 946 ; (2) from pyrites in England, 947 ; in France,
948 ; in Germany, 949 ; statements by l^iedenfiihr and Luty for
the old system, 950; for a combination of chambers and towers,
952; for the contact-process, 953 ; for concentrated acid, 953;
comparison of both processes, 955 ; cost in America, 956.
CHAPTER XI.
The Manufactdee op Norbhausen or Fuming Oil of Vitriol and
OF Sulphuric Anhydride 959
Introduction, 959 : the manufacture of fuming oil of vitriol in
Bohemia, 960.
The production of sulphuric anhydride and fuming of vitriol from
artificially-prepared sulphates, 965 ; from ferric sulphate, 965 :
by the vacuum process, 966 ; from acid-tar, 967 ; from ferrous
sulphate, 968 ; from magnesium sulphate, 968 ; from sodium
pyrosulphate, 968.
Utilization of SO^, contained in pyrites-kiln gases, 971 ; dehydration
of sulphuric acid by metaphosphoric acid, 971 ; SOg from
chamber-crystals, 972 ; SO, by electricity, 972.
The manufacture of sulphuric anhydride (also of fuming and ordinar}-
sulphuric acid) by " contact '^-processes, 973 ; historical, 973 ;
first observations on contact-reactions, 974 ; generalization of
Berzelius, 974 ; discovery of the synthesis of SO^ by platinum by
Peregrine Phillips, 975 ; following investigations, 975 ; Schnei-
der's process, 976 ; processes of Jullion, Petrie, Piria, Laming,
977 ; Blondeau, Woehler & Mahla, 978 ; Robb, Trueman,
Schmersahl & Bouck, 979 ; employment of silica by Petrie and
Hunt, 979 ; experiments made by Plattner and Reich, 980 :
process of Deacon, 980 ; common shortcomings of former in-
ventors, 981 ; demand springing up for fuming acid, 982.
New epoch inaugurated by Winkler's paper in 1875, 983 ; Squire
& Messel's process, 983 ; Winkler's paper, 984 ; efifects of it,
985 ; manufacture of fuming acid by Jacob, 988 ; at Thann,
CONTENTS. XXV
Pago
989 ; in Germany, by Winkler's process, 990 ; modifications
of this, 991 ; process of Schroeder & Haenisch with pure
sulphur dioxide, 993 ; processes employing pure oxygen, 994 ;
employment of burner-gases, 994.
Theory of the contact-processes for the manufacture of sulphuric
anhydride, 996 ; views of Berzelius, Liebig, 996 ; Bunsen,
Ostwald, 997; modem views on catalyzers, 998; views of
Sackur, 999 ; Kassell & Smith, 1000 ; experiments and views
of Knietsch, 1001 ; of Brode, 1007 ; paper by Lunge & Pollitt,
1008 ; maximum yield theoretically obtainable, 1009 ; vapour-
tensions of sulphates in contact-processes, 1011.
Present state of the Manufacture of Sulphuric Anhydride and
Sulphuric Acid by the Contact-process 1011
I. The process of the Badische Anilin- und Sodafabrik at Lud-
wigshafen, 1012; statements of Knietsch and of the patent
specifications, 1012 ; contribution by the firm, 1026.
II. Process of the Farbwerke, vorm. Meister, Lucius, & Briining, at
Hochst, 1039.
HI. The Schroeder-Grillo process, 1057.
IV. The process of the Mannheimer Verein, 1067.
V. The Freiberg process, 1073.
VI. Babe's process, 1078.
Recent proposods for impt^oving the Manufacture of Sulphuric Acid
by the Contact-process • 10S2
A. Purification of the gases, 1082. B. Contact^apparatus, 1083.
C. Absorbing-apparatus, 1085. D. Contact-substances, 1086.
E. Manufacture of 8O3 from by-products, 1091.
CHAPTER XIL
Other Pbocssses fob Manufactukinq Sulphuric Acid 1092
I. Oxidation of sulphurous acid by means of nitrous acid, but
without lead-chambers, 1092.
II. Processes dispensing with nitre, 1096.
III. Sulphuric Acid from sulphates, 1097 ; from calcium sulphate,
1097 ; production of SO^ from gypsum, 1098 ; sulphuric acid
direct from gyjwum, 1099.
IV. Calcining pyrites with salt, 1100.
V. By electricity, 1100.
VOL, I. c
XXVI CONTENTS.
CHAPTER XIII.
Page
Br-rB0DucT8 of the Manufactctbe of Sulphubic Acid 1102
Utilization of cinders from non-cupreous pyrites as ballast, 1102 ;
for absorbing H,8, 1102; for absorbing HCl, 1103; for
manufacturing iron compounds, 1103. Manufacture of iron
from pyrites-cinders, 1104.
Recovery of mw? from pyrites-cinders, 1107.
Recovery of ihallium^ 1110 ; oi selenium, 1113.
The extraction of copper from pyrites-cinderSy 1116 : general, 1115;
lirst attempts, 1117 ; recent process, 1117 ; composition of
cinders, 1118 ; grinding, 1121 ; calcining-fumacesy 1121 ;
reverberatory furnaces, 112L; blind roasters (muffles), 1126;
mechanical furnaces, 1129 ; mode of operating the calcination,
1133; tests, 1138; condensation of the calcination -gas,
1138. Lixiviation of the calcined mixture, 1139; residue
(purple ox% blue billy), 1141; extraction of lead, 1142;
copper-liquors, 1143; process at Oker, 1144, Precipitation
of the copper from the liquors, 1146; Gibb's H^S process,
1146; use of scrap-iron, 1147; use of spongy iron, 1148;
furnace for making spongy iron, 1148 ; operating it, 1151 ;
composition of spongy iron, 1153; composition of coppcr-
prccipitate, 1154; furnaces for smelting it, 1156; pure
copper, 1157. Extraction of tJie sHver contained in burnt
p}'rites, 1158; Claudet*s process, 1158; other processes,
1160. Waste-liquors from the copper extraction^ 1162.
Working-results, 1165.
Other copper-extracting processes, 1165. Treatment of very poor
cupreous cinders, 1167. General, 1168.
CHAPTER XIV.
Afplicatioks of Sulphubic Acid and Statistics 1169
Principal applications, 1169.
Statistics for the United Kingdom, 1170 ; Germany, 1172 ; France,
Austria, Belgium, Italy, 1174; Russia^ United States, 1175;
Japan, 1177.
CONTENTS. XXVll
ADDENDA.
Brimstone as raw material for sulphuric-acid making, Boiling-
point of sulphur. Exportation of sulphur from Sicily, 1178.
Sulphur in Transylvania, Greece, Japan, North America, 1179.
Sulphur from sulphides, 1179. World's production of sulphur,
1180. Analysis of crude sulphur. Detection of arsenic and
selenium, 1180. Grinding and sifting sulphur, Examination
of sulphur for fineness, &c., 1181.
Pyrites in Germany, Austria, 1181 ; in America, 1182. Leached
ores from Spain, 1182. World's production of iron-pyrites,
1182. Estimation of carhon in pyrites, 1183.
Snlphuric acid from hydrogen sulphide, 1183.
Nitrate of soda in California, 1183. Statistics, 1183.
Hydrates of nitric acid, 1183; nitric-acid retorts, 1183; con-
densation of nitric acid, 1184 ; clear concentrated nitric acid,
1185; utilization of nitre-cake, 1185, 1186; recovery of
nitric acid from waste acids, 1186; production of nitric acid
from atmospheric air, 1186 ; from ammonia, 1187*
Scsquioxide of sulphur, 1187.
Acid-smoke, 1187.
Specific gravities of sulphuric acid, 1187; influence of impurities
on specific gravity, 1188 ; action of sulphuric acid on platinum,
1188 ; analysis of fuming sulphuric acid, 1188 ; detection and
ostimation of arseuic in sulphuric acid, 1189.
Herreshoflf furnaces, 1192.
Mechanical furnaces for hlende, 1193.
Cooling the gases in chamhers, 1193.
Keaction-towers, 1193 ; pipe-towers, 1193.
Mechanical calculation of the weight of acid in the chamhers, 1193.
lutrodttction of water as spray, 1194.
Fan-blasts for producing draught in chambers, 1194.
Measurement of the velocity of a current of gas, 1194.
Path which the gases travel in the chambers, 1194.
Page
IHDEX 1195
FIRST BOOK.
STJLFHUBIC ACID.
VOL. 1. »
INTRODUCTION;
Formerly the term ^^ alkali manufacture'^ nearly always com-
prised a cycle of operations^ beginning with the manufacture of
sulphnric acid and proceeding to that of sulphate of soda (salt-
cake), hydrochloric acid, soda-ash (with caustic soda, soda-
crystals, fee), and bleaching-powder. This cycle is not completed
in all factories, but frequently (at the present day even more so
than formerly) the operation stops at sulphuric acid or sulphate
of soda ; but we may embrace all this under the general term of
" Sulphuric Acid and Alkali Manufacture/'
In this wider meaning the products of alkali-making are
necessary materials for many metallurgical processes, for the
manufacture of artificial manures, soap, fatty and mineral oils^
glass, paper, many inorganic and organic colouring-matters
(especially nearly all coal-tar dyes), and even of many articles of
food, — that is to say, for nearly all branches of manufacturing
chemistry. In fact, among all branches of chemical industry
the cycle of technical operations connected with alkali-making is
preeminent, not merely from the magnitude of the works and the
absolute bulk of the raw material used and the quantity produced,
but also from the fact that most other chemical products require
one or more branches of alkali-making as the conditions of their
own existence. It can be truly said that the manufacture of
acids and alkali is the foundation upon which the whole chemical
industry of our times is built up, and that such industry cannot
b2
INTRODUCTION.
be much developed in any country not possessing a flourishing
alkali trade^ or not being at least specially well situated for buying
the produce of the latter. It is thus evident how great is the
importance of the alkali trade in its wider meaning to the
civilization of mankind^ though we should certainly be going too
far if we measured^ as some have done^ the civilization of a
country by the development of this special industry.
Formerly the whole cycle of processes here described was
intimately connected with the great invention of Leblanc, now
rather more than a century old. Twenty years ago, although the
ammonia-soda process had then already more than proved its right
to be considered a full success, it had not yet shaken in any tangible
degree the supremacy of the Leblanc process, at least not in Great
Britain. In the latter process the different branches of '^ alkali-
making ^' mentioned above are connected in such a manner that
only under special local conditions can one or more of the
principal substances be omitted. Formerly this was the case even
less than now, as the competition of ammonia-soda ash has
completely altered some of the conditions of trade, making it
unremunerative in many cases to convert the sulphate of soda
into the carbonate. Thus there exist now many works stopping
at the manufacture of sulphuric acid, and others which go as far
as sulphate of soda, tpgether with chlorine products ; but very
many proceed still further, going on to the manufacture of soda in
its various branches.
The manufacture of sulphuric acid is in reality a very large
industry, quite apart from its connection with the Leblanc
process. Enormous quantities of it are required for the manu-
facture of artificial manures* (fertilizers), and therefore every
large manure factory makes its own sulphuric acid. This is done
also by the largest sulphate-of-ammonia works, petroleum refiners
coal-tar dye manufacturers, and in a few other cases. Some works
in England, aud many on the Continent, make sulphuric acid to
a great extent, or even entirely, not for their own use, but for
sale. Since this acid is no longer sent out in any considerable
INTRODUCTION. 5
quaBtity in glass carboys, but in iron tank -waggons, it can be
carried to considerable distances at moderate cost.
As sulphuric acid is mostly made from pyrites, its manufacture
is intimately connected with the recovery of copper from the
cinders, in which process iron oxide, silver, and other by-products
are obtained.
A very large (formerly even the largest) quantity of sulphuric
acid is used up at the works themselves for the manufacture of
stdphaie of soda {salt-cake) and sulphate of potash, in which hydro^
chloric add is a necessary by-product. Sometimes salt-cake is
obtained without previously manufacturing siQphuric acid, as a
by-product of other manufactures or by the '' direct process '* of
Hargreaves and Robinson. There is only one use of salt-cake
on a large scale, except for alkali-making, namely, for the manu-
facture of glass ; but a much larger quantity of salt-cake still
enters into the Leblanc process for manufacturing soda. This
article is mostly the final product, either in the calcined or
crystallized or caustic state^ and the series of operations is thus
brought to a close in this direction.
Hydrochloric acid {muriatic add) is, of course, sold as such to
some extent, but in nothing like such large quantities as sulphuric
acid, as its carriage is impossible in metallic vessels^ and there-
fore more expensive and troublesome. Most of it is at once,
sometimes even without condensation to liquid acid, converted
into chlorine, which, being a gas, is immediately worked up into
bleaching-powder or chlorate of potash, or occasionally into other
products ; a comparatively small quantity of it is sold in the state
of liquid chlorine. The time when the hydrochloric acid was
condensed merely to satisfy the exigencies of laws made for
protecting the health and vegetation of a neighbourhood, and was
then run to waste into the Acarest watercourse, is now past, since
the process of decomposing salt by sulphuric acid is only profitable
if the hydrochloric acid is fully utilized : this acid has thus risen
from the rank of a by-product to that of the best-paying principal
product.
6 .INTRODUCTION.
Once more the conditions of the Leblanc process were changed
by the solution of the problem of the recovery of sulphur from the
alkali waste^ and it has now been made into a real cycle, into which
common salt and coal enter at one end, alkali and chlorine issuing
at the other, whilst sulphur and possibly even lime are made to do
service over and over again. This, however, appears most clearly
in Vols. II. & III. of this Treatise ; to which we also refer for such
general observations as the ammonia process of soda manufacture
calls for.
HISTORY OF SULPHURIC ACID.
CHAPTER I.
HISTOKICAL^AND GENERAL NOTES ON THE MANUFACTURE
OF SULPHURIC ACID.
History of the Manufacture of Sulphuric Add.
According to Rod well (^ Birth of Chemistry^) it is very probable
that sulphuric acid was already known to the ancients ; but usually
its firsts although indistinct^ mention is ascribed to the Arab
Geber, who speaks of the '^ spirit ^' which can be expelled from
alum and which possesses solvent powers. Geber is, however, a
mythical personage, and many of his alleged numerous discoveries
have wrongfully crept into the Latin " translations '^ of his pre-
tended writings, as proved by Berthelot and Steinschneider (comp.
Lippmann, Zsch. angew. Chem. 1901, p. 646), who show that
sulphuric acid was unknown to Arabian writers about 975 a.d.
Others give the honour of its discovery to the Persian alchemist
Abu-Bekr-Alrhases, who is said to have died in 9iO. Vincentius
de Beauvais (about 1250) alludes to it; and Albertus Magnus
(1193-1280) speaks of a spiritus vitrioli Romani, which can only
have been sulphuric acid ; his '^ sulphur philosophorum '^ is the
same thing.
With all distinctness Basilius Valentinus, in his ^ Revelation of
the Hidden Manipulations/ describes its preparation from cal-
cined copperas and silica, and, in his ' Triumphal Car of Antimony/
also its preparation by burning sulphur with saltpetre (Kopp,
' Geschichte der Chemie,^ iii. p. 303) ; but he took the two to be
different substances.
Gerhard Domaeus (1570) described its properties accurately;
Libavius (1595) recognized the identity of the acids from different
processes of preparation ; the same was done by Angelus Sala
(1613), who pointed out the fact, which had sunk into oblivion
since Basilius, that sulphuric acid can be obtained by burning
sulphur in moist vessels (of course with access of air) ; after that
8 HISTORICAL AND GENERAL NOTES.
time it was prepared by the apothecaries in this way. An
essential improvement, ^iz. the addition of a little saltpetre, was
introduced in 1666 by Nicolas le Fevre and Nicolas Lemeiy.
This caused a sort of manufacture of vitriol which is said to have
been introduced into England by Cornelius Drebbel : this only is
certain — that a quack doctor of the name of Ward first carried
on sulphuric-acid making on a large scale at Richmond near
London, probably a little before 1740. Ward employed large
glass vessels up to 66 gallons capacity, which stood in two rows
in a sand-bath, and which were provided with horizontally pro-
jecting necks ; at the bottom they contained a little water. In
each neck there was an earthenware pot, and on this a small red-
hot iron dish, into which a mixture of one part saltpetre and eight
parts of brimstone were put ; then the neck of the bottle was
closed with a wooden plug; on the combustion being finished,
fresh air was allowed to enter the vessel, and the operation was
repeated till the acid had become strong enough to pay for con-
centrating in glass retorts.
Ward called the product '^oil of vitriol made by the bell''
(already Basilius Valentinus had used the expression "percam-
panam'' in this sense), in order to distinguish the spirit of vitriol
made from brimstone from that distilled from sulphate of iron,
the latter having been made on a kind of manufacturing scale in
England previously : an exact description of this is given by
J. C. Bernhardt in his ^ Chemische Versuche und Erfahrungen,'
17.55. Ward's process, troublesome as it is, reduced the price
of the acid from 2*. 6d. per ounce (the price of the acid from
copperas or from burning brimstone under a moist glass jar) to
2$. per lb.
An extremely important improvement in this process was the
introduction of the lead chambei's, which by general consent is
ascribed to a Dr. Roebuck of Birmingham, who in 174f6 erected
such a chamber 6 feet square, and in 1749, in partnership with
Mr. Garbett, built a factory, founded thereon, at Prestonpans in
Scotland, in order to supply acid for the bleaching of linen. The
mixture of brimstone and saltpetre in the above proportion was
put into small iron waggons which were run into the chamber
on a railway : the chamber was closed, and the process carried on
intermittently in this way. Guttmann (Journ. Soc. Chem. Ind.
1901, p. 5) gives a detailed description and some drawings of such
LEAD CHAMBERS IN ENGLAND. 9
*' lead-houses '' and their style of working from a manuscript by a
Birmingham chemist^ W. E. Sheffield^ written between 1771 and
1 790. The cost of acid of spec. grav. 1'844 per ton was £22 6*. 4rf.
without labour.
Soon other works followed at Bridgenorth^ and at Dowles in
Worcestershire, where the chambers were already made 10 feet
square; in 1772 a factory was erected in London with 71 cylin-
drical lead chambers, each 6 feet in diameter and 6 feet high.
In 1797 there were already six or eight works in Glasgow alone.
According to the statements given in Mactear's ^ Report of the
Alkali and Bleaching-Powdcr Manufacture in the Glasgow
District^ (p. 8), the acid at that time cost the Glasgow manu-
facturers £32 per ton, and was sold at £64:. At Radcliffe^ near
Manchester, it cost, in 1799, £21 10*. per ton, without interest
on capital. In the latter place there were six chambers 12 feet
long, 12 feet wide, and 10 feet high, with roofs like those of
houses, and valves opened between each operation; on their
bottom were 8 or 9 inches of water ; every four hours a
mixture of 1 lb. saltpeti*e and 7 lb. brimstone was burnt in each
chamber on iron shelves, of which each chamber contained four,
4 inches distant one from another. The shelves were made of
very thin iron, in order to get heated very quickly, and rested
on iron frames, by means of which they could be slid in and
out ; a quarter of an hour before each operation the valves and
doors were opened in ordei* to allow air to enter. Thus, weekly,
1386 lb. of brimstone and 198 lb. of saltpetre were burnt, yielding
1800 lb. of oil of vitriol — that is, 130 per cent, of the sulphur with
a consumption of 14''28 per cent, saltpetre on the same. In six
weeks the strength of the acid attained only 1*250 sp. gr. ; it was
thus run off and concentrated up to 1*375 sp. gr., in which state
it was used and sold. At Prestonpans, in 1800, a yield of only
111 per cent, on the sulphur was attained, with a consumption of
13 per cent, saltpetre on the brimstone ; in 1813 there were in
that place 108 chambers of 14 feet length, 10 feet height, and
4J feet width. In 1805 there existed at Burntisland a factory
with 360 chambers of a capacity of 19 cubic feet each. 11. Forbes
Carpenter and W. F. Reid also give some interesting notes on the
early manufacture of sulphuric acid (Journ. Soc. Chem. Ind. 1901,
p. 7). The former mentions acid-charabers erected in Cornwall
from dressed granite, with lead top and bottom.
10 HISTORICAL AXD GENERAL NOTES.
Ill the meantime the first lead chamber in France had been
erected at Rouen by Holker in 1766. In 177 J*, in that place, on
the advice of De la FoUie, an important improvement was intro-
duced, viz. the introduction of steam into the chambers during
the combustion of brimstone. In 1793 Clement and Desormes
showed that the acid- chambers can be fed by a continuous current
of air, by which a great deal of saltpetre could be saved. They
showed that the oxidation of sulphurous acid takes place to the
extent of nine tenths at the expense of atmospheric oxygen, and
that the saltpetre plays only the part of intermediary between the
air and the sulphurous acid. By this the modem theory of the
essence of the sulpliuric-acid-making process was established ; but
it took a remarkable long time before the difficulties were over-
come which stood in the way of introducing the continuous
system into practice. Usually the introduction of the continuous
burning of brimstone is ascribed to Jean Holker (a grandson of
the first Holker), in 1810; but, according to Mactear, a con-
tinuous system had been introduced at St. RoUox, at least
partially, already in 1807 : steam was first introduced there in
1813 or 1814.
In Germanv the first lead chambers seem to have been those
at liingkuhl, near Cassel. One of the oldest chambers was that
erected by Dr. Richard at Potschappel near Dresden in 1820; as
he had no plumber at his disposal, he had to solder the chamber
himself with soft solder and a smoothing-iron (Bode, in his transla-
tion of H. A. Smith's ^ Sulphuric Acid Manufacture,' p. 96). This
chamber was still charged intermittently, 100 lb. of brimstone
yielding only 150 lb. of vitriol.
Lampadius, in a treatise published in 1815 *, speaking of
sulphuric acid as then made in England and at Schwemsal near
Leipzig, describes the lead chambers as '^ rooms," about 25 feet
square with a stone floor, lined throughout with lead, with two
doors through which sulphur is introduced and burned on iron
dishes, holding J cwt. of brimstone, mixed with 20 per cent,
nitrate of potash, steam being introduced from a copper outside
the room. Through four pipes and taps, during the later stages
of the combustion (which lasted three hours), a little air was
admitted. The '' room '' was exposed to great strain through the
* Quoted by CI. Wiukler, Zsch. angew. Chem. 1900, p. 731. Some of the
following statements are also taken from this paper.
GENERAL PRINCIPLES. 11
strong expansion and subsequent contraction of the atmosphere
inside. The dilute acid formed was boiled down in glass retorts
to spec. grav. 1*800.
The invention of soldering lead with the same material or
" burning ^' by means of the hydrogen blowpipe is due to
Debaissayns de Richemond, in 1838. As late as 1846 PrechtPs
'Technical- Encyclopaedia' (xiv. p. 246) mentions the chamber-
sides as being sometimes covered with a crust of chamber-
crystals, i or 1 inch thick^ which proves the want of under-
standing the process at that time.
Kestner, of Thann in Alsace, was the first to collect the products
of condensation at the chamber-sides in order to regulate the
working of the chambers thereby. This innovation was at once
considered of such importance that Kestner was called to Glasgow
in order to introduce his plan into Tennant's works.
In 1827 Gay-Lussac's condensing-apparatus for the nitre-gas
escaping from the cjiambers was invented : at Chauny this appa-
ratus was erected in 1842, at Glasgow in 1844. But we have now
come so near the present time that we may conclude the historical
part of our task.
General Principles of the Manufacture of Sulphuric Acid.
Sulphuric acid can be obtained on a large scale in one of two
ways — viz., first, by burning sulphur or sulphides into sulphur
dioxide and further oxidizing the latter, or, secondly, by decompos-
ing natural or artificially prepared sulphates. The latter process,
apart from several proposals so far not carried out practically,
until quite recently served for making fuming oil of vitriol, which
will be treated of hereafter ; by far the greater portion of sulphuric
acid has always been obtained by the former process, which will
occupy us in the first instance. We shall, at first, only describe
those apparatus and processes which actually serve for manufac-
turing on a large scale ; and we shall close by mentioning the
alterations whicb have merely remained as proposals, as well as
the processes founded on totally new principles.
By the combustion of sulphur, either free (as brimstone, gas-
sulphur, &c.) or combined with metals or with hydrogen, sulphur
' dioxide (SO2) is always formed at first. Sulphuretted hydrogen, even
when mixed with as much as 70 per cent, of inert gases (nitrogen),
can be lighted like illuminating-gas and continues burning without
12 HISTORICAL AND GENERAL NOTES.
any difficulty^ aqueous vapour being formed at the same time as
SO2. Brimstone ignites in the air at a temperature rather below
300° C. ; and when once it has begun to bum, the heat generated
suffices to raise the whole of the sulphur to the point of ignition^
provided that sufficient air be present. A number of metallic
sulphides behave similarly : the most important of these for our
purpose is the iron disulphide, FeSg ; but here special precautions
must be taken, so that the whole mass may be completely burnt
(roasted). In both cases, along with sulphur dioxide^ SO2, a little
trioxide (sulphuric anhydride), SO3, is always formed, and, in the
presence of water or steam, also sulphuric acid, SO4H2, more or
less diluted with water. Moreover an aqueous solution of sulphu-
rous acid in contact with air gradually changes into sulphuric acid.
In both cases it is, of course, the oxygen of the air which converts
the SO2 into SO,? or SO4H2 ; but this reaction at the ordinary or
only moderately elevated temperature goes on far too slowly to be
applicable for technical purposes.
There are two ways of increasing the velocity of the oxidation
of sulphur dioxide. One of these^ which is principally applicable
to dry gases and therefore leads to the preparation of sulphur
trioxide in the anhydrous state, is the employment of '^ catalytic ^^
substances. We shall discuss this in Chapter XI., devoted to the
" Contact processes"
The second way, which is exclusively applicable to the produc«
tion of real sulphuric acid, H2SO4, is founded on the property of
the acids of nitrogen to serve as carriers of oxygen from atmo-
spheric air upon sulphur dioxide and water, the original nitrogen
oxide being always re-formed. This process will be explained in
detail when we treat of the theory of the formation of sulphuric
acid ; it is called the vitriol-chamber process.
The reaction between nitrogen acids and sulphur dioxide only
goes on in the presence of water ; and we must add at once that,
in practice, much more water is needed than suffices for the
formation of SO4H.2 ; the sulphuric acid formed is therefore
always dilute, and must be concentrated for most purposes.
For some purposes the acid must also be deprived of certain
foreign substances which get into it from the raw materials and
the apparatus ; and in such cases the sulphuric acid has to be
purified.
DIVISION OF THE SVliJECT. 13
Thus our subject subdivides itself into the following headings : —
1. An examination of the raw materials oi the sulphuric- acid
manufacture, and an account of the properties of the oxides and
acids of sulphur »
2. The generation of sulphur dioxide. Since the processes
diflfer very much, both as to apparatus and the method of pro-
ceeding, we have to distinguish between (a) sulphur dioxide
from brimstone, (A) from metallic sulphides, (c) from sulphuretted
hydrogen.
3. The conversion of sulphur dioxide into sulphuric acid in tlic
so-called vitriol-chambers.
4. The purification,
5. The concentration of sulphuric acid.
6. The contact processes,
7. The utilization of the by-products.
14 RAW MATERIALS OF MANUFACTURE,
CHAFTER II-
THE RAW MATERIALS OF THE SULPHURIC-ACID
MANUFACTURE (including NITRIC ACID).
1. Natural Sulphur (Brimstone).
Brimstone, owing to its being found in nature in the free state,
has been known to mankind since very ancient times. It is hardly
necessary to point to its being noticed in the Bible ; it is also men-
tioned several times in the Homeric poems. The Romans evidently
obtained it in the same way as is done now, by melting it out of
its mixture with marl, &c. The ancients employed it principally
for fumigating purposes, both on account of its disinfecting pro-
perties and as a religious rite (compare the well-known passage
from the Odyssey, where Ulysses purifies his house after slaying
the intruders), but also for many of the uses to which it is put
at the present day, as for cleaning wine-casks, for destroying
fungus-growths in vineyards and orchards, for plasters in skin-
diseases, for lighting fires and preparing torches, for cementing
glass, for bleaching, for " niello '* work on metals ■'^.
In modern times brimstone has been used for most of the just-
mentioned purposes and for many others ; but we are here con-
cerned only with its use for the manufacture of sulpliuric acid.
Brimstone is undoubtedly the most convenient raw material for
this manufacture, and for a long tiibe all the sulphuric acid of
commerce was made from it ; but its use in this respect has been
almost entirely abandoned in most localities, and is not likely to
be revived, since iron-pyrites, and more especially that containing a
few per cent, of copper, supplies sulphur for the above purpose far
more cheaply than natural brimstone ever can do. In spite of this»
brimstone is still a principal raw material for the manufacture of
sulphuric acid in America (where, however, it has lost its former
* I owe these historical notes to a treatise by Professor Blumner, of Zurich
in the * Festschrift zur Begriissiing der Philologen-Versammlung ' ^Ziltich, 1887),
p. 23 et seq.
PROPERTIES OP SULPHUR. 15
exclusive sway), and it is also used to some considerable extent in
England for that purpose, but very little indeed in other European
countries.
A somewhat considerable quantity of brimstone is also consumed
in the manufacture of sulphurous acid^ principally in order to
prepare bisulphate of lime for the manufacture of wood-pulp.
Sulphur is an element whose atomic weight is now assumed to
be 32'06 (oxygen = 16). It is very brittle ; its hardness is from
1*5 to 2'5 of the ordinary mineralogical scale ; its specific gravity
is 2'07. As usually occurring, it is semi-transparent at the edges
and of the well-known bright yellow colour, which darkens with
an increase of temperature ; at —50° it is nearly devoid of colour.
Its taste and smell are very slight. It does not conduct electricity,
but itself becomes electric by friction ; and it is therefore difficult
to powder finely, as it adheres to the mortar and pestle.
Sulphur melts at lll°-5 C, and forms a thin, light-yellow liquid,,
which, on being more strongly heated, becomes darker and thicker;
at 250° to 260° C. it is nearly black, and so viscid that it does
not run out when the vessel is upset ; at a still higher temperature
it becomes thinner again, keeping its brown colour ; and at 440° C.
it boils, forming a brownish-red vapour ; but it begins to volatilize
before boiling ; and its action on silver seems to show that it doea
so to a slight extent even at the ordinary temperature (unless there
is formation of H^S in this case).
Sulphur exists in different allotropic conditions. That occurring
in nature often appears in rhombic crystals, mostly pointed
rhombic octahedra, whose physical properties have been described
above ; this a-modification is also obtained by crystallizing sul-
phur from its solution in carbon disulphide. The /8-modification
is obtained by slowly cooling melted sulphur, and pouring off the
liquid portion when another portion has crystallized ; it consists,
of long thin oblique rhombic prisms, belonging to the monoclino-
hedric system, of a brownish-yellow colour, transparent, spec. grav.
1'982 ; they gradually pass over into the a-modification, com-
pletely so after a few days, even at the ordinary temperature —
suddenly by shaking or scratching ; the colour then becomes
light yellow; and the crystals lose their transparency, but remain
as pseudomorphs of the )3-sulphur. The sulphur in rolls consists,
when fresh, of /8-sulphur — after a short time, of a-sulphur.
When sulphur has been heated up to the point of viscosity, and
16 RAW MATERIALS OP MANUFACTURE.
is then poured into very cold water, the 7- modification is formed,
viz. amorphous, soft, tough, reddish-brown sulphur, of 1'957 spec,
grav. ; this also is gradually converted into a-sulphur ; but it takes
some time before this conversion is complete. The tough state
lasts very much longer if resinous substances, iodine, &c. are mixed
with the sulphur, even in very small quantity. This modification
is partly contained in the " flowers " of sulphur.
Heated in the air to 250° C. *, sulphur inflames and burns with
a purplish-blue flame, forming sulphur dioxide (SOo), and giving
out 2221 metrical units of heat per gram of sulphur. More exactly,
.according to Thomsen (Berl. Ber. 1880, p. 959), the heat evolved
in burning the different modifications of sulphur, expressed in
atomic calories (that is, applied to 32 grams of sulphur), is: —
S (rhombic, octahedric) +02 = S02= +71,080 cal.
S (monoclinic) . . . +0^=802= +71,720 cal.
Hence the conversion of 32 parts by weight of monoclinic into
rhombic sulphur is accompanied by the evolution of 640 calories.
Berthelot, however (Compt. rend. xc. p. 14=19), states the figure for
octahedric sulphur = +69260 atom. cal. = 2164 gram- calories.
Hydrogen combines with sulphur very slightly at a temperature
of 120°, very sensibly so at 200°. On boiling sulphur with water,
hydrogen sulphide is evolved and sulphuric acid is found in the
residue (Cross & Higgin, J. Chem. Soc, xxxv. p. 249; comp.
also Colson, Bull. Soc. Chim, [2] xxxiv. p. 66 ; Bohm, Jahresb.
1883, p. 225).
Sulphur is insoluble in water, a very little soluble in alcohol and
in glycerine, rather more so in essential oils, but easily soluble
(excepting the 7-modification) in disulphide of carbon and in
chloride of sulphur.
The sulphur occurring in commerce as refined sulphur, in rolls
or as '^ rock-sulphur," is frequently almost chemically pure.
^' Flowers of sulphur " always contains a little sulphurous acid,
also some sulphuric acid, persistently retained in spite of prolonged
washing ; it owes its greater efficacy agaiust diseases of the vine
(oidium) and other cases principally to this property.
O. Roessler (Arch. d. Pharm, 1887, p. 845) states that sulphur
ill rolls is practically free from acids of any sort. Flowers of
sulphur contains a somewhat considerable quantity of sulphurous
* Comp. J. R. Hill, Chem. News, 1890, Ixi. p. 126.
NATURAL OCCURREXCE OF SULPHUR. 17
acid (100 grams up to 3*14 c.c. of SOj), which may be partially
oxidized to sulphuric acid ; thiosulphuric acid is not found in it^
but in milk of sulphur (up to 015 per cent.).
Natural Occurrence of Sulphur.
Sulphur occurs in nature in very large quantities, both in the
free state and in combination with other bodies as sulphides and
sulphates. Deposits of sulphur are forming at the present day,
especially in volcanic countries, by the decomposition of sulphur-
etted hydrogen and of sulphurous acid.
Daubree (Compt. rend. xcii. p. 101) noticed a recent formation
of sulphur in the subsoil of Paris, from the action of organic sub-
stances on sulphates ; and the same action has been observed in
many mineral springs, where it is especially attributed to the
action of Algse, such as Beggiatoa, Osdllaria^ and Ulothrix (Compt.
rend. xcv. pp. 846 & 1363).
But of far more consequeuce are the beds of sulphur deposited
in former geological periods. The most important of all are those
of Sicily, in the chalk ; next, those in the Romagna and in other
parts of Italy. The Sicilian sulphur industry is described in
detail by Angelo Barbaglia in Hof mann's Official Report on the
Vienna Exhibition, i. p. 144, and by Parodi (Berichte d. deutsch.
chem*. Ges. 1874, p. 358), According to the latter, the disposable
stock of sulphur in Sicily is said to amount to ten millions of
tons ; so that it would be exhausted about 1950. Other calcula-
tions gave more than twice that amount. The apparatus for melting
the sulphur out of the " sulphur-earth *' in Sicily is also described
by Barbaglia.
Further descriptions of the Sicilian sulphur industry are found
iQ the ' Chemiker-Zeitung ' of 1882, pp. 1389, 1405, 1421 ; also
in ' Zeitsch. f . angew. Chemie,* 1890, p. 56, in the ' Journal of the
Society of Chemical Industry,' 1890, p. 118 ; by Bechhold, ' Zsch.
angw. Ch.' 1894, p. 33 ; by Frank, ibid. 1900, p. 843 ; by Inng-
fleisch (Monat. Scient. 1901, p. 511).
Probably the most exhaustive description of the modern Italian
sulphur industry is that given by Aichino, in ' Mineral Industry,'
viii. p. 592.
Here we quote only' a few brief notes on more recent processes
or proposals for extracting sulphur from its ores (whether intended
for Sicilian sulphur or that from other localities).
VOL. 1. c
18 RAW MATERIALS OF MANUFACTURE.
In recent years the " Gill kilns " have rapidly gained ground in
replacing the calcarone methods, and they incrjease the output of
sulphur by 50 per cent. These kilns consist of an oven covered
by a cupola, called a " cell ; " inside there is a smaller cupola,
within which a coke fire is burning. Each cell holds from
5 to 30 cubic metres of ore. There are generally six cells working
in an angular battery. The gases generated in the first cell pass
by lateral channels into the next; by the time the fusion is
completed in the first cell, the contents of the second cell are
already heated up to the igniting-point by the gases, and so on.
The gases heavily charged with sulphur are not lost as in the
calcarone method ; the yield is much larger, the time shorter
(three or four days for each cell), and as the quantity of smoke is
much less, the work can be continued almost all the year round
without danger to the crops. In 1888 there were only 365 cells
in 40 mines, in 1894 the number had increased to 1821 cells in
225 mines, that is over two-thirds of the total.
E. F. White (Eng. Min. Journ. Ixii. p. 536) extracts sulphur from
sulphur-ores by passing the latter downwards between two conical
steam-heated vessels, where the sulphur is fused and runs out at
the bottom.
De la Tour de Breuil (Compt. rend, xciii. p. 456) employs a
66-per-cent. solution of calcium chloride for the same purpose.
C. Vincent recommends this process (Bull. Soc. Chim. xl.
p. 528).
K. Walter, in Milan, has patented an apparatus for continu-
ously melting out sulphur by combustion of part of the sulphur
itself, and converting the sulphur dioxide formed into sulphuric
acid in lead chambers (Chem. Zeit. 1886, p. 1199).
The American patents of P. Dickert, Nos. 298,734 and 301,222,
describe an apparatus for melting sulphur, consisting of a jacketed
pan connected by a perforated diaphragm with another jacketed
pan turned bottom upwards. The melting is produced by means
of steam introduced into the jacket.
Very ingenious is the process of Herman Frasch (U.S.
pat. 461429, 461430, 461431 of 1891), which is applied to the
Louisiana sulphur, where quicksand makes it impossible to use
the ordinary methods. A well is sunk by drilling in the same
way as for petroleum. This well is cased with an iron pipe
10 in. in diameter, reaching 10 feet into the rock overlying the
FRASCH PROCESS FOR SULPHUR-EXTRACTION. J 9
sulphur-bed. Inside this there is a 6-in. pipe, inside the latter
a 3-in. pipe, and, finally, in the centre a 1-in. pipe. The well is
carried down to the bottom of the sulphur-bed, and the 6-in. pipe
and smaller pipes are dropped nearly to the bottom of the lode.
Water heated to 166° C. is forced down the 10-in. and 6-in. pipes
under corresponding pressures, and on coming in contact with
the sulphur it is melted and collects in the well. Ordinary
pumping not haying answered, compressed air is forced down
through the 1-in. pipe, which mixes with the molten sulphur and
reduces its specific gravity so that it rises in the 3-in. pipe to the
top with great rapidity. As much as 150 tons of sulphur has been
pumped in 24 hours. About 50 tons of coal per day are burned
under the boilers.
The loss of heat is, of course, very great, and there is an
uncertainty about the way in which the impurities present may
choke up the pipes. In 1896 one hole, sunk in very pure sulphur,
produced 2100 tons; this was continued in 1898, bringing up the
total to 4509 tons. However, according to vol. vii. of ' Mineral
Industry,^ p. 643, in 1898 the conclusion was reached that the
process was unprofitable ; but this is entirely wrong. In
February 1902 I received from the inventor (Mr. Frasch) a set
of photographs, taken on the spot a few weeks before, exhibiting
the manufacture on a large scale, the pumps discharging a full
3-in. stream of liquid sulphur all through. The boiler-power is
being increased from 2100 to 4300 H.P. With the present
(lower) power the average production is 100 tons per day from
one well; up to 250 tons per day has been realized before.
The liquid sulphur is discharged into bins up to a size of
72 ft. X 60 ft. X 5 ft., where it solidifies and is broken up for
shipment. Many thousand tons have been pumped and shipped,
and there are large stocks on hand shown in those bins.
Analyses of Sicilian sulphur (Mene, Monit. Scient. 1867,
p. 400) :—
Sulphur soluble in OS2 96*2 921 921 91-3 901 900 887
Sulpbur ingoluble in CSj 15 2 0 21 17
Carbonaceous matter 05 10 11 07 10 11 10
Sand 1-5 23 28 33 23 28 55
limestone and strontium! j.g ^.j 3.Q 0.5 4.1 ^.q 2*8
Bidpbate J ^ ^
10000 99-5 990 993 99*5 990 997
[Mine's samples are evidently of a very low quality.]
c2
20 RAW MATERIALS OF MANUFACTURE.
The average price of sulphur in Sicily in 1881 was 115 lire, in
1882 105 lire, in 1887 only 69 lire (say £2 15«.) per ton.
A Report by the Italian Minister of Agriculture for the year
1894 (Chem. lud. 1895, p. 182) describes the depressed state of
the Sicilian sulphur industry.
In 1896, by the formation of the Anglo-Sicilian Sulphur
Company, working with English capital, the Sicilian sulphur
industry was at last placed on an economically sound basis, ami
at the same time the ruinous cutting down of prices ceased.
During 1900 the average price for '^ best thirds '^ was £3 13«. 6rf.
per ton f.o.b. Catania.
The total exportation of sulphur in Sicily during the year
ending June 30, 1900, was 517,741 tons; during the year ending:
June 30, 1901, 521,497 tons. This quantity was exported ta
different countries as follows : —
1900-1901. 1899-1900.
Tons. Tons.
United States 147,094 138,846
France 98,455 98,393
Italy 85,210 101,624
Germany 30,549 26,290
Norway, Sweden, Denmark ... 27,373 18,313
Greece and Turkey 22,304 19,795
Great Britain ....! 19,923 25,933
Russia 19,878 16,815
Austria 19,647 23,067
Holland 15,813 11,781
Portugal 11,315 11,462
Belgium 9,316 8,845
Spain 3,566 6,298
Other countries 11,059 10,259
(The quantities for a number of years previously to the above are
given in our second edition.)
Northern Italy formerly yielded a very large quantity of brim*
stone, especially the Romagna ; but many of the beds are now
exhausted, and the production is slowly receding, from 23,274
tons in 1886 to 21,663 tons in 1887 (details in the Chem. Zeit.
xii. p. 1659; abstr. J. Soc. Ch. Ind. 1889, p. 142). The whole
OCCURRENCES OF SULPHUR. 21
t>f this brimstone is used for inland consumption, as a remedy
against the vinendisease.
Salphur containing selenium is found in the Lipari Islands and
near Naples, but not in quantity.
In many other parts of the world deposits of sulphur have been
found, and have been sometimes declared to be very important;
but bitherto these sulphur-mines have not made any sensible
impression on the sulphur trade, of which Sicily has still almost
the monopoly. The more important of these sulphur-mines are
the following : —
Andalusia produces sulphur, some of which is refined in
Almeria.
In Germany sulphur has been found at Stassfurt (Berl. Ber.
1890, p. 192) and in Upper Silesia, near Ratibor, where beds up
to 20 ft. thick exist (Chem. Ind. ii. p. 136 ; Fischer's Jahresb.
1882, p. 223). It has been worked by extracting it from the ore
by means of carbon disulphide.
The production of. sulphur in Germany, as indicated by the
official statistics, does not distinguish between natural sulphur
and that which is recovered in the Leblanc process. It amounted
in 1897 to 2317 tons, in 1898 to 1954 tons, in 1899 to 1663 tons.
In 1899 the importation was 31,196 tons, in 1900 40,689 tons.
The exportation is very small.
In Austria, at Swoscowice near Cracow, an old sulphur-mine
exists which for some years was carried on with great vigour,
the sulphur being extracted from the marl by means of carbon
disulphide. This bed is now practically exhausted, and the mine
has ceased working (Wagner's Jahresb. 1878, p. 333, 1879,
p. 272; Fischer's Jahresb. 1885, p. 204).
Russia possesses considerable stores of sulphur. A bed of
brimstone has been found by GluschkofE in tlie Astrachan
Government, on the east bank of the Baskuntschak salt-sea ; it
is a mixture of «andy rock with 30 to 35 per cent, of pure sulphur
(Fischer's Jahresb. 1884, p. 264). In West Siberia considerable
beds of sulphur are said to exist. In the Vistula district a
sulphur-bed is being worked at Czarki, which in 1883 yielded
60,000 pud. In the sandy steppe of Karakum a large number of
conical hills have been discovered consisting of sulphury rock of
50 per cent. (Chem. Zeit. 1884, p. 478). In the north of the
Caucasus, in the Grodno district, sulphur has been found by
22 RAW MATERIALS OF MANUFACTURE.
Baron Heyking (Chem. Zeit. 1887, p. 1620). According to
Chem. Ind. 1892, p. 443, the sulphur is obtained at Tschirkat in
Daghestan to the extent of 300,000 pud (at 40 lbs.) per annum ;
the selling price is 1'18 roubles per pud.
According to Chem. Zeit. 1894, p. 2002, a large bed of brim-
stone has been found in Transcaspia, 60 versts from the port of
Usun-Ada on the Caspian Sea, only 2 versts from a railway-line.
The bed is just below the surface of the ground and is worked by
open quarrying. The ore contains from 35 to 40 per cent,
sulphur, and costs 20 to 25 copeks at Usun-Ada. If the results
expected were realized, Russia would become entirely independent
of Sicilian sulphur. Possibly this is the same occurrence which
is reported in Journ. Soc. Chem. Ind. 1900, p. 867, as '' the
richest in the world," 100 miles from Khiva, extending over
23 square miles.
Chonski (Chem. Zeit. 1895, Repert. p. 411) reports on the
production of sulphur, which, he says, has been tried in several
places in Russia, but has been everywhere discontinued. He
enumerates the drawbacks connected with the various methods for
extracting sulphur from the raw ore. This paper should be
consulted by those who consider any newly discovered vein of
sulphur-ore as being equal to ready money.
Comp. on Russian brimstone also Zsch. angew. Chem. 1897,
p. 36 ; Chem. Ind. 1898, p. 241 ; Niedenfuhr, Chem. Zeit. 1897,
No. 30; Machalski, Engin. Min. Journ. Ixx. p. 216; Chem. Trade
Journal, 1900, xxvii. p. 220.
Brimstone is got near Mossul in Mesopotamiay near Cairo, and
in Tunis. At Djemsali and Ranga, on the coast of the Red Sea,
the " Compagnie Soufriere '* is said to get 300 tons monthly.
Very large quantities of sulphur are said to exist in Iceland,
even more important than those in Sicily ; the deposits at Guld-
bringe Syssel, in the south-west of Iceland, were some time ago at
work with satisfactory results (Chem. News, xl. p. 31).
Japan possesses veiy large stores of sulphur, but the absence
of facilities for increasing its output and for shipping it has
hitherto very much restricted the development of the sulphur
industiy in that country. Formerly (comp. J. Soc. Chem. Ind.
1890, p. 344) there was only one place where sulphur was worked
on any scale, namely Atosanobori, near Kushiro, on the south-east
coast of the island in which the port of Hakodate is situated.
OCCURREXCES OF SULPHUR. 23
The Atoaanobori mine is part of an extinct volcano^ whose crater
and slopes are partially covered with a 50-per-cent. sulphur-ore.
According to the lowest estimate this mine contains a million tons
of good ore^ but there is probably five times as much. The output
is at present about 9000 tons per annum ; the cost of the sulphur
free on board at Kushiro is about SOs. 10d.y but it has hitherto
been shipped from Hakodate^ where it costs £2 Ss, 2d. per ton,
which is not remunerative. The exportation was 1541 tons in
1885, 4972 tons in 1886, 7096 tons in 1887, 3609 tons in 1888,
15,700 tons in 1895. The shipments generally go to San
Francisco.
Apart from the ordinary yellow sulphur, which sometimes con^
tains traces of selenium and tellurium, there occurs in Japan an
orange-red variety, containing 0*17 per cent. Te, 0*06 Se, 0*01
As, traces of molybdenum ^and earthy matters (Divers and
Shimidzu, Chem. News, 1883, No. 1256).
A considerable source of sulphur has recently been discovered on
the volcanic island of Etrof u, about halfway between the northern
extremity of Japan and Kamschatka. In 1900 already 10,000
tons sulphur were mined, and it is expected to reach 300 tons
per month (Joum. Soc. Chem. Ind. 1901, p. 300).
A small island belonging to New Zealand, evidently the crater
of a huge submerged volcano, which contains large deposits of
sulphur, has been described by Mclvor (Chem. News, 1887, Ivi.
p. 251). It will probably be very soon submerged as well.
1692 tons of sulphur were produced in that colony in 1900.
In the United States of America sulphur has been found in
many places, particularly at Cove Creek, Millard County, South
Utah ; at Rabbit-Hole Springs, Humboldt County, Nevada ; near
Lake Charles, Louisiana ; in Wyoming and other Rocky Mountain
Territories ; in Texas, California, and so forth. Very few of these
occurrences are worked, either owing to the smallness of the stock
or to the diflSculties of carriage. The Rabbit- Hole mines have
been worked off and on, but never at a considerable rate. For the
purpose of working the best deposit, the Cove Creek mine, a Com-
pany (the Dickert and Myers Sulphur Company) was formed
with a capital of $2,000,000, which acquired Dickert's patents,
mentioned on p. 18 ; but it did not produce more than about
2000 tons of brimstone in 1886, and 2700 tons in 1887, so that it
could not even supply all the needs of the Pacific coast, which
24
KAW MATERIALS OF MANUFACTURE.
amount to 4500 tons^ and have to be made up from Sicily and
Japan (details in the ' Mineral Resources of the United States,'
Calendar years 1885, 1886, and 1887).
A very large deposit of sulphur has been discovered in the
Grand Gulf Basin, near Lake Charles City, in the parish of
Calcasieu, State of Louisiana. Two wells have been bored at
different localities, which struck sulphur at a depth of about
^50 feet. The first and thickest bed of sulphur is from 108 to
112 feet thick ; it contains near the top 62 per cent., in the centre
90 per cent, of pure sulphur. One of the wells has been continued
lower down, and proved the existence of several smaller beds of
rich brimstone-ore, and several hundred feet of gypsum containing
sulphur ; in the other well no gypsum, but calcareous substance
containing sulphur is found. The working presents a difBcnlty,
as there are several hundred feet, of quicksand on the top of
the brimstone beds, which is to be overcome by the Frasch
process (p. 18).
In Lower California, not far from the mouth of the Colorado
River, a new bed of sulphur has been found. The crude sulphur
contains 70 to 80 per cent, pure S (Fischer's Jahresb. 1897, p. 420).
According to the volumes of ^ Mineral Industry,^ the production,
imports, and consumption of sulphur in the United States during
the years 1896 to 1900 have been as follows (in long tons) : —
Year.
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
Productiou.
Tons.
1,071
1,0^0
1,200
441
1,(»50
3,800
1,690
2,726
1 ,5(i5
4,630
Crude.
110,971
100,938
107,601
124,467
12.5,950
145,318
138,846
159,790
140,841
166,457
Importation.
1
1
/
Consumption.
\
Flowers.
Refined. ^
1
Total.
206
10 i
117,187
118,208
158
26
101,122
102,752
241
1 41 ,
107,885
109,090
165
41
124,073
r2,M 14
581
' 229
128,410
128,410
665
447
146,430
149.746
319
! 148
139,313
140,849
507
163
160,460
161.772
336
' 184
141,361
142,449
628
142
167,328
i
171,418
In British Columbia, on the Skeena River, a large deposit of
sulphur has been found, and it was expected to supply the Puget
Sound Country, and eventually the whole Pacific Coast, from this
source (Journ. Soc. Chem. Ind. 1901, p. 1040).
SULPHUR FROM OTHER MATERIALS. 25
In America sulphur is found also in many other places^ as in
the volcanic regions of Ecuador and Mexico, at Chilian in Cliili^
and elsewhere.
It has long been known that the crater of the Popocatepetl, in
Mexico^ contains large quantities of brimstone ; and a railway has
been projected in order to get at it. An annual production of
50,000 tons of brimstone was expected to begin with, and vitriol-
works were to be erected all along the line (Fischer^s Jahresb.
1884, p. 265). So far this project does not seem to have been
carried out.
On the island of Saba, one of the Antilles, there is a bed ot
brimstone, yielding on an average 45 per cent, of sulphur, which
was worked for some time, but had to be abandoned as not
paying the expense (Dingl. Journ. 1886, cclix. p. 43).
lu Chili brimstone is found at a height of from 12,000 to
13,000 feet above the sea. The mines, which were started in
September, 1900, by an Iquique Company, are 54 miles from the
nearest railway station. The product is said to be of very
good quality ; it is used in the provinces of Atacama and Tarapaca
for the recovery of iodine and for making blasting-powder for the
nitrate works. This seems to be the Taltal deposit mentioned in
a Consular Report, quoted Journ. Soc. Chem. Ind. 1901, p. 1039.
In 1901, the working was stopped as being at present unremune-
rative (Zsch. angew. Ch. 1901, p. 1243).
In Venezuela large beds of sulphur have been found, 20 or 30
miles distant from the coast, with which they are to be connected
by a wire-rope line. The beds are to be worked by the Deutsch-
Venezolanische Schwefelgruben A. G., at Cologne.
Sulphur Manufactured from other Materials.
Free sulphur is produced in large quantities in the manufacture
of coal-gas, and is contained in the spent oxide of iron. We shall
treat of this later on, as material for producing sulphur dioxide ;
in this place we mention it only as a source of obtaining free
sulphur. This is sometimes done by extracting it with carbon
disulphide ; but this process does not generally seem to pay (comp.
Journ. Soc. Chem. Ind. 1883, p. 491). Some of the processes
described below are specially intended for this kind of material.
Broadberry (Gas World, 1895, xxiii. p. 643) extracts the
sulphur from spent oxide of gas-works by means of benzol at a
26 RAW MATERIALS OF UANUFACTUBE.
temperature of 70° or 80° C, employing a circulating-apparatus.
1 gallon of hot benzol yields on cooling 2'5 to 2*75 lbs. of solid
sulphur, and retains 0*25 lb. in solution^ which is obtained in dis-
tillatioD. From an experiment with 20 lbs,, he calculates a profit
of £1 6s. Sd. per ton of spent oxide.
The sulphur made from gas oxide is mostly of a dark colour,
owing to the presence of a very small quantity of tarry substances,
and this makes it very difficult to sell, so that it is nearly always
burned without extracting it, as we shall see infra, No. 5.
A certain quantity of sulphur is obtained by the distillation of
pt/rites. This, however, pays in very few places, except under
special circumstances — for instance, at the works of J, D, Starck
in Bohemia, which formerly supplied nearly all the fuming oil of
vitriol, and where the distillation of pyrites is practised in order
to obtain a material for the manufacture of sulphate of iron. Be-
tween 1863 and 1872, 2440 tons of sulphur were thus made.
The distillation takes place in earthenware tubes 3 ft. 3 in. long,
5 in. high, and 5^ in. wide, quite open at the back, and in front
narrowed to an opening of | in. diameter; they are glazed with
common salt ; and three tiers of seven tubes each are placed in
each furnace. For each tube there is a small receiver of sheet-
iron half filled with water, and attached to the tapering end of the
tube. The charge of pyrites is put into the open end ; a slanting
piece of sheet-iron is placed in front ; and the opening is closed
with sand or pyrites cinders, as shown in fig. 1.
Here only one-third of the sulphur contained in the pyrites is
gained ; but in Sweden half (?) of the sulphur is said to be ob-
tained in a furnace built similar to a lime-kiln, and continued at
SULPHUR FROM PYRITES. 27
the top in a wooden chimney serving as a condensing-space. The
kiln, having been first made red-hot, is charged with pyrites, of
which one portion is burned whilst the other portion volatilizes
and is condensed at the top. The work is carried on continuously^
fresh pyrites being from time to time introduced through an
opening near the top, and the cinders being removed at the
bottom.
P. W. Hofmann (Dingl. Joum. ccxx. p. 232) proposed to utilize
the sulphur of the pyrites smalls. If sulphur dioxide is conducted
over red-hot calcium sulphide (from alkali- waste), it is at first
completely absorbed ; afterwards sulphur distils over, and the
calcium sulphide is transformed into sulphate. Tiie latter, by
conducting ordinary coal-gas over it in a red-hot state, or by
mixing it with coal and igniting, is reconverted into sulphide,
which can be used over again. Hofmann proposed to burn the
pyrites smalls at the mines, to obtain their sulphur by means of
calcium sulphide in a state fit for sale, to treat the cinders by a
process to be described in the 13th Chapter for zinc chloride and
sodium sulphate, and to work the residue for pig-iron in a blast-
furnace. This proposal has not found any application in practice,
and is not likely to do so, since the burning of pyrites dust by the
shelf-furnace is quite as advantageous for acid-making as that of
lump ore.
Gerlach (German patent 229, 1877) proposed to obtain sulphur
from sulphur-ores, and especially from the spent oxides of gas-
works, by heating them in iron or fireclay retorts whilst at the
same time superheated steam is passed through. The sulphur is
said to distil very rapidly. A description of this process, with
diagrams, is found in Wagner's Jahresb. 1879, p. 268. It was
tried in Upper Silesia with sulphur marl, but did not auswer
(Fischer's Jahresb. 1882, p. 234).
O. C. D. Ross patented a process in every way similar to
Gerlach's (E. P. No. 713, 1879) . Other processes of the same
kind are described in the ^ Scientific American,' xxxix. p. 276, and
in the ' Chemiker-Zeituug,' 1879, p. 241, by Dubois (E. P.
No. 13,108 of 1885 and No. 7129 of 1886 ; G. P. 41,718: the
last patent describes a revolving retort).
According to a French patent of the Societe de St. Gobain
(No. 107,820, 28th April, 1875), on the top shelf of an ordinary
shelf-burner for pyrites smalls (see Chapter IV,) a fireclay retort
28 RAW MATERIALS OF MANUFACTURE.
is to be placed, in which the pyrites is first deprived, by distillation,
of a portion of its sulphur, which is collected in the well-known
condensing-chambers as flowers of sulphur. After some time the
partially desulphurized pyrites is let down to the next lower shelf,
and so on to others lower down. Here the remainder of the sulphur
burns, and the gas goes into acid-chambers, whilst its heat causes
the distillation of further portions of pyrites in the retort. This
process does not answer : the flowers of sulphur obtained is very
acid; and both the burner and the chambers work very badly. A
similar process has been again patented by Labois (E. P. No. 9761
of 1884).
Buisine (Germ. pat. 73223) heats' half-roasted pyrites with
sulphuric acid, to recover sulphur, ferrous sulphate being ob-
tained as by-product. According to pat. 79706, the pyrites is to
be distilled in closed vessels at 700° C. and the residue treated as
above with sulphuric acid. The residue, consisting of sulphur and
ferrous and cupric sulphate, can be applied as it is to vines for
. certain diseases ; or else it is extracted with water, the residue is
worked for sulphur, and the solution, by treating it with metallic
iron, yields metallic copper and ferrous sulphate.
Holloway's process (E. P. No. 500 and 1131, 1878) at one time
excited much attention. He blows heated air through melted iron
sulphide, thus decomposing it into a cupreous matt and a slag,
together with free sulphur, which distils oflF and can be collected.
The principal aim of this process was the concentration of poor
copper-ores in countries where fuel is expensive. It has been
described by the inventor in a paper read before the Society of
Arts, which has been published together with the discussion
following the reading of the paper. Bode (Dingl. Journ. ccxxxii.
p. 433) has criticised it. Dr. Angus Smith (Alkali Reports,
1877-78, p. 47) expected important results from this process,
which is decidedly very interesting ; but it has found no practical
application.
Stickney (Eng. & Miu. Journ. Ixv. p. 674) heats pyrites to red
heat by means of producer-gas. There is an escape of HoS and
SO2, which are converted into free sulphur by means of a spray of
salt solution (comp. Schaff^ner^s and Helbig's process, vol. ii. 2nd
edition of this work, p. 857) ; the reaction is to be promoted by
electric sparks.
Frohling, Fleming, and Whitlock (E. P. 10,295, 1900) intend
SULPHUR FROM OTHER SOURCES. 29
to obtain practically all the sulphur from FeSg by heating the ore
in a retort in a stream of carbon dioxide or nitrogen with the
addition of a small regulated quantity of oxygen^ by which only
the iron is burnt to Vefi^, the sulphur being set free.
It has often been proposed to prepare sulphur by y^a^^in^ gases
containing sulphur dioxide through red-hot coal. A special appa-
ratus for this purpose has been proposed by Haenisch and
Schroeder (E. P. 6404, 1885). They pass the first gases through
fire-clay cylinders filled with coke and heated from the outside by
producer-gas ; the products of combustion travel through another
cylinder, filled with open brick- work and heated by the waste fire-.
gases of the first operation ; here the undecomposed SO2, the
carbon monoxide, carbon disulphide, and carbon oxysulphide act
upon one another, so that, if the current has been properly
regulated, ultimately only CO2 and S are formed. Or else the
SOo is at once treated with CO, according to the equation
S02 + 2CO = 2C02 + S.
This process has been tried on a large scale at Oberhausen
(comp. Chem. Zeit. 1886, p. 1039, abstracted in J. Soc. Chem.
Ind. 1886, p. 534), but evidently not successfully.
J. and F. Weeren obtain the SO2 for this purpose by calcining
sulphates with silica (G. P. 38,041). They describe a special
apparatus for this reduction and the reduction of the SO2 by in-
candescent coke to S.
Heinrici (Zsch. angew. Ch. 1898, p. 525) employs this reaction
for the purpose of utilizing the acid tar formed in the purification
of mineral oils. By heating this SO2 is evolved, which is reduced
to free S by red-hot coke.
Sulphuretted hydrogen has been very frequently proposed as a
material for the production of free sulphur, which has become an
economical possibility through the Glaus process. We cannot
describe this industry in this place ; it is fully dealt with in vol. ii.
of the 2nd edition of this work, page 891 et seq., and we subjoin
only a brief notice of some improvements proposed since 1896.
CaruUa (J. Soc. Chem. Ind. 1897, p. 980) prefers burning H2S
in an ordinary pyrites-kitn, in lieu of the Glaus kiln, as the yield
is much larger.
Baranoff and others patent (E. P. 7269, 1898) the production of
S from H2S, obtained by the reduction of native sulphates and
30
RAW MATERIALS OF MANUFACTURE.
treatment by COg. This H2S is passed over red-hot sulphates^
which are thereby reduced and yield free S and SOg.
Falaschkowski (Russ. pat. 5464 and 5477, 190L ; Chem. Zg.
1902, p. 15) describes the following modifications of the process of
Baranoif and Hildt for obtaining S and SOo from sulphates.
Instead of simply mixing the sulphates with coke, he moulds them
into briquettes by means of coal-tar, &c., which shortens the time of
reduction. The sulphide is decomposed by CO2 at a pressure of
2 or 3 atmospheres. The HgS is best not passed at once through
red-hot sulphates, but first through a solution of the sulphides,
which forms Ca(SH)2 and NaHS. The former is converted by
eneans of Na2S04 into NaHS, which with CO3 gives HgS and
NaHCOa. Only this HgS is employed for being oxidized by
sulphates to S and SO2.
Davidson and the United Alkali Co. (E. P. 17476, 1897, and
7088, 1898) describe improvements in the subliming of sulphur.
Albright and Hood (E. P. 11988, 1894) purify sulphur by
heating it for some time just below its boiling-point.
The tvorld's production of sulphur is stated as follows in Min.
Ind. ix. p. 611, from official sources, in metric tons : —
Country.
1895.
Austria 830
France 4,213
Hungary , 102
Germany 2,061
Greece 1,480
Italy 370,766
Japan 15,557
Bussia 190
Spain 2,231
Sweden nil.
United States 1,676
1896.
643
9,720
i 138
! 2,263
1.540
! 426,a53
12,540
437
1.800
77
3,861
1897.
530
10,723
112
2.317
358
496,658
12,013
574
3,500
nil.
1,717
1898.
496
9,818
93
1,954
135
502,351
10,339
?
3!l00
50
2,770
1899.
.555
11,744
116
1,663
1,150
563,697
10.241
?
1,100
nil.
1,.590
The sulphur produced as a by-product of the Leblanc process
by the Chance-Claus process in Great Britain is not included.
This is estimated for 1898 at 31,000 tons.
ANALYSIS OF SULPHUR. 31
Technical Analysis of Sulphur,
The first quality of Sicilian sulphur (prima Lercara or prima
Licata) is o£ a fine amber colour, in large shining pieces^ and does
not contain more than 1 per cent. ash. The second quality (seconda
vantaggiata) is still fine yellow, but not so shining, with ash up to
2 per cent.; the third (terza vantaggiata), which is that generally
used for sulphuric-acid making, contains up to 5 per cent, ash
and is coloured brown, partly by bituminous substances, partly by
amorphous sulphur. " Zolfo ventilato '^ (manufactured by Walter
aud Trewhella at Naples, Fischer's Jahresb. 1901, p. 406) is
sulphur, finely ground, raised by a chain of buckets and sifted by the
action of a current of cooled furnace-gases (which produce less
danger of fire than air would do, as they contain less oxygen) .
The ordinary technical assay of brimstone is performed by burn-
ing a weighed quantity, say 10 grams, in a small porcelain capsule,
and estimating the quantity of ash left behind. It is, however,
advisable to estimate also the moisture, for which purpose, in
order to prevent any evaporation during grinding, an average
sample of the unground or merely roughly-crushed material,
weighing 100 grams, is dried at 100° C. for some hours in an
oven or water-bath.
Bituminous substances are most easily recognized by the colour
of the sulphur j they occur chiefly in sulphur made from spent
oxide of gas-works, which is sometimes coloured quite black.
Arsenic sometimes occurs in brimstone, more especially in that
recovered from alkali-works, always in the shape of arsenious sul-
phide. On extracting the brimstone with disulphide of carbon,
the As^Sa remains behind aud can be estimated. Schaeppi (Chem,
Ind. 1881, p. 409) prefers extracting the AsgSg with dilute
ammonia (as described by the German Pharmacopoeia) at 70P or
80° C. In the solution the sulphur can be precipitated with silver
nitrate as Ag^S, which is filtered, washed, dissolved in hot nitric
acid, and estimated as chloride ; 6 molecules of AgCl correspond
to 1 of A83S3. It is, however, more expeditious to work volume-
trically. The ammoniacal solution of arsenious sulphide is exactly
neutralized with pure nitric acid, strongly diluted and titrated with
decinormal silver nitrate, till a drop of the solution, on addition
of neutral potassium chromate, produces a brown colour. When
.employing 10 grams of sulphur for extraction with ammonia, each
32 RAW MATERIALS OF MANUFACTURE.
c.c. of decinormal silver solution indicates 0*041 per cent. AS2S3.
Sometimes arsenic occurs in recovered sulphur in the shape of
ferric or calcium arsenite, which cannot be found by the above
process, but only by extracting the sulphur with carbon disulphide
and digesting the residue with aqua regia.
A qualitative reaction for arsenic consists in agitating 1 gram
sulphur with 15 drops liquor ammonia and 2 c.c. water for half an
hour, filtering, adding to the filtrate 30 drops of hydrochloric acid
and 15 drops solution of oxalic acid, placing a bright piece of brass
in the solution and heating to 60"-100° ; in the presence of arsenic
the brass is stained grey or black (Hager, Pharm. Centr. 1884,
pp. 263 & 443) .
Selenium is discovered by deflagrating the sulphur with nitrate
of potash, dissolving in hydrochloric acid, and treating with SO2,
which precipitates the selenium as a red powder.
Keed (Chem. Zeit. 1897, Rep. p. 252) describes the American
test for selenium. Boil 05 gram sulphur with a solution of
0*5 gram potassium cyanide in 5 c.c. water, filter and acidify the
filtrate with H CI ; a red colour, appearing within an hour, shows
selenium. Or else boil 1 gram sulphur with a solution of 2 grams
potassium cyanide for an hour, add another 0*5 gram KCy and
boil for another half-hour. Of course any iron present will react
with the sulphocyanide formed.
Macagno (Chem. News, xliii. p. 192) proposed the direct esti-
mation of sulphur by means of its solubility in carbon disulphide *.
Pfeiffer (Zsch. anorg. Ch. xv. p. 194, and Lunge, Chem. techn.
Untersuchungs methoden, i. p. 240) has re-investigated this matter
and given the following prescriptions for this method : — Shake a
weighed sample of powdered sulphur with exactly four times the
quantity of pure carbon disulphide, filter, reduce the temperature
to 15° C, and ascertain the specific gravity of the solution. The
following table (abridged from the original) shows the number of
parts of sulphur dissolved by 100 parts by weight of CS2 at 15° C,
for various specific gravities found : —
♦ F. B. Carpenter (Journ. Soc. Chem. Ind. 1902, p. 832) points out that crude
sulphui* sometimes contains a large quantity of gypsum which interferes with
the carhon-disulphide treatment. In such cases the calcium sulphate must be
previously removed by boiling with hydrochloric acid.
ANALYSIS OF 8ULPHUK.
33
Specific
Sulphur
Specific
Sulphur
Specific
Sulphur
graTity.
diaaolved.
gravity.
dissolved.
gravity.
dissolved.
1-2708
0
1-3087
8-5
13426
17-0
1-2731
0-5
1-3108
9-0
1-3445
17-5
1-2754
10
1-3129
9-5
1-3463
18-0
1-2779
1-5
1-3150
10-0
1-3481
18-5
1-2800
20
1-3170
10-5 ,
1-3500
190
1-2833
.7.5 1
1-3190
110
1-3517
19-5
1-2^7
30
1-3211
11-5
1-3536
20-0
1-2870
3-5
1-3231
120
1-3553
20-5
1-2894
40
1-3251
12-5
1-3571
21-0
1-2916
4-5
1-3271
130
l-a587
21-5
1-2938
50
1-3-291
13-5
1-3605
220
1-2960
5-5
1-3311
140
1-3622
22-5
1-2982
60 ,
1-3330
14-5
1-3640
23-0
1-3003
6-5 '
1-3350
150
13()57
23-5
1-3024
70
1-3369
15-5
l-ii(>74
24-0
1-3045
7-5
1-3388
160
1 3692
24-5
1-3066
80
1-3408
16-0
1-37(K)
25-0
The same subject is again treated by P. Puchs (Zsch. angew. Ch.
1898, p. 1189). His figures are slightly different from those of
Pfeiffer; but the latter^s results (which were obtained in my own
laboratory) seem to me the most accurate obtainable.
For the purpose of preventing the grape-disease (the Oidium) the
decree of fineness of ground sulphur is very important. In Prance
this is ascertained by Chancel's sulphurinieier, shown in fig. 2 on a
scale of 1 : 2. This is a glass tube sealed at the bottom and closed
at the top by a ground-in glass stopper. It is 23 cm. long and
15 mm. wide, and divided into 100 degrees of ^ c.c. each, beginning
from the bottom ; the 100 degrees occupy a space of 100 mm.
If ground sulphur is shaken up with anhydrous ether it forms,
after settling down, a layer, the height of which corresponds to the
fineness of the powder. The sulphur to be tested is passed through
a sieve with meshes 1 mm. wide ; 5 grams of it are put in the
tube, together with ether of 17°*5 C, or very nearly that tempera-
ture, filling half the tube. The tube is well shaken to break up
all small lumps, more ether is poured in up to 1 cm. above the
degree 100, the tube is again shaken and placed in a vertical
position. When the layer of sulphur ceases to subside, its height
is read off and stated as " degrees Chancel.'^
Ordinary ground and sifted sulphur shows 50° to 55° Chancel,
finer grades 70° to 75°. The finest grade is obtained not by
VOL. I. n
34
RAW MATERIALS OF MANUFACTURE.
•^o
sieving, but by fanning— this " zolfo ventilate " shows 90° to 95
Chancel (Walter, Chem. Zeit. 1901, p. 459 ; supra, p. 31). For
use in vineyards the sulphur ought to show at least
60° Chancel ; some go as far as 75° Chancel. 1% -•
Analysis of refined sulphur, — Sulphur in blocks
or rods is almost always practically pure ; it may be
tested for ashes, arsenic, and selenium as above.
Flowers of sulphur is never quite pure unless
specially washed ; it always contains some acid,
landa (Fischer's Jahresb. 1898, p. 421) found up to
0*283 per cent, ashes ; the average of 30 samples was
0*063 per cent. He also tests the solubility i n boiling
solution of caustic soda of spec. grav. 1*2. This
varied from 88 per cent, (in one case 68 per cent.)
to 99*99 per cent., average 98*04 per cent.
2. Pyrites.
^ITi
-n
1^
What is called pyrites or iron-pyrites, in a techni-
cal sense, is hardly ever pure iron disulphide, PeSo,
but either a mixture of this with gangue, or more
frequently at the same time with other sulphuretted
ores, as shown by ftumerous analyses. The iron di-
sulphide may be present either as iron-pyrites proper
or as marcasite. The iron-pyrites proper crystallizes
in the regular system, with parallel hemihedry,
proved even on the faces of the cube by striation
(fig. 3).
Besides the cube (I), the octahedron, a (IT), is
frequently found, often modified by the pyritohedron,
b (III), and, if both are equally developed, forming
the icosahedron (II). The combined forms IV, V,
VI, as well as twin crystals, frequently occur. The
crystals are often well developed ; but in the pyrites
used on the large scale they are mostly quite indistinct.
The colour of pure pyrites is greyish yellow, and
easily distinguished from that of copper- pyrites. The microcrys-
talline pyrites of trade often shows other colours, even a slate^
grey ; the powder is brownish black. Its cleavage parallel to the
faces of the cube is not very pronounced ; the fracture is con-
FVHITEf. 35
choidal or irr^ular. Hardness 6 to 6^, specific gravity 4-83 to 5-2.
Pure FeS, contains 46-67 per cent, of iron and 53'33 per cent, of
sulphur.
According to Meue, the pyrites from volcanic formations con-
tains DO water, but that from sedimentary strata both water and
Fig. 3.
clay. Among the first he classes tlie yellow octaliedral pyrites,
which is stable in the air; among the latter the grey cubical
pyrites, which is easily transformed into ferric sulphate (Compt.
rend. 29th April, 1867).
Marcasite crystallizes in the rhombic system, in rhombical
prisms wi = 106° 2' with longitudinal faces /=100^ and r, and the
end face p inclined to r at an angle of 158^ aff. Tiiey frequently
d2
S6 KAW MATERIALS OF MANUFACTUKE.
occur in twin crystals, partially united along one of the faces m,
also in triplets aud quadruplets, &c. (fig. 4), in fibrous, bulbous, &c.
varieties ; cleavage indistinct ; hard-
ness 6 to 6-5 ; specific gravity 4-65 *"'?■ *■
to 488; colour grey to yellow or
greenish yellow, lighter than iron-
pyrites proper; powder greenish
dark grey. Marcasite ia moat fre-
quently found in bituminous slate
and coal, and decays even more
quickly than pyrites, with the formation of ferrous sulphate and
basic ferric sulphate.
Detailed investigations on the different behaviour of pyrites and
marcasite have been published by A, P. Brown (Chem. News,
1895, Ixxi. p. 139, and following numbers) aud by Stokes (Bulletin
U.S. Geol. Surv. No. 186, 1901).
In the oiea of commerce there is often present, mixed with FeSj,
^magnetic pyrites (pyrrhotite) of the formula FejSg, with 60'5 iron
and 39'5 sulphur; colour between brass-yellow and copper-red ;
hanlness 3"5 to 4'5 ; specific gravity 4'4 to 4* 7. Pyrrhotite occurs
by itself in large quantities, which have been scarcely utilized up
to the present.
The copper-pyrites so often mixed with iron-pyritea is distin-
guished from it by its colour, yellow as brass, sometimes as gold ;
this colour modifies that of the iron-pyrites according to the degree
of admixture. It crystallizes in the tetragonal system, but in the
ores which concern us always occurs in a microcryatalline form.
Its hardness is 3'5 to 40, specific gravity 41 to4-3. Pure copper-
pyrites of the formula FeCuSj should contain 3053 per cent, iron,
34'58 per cent, copper, and 34'88 per cent, sulphur; but the ores
serving for the manufacture of sulphuric acid rarely contain beyond
4 per cent, of copper.
The first application of pyrites for sulphuric -acid making is gene-
rally ascribed to a Mr. Hill, of Deptford, who in 1818 took out a
patent for it ♦. In France, Clement- Desormes seems to have made
the first proposals and experiraenta in that way. His trials did
not, however, succeed, because he believed it necessary to increase
the combustibility of pyrites by an addition of coal. This is both
unnecessary, with properly constructed kilns, and very injurious
* Sotol attributes this honour to his couatryuian, d'Artiguaa, in 1793.
FIRST APPLICATION OP PYRITK8, 37
to the process, from the carbonic acid getting into the chambers.
A great difficulty was experienced in lighting the kilns. So long
as it was attempted to do this from below, like an ordinary fire,
the thing would not work. According to communications from
Mr. Todd (Government Inspector of Alkali Works), a workman
of his father's in Cornwall, in 1830, accidentally discovered the
way of lighting the kilns from the top, such as is practised to this
day. Generally, however, the principal merit of introducing
pyrites is ascribed to Messrs. Perret and Son, of Chessy, who were
led to it by the necessity of desulphurizing their cupreous pyrites
previously to getting the copper, and of condensing the gas evolved.
There was no question then of supplanting the Sicilian sulphur
generally. Perret and Son, with the greatest care, searched into
all the conditions necessary for a proper combustion of the ore ;
and to them the honour is due of having overcome all the difficulties
opposed to the solution of this problem, and of having rendered
the process technically useful. Already in 1833 they had burnt
iron-pyrites successfully ; and in a patent dated November 20th,
1835, they described their process, to which their partner, Olivier,
is said to have given the first impulse. In 1837, Messrs. Wehrle
and Braun, in Bohemia, used pyrites for generating sulphurous
acid (Otto, ' Lehrbuch der Chemie,' 4th ed. iii. 1, p. 545) ; but,
according to Bauer (/. c. p. 6), I. Brem had introduced this process
into Austria (at Lukawetz) already in 1833 — that is, at the same
time as Perret.
These trials at making sulphuric acid from pyrites possessed,
however, only local interest ; and probably for a long time no
general attempt to do away with Sicilian brimstone would have
been made, but for the Neapolitan Government, in 1838, being
induced to grant a monopoly for the exportation of Sicilian sulphur
to the Marseilles firm of Taix and Co. This firm at once raised
the price of brimstone from £6 to £14 per ton, and thus would
have given a fatal blow to the manufacture of artificial soda-ash,
just then beginning to take a firm hold, if no other source of
sulphur had been known, and if such an unnatural measure
could have been carried out at all. But the attempt came
too late — just after Perret and Son had proved that the pyrites
occurring in most industrial countries could be used for vitriol-
making without any difficulty. The result was to be foreseen.
Once the discovery of a new source of sulphur had become a
38 RAW MATERIALS Of MANUFACTURE.
matter of necessity^ there were, within twelve months from the
rise in the price of brimstone, fifteen patents taken out in England
for burning pyrites for the purpose of making sulphurous acid,
and even a larger number for making sulphur from pyrites,
gypsum, &c.
According to Muspratt^s ^ Chemistry^ (vol. ii. p. 1023), Dr.
Thomas Thomson was the first in England to point to pyrites in
this crisis ; but most probably many others at the same time
turned to it. So much is proved — that Thomas Farmer, of Lon-
don, was the first in England who employed pyrites on a large
scale (in 1839) for the manufacture of sulphuric acid (Hofmann,
' Beport by the Juries, International Exhibition, 1862, Class II.
Section A/ p. 12).
Mr. E. K. Muspratt states that his father, in 1839 and 1840,
used large quantities of Wicklow and Welsh pyrites at Liverpool
and Newton, the cupreous cinders being sent to the Sutton Copper
Company at St. Helens.
In Germany the Oker works, at the Rammelsberg, near Goslar,
seem to have been the first who calcined the local ore in such a
way as to convert the sulphurous acid given off into sulphuric acid
in acid-chambers; and other smelting- works, such as those at
Freiberg, followed their example. In these cases the reasons
were, not only that at a comparatively small expense sulphuric
acid could be obtained as a by-product from the sulphurous acid
otherwise lost, but also that the acid fumes destroyed the vege-
tation of a large district round the works, and that means had to
be taken for preventing this, apart from any consideration of
profit.
The Sicilian brimstone monopoly certainly did not last long ;
diplomatic pressure on the part of England soon led to its
abolition ; and with the lowering of the price of brimstone most
works at first returned to its employment. But the ice was now
broken ; the conviction had been gained that it was possible to
make acid without Sicilian brimstone ; the owners of pyrites-
mines took pains to advance the use of pyrites by low prices;
and thus, in the course of the next twenty years, brimstone was
gradually, but steadily, driven from its place in sulphuric-acid
making, in proportion as, on the one hand, it became dearer
from the causes above mentioned (disease of the vines &c.), and as
f
FIRST EMPLOYMENT OF PYRITES. 39
more pyrites-mines were opened out. In 1852 brimstone was no
longer used in Lancashire, but on the Tyne 7580 tons of it were
still burned.
The decisive point in favour of the use of pyrites was this^ that
continually increasing quantities of cupreous pyrites came into the
market^ from Spain especiaUy (first in 1859), but also from other
countries. These had at any rate to be burnt and their sulphur
expelled before they could be worked for copper. Already with
Perret and Son this had been the stimulus for their efforts ; but
this has been done on a much more colossal scale in consequence
of the opening out of the cupreous-pyrites mines in the south of
Spain^ in Portugal, and in Norway. In England iron-pyrites
is now all but out of the fields and has been supplanted by
cupreous pyrites. This has been the case to a less extent in
France and Germany^ because they possess mines of rich iron-
pyrites, which in England are not numerous ; but even in those
countries more cupreous pyrites is now used. In 1867 pyrites
had almost entirely supplanted brimstone in France as a raw
material for acid-making in the large industrial centres (' Rapport
du Jury International^ Expos. Univ. de 1867/ vol. vii. p. 19).
In Germany this state was brought about somewhat later, but
quite as completely. Only during the feverish years 1871 to 1873,
when the price of pyrites had risen very high, a few factories in
Hanover, in Hamburg, and Stettin temporarily returned to the use
of brimstone (Hasenclever, /. c. p. 155) ; but with lower prices of
pyrites this was given up again. Spanish pyrites began to be used
in (iermany in 1877.
Tims, starting from tlie use of iron-pyrites, that of cupreous
pyrites has followed ; and to this was added the employment of
other sulphurous ores and of the intermediate products of smelting
— for instance, copper-regulus {'' coarse metal '') at Mansfield and
Swansea. The first chambers working with SO3 from the metal-
lurgical treatment of copper and lead ores were started at Oker in
1859, at Freiberg in 1861. The Freiberg works employ for this
purpose even poor lead-matts with about 22 per cent, of sulphur,
which formerly was thought impossible (Bode, ' Beitriige zur
Theorie und Praxis der Schwefelsiiurefabrikation,' 1872, p. 1).
We shall further on describe the application of other sulphur^ores
apart from pyrites proper.
40
RAW MATERIALS OF MANUFACTURE.
The principal sources of pyrites will now be mentioned without
separating the cupreous from the non-cupreous, because no strict
limit can be drawn. Some kinds of pyrites contain so little copper
(below 1 per cent.) that it cannot be utilized ; and these go with
the totally non-cupreous ores.
Great Britain possesses several deposits of pyrites, but none of
very great importance. In Cornwall and Devonshire pyrites of
the following composition is found : —
rattinson.
Sulphur 27-00
Iron 22-69
Copper 2-00
Lead trace
Zinc 1-23
Lime 0-22
Carbonate of Lime
Magnesia 0*12
Arsenic 032
Insoluble (silica) . . , 45*60
Oxygen as FcgO;; . , 0* 1 3
Moisture 0*64
Clapham (8 analyses).
24-013-34-880
27-076-60-676
0-400- 4-600
0- - 7-446
0- - 9-086
Gypsum 0* - 0-596
0- - 3-579
0- - 1-160
2000-38-676
99-95
Cleveland pyrites (from the north of: Yorkshire) is only used in a
local factory ; in 1874, 500 tons of it were obtained. Composition
(according to Pattinson^s analysis) : —
Iron disulphide 52*12
(corresponding to 27' 18 sulphur) .
Iron protoxide 11-92
Alumina 8*10
Lime , 0*27
Magnesia TOO
Carbonic acid 2-40
Insoluble in acid 11*12
Water ../. 12-80
99-79
PYRITES IN GREAT BRITAIN.
41
In Ireland there are large beds of pyrites, especially in the
county of Wickloir ; and up to about 1862 this Irish ore supplied
a very large portion of the pyrites used in England. In 1860
still 40,000 tons of it were imported into the Tyne River ; but in
1863 the importation had fallen to 4000 tons, and has long since
ceased entirely. The same has been the case in Lancashire j and
only a few works in Ireland itself burn this kind of pyrites. It is
found in the county of Wicklow in beds from 6 to 50 feet in
thickness, which overlie siliceous clay-slate. The beds go down
to depths of 80 to 100 fathoms. The bulk of the ore contains
only 30 to 35 per cent, of sulphur. A small quantity only
of richer ore (analyses a, b, c) has been found in the valley of
Ovoca. The Irish ore is too hard and slaty and does not bum
well ; it requires a great heat, and consequently deep kilns. It
nearly always contains copper, but rarely suflScient to pay for
extracting it, from which standpoint the following analyses must
be judged : —
Pattinson.
a.
Clapham.
b.
r.
d.
e.
Sulphur
.. 44-20
40-410
42-128
37-975
34-()76
Iron
.. 40o2
32-222
a'l-ooo
34-650
42-400
Copper
.. 0-90
4-133
2-400
2-400
1-333
Lead
.. 1-50
2-963
1-600
1-080
1-593
Zinc
.. 3-51
• • ■
• • •
• ■ •
• • •
Arsenic
.. 0-33
• • ■
0-602
0-40(»
0-183
Lime
.. 0-24
• • ■
• • •
• • •
• ■ •
Inifoluble
8-80
17-676
18-676
22-500
20-000
Moisture
.. 0K)9
• • •
• ■ ■
• ••
• • ■
Oxygen as Fe-^O,
.. 0-25
• • •
• ■•
■ • ■
• • •
100-34
97-404
100-406
99-005
100-185
In JVales (in the Cae Coch Mine) pyrites is found entirely free
from arsenic^ according to Drink water (Journ. Soc. Chem. Ind.
1885, p. 533). It contains 48*3 per cent, of sulphur, and is used
for manufacturing very pure oil of vitriol.
A certain quantity of pyrites is picked from coals, and is known
as coal'brasses. If quite free from coal, they are very rich —
according to R. D. Thompson (in Richardson & Watts's ' Chemical
Technology,' vol. i. pt. iii. p. 15), 53*55 sulphur, 45*07 iron, 0*70
manganese, 0*80 silica; but practically they cannot be obtained
42 RAW MATERIALS OF MANUFACTURE.
in this state^ and the carbonaceous matter adhering to them causes
very much trouble in the vitriol chambers. Their principal use
in acid-making is for lighting the burners, or for getting up their
heat if it has gone down. Mr. G. E. Davis informs me that coal-
brasses with 44 per cent. S and some carbon can be burnt in
ordinary pyrites-kilns, and, if proper arrangements are made, yield
perfectly clear acid from the Glover tower. Apart from this they
are worked up by weathering for copperas and Venetian red, and
this is probably the way in which most of the quantity obtained
in England (10,000 tons in 1874) is consumed.
The total production of pyrites in the United Kingdom was
25,401 tons in 1882, 27,829 tons in 1886; in 1887, 22,079 tons;
in 1888, 23,507 tons ; in 1889, 17,719 tons; in 1890, 16,018 tons ;
1895, 9193 tons; ^1896, 10,177 tons; 1897, 10,752 tons; 1898,
12,302 tons; 1899, 12,426 tons; 1900, 12,484 tons.
The importations of foreign pi/rites into Great Britain during the
years 1862 to 1887 are quoted in our second edition, vol. i. p. 32.
Here we shall quote (from the volumes of ^ Mineral Industry ')
the importations of pyrites (iron and cupreous) into Great Britain
since the year 1888. (The tons are metrical tons at 1000 kil8.=
0-9812 English ton.)
Year. Pyrites imported. ' Year. Pyrites imported.
Tons. Tons.
1888 629,056
1889 654,872
1890 667,625
1895 591,782
1896 598,480
1897 633,009
1891 662,297 1898 665,544
1892 614,238 I 1899 712,393
1893 622,63i i 1900 752,605
1894 625,968 |
The most important German pyrites-bed is that of Meggen in
the Siegen district, in Westphalia, 3 miles from the Altenhunden
station on the Ruhr-Sieg railway. This bed occurs along with
heavy spar in the so-called " Kramenzel;^^ it is known for a length
of 2000 fathoms, and its thickness changes from ^ to 3 fathoms
(Wagner^s ' Jahresbericht,' 1865, p. 221). The same authority
states it to be *'grey iron-pyrites," quite uncrystalline, free from
arsenic [?] . The mass of ore above the bottom of the valley is
PYRITES IN GERMANY. 43
estimated at 4J million tons ; how far the ore descends below the
bottom of the valley is as yet unknown.
The ore has not an attractive outward appearance ; its colour
is slate-grey ; but it burns very well in the kilns^ and it would be
even more valuable if the zinc contained in it did not prevent its
burning completely. Older analyses of it are given iu our 2nd
edition^ vol. i. p. 35. Here I quote the most recent analyses
made at the Rhenania Chemical works^ as furnished to mc by
Dr. R. Hasenclever in 1902 : —
a,
Gangue... 1202
Sulphur 41-94
Iron 34-92
Zinc 7-56
Lead 038
Lime 0*50
Arsenic trace
b.
Average.
12-96
120
43-42
400-43-0
35-56
35-0
5-81
7-0
not estim.
0-3-0-5
ij
0-1-0-5
005
trace-005
97-32 97-80
The average represents 12 per cent, gangue, 75 per cent, iron-
pyrites (FeS2), 10*5 per cent, zinc-blende (ZnS), together 97*5.
Jurisch (Schwefelsaure-fabrikation^ p. 14 et seq.) quotes a
number of analyses of Westphalian pyrites, by F. Quincke, for
the year 1892. In these the sulphur varied from 41 to 46-75
per cent., iron from 2955 to 36*16 per cent., zinc from 8*2 to
19-41 per cent. (I exclude two samples containing 27*58 and
4205 per cent, zinc, which are more correctly classed as blende
than as pyrites), lead from 0-3 to 1*7 per cent., arsenic from 01 to
0-2 per cent., besides silica, alumina, lime, magnesia, manganese,
alkalies, and small quantities of other substances.
The same source quotes analyses of pyrites from Bensberg
(46'86 per cent. S), Aachen (46*0 per cent.), Rammelsberg
(44-65 to 48-4 per cent.), Freiberg (52-20 per cent.).
The following is a complete analysis by Fresenius of ore from
the Philippine pit belonging to the Sicilia Mining Company at
Meggen, made in 1898 : —
44 RAW MATERIALS OF MANUFACTURE.
Iron 34-89
Zinc 8-38
Manganese 0"155
Cobalt and Nickel 0024
Lead 0*298
Alumina trace
Lime 1 '41
Magnesia 0*75
Sulphur 44*55
Arsenic 0*07
Carbon dioxide 1*90
Phosphorus peroxide trace
Gangue 5*83
Oxygen as sulphate^ thiosulphate^ &c.^
and traces of other substances 1* 743
100,000
There are smaller beds of pyrites in several other places in
Germany, such as those near Goslar, near Schwelm in West-
phalia, near Merzdorf in Silesia, &c. Their production is only
small compared with that of the Meggen pyrites. The pyrites of
the Rammelsberg in the Harz, according to Mene, contains 48*4
per cent, of sulphur (probably only picked lumps). The cupreous
pyrites of the same place is stated by Hilgenfeld to contain : —
Copper 12-22
Lead 2*43
Iron 39-10
Zinc, Manganese, Cobalt, Nickel 1*23
Arsenic 018
Antimony 0'16
Sulphur 44'65
Selenium, Thallium, Indium, Bismuth traces
99-97
The bed of Schwelm in Westphalia, in the Devonian formation,
has a thickness of from 10 to 33 feet, over a surface of nearly
150 acres, and is covered by rich iron-ore; the pyrites itself
consists of two-thirds powder mixed with well-crystallized pieces.
PYRITES IN GERMANY AND AUSTRIA. 45
The ore contains about 40 per cent, sulphur^ and more or less
clay, which is removed by washing ; after this it is sold to the
vitriol-makers, who like it on account of its freedom from arsenic
(Dingl. Journ. ccxxviii. p. 283) ; Hjelt, however, found more
arsenic in it than in Meggen pyrites.
On the island of Wollin a pyrites-bed is found in a bed of marl
belonging to the chalk formation.
A great drawback to the German ores is their proportion of
zinc, which retains a considerable quantity of sulphur in the
state of sulphate. In Silesia 14 per cent, of zinc is allowed as
a maximum; upwards of this for each per cent, of zinc the same
quantity of sulphur is deducted from the invoice (Kosmann, in
Fischer's Jahresb. 1886, p. 268).
The production of German pyrites during the years 1853 to
1886 is quoted in the second edition of this work, vol. i. p. 38;
that between 1891 and 1897 by Hasenclever, Chem. Ind. 1899,
p. 25.
The maximum production in 1874 was 143,476 tons ; the
average of the last years is about 130,000 tons. The production
was 133,302 tons in 1897, 136,849 tons in 1898, 144,602 tons in
1899. Hasenclever estimates that about 65,000 tons was con-
sumed in the manufacture of wood pulp. The importation of
foreign (chiefly Spanish) pyrites into Germany increased from
238,643 tons in 1891 to 357,017 tons in 1897, all of this for
sulphuric acid. In 1898 it was 376,817 tons, in 1899 437,732
tons, in 1900 457,679 tons.
The exportation is small : in 1898 19,219 tons, in 1899 16,985
tons, in 1900 24,936 tons.
Austria-Hungary possesses large beds of pyrites at Schemnitz
and Schmolnitz in Hungary, in Styria, and Tirol. The pyrites from
Schemnitz contains on an average 47 to 48 per cent. S, 39 to 40
per cent. Fe, 0*58 per cent. Cu, 1*5 to 2 per cent. Zn, besides lead,
silver (81 grams per ton), and gold (2*2 grams per ton). Schmol-
nitz ore contains 44 to 48 per cent. S, 0*4 to 06 per cent. Cu,
2 to 3 per cent. Zn. Jurisch (from whose * Schwefelsaure-
fabrikation,' p. 18, the above is taken) quotes an analysis of
Schmolnitz pyrites with 48*89 per cent. S, 0'32 per cent. Cu,
0'14 per cent. As; 70,000 or 80,000 tons of this pyrites are
obtained per annum.
In Tirol pyrites is found testing 40*5 to 41*3 per cent. S.
In Styria (Riedl, Zeitschr. f. d. chem. Grossgewerbe, ii. p. 567),
46
RAW MATERIALS OF MANUFACTURE.
in the Saun valley, a number of beds of very pure but easily decom-
posable pyrites occur in the clay porphyry^ with a percentage of
48 to 52 of sulphur. It is used in the chemical works at Hrastnigg^
and in Bohemia. The production is about 3000 tons per annum
(Fischer, Jahresb. 1886, p. 255).
In Switzerland, in the Canton of Wallis, a pyrites is obtained
which does not seem to have found any technical application
as yet. Its composition, according to an analysis made in my
laboratory, is : —
(of this 0*05 as lead
sulphate in the in-
soluble residue).
Sulphur ....
50-46
Iron
44'55
Lead
0-37
Lime
1-13
Maffnesia ....
Oil
Carbonic acid
1-01
Silica, Iron
Alumina .
Moisture . . . .
, ,, ,
peroxide,
• •••••••••a
}
. . .
1-70
0-40
(insoluble).
99 73
In Poland pyrites occurs together with blende, containing some
thallium (Antiporo, Fischer's Jahresb. 1897, p. 421).
lu Belgium a rich pyrites is met with, the great friability and
softness of which do not tell in its favour. The following are
analyses of this pyrites : —
0.
Sulphur 42-80
5ITic 1
36-70
7-23
0-92
0'4(>
Iron
Ferric oxide
Oxygen in ferric
oxide...
Leud
Zinc
Arsenic 0*20
Thallium
Alumina trace
Silica 8-8(>
Carbonic acid
Calcium carbonate . 0*84
Lime
Water I'-IG
b.
35-50
38-60
4-24
c.
46-20
40:)0
2-20
0(ii>
5-26
031
14-90
trace
0-5(>
0-41
0-22
0-41
d.
40-01
39-68
0-32
0-37
1-80
trat^e
trace
9-10 12-23
042
025
025
e.
50W
43-61
0-18
1-75
trace
2-85
0-73
45-60
38-52
6-0O
0-92
0-10
trace
9-00
Oil
0-36
99 U 10002 lK)-46 9991 10014 9959
PYRITES IN BELGIUM AND FRANCE. 47
{a, b, and c bv Clapham in Richardson and Watts's ' Chemical
Technology/ vol. i. part iii. p. 14 ; d, pyrites from llodieux near
Spa ; e, from Santon's pit on the Meuse^ both by Pattinson, /. c. ;
f, by MacCuUoch^ Chem. News, xxvii. p. 125.)
The Belgian pyrites is usually only got as a by-product in
obtaining lead- and zinc-ores in the provinces of Liege and
Namur ; it is either microcrystalline or crystalline or in bulbous
pieces with a concentrically fibrous structure. Its quality is
uneven. It is mostly used locally and in the north of France.
In I860 as much as 42,513 tons of pyrites was got, but this was
the maximum attained. In 1879 still 15,577 tons were procured,
but in 1880 only 7913 tons, and in 1881 2965 tons. The pro-
duction remained about 2000 or 3000 tons for a number of years,
but since 1898 it has been altogether insignificant.
In France the principal deposits of pyrites are those of the
Rhone (Chessy and Sain-Bel) and of the South (Gard and Arduch).
The Rhone beds exist on both banks of the Brevenne, a tributary
of the Saone, on a widtii of 4 or 5 miles. The bed on the left bank
is that of Chessy, about 6 miles long and several yards thick.
This pyrites is bright yellow, very crystalline and friable. When
first got it contained 4 or 5 per cent, of copper ; but the cupreous
vein has run out, and the non-cupreous ore on this side has nearly
ceased to be worked. The beds on the right bank are those of
Sain-Bel or Sourcieux. The northern part furnishes a more com-
pact ore than that from Chessy. Most of it is non-cupreous, but
there is also a vein with 4 or 5 per cent, copper ; the gangue is
mostly sulphate of baryta. Much more important is the southern
part of this bed, the " masse de Bibost.'' The ore is here very
rich in sulphur, green with yellow reflexion, and very friable, so
that there is almost as much smalls as lumps ; the gangue is almost
entirely siliceous. The beds in the South of France are more
numerous, but much less important. The most considerable mine
is that of Saint-Julien-de-Valgalgues^ in the Dc'partement du
Gard; there is another mine at Soyons, in the Ardeche. The
other French mines are of little importance. A detailed descrip-
tion of the French pyrites-mines has been given by Girard and
Morin (Compt. rend. 1875, Ixxxi. p. 190; Annales de Chimie
et de Physique, [5] vii. p. 229) ; from this paper the following
analyses are extracted, which seem more trustworthy than those
48
RAW MATERIALS OP MANUFACTURE.
J
%
CD
<
§1
22
5 c
Cico
c
o
S^s
*^ =5 5C 2? 30
TT 1
6 e-rco
oo
i
=3 i-J
o
a
2p6
c
X
3
rc »5 XCiX
— X p « o
c o
g
O
2
X0i?O
§
g^is
•^«
=r,®
c
Ha
i
a
a
CQ
XOi-i X
CO
-I 2
S3 ■ « •
»- rO r-i •-H
1
s
'Tl t^
ei-Tit^
o
8
coca :
. • ft •
:ccoo
cu
<^co
r^
sa
o
V
eS
s
^?5
CO
>^
»
00
o
-
*
J3
O
•
s
'^'?i L*:
s
3
'M c: X
• >^ '^ ^K
• 1^ w' -V
3 : ^
3 2-0^
XmQNI«<1
5§l
'.S £ S =>
cSs
•
'3
s g
i o
1-1 "M o »o 2
e?i
32 m
nderg
5gou^e
: 26
> 1-^
I ^
c '5
1-H -^ »-i ic
1— I C^l »-i kC
©X
S2
-t" "^
•— « *?! . »-*
•2 ^3*0
«9 E 3 3
S w
3 : c» c » s s ^
= *2 3 ® 3 3 S
^ 2 ?^i2 ^^ c
«
3
e 3
esc to
PYRITES IN FRANCE. 49
•
by ilene, given in the 1st edition of this book^ p. 92, from the
Monit. Scient. 1867, p. 410.
These analyses may be summed up as follows : — ^The pyrites from
the Rhone and Sain-Bel, on an average, contains 46 to 48 per
cent, of sulphur and 10 to 12 per cent, of gangue, consisting of
clay, sand, and barytes. In the southern part of the district
of Sain-Bel the percentage of sulphur rises to 50 or 53, and the
gangue is inconsiderable and free from barytes ; arsenic is not
present in quantities sufficient for determination. In the district
of St. Julien (Le Card) pyrites is not found in the clay-slate, as
at Sain-Bel, but in the Lias or Trias ; the sulphur varies from 40
to 45 per cent. ; the gangue is calcareous, and varies from 3 to 6
per cent. ; arsenic is present up to O'l per cent., sometimes also
fluorspar in quantities sufficient for estimation. The pyrites
from Ardeche contains from 45 to 50 per cent, of sulphur ; the
gangue is clay, free from lime ; arsenic occurs up to 0*3 percent.:
fluorspar sometimes occurs in injurious quantities ; the hydro-
fluoric acid given off from it in one case destroyed the glass
apparatus for spreading the nitric acid, and the latter thus got
to the chamber-bottom and corroded it.
In 1874 there were used in France 178,400 tons, of the value
of £240,000. Of this the beds of Sain-Bel, which supply two-
thirds of the French factories, contributed 120,000 tons, those
of St. Julien (in the department Gard) 24,600 tons, those of
Le Soulier (Gard) 6000 tons, those of Soyons (Ardeche) 900 tons.
Girard and Morin give twenty-three analyses of French and five
of foreign pyrites used in France.
According to Scheurer-Kestner (Wurst, Diet, de Chimie, ii.
p. 138) the pyrites from Chessy and Sain-Bel contains 45 to 48 per
cent, of sulphur with very little arsenic and selenium, that from
Chessy also 1 or 2 per cent, of copper and zinc. The copper is
obtained from the cinders, at least at the Chessy works, by allowing
them to lie for a time and moistening them : the liquid running oS
contains copper and zinc sulphates ; and the copper is got from
it by cementation. Nearly all French works^ as well as those
in Alsace and Switzerland, obtain their ore from those two
pits ; only the works at Gard and Marseilles get it from Alais,
where the pyrites contains 38 to 42 per cent, of sulphur ; a few
factories in the north of France use Belgian pyrites, those in the
south use Spanish pyrites.
VOL. I. E
50
RAW MATERIALS OF MANUFACTURE.
The production of pyrites in France was : —
1891.
243,030 tons.
1896.
295,325 tons
1892.
226,804 „
1897.
298,571 „
1893.
227,288 „
1898.
306,002 „
1894.
278,452 „
1899.
313,087 „
1895.
248,934 „
France imported and exported pyrites as follows : —
1890. Imported 39,552 tons. Exported 15,907 tons.
1891. „ 45,457 „ „ 12,120 „
Italy possesses beds of pyrites in several places. Those
occurring in the province of Bergamo, tested in Vienna (Wag-
ner's Jahresb. 1879, p. 272), are composed as follows : —
Iron
Copper
Zinc
Lead
Silver .'
Sulphur
Arsenic
Alumina
Lime
Magnesia
Silica
COj, O and H^O (by diff.)
Redolta
quarry.
lOO-OO
Fasseyra
quarry.
trace
41-72
trace
trace
■ • •
• « •
39'32
44-'36
0-53
014
2'37
1-28
5-89
0-88
0-66
0-39
7-16
9-68
7-78
l*5.i
10000
S. Guiseppe
pit.
48-35
007
018
30-97
• ■ ■
1-86
1-70
014
10-45
6-28
100-00
Vallantica
pit.
36-79
1-69
• • •
trace
0014
41-56
0-18
1-25
0-37
010
16-40
1-646
100-00
In the Val d^Aosta there are several mines, some of which
contain a strougly arsenical pyrites.
Those at Brosso, near Ivrea, belonging to Messrs. Sclopis & Co.,
yield pyrites containing very little arsenic : present output about
20,000 tons per annum. One quality contains 48 or 49 per cent,
of sulphur and 0*2 of arsenic, the other nearly 50 of sulphur and
only traces of arsenic. It is too explosive for burning in lumps,
but excellent for burning as smalls. Another mine is at
Pr^ St. Didier in the same valley. This Aosta pyrites mostly
requires special contrivances for getting rid of the arsenic in
burning ; the cinders are worked for copper, silver, and gold.
PYRITES IN ITALY, SWEDEN, AND NORWAY. 51
Pyrites in quantity is also found at Agordo (Cadore), Sestri
Levante, and of very good quality in Sicily. The whole of the
24 Italian sulphuric-acid works burn pyrites, partly imported
from Spain (Caudiani, Chem. Ind. 1895, p. 153) ; none of them
bum brimstone.
The production of pyrites in Italy was : —
1897 57,383 tons.
1898 66,120 „
1899 75,308 „
Swedish pyrites, from Fahlun, varies between 43 and 48 per
cent, of sulphur. This ore is obtained as a by-product in the
getting of copper-ores, and is said to exist in enormous quantity ;
but, owing to the diflSculty of transit, its exportation does not pay.
It is said to burn well.
Pattinson. Browell and Marreco.
Sulphur 43-70 38*05
Iron 3901 42 80
Copper 0-60 1*50
Lead 0-12
Zinc 2-57
Lime 0*85
Magnesia 0*69
Arsenic trace
Insoluble 11-66 12-16
Oxygen, as Fe A 0-22 fjS)^'^^
Water 0-20
99-62 100-00
In Norway there exist, very lai'ge beds of pyrites, both free
from and containing copper. Of the many pits formerly worked
there, all those had to stop which produce ores with from 35 to
40 per cent, of sulphur. The richer ores, even those free from
copper, have maintained their position to some extent, because they
bum well, are easily lighted, keep the heat well, do not " scar,'' &c.
They are mostly hard and difficult to break. The most consider-
able pits are those of Ytteroen, which export vid Drontheim; they
supply 6000 to 8000 tons per annum. A large mass of cupreous
£2
52 RAW MATERIALS OF MANUFACTURE.
pyrites, with 45 per cent, of sulphur and 3 of copper, at Vigsnaes,
was worked by an Antwerp Company, but is now exhausted.
Norwegian pyrites contains very little arsenic. Other pits exist
thirty miles from Drontheim, on the Hardanger Fjord, near
Bergen, &c. The Norwegian pyrites is more in favour as a
sulphur-ore (excepting its difficult breakage) than as a copper-ore ;
its cinders do not very well suit the copper-works.
Analyses of Norwegian Pyrites.
Faltinson. MacCuUoch.
/^
Ytteroen ore. Drontheim ore. I. II.
Sulphur 44-50 5060 46-15 38-17
Iron 39-22 44*62 4420 82-80
Copper 1-80 trace 120 110
Zinc 1-18 1'34 210 2-32
Lead trace
Lime 2-10 trace
Calcium carbonate ... 2'55 11*90
Magnesia 0*01 trace
Magnesium carbonate ... ... 1*08
Carbonic acid 1*65 ... ... ...
Arsenic ... ... trace
Insoluble OOS 315 3-20 1220
Oxygen, as Fe.^O., 0*45
Moi.sture ....'... 0-17 020 040 0-25
100-16 99-91 99-60 99*82
The following information is due to Mr. Knudsen^ manager of
the Sulitjelma mine (through Dr. Hasenclever). Norway exported
in 1901 about 90,000 tons pyrites from the following pits : —
Sulitjelma, near Bodoe . . . 35-36,000 tons with 45 p.c. sulphur.
Killingdal 25,000 „ 43-44 p.c. „
Roros, nearTrondhjera... 15,000 „ 43-44 „ „
Bossmo „ ... 15,000 „ 48-50 „ „
By far the largest mine is the first-mentioned, which is expected
to yield an additional 20,000 or 25,000 tons in 1902 and is
reckoned to last for centuries to come. It contains very little
arsenic. Much ore, richer in copper and poorer in sulphur, is
also smelted on the spot. The ore from Killingdal and Roros
is also cupreous ; that from Bossmo is free from copper, with
PYRITES IN SPAIN AND PORTUGAL. 53
traces of arsenic. The following mines are not yet worked for
want of railway communication : — Foldal, Vaarteigen, and Nudal.
They test 43 or 44 per cent. S, and at most 2*5 per cent. Cu, and
might yield from 60,000 to 90,000 tons per annum.
Spain and Portugal possess the largest known beds of pyrites.
Much of it is cupreous, and all of it is distinguished by its very
good behaviour in burners ; so that the burners have been built
very much lower for it, and much labour is saved. This pyrites
has only been worked again since 1855 ; but the Romans, and
before them the Phoenicians and Carthaginians, knew it very
well, as is shown by many traces. The bed, however, was at that
time only worked where it was richest in copper. According to
Schonichen (Dingl. Journ. clxx. p. 448) all the beds are within a
belt of five leagues width, reaching, parallel to the Sierra Morena,
from the western frontier of the province of Seville, across the
hilly country situated to the south of this, right through Portugal
to the Atlantic Ocean — a distance of 30 leagues. The prevailing
rocks in that country are clay- slate and crystalline slates; but
parallel to the granitic tract of the Sierra Morena felsite-porphyry
and quartzite have broken through the slate, and only in the
neighbourhood of such dykes are the pyrites-beds found. Their
shape is that of large lenticular pockets in the metamorphic clay-
slate, from 20 to 36 fathoms thick, and extending to a length of
170 to 260 fathoms. The whole bed is filled with pure pyrites,
without any visible gangue. The ore is in a few places found
at only 1 or 2 fathoms below the surface, undecomposed, and
in a sandy state, so that it can be got by daylight work. In other
places the zone of decomposition reaches from 10 to 50 fathoms
downwards. The percentage of copper varies from 2^ to 40 ; but
ores with more than 10 per cent, of copper are only contained in
small vertical zones within the large masses. Only these " black '^
ores were the object of the mining-operations on the part of the
Phoenicians and Romans. The quantity of pyrites existing there
is almost inexhaustible, and can certainly supply the requirements
of mankind, both of copper and of sulphur, for thousands of years
to come.
Special highways, and latterly also railways, have been made, in
order to facilitate the communication with the ports of Huelva,
San Lucar du Guadiana, and Pomaron ; but a great deal of the
ore is still conveyed for some distance on mule-back.
54 RAW MATERIALS OF MANUFACTURE.
Of the many companies which had been formed for working
this ore most have ceased to exist ; and only four or five remain,
all of them in English and French hands. The smallest of
these is the Buitron Pyrites Company, which works the mines
of Buitron and Poderosa. The Tharsis Sulphur and Copper
Company possesses much more extensive mines, a railway of its
own, a wharf at Huelva, and also a number of works in England
and Scotland for the wet extraction of the copper from the cinders
returned to them. The Tharsis ore is very good, but very soft,
and makes much dust in breaking. The San-Domingo mine lies
in Portuguese territory ; its ore is known as Mason's ore, and is
considered superior to all others^ so that it commands a better
price. The last, but largest, of these companies is the Rio-Tinto
Company, which has thrown such large quantities of pyrites into
the English market that, from 1875 to 1876, prices receded by
more than one-third. Its ore is also of excellent quality. The
mines of Carpio and Lagunazo, in the province of Huelva, are
not yet worked for exportation.
The Spanish (and Portuguese) pyrites never contains less than
46, and up to 50 per cent, of sulphur, besides 3 to 4^ per cent,
of copper, which, however, by most of the English buyers, is not
bought, but returned in kind to the seller in the shape of cinders
from the pyrites-burners. The value of the copper (if bought) is
still fixed by the so-called " Cornish assay " — that is, a process of
dry assaying known only to a few assayers living at Redruth and
other places in Cornwall, the great inaccuracy of which is per-
fectly well known to all parties concerned : it shows, for instance,
only 2 per cent, if 4 per cent, is actually present ; and from this
diflference the buyer must pay the cost of extracting the copper
and his own profit, since the price to be paid for the copper in the
ore by Cornish assay is sometimes higher than the value of a similar
quantity of copper metal. This remarkably irrational system has
not hitherto been done awav with for sales.
In Germanv also in 1877 a number of manufactui*ers united
r
in working Spanish pyrites (especially Rio-Tinto ore), and in
delivering all their cinders to the Duisburg copper-works. The
Oker works also use similar ores, which they work up themselves
for copper.
The ore of the three principal companies is very similar in
composition ; its analysis is as follows : —
SPANISH PYRITES. 55
Ciaudet MacCuUoch
Pattinson. (San Domingo (San Bomingo
t * s ore). ore).
Sulphur 4800 49*60 4460 49-30 49O0 4980
Iron 40-74 4288 38-70 41-41 43-65 4288
CJopper 3-42 2-26 380 581 3-20 226
I/ead 0-82 0*52 0-58 066 OOa
Zinc trace 0*10 0*30 trace 0-35 0-10
Idme 0-21 0-18 0*14 014 0-10 018
MagnesiA 0*08 trace trace trace
ThaUinm trace trace trace trace
Araenic 0-21 0-28 026 031 047 028
Injoluble 5-67 294 1110 2-00 0-63 2*94
Oiygen (aa Fe^Oa) 0-09 015 0*23 0*25 1*07
Moisture 091 0-95 0-17 0-05 070 095
100-15 99-86 99-88 99-93 10000 9939
The following analyses represent the average quality : —
Bio Tinto. S. Domingo.
Cumenge. Caron. ^^^ Pattinson. Bartlett. 5]^^"^
Sulphur 48-00 607 4900' 49-90 4980 4750'
Iron 4000 41-3 43-55 41-41 43-55 41-92
Copper 3-42 35 320 246 8-20 421
Lead 0*82 ... 0-93 0-98 0-93 1-52
Zinc trace ... 0-35 044 035 022
Araenic 0-21 ... 047 055 047 038
Jarisch (Schwefelsaure-fabrikation, p. 30 et seq,) quotes further
analyses. In 1879 the sulphur in Rio-Tinto ore averaged 48*55
per cent., varying between 48-18 and 48*77 per cent. S.
A new pyrites-mine has been opened in Spain^ called St. Mardy
Tinto Santarossa. Its product has been found by Lunge and
Banziger (Zeitsch. angew. Chem. 1896^ p. 421) to contain 0*85
per cent, moisture, 5*20 insoluble, 43*87 sulphur, 42" 12 iron,
1*09 arsenic, 2*15 antimony, 3*17 copper.
Some kinds of Spanish pyrites are in bad repute with the
manufactnrers as '^ explosive" or '^detonating/' because they
decrepitate in the kilns shortly after lighting with loud detona-
tions, and thereby make so much fine powder that the burners
are stopped up and "scars'' are formed. The reason of this
detonating property is probably to be sought in the presence of
hydrated silicates (zeolites) in the ore *.
* The explosive pyrites from Goshen Copper Mine, near Scull Harbour,
County Cork, is said to contain confined carbon dioxide and water (Blount.
J. Soc. Chem. Ind. 1885, p. 074).
56 RAW MATERIALS OF MANUFACTURE.
According to Hjelt the average percentage of arsenic in Spanish
ores is 0*91.
Pyrites with very little or no copper is also found in Spain.
One of the best descriptions is that of the Agnas Teuidas mine,
containing iron 4660 per cent., sulphur 53' 15 per cent., silica
0*20 per cent., arsenic, copper, selenium, silver and gold traces.
It is sold both in the state of lumps and smalls. It burns A'ery
easily down to 1*0 or 05 per cent, of sulphur, so that the cinders,
which contain 685 per cent, of metallic iron, and no copper,
phosphorus, Jead, or zinc, are very valuable for blast-furnaces.
The annual sales have exceeded 200,000 tons, but recently very
little has come into the market owing to an accident at the mine.
H. J. Davis of New York, one of the principal importers of
pyrites to the U.S., gives the following analyses of very good,
hard Spanish ores containing but little copper : —
Araceua. Balmacca. San Tolmo.
S 51-77 50-19 46-40
Fe 45-53 45-61 40*11
Cu 0-29 0-20 1-90
Si l'9D 300 11-27
As ? ? none
The production of iron-pyrites (non-cupreous) in Spain, accord-
ing to ' United States Mineral Resources ' for 1900, p. 826, was
as follows 1 —
1891.
279,161 tons.
1896.
98,393 tons (?).
1892.
435,906 „
1897.
217,545 „
1893.
393,453 „
1898.
255,896 „
1894.
511,769 „
1899.
316,212 „
1895.
480,255 „
•
The production of cupreous pyrites and its exportation by the
three principal firms in Spain and Portugal was, according to
the same authority, p. 186 : —
1. Rio Tin to Company.
1898. Production 1,465,380 tons (2,852 p.c. Cu).
1899. „ 1,649,841 „ (2,719 „ „ ).
1900. „ 1,894,504 „ (2,744 „ „ ).
Exportation in 1900 : 704,803 tons.
PYRITES IN NORTH AMERICA. 57
2. Tharsis Sulphur & Copper Co.
1899. Production 572,854; exportation 222,475 tons.
1900. „ 468,738) „ 220,019 „
3. Mason & Barry, Ltd. Exported :
1899. 339,298 tons.
1900. 394,740 „
Enormous quantities of cupreous pyrites are roasted and worked
for copper in Spain, without utilizing the sulphur, as is apparent
from the above figures.
The United States of North America are very rich in pyrites.
The principal mines worked at present are the following : —
In New Hampshire : the Milan mines, Coos County. The ore
is of excellent quality, and is now sorted into two grades, of the
following composition : —
No. 1. No. '2,
Sulphur 46-0 350
Copper 3-7 50
Iron 40-0 305
Silica 6-2 25*5
Zinc 4-0 8-0
Arsenic 0 0
No. 1 is in special favour, but No. 2 burns very well and is
readily bought. Smelting-works exist for extracting the copper
and silver.
New York : 2000 tons were raised at Hermon, County
St, Lawrence ; sulphur contents, 38*0 per cent. Another mine
at Ulster County, with 39 per cent, ore, was worked till stopped
bv an inrush of water.
Massachusetts : the mines of the Davis Company, at Charle-
mont (Mass.), are about 130 miles from !New York, in the
centre of a network of railways. The ore contains 485 per cent,
sulphur, no trace of arsenic, antimony, or cobalt, little or no zinc,
lead, or calcium, 1*5 copper, 45*3 iron, and less than 3 per cent,
of silica. It is granular, easily broken by hand, and burns down
to 3 per cent, of sulphur. In 1884, about 30,000 tons were got;
it was burned in four works.
Virginia: Arminius Copper Mines Company and Sulphur Mines
Company, both in Louisa County. The ore contains 49*5 per cent.
58
RAW MATERIALS OF MANUFACTURE.
sulphur, 0*5 copper, 43*5 iron, 6*4 silica, &c. Annual output,
13,000 tons (in 1885).
Georgia: Dallas mine, Paulding County. The ore contains:
sulphur 40 per cent., copper 2'75 (sometimes up to 11 per cent.),
silica 8 per cent.
The absence of arsenic in most American pyrites (as far as it is
now worked) is a remarkable feature.
K. F. Stahl (Zsch. f. angew. Chem. 1893, p. 54) quotes analyses
of American pyrites. No. 1 is from Tallapoosa Mine, Georgia,
1882 ; No. 2 from Rogers Mine, Paulding Co., Dallas, Ga. ; No. 3
from Sulphur Mines Co. of Virginia, Louisa Co., 1884 ; No. 4,
Peru Zinc Co., La Salle, 111. ; No. 5 from Dodgeville, Wis. ;
No. 6, from the same mine as No. 3, 1891 ; No. 7, Davis Sulphur
Ore Co., Franklin Co., Mass., 1891.
1.
2. 3. 4. 5.
6.
7. ;
Water
451
31
30
01
2-9
?
— 1 2-9
37-6 371
40-6 41-5
5-2 O-fl
50-2 1 43-7
1-3
40-6
37-3
10
1-9
10-5
iraoe
0-8 t
Sulphur
42*4
Irou
354
Copper
1"4
Ziiie
4-5
0-01
9-5
?
0-8
1-4
5-5
Cadmium
Insoluble
Arsenic
?
14-7 -
002 —
?
51
trace i
1
We quote from the volumes of ' Mineral Industry ' the pro-
duction, imports, and consumption of pyrites in the United States,
not counting the auriferous pyrites used for the manufacture of
sulphuric acid in Canada, expressed in long tons of 2,240 lbs. : —
Year.
Production.
Imports.
1
Consumption.
1891
100,319
106,250
95,000
107,462
81,000
109,282
133,368
191,160
178,408
201,317
130,000
210,000
194,000
146,023
190,436
199,678
259,546
171,879
310,008
329,449
i 239.319
1892
316,250
289,000
1 253.485
1 271,436
1 308,960
392,914
363,039
488,416
' 530.766
1893
1894
1895
1896
1897
1898
1899
1900
1
Most of the American pyrites, including the Newfoundland
PYRITES IN CANADA^ NEWFOUNDLAND^ AND AUSTRALIA.
59
ore^ is granular and more fit for fines' burners thau for lump-
burners.
In Canada there are two mines : the Albert mine and the
Crown mine. They supplied the first pyrites used in the United
States for making oil of vitriol. Sulphur 40*0 per cent., iron 350,
copper 4'0, silica 20*0. Output :
1891. 60,474 tons. 1894. 36,185 tons. 1897. 34,471 tons.
1892. 53,372 „ 1895. 30,534 „ 1898. 24,721
1893. 52,270 „ 1896. 30,103 „ 1899. 35,742
}>
>>
In Newfoundland, on Piney's Island, there is a very large bed
of pyrites, accessible by a shaft 60 feet deep. The bed is 72 feet
-wide, and 28 feet of it contains 3 or 4 per cent, copper. This ore
is veiy easy to roast ; an analysis of the non-cupreous ore showed
Cu0b7, S 51-16, Fe 48-35, SiO^ 013, CaO 022, As 002; no
Sb, Pb, Zn, Bi (Eng. Min. Joum. 1892, p. 467).
In South Australia pyrites is found with 48* 7 per cent. S and
2-8 per cent. Cu (Mene, Mon. Sc. 1867, p. 411 ).
The World^s production of iron-pyrites (in metric tons) is
stated in Min. Ind. ix. p. 615, from official sources (I cannot
explain the difference between these and the other " official '^
figures given in various places supra) : —
■ ■ ""~
"
1
i8'j6. :
1897.
1898.
1899.
2,560
1,828
147
283
2,000
3,670
240
430
30,586
35,299
29,228
25,117 j
282,064
303,448
310,972
318,832
129.168
133,302
136,849
144,623
52,697
44,454
58,079
79,519 1
45,728
58,320
67,191
76,538
27,712
33,316
33,100
31,500
60,rj07
9i,484
89,763
90,00() 1
200,000
210,265
248,218
275,(x>8 i
11,550
19,380 .
20,000
25.(KJ0
100,000
217,545 ;
2<50,016
319,285 1
1,<X)9
517 1
386
150 1
10,178
10,753
12,302
12.426
117,782
128,468
1,295,049
1H219
181,263
1,073,.541
1,460,710
1,580,624 ;
1895.
Belgium 3,510
Sosnia
Canada 31,024
France 253.416
Germany 127,03(5
Hungary 69,195
Italy 38,586
Newfoundland 34,879
Norway 61,994
Portugal (r«) 200,000
Buasia 11,042
Spain (6) 60/J67
Sweden 221
Great Britain 9,193
United States 107,371
TotalB 1,007,734
(a) For Portugal the output for 1895 and 1896 is roughly estimated. For this
country only pyrites with less than 1 per cent, copper is included.
(6) For Spain the cupreous pyrites, from which copper is extracted, is not
included, oomp. p. 57.
60
RAW MATERIALS OF MANUFACTURE.
C
O}
cb -^ 00 w »b t* 'fi 00 i^i 6i 'fi cb « 6c 'fi c? '^
^ C cm
cS O
S 5 OC Ci O S S
X O O O lO -fl O
»"
52.)
t
s
o
a
■
•
a
•
•
•
•
•
•
•
•
•
%3
: 0
•
0
•
•
•
•
•
•
a
't* 01 c
: I- c: ^ :
• c c c •
«o
-
2 '^
•C CO
-r j
•
5
s
•
•
■
•
•
•
•
•
•
•
•
•
0
•
■
•
•
•
■
■
. S S 13 ^
• ^^ • • •
* ■;£ C ^ 01 '
1
.«« «
•^ 5^
•
•1 2
^ o
a
4^
:3
•
•
•
•
•
•
11
1^
■
0
6
0
0
2
m c »-< •-'
: c: ?i c: ^
* • a • •
• 0 c c 01
:^ 90
•<
0^
1
^<> ^-4
u
3» of
inera
^
ns
I'*
?
•
s?
0
•
2 r,
0
•
to
Cl
•
0
a
2 Q CO
0 • • T^ t
6
CO
1
0
m^
CO
0^
«M
kO -f
^
CO
i-H
<M
1
1
1
ed State
G
2
CI
s s s
• • ■
"* ?3 3:
"^ -^ '+
0 CI 0 Q TO
t^ »« t^ CI I-
^ » s 5.' ^ '
■
5S -S
tm
3 a
—
s s
s
8 8 8
?l 25
s
g
. 0
0
10
0 C — "T "^
S CC X Ot "*
^n
s
C5
I CO
s
to
0 00
CS
CS
s 5
5C i^ CI -* l-
't« -r -* CO "^
a
1
o
^i
CI
•
•
•
a>
u
•
•
•
^
bD
^
0
•
•
•
*
0
: <
<
1
1
1
1
"*
,
s>
• • " ■ *
•^J
X.
S4
c
«•
•*:
s
; 0
K
V
, g
s
g
a
1
1 :
» 1
1
3
1 **
^
1
li
C
c
1 es
N
1 ^
E
0
8
t
•
CO
a
• ^
E
■ ^
■
■ ■ ■
Ill
5 Q «
l
&
u
: c
: P-
h
1 C
< a
D
;
i i
1
1
S
! 'S
! S
1 ;z
• • •
• • • •
) u a» cj a 'r
i fc e =: K ^
- 1
Li
1
1
•
PRICES OF PYRITES ; POOR AND RICH ORE.
61
Prices obtained in the Liverpool district for pyrites, when sold
for sulphur only : —
Prices per
Year. unit of sulphur.
1861 10-85rf.
1862 9-625
1863 8
1865 • 9
Jan. 1866 9i
Mar. 1866 1102
Jan. 1868 1004
June 1868 9^
Oct. 1868 8
Jan. 1869 7
June 1869 8
Year.
1870
1871
1872
1873
1876
1877
1878
Jan. 1879 to
Dec. 1884
1885 to
1887
}
}
Prices per
unit of sulphur.
6irf.
-i
/ ■>->
7i
4
^
H
H
3 to 34
(E. K. Muspratt in J. Soc. Chera. Ind. 1886, p. 406.)
Jurisch (Schwefelsaure-fabrikation p. 58) quotes the prices of
pyrites in Germany from 1853 to 1892 ; but this is of very little
use, as there is nothing said about the percentage of sulphur.
Previous to 1883 the price was always above 10 marks per ton,,
from 1884 to 1892 it ranged about 7^ marks per ton. Only one
definite statement is made, viz. that 100 kil. of sulphur actually
burned from Rio Tinto, deducting the drawback for copper, in
1881 cost 4-65 to 5-10 marks (?).
In 1900, the ordinary price of German pyrites was 12 marks
per ton at the mine ; for large contracts probably less.
Proportional value of poor and rich Pyrites.
It is no matter of surprise that the rich and at the same time
well-burning Spanish ores, and those ores similar to them else-
where, where they could be imported, have driven the poor ores
out of the field. An ore of 35 per cent., like that from Wicklow,
even for the same weight of sulphur, has much less value than a
45-per-cent. ore. The wages for breaking and burning the ore in
both cases must be ruled by the gross weight of the ore, and con-
sequently for equal weights of sulphur come to much more with
poor than with rich pyrites; moreover, under conditions otherwise
equal, the unburnt sulphur in the cinders is the same by weight.
If, for instance, 5 percent, of sulphur are left in the cinders.
62 RAW MATERIALS OF MANUFACTURE.
this amounts with 35-per-cent. ore to.!l = |; with 45-per-cent. ore
to only 75= i; *he proportion to be kept in view is accordingly
not 85 : 45 = 7 : 9, but 30 : 40=3 : 4. Furthermore, the same
holds good of cost of plant and repairs and of wages, and, lastly,
since the poor ores generally contain no copper, also of the cost
of removing the cinders.
Therefore, unless the burning is nothing but a preparation for
the metallurgical treatment, where the sulphur is quite of secondary
importance, the ores poor in sulphur are always avoided as much
as possible.
Analysis of Pyrites.
In the analysis of pyrites for technical purposes, in the first
instance their percentage of sulphur is taken into account ; and it
is therefore mostly the custom to estimate only this and, perhaps,
also the moisture. If the ore has afterwards to yield copper, this
must of course also be considered ; but where the copper, as is
mostly the case, is not bought by the vitriol-maker, but the cinders
are returned to the seller, the estimation of copper is generally
omitted in the chemical works as unnecessarv. This restriction
to the estimation of sulphur ought, however, only to take place
with pyrites from well-known localities, whose general composition
and properties are well known, and where the salient point is only
the percentage of the most important constituent, viz. the sulphur.
Each cargo, even each portion of a cargo going to a separate
buyer, is sampled in the presence of both the buyer's and seller's
agents, according to well-understood rules : the sample is broken
up. reduced, and sealed up in bottles, which are sent to an
analytical chemist (generally mentioned in the contract note) ;
this chemist's certificate rules the price to be paid for the pyrites
down to i per cent. If, for instance, a sale has been made at 6d,
per ^' unit," this means that for each per cent, of sulphur found
the sum of 6d. per ton is paid; thus for 48;^^ per cent, of sulphur
48ix6rf. = 245. lirf. per ton. The ton is generally calculated
= 21 cwt. ; that is to say, the buyer receives an allowance in
weight of 5 per cent.
The first treatment of the pyrites in analyses, in the majority
of cases, is by the wet way, by fuming nitric acid or aqua regia ; but
the prescriptions differ in details. The decomposition was formerlv
ANALYSIS OF FYRIT£S. 63
frequently made^ irom Fresenius's prescription, by means of red
filming nitric acid, which it is sometimes difficult to obtain free
from sulphuric acid, and which is unpleasant to handle. In lieu
of this sometimes chlorate of potash along with hydrochloric acid,
or, still better, with nitric acid of 1*36 sp. gr., have been used.
The author has always found the best, safest, and cheapest way to
be that by aqua regia, made from 1 part of fuming hydrochloric
acid and 3 or 4 parts of nitric acid of 1*36 to 1*4 sp. gr., and this
mixture is now used in most places. The mineral is converted
into an impalpable powder and passed through the finest silk
gauze ; the triturating ought to be done first in a steel mortar or
by wrapping it up in paper and smashing with a hammer, and
then in an agate mortar, no/ in a porcelain or Wedgwood mortar.
The powder is treated with about 50 parts of aqua regia ; if no
reaction takes place at once, it should be gently heated on a
water-bath till the reaction sets in ; but then the beaker should
be removed instantly from the water-bath, and only replaced when
the reaction slackens : thus the decomposition is generally complete
in 10 minutes. The operation should be performed in a large
beaker, or, still better, in an Erlenmeyer's flask covered by a
funnel or a watchglass, lest any loss should take place by spurting;
and the work must be done in a draught-place, on account of the
mass of acid vapours given off. If the decomposition should not be
perfect after heating some time, some more aqua regia has to be
added and the heating continued ; but mostly this will be caused
by the powder not being sufficiently fine, and the analysis in this
case cannot easily be finished. In this way of decomposing the
ore, which is both quick and safe, the disagreeable separation of
sulphur happens very rarely. If it does, the sulphur is oxidized
by cautiously adding a little chlorate of potash.
The residue from the solution will contain silica and silicates,
perhaps a little lead or barium, both as sulphates. Although their
sulphur is thus not estimated, no harm is done, as it is anyhow
valueless to the manufacturer. Lead sulphate is pretty soluble in
concentrated acids, but it is almost entirely precipitated again by
the immediately following treatment.
At all events the whole of the nitric acid present must be
destroyed or removed, because the estimation of sulphuric acid
by barium chloride in the presence of nitric acid or its salts gives
results much in excess of the truth. Tlie whole is therefore
64 RAW MATERIALS OF MANUFACTURE.
evaporated to dryness on the water-bath* with an excess of
hydrochloric acid^ by which at the same time all silicic acid
dissolved is made insoluble. The mass is again moistened with
strong hydrochloric acid; and if on gently heating no yellow
vapour and no smell of nitrous products are perceived, it is diluted
with hot water and the solution filtered from the residue. Care
must be taken not to employ too much hydrochloric acid, as
barium sulphate is not quite insoluble in hot concentrated acids ;
on the other hand, enough acid must be present to dissolve all
salts of iron, which cannot be doubtful if the colour and behaviour
of the residue are observed.
Some chemists prefer to the aqua regia above described a
solution of bromine in hydrochloric acid ; but I have not found
this to answer very well. Drown (Chem. News, xliii. p. 89) heats
the pyrites with a solution of caustic soda of spec. grav. 1*25, adds
cautiously bromine in excess, acidulates with hydrochloric acid,
and evaporates to dryness.
Noaillon (Zsch. angew. Ch. 1897, p. 351) employs a mixture ot
sodium chlorate and nitric acid for decomposing the pyrites, in
order to avoid loss of sulphuric acid when drying the resulting
mass. (Such a loss never occurs when working according to my
prescription.)
Where the utmost accuracy is not desired the solution may
now be at once treated with barium chloride, as follows : —
The clear solution is brought to full boiling ; and to it during
boiling a hot solution of barium chloride is slowly added. Lest
too great an excess of this be used, it is preferable to use a
measured quantity of a concentrated solution of known strength,
but more than sufficient for precipitating all the sulphuric acid
present. It is best to pour the hot solution of barium chloride
rather slowly into the boiling solution of sulphate, with constant
stirring, but it is quite unnecessary to do this drop by drop as
prescribed by Gladding (Chem. News, Ixxx. p. 181) ; comp.
Lunge, Zsch. ang. Chem. 1895, p. 69, and Journ. Amer. Chem.
Soc, March 1895. If the process is carried out as described
here, the barium sulphate settles down completely in a few
seconds, leaving a perfectly clear liquid, and nothing of the
♦ The last evaporation may be hastened by employing a saud-bath or asbestos
cardboard, taking Ctire to remove the vessel at the moment when the mass has
become drv.
PRECIPITATION OF BABIUM SULPHATE.
65
Fig, 5.
precipitate passes through the filter. It is quite unnecessary to
allow a long time for the settling. If the operation is carried on
as described^ the filtrate never becomes cloudy afterwards ; on the
other hand, the work ia greatly expedited by filtering the liquid
in the boiling-liot state, say 15 or 20 minutes after precipitation.
A Bunsen's filter-pump acts rather too strongly in
this case ; but it is very useful to employ a simple
contrivance indicated many years ago by Piccard,
which does excellent service in other analytical
operations^ viz.^ a glass tube attached to the funnel
by means of an elastic joint, with a loop causing
a continuous jet of liquid to issue at the bottom
(fig. 5). The straight part below the loop may
be 8 or 10 inches long ; the filter must be pressed
tight to the sides of the funnel to prevent any air
being sucked in. When this contrivance is used,
which does not act so violently as a Bunsen pump,
the liquid, so long as there is not too much
precipitate in the filter, runs through in a con-
tinuous jet.
At first only the clear liquid is poured off as
completely as possible from the dense granular
precipitate; this is covered with boiling water,
acidulated with a few drops of hydrochloric acid ;
the liquid is boiled for a few moments, and can
be decanted in about two minutes' time. This
operation is again repeated twice or three times,
but without adding any more hydrochloric acid ;
the precipitate is washed onto the filter, and after
very little washing the filtrate will be found per-
fectlv neutral and free from dissolved matters.
The filter is dried, the precipitate taken out, and
the filter burnt, preferably in a platinum crucible
laid on its side ; then the precipitate is put in and ^^
Ignited, not too strongly : and for each 100 parts of barium
sulphate found 13' 733 parts of sulphur are calculated. The
ignited barium sulphate must not cake together; on moistening,
it should not give an alkaline reaction; and on heating with
dilute hydrochloric acid and filtering, no barium salt ought to be
found in the solution.
VOL, I. F
66 . RAW MATERIALS OF MANUFACTURE.
It generally happens^ even . if the solution before precipitation
was rather strongly acid^ that t^e precipitate is stained yellowish
by precipitated ferric oxide or basic ferric sulphate, which cannot
be removed even by boiling with dilute hydrochloric acid.
Although this proves the presence of a foreign substance in the
barium sulphate^ which ought to make the result too high^ yet it
is found in practice that the results are always too low. The
cause of this apparent anomaly has been studied by Jannasch and
Richards (Journ. f. prakt. Chem. [2] xxxiv. p. 321), who found
that in the presence of iron a barium-ferric sulphate is precipitated,
which on ignition slowly loses a portion of its sulphuric acid. If
the ignition is carried on very persistently, the error thus pro-
duced may amount to a full per cent, of sulphur or upwards ;
but I have shown, in the paper quoted below, that with the
ordinary mode of ignition it does not exceed 018 per cent. ;
hence the above-described method is always available where
perfect accuracy is not required and a speedy completion of
the test is a consideration. Westmoreland (J. Soc. Chem. Ind.
1887, p. 84) even contends that its results entirely agree with
those obtained by my new method (an opinion to which I must
demur).
Where, however, the greatest possible accuracy and freedom
from error is required (and this is the case when testing an average
sample of pyrites representing a whole cargo, or a large portion
of such), it is necessary to remove the disturbing influence of
the iron. This can be done in two different ways, by the dry and
by the wet method. The dry method is that recommended by
Fresenius (Zeitsch. f . anal. Chemie, xvi. p. 335) . He prescribes
to decompose the pyrites by fluxing it with 20 parts of a mixture
of 2 parts of dry sodium carbonate entirely free from sulphate,
and 1 part of potassium nitrate, passing carbonic acid into the
solution for the precipitation of lead, boiling the residue with a
solution of sodium carbouate and then with water, acidulating with
hydrochloric acid, and repeatedly evaporating for the expulsion of
nitric acid^ after which the process is carried on as usual, the
precipitation taking place by barium chloride. This process is
much more troublesome and tedious than that to be described
below, which employs the wet method. Another objection to it is
that it estimates not merely the sulphur of the iron and copper
sulphides, but also that of galena and of barium sulphate, which
REMOVAL OF IRON IN TESTING PYRITES. G7
are entirely useless to th^ manufacturer of sulphuric acid^.
Moreover, the platinum crucibles are strongly acted upon, and coal-
gas cannot very well be employed in the operation of decomposing
the pyrites, as its sulphur would cause an error in the test. Hence
it is advisable to use spirit-lamps of a special shape, calculated
to yield the necessary heat. [The use of spirit-lamps can be
practically avoided by placing the crucible in a round hole made
in a piece of asbestos cardboard, in which case the products of
the combustion of coal-gas are carried off sideways.]
Hayes (* American Chemist,' v. p. 271) describes a method of
decomposing pyrites with alcoholic soda and lime, the advantages
of which are in no way evident.
Fahlberg and lies (Ber. d. deutsch. chem. Ges. xi. p. 1187)
recommend fluxing the sulphur-ore with caustic potash (25 grams
to 0*1 gram of S) in a silver crucible for 15 to 20 minutes, lixivia-
ting the mass, oxidizing the lower oxides of sulphur by bromine-
water, and precipitating by barium chloride.
Clark (J. Soc. Chem. Ind. 1885, p. 329) heats the pyrites with
a mixture of sodium carbonate and magnesia to a dark-red heat ;
the resulting mass is lixiviated with water whilst passing in
carbonic acid, and the sulphuric acid estimated in the usual way.
J. Pattison {ibid, p. 724) shows that this method gives exactly the
same result as mine (of course, only in cases where barium
sulphate and galena are absent, which militates against Clark's
process).
Looking at the great desirability of retaining the decomposition
of pyrites in the wei way, I have worked out a method for doing so
without incurring the error caused by the presence of iron. This
method was first described in the Zeitsch. f. anal. Chemie, 1880,
p. 419, and has been very generally accepted for the assaying of
pyrites between buyer and seller. Objections made to the accuracy
of that process by Jannasch and Richards (Joum. f. prakt. Chem.
[2] xxxix. p. 321) have been withdrawn by them {ibid, xl.
p. 326), and have been completely refuted by experiments made
in my laboratory by two independent investigators (Zeitsch. f.
angew. Chemie, 1889, p. 473). The process, as it will now be
described, may hence be regarded as the most accurate known for
the estimation of sulphur in pyrites, where it is desirable nol
* Concerning the action of nitric acid on lead sulphate, see my experiments
detailed in J. Soc. Chem. Ind. 1887, p. 96.
f2
68 RAW MATERIALS OF MANUFACTURE.
to include galena and heavy -spar^ and it is at the same time very
easy and speedy of execution^ if the following directions are
accurately observed.
About 0*5 grm. of pyrites is heated with about 10 c. c. of a
mixture of 3 vols, nitric acid- (spec. grav. 1*4) and 1 vol. strong
hydrochloric acid, both ascertained to be absolutely free from
sulphuric acid. The operation is performed as described above,
in such manner that no loss by spurting takes place. The mixture
is heated up now and then^ till the decomposition is complete^ and
is then evaporated to dryness in a water-bath. Now add 5 c. c.
hydrochloric acid^ evaporate once more (no nitrous fumes ought
to escape now)^ add 1 c. c. concentrated hydrochloric acid and
100 c. c. hot water; pass through a small filter and wash with hot
water. The insoluble residue may be dried, ignited, and weighed.
It may contain^ besides silicic acid and silicates, the sulphates of
barium, lead, and even calcium, whose sulphur, as being useless,
is purposely neglected. (If this residue is not to be estimated, it
need not be filtered oflp, and the next step, the elimination of the
iron, may be taken without removing the silica, &c.) The filtrate
and washings are saturated with ammonia, avoiding a very large
excess of it ; the mixture is kept at a moderately warm tempe-
rature for about 10 minutes (at the expiration of which time it
ought still to smell of ammonia very distinctly, not merely faintly),
and the precipitated ferric hydrate is filtered off while the liquid
is still hot. For this purpose funnels must be used, made at an
angle of exactly 60°, whose tube is not too wide and is completely
filled by the liquid running through; or else a Piccard's tube
(comp. above, p. 65), or a filter-pump is employed. The filtering-
paper must be sufficiently dense, but should act rapidly ; the filter
must be adapted to the funnel so that no air-channel is left be-
tween paper and glass. The hot liquid is first decanted from tlie
ferric hydrate, and the latter is then washed onto the filter with
boiling water. The washing is continued with hot water in such
manner that each time the whole precipitate is thoroughly churned
up and no channels are formed in the mass. When acting on
these instructions, the whole operation can be performed in from
half to one hour, and no trace of sulphuric acid is retained in the
precipitate. The total bulk of the filtrate and washings need not
exceed 200 or 250 c. c, which saves concentrating the liquid b}'
evaporation. The end of the washing is indicated by the fact that
REMOVAL OF IRON IN TESTING PYRITES. 69
1 c. c.^ on adding barium chloride^ shows no opalescence even
after a few minutes. (It is, however, best in important cases to
test the iron precipitate for sulphur by drying it, detaching it as
well as possible from the filter, fluxing with pure sodium carbonate,
dissolving in hot water, acidulating and adding BaCla.)
The clear liquid, which now contains all the sulphuric acid
combined with ammonia, is acidulated with pure hydrochloric
acid in very slight excess, heated to boiling, the burner removed,
and 20 c. c. of a 10-per-cent. solution of barium chloride, pre-
viously heated, is slowly poured in. This quantity, which suffices
in any case for 0*5 gram pyrites, is roughly measured oflP in a test-
in be provided with a mark, and is heated in the same tube. After
precipitation, the liquid is left to stand for half an hour, when the
precipitate should be completely settled. The clear portion is now
decanted, and the washing continued by decantation with boiling
water, as mentioned on p. 65, where the mode of igniting the
precipitate is also described. The ignited precipitate should be a
perfectly white and loose powder, one part of which is equal to
013733 sulphur.
The accuracy of this method and its complete accordance
(in the case of pure ores) with the fusion method of Fresenius
have been proved by Pattinson (Joum. Soc. Chem. Ind. 1890,
p. 21), who points out how much more convenient is the former
than the latter.
Kiister and Thiel (Zsch. anorg. Ch. xix. p. 97) erroneously
assume that the ferric hydrate cannot easily be washed by my
method so as to remove all sulphate (which is refuted by hundreds
of students who have carried out this method in my laboratory,
and thousands of chemists elsewhere). They therefore propose to
precipitate the barium sulphate without filtering off the ferric
hydrate and to remove the latter subsequently by several hours'
digestion with hydrochloric acid ; or else to prevent the preci-
pitation of ferric salts along with the barium sulphate by the
addition of a large quantity of ammonium oxalate. Both methods
take much more time than mine, without any gain in accuracy,
•as I have shown in Zsch. anorg. Ch. ix. p. 454, and again
xxi. p. 194. This is confirmed by Herting (Zsch. angew. Ch.
1899, p. 274).
According to Heidenreich (Zsch. anorg. Ch. xx. p. 233) the
contamination of barium sulphate by iron salts can be avoided by
70 RAW MATERIALS OP MANUFACTURE.
reducing the ferric sulphate to ferrous sulphate by means of zinc
and preventing the access of air and light during the precipitation.
Herting and Lehnardt (Chem. Zeit. 1899, No. 75) effect the same
purpose much more quickly by employing stannous chloride in
lieu of zinc.
Noaillon (Zsch. angew. Ch. 1897, p. 351) proposes to avoid the
filtration of the ferric hydrate by diluting the liquid to a certain
mark, filtering, and employing a portion of the filtrate for the
precipitation with barium chloride ; but this method introduces
more than one error and must be rejected.
Sodium peroxide is proposed for the decomposition of pyrites by
Hempel, Hohnel (Arch. Pharm. 1894, p. 222), and Glaser (Chem.
Zeit. 1894, p. 1448).
In lieu of the estimation of sulphuric acid by weight, some
chemists prefer titration by means of a standard solution of
barium chloride. This was first proposed by Wildenstein (Zeitsch.
f. anal. Chemie, i. p. 432), and afterwards, especially for the
analysis of pyrites, by Teschemacher and Smith (Chem. News,
xxiv. pp, 61, 66 ; comp. also Glendinning and Edger, ib. p. 140).
Although this process, notwithstanding some assertions to the
contrary, is most certainly no more accurate than the gravimetric
process, and in most hands is less so, and is not used by many
chemists in important cases, we shall take this opportunity of
describing the estimation of sulphates by titration with barium
chloride in its simplest form, such as is used at some works in
testing black-ash, &c. ; it is also used sometimes in testing pyrites,
blende, burnt-ore, &c. for purposes where no great accuracy is
required.
The liquid is brought to . the boil in a porcelain dish, barium-
chloride solution is added from a burette ; from time to time a
few drops are taken out with a glass tube, passed through a
miniature filter onto a glass plate resting on a black background,
upon which a number of drops both of dilute sulphuric acid and of
barium chloride have been put. If the filtrate still gives a
cloudiness with a barium-chloride drop, easily visible on the black
ground, the little filter is thrown back into the dish, more barium-
chloride solution is added from the burette, another test is made,
and so forth. The end is attained when a filtered, drop gives an
extremely slight cloudiness both with a drop of barium chloride
and with one of sulphuric acid. The work is very much expedited
ESTJUATION OF SULPHATES BY TITRATION. 71
by the following contrivance, proposed by Wildensteio and shown
ID fig. 6. The acidulated solution is poured into a vessel, A, made
of a bottle by removing the bottom, or a small tubulated jar,
through whose cork posses a bent tube B, provided at the lower
end with the pinchcock /, at the upper end with a bent-down funnel
/. The latter (which must be bell-shaped) is closed by two disks
of filtering-paper and a piece of linen gauze tied over all ; and the
liquid must cover the whole tube. This arrangement permits of
withdrawing at will small filtered samples of a few drops each,
Fig. 0.
which ai'e run into a test-tube and tried with a drop of barium-
chloride solution. It must, of course, not be omitted first to run
a few c. c. out of the tube B and back into the jar A before taking
the sample for testing ; and the contents of the test-tube must be
always put back into A, not to waste too much substance. If by
accident the point of finishing the reaction has been overstepped,
one or more c. c. of titrated sulphuric acid are put in and are
afterwards deducted from the result.
C. and J. Beringer effect the titration by barium chloride after
addition of sodium acetate and acetic acid (Chem. News, lix.
p. 41).
Various other volumetric methods for the estimation of sul-
phates, by Carl Alohr, Ad. Clemm, Wildenstein (2nd method],
Schwarz, and Fappenheim, are described in the treatises of Fresenius
and Mohr ; but they are more complicated and not more accurate
than the direct titration with barium chloride as just described.
We, therefore, mention only one method of this kind. This is the
method proposed by Wilsing (Chcm. Iiul. 1886, p. 25), a slight
72 RAW MATERIALS OF MANUFACTURE.
but apparently useful modiHeation of those just mentioued. He
precipitates a neutral solution of the sulphate or solution con-
taining such^ boiling in a porcelain dish^ with a 4-per-cent.
barium-chloride solution of known strength ; he then adds a few
drops of an alcoholic solution of phenolphthalein and a 2-per-
cent, solution of sodium carbonate : as soon as the last part of
barium chloride has been precipitated as carbonate the colour
turns red^ so that the soda used is a measure for the sulphate
originally present. If the solution to be tested is either acid or
alkaline^ it must be made neutral by sodium carbonate or hydro-
chloric acid^ phenolphthalein being used as indicator here as well.
L. W. Andrews (Amer. Chem. J. 1889, p. 567, & Chem. Zeit.
1889, Rep. p. 39) proposes the following method for estimating
sulphuric acid combined with bases : — The solution is diluted till
it contains no more than 2 per cent. SO3, is almost neutralized
and brought to a boiling heat. !Now a solution of pure barium
chromate in hydrochloric acid is added, and then ammonia, the
excess of which is removed by boiling. The solution is filtered
while hot and is washed at once. Now a quantity of chromic
acid, equivalent to the sulphuric acid originally present, will he in
solution ; this is estimated by adding potassium iodide and strong
hydrochloric acid, and titrating with decinormal thiosulphate-
solution (1 c.c. = 12*85 mg. I = 2'662 mg. SO3). It is claimed
that this operation is more quickly performed than the gravimetric
process, and is at least as accurate ; but both of these assertions
are doubtful.
Similar methods are described by Renter (Chem. Zeit. 1898,
p. 357) and Marboutin and Moulinie (Chem. Cbl. 1898, i.
p. 218) .
Several methods have been proposed for estimating the available
sulphur of sulphur-ores — that is, that portion of it which passes
into the chambers in the shape of SOg and SO3. Thus W. G.
Mixter (Amer. Chem. J. ii. p. 396) burns the pyrites in a current
of oxygen, and passes the vapours into bromine-water containing
some hydrochloric acid and an excess of bromine. A similar
process is recommended by Zulkowsky (Wagner's Jahresb. 1881,
p. 160), both for testing pyrites and spent oxides of gas-works.
The latter is always contaminated with sawdust, tarry matters, and
variable quantities of lime, which retains part of the sulphur in
burning, whence an estimation of the total sulphur is quite useless
ESTIMATION OP AVAttABLB gULPHUK. 73
for practical purposes. The process takes place in a combustion-
tube (fig. 7), 3 feet long, narrowed at a, aod dravn out at tbe end
iato a long tube, not too thin, and bent downwards. Between a
snd b there is a layer of platinized asbestos (see below), 8 to 10
inches long, and at a distance of 3 or 4 inches from this a porcelain
Fig. 7.
boat with about 0'4 grm, of spent oxide or pyrites. The end of
the tube at k is connected with an oxygen gas-bolder. The absorp-
tion of tbe vapours takes place in tbe two 3-bull) tubes c and d
(5| inches high) and the tube e, filled with glass-wool. The
absorbing-liquid is made by dissolving 180 grama caustic potash
(purified with alcohol from sulphate) in water, adding 100 grams
bromine, taking care to keep tbe mixture cool, and diluting to
1000 c. c. 30 c. c. of this suffice for estimating 0'5 grm. sulphur.
The tube e ought also to be moistened with it. First heat the
portion of tbe tube between a and b, passing moist oxygen through
it at the same time ; then heat the boat from the right to tbe left ;
lastly the tube, up to the place /. The current of gas must l>e
much stronger than for an organic analysis, lest any sulphur should
escape unbumt, but not strong enough to draw off any SOi uu-
absorbed. So long as any dew appears at h, it must be driven into
the receiver with a fiunsen burner. When this ceases (usually in
about an hour), the experiment is finished. The receivers are then
taken off, washed out, and the acid remaining in h is recovered by
aspunting water several times through it. All the liquids are
united, supersaturated with hydrochloric acid in order to de-
compose the potassium hypobromite, heated, concentrated if
necessary, and the sulphuric acid is precipitated by barium
chloride in tbe usnal manner (or, more conveniently, the receivers
74 RAW MATERIALS OF MANUFACTURE.
are charged with hydrogen peroxides free from sulphuric acid and
are retitrated after the end of the operation).
Jannasch (Journ. f. prakt. Ch. [2] xl. p. 237) heats pyrites in
a mixture of air or oxygen and nitric-acid vapours ; the vapours
of SO2 and SOg are absorbed in bromine- water. [It must not be
forgotten that in the presence of nitric acid the barium sulphate
is never free from nitrate^ and that hence all nitric acid must be
removed previously. I would, therefore, propose to use in that
class of processes hydrogen peroxide as an absorbent, which is very
efficient and requires no special precautions; in this case the
acids absorbed can be estimated by titration with the standard
alkali.]
Graeger (Dingl. Journ. ccxli. p. 53 ; Fischer's Jahresb. 1881,
p. 161) beats pyrites with reduced metallic iron, decomposes the
FeS formed by dilute hydrochloric acid, and titrates the H^S
evolved by passing it into a solution of iodine. This method has
been again proposed by Treadwell (Berl. Ber. xxiv. p. 1937, and
xxiv. p. 2377) and by Eliasberg (Zsch. anal. Ch. 1899, p. 240),
but it does not seem to be practically employed.
Expeditious assays of pyrites have been proposed in many ways,
but none of them is sufficiently accurate to be employed for esti-
mating the sulphur in fresh pyrites, and some of them are not
even accurate enough for testing the sulphur remaining in burnt-
ore (pyrites cinders).
The so-called mechanical pi/rites assay of Anthon (Dingl. Journ.
clxi. p. 115) is too rough and unreliable even for very simple pur-
poses. (Comp. 1st ed. of this work, i. p. 108.)
In the Freiberg works 1 gram of finely ground pyrites is mixed
with 2 grams sodium carbonate and 2 grams saltpetre : the mixture
is fluxed in a small iron dish in a red-hot muffle-furnace, dissolved
in hot water and filtered into a beaker in which there is hydro-
chloric acid by saturating the soda in excess. Then the filtrate is
brought to boiling, and the sulphuric acid estimated by a standard
solution of barium chloride, preferably by Wildenstein's method
(supra, p. 70). Liebig (Post's Tech. chem. Analyse, 2nd ed. i,
p. 677) recommends this method as a quick and easy one, where
no great accuracy is required.
In the method of Pelouze (Compt. Bend. liii. p. 685 ; Ann. de
Chim. et de Phys. [3] Ixiii. p. 415) the finely powdered pyrites
is mixed with chlorate of potash, common salt, and an exactly
TESTING PYRITES CINDERS FOR SULPHUR. 75
weighed quantity of sodium carbonate^ and ignited^ which can be
done in an iron spoon. The fluxed mass is dissolved in water^
filtered J the residue washed^ and the soda not converted into sul-
phate estimated alkalimetrically. This process continued to be
recommended in French treatises till very recently^ although its
inaccuracy was established and the sources of error partly demon-
strated by many chemists, such as Barreswil^ Bottomley^ Bocheroff^
Lunge^ and especially Kolb (' Notes sur PEssai des Pyrites de
Fer '). Kolb found the sources of error on the one hand in the
formation of sodium silicate^ on the other hand in the decompo-
sition of potassium chlorate in the presence of ferric oxide into
chlorine^ oxygen^. and caustic potash. New experiments made in
my laboratory by Mr. Rey have equally proved the method
to be wrong, even if the "constant error '^ of 1 to 1^ per cent.,
admitted by Pelouze, is taken into account. A principal objection
is the difficulty of evading the mechanical loss by spurting iu the
fluxing process.
This fault is avoided in the plan proposed by Kolb (J. pharm.
Chim. [4] X. p. 401), which, however, is only intended for testing
burnt-ore. Kolb mixes 5 or 10 grams of this with 5 grams pure
sodium carbonate and 50 grams cupric oxide, heats about 15
minutes in an iron capsule to a dark-red heat, with stirring, lixi-
viates the melted mass, and estimates the unconsumed soda volu-
metrically. The trials made in the autboi*^s laboratory showed
that there is no spurting, but that the heating must not be pro-
longed too much, in order to avoid the formation of silicates. The
lixiviation of the large bulk of cupric and ferric oxide is tedious,
and the method is somewhat costly, as it requires 50 grams of
cupric oxide for each test, nor are the results very satisfactory
(see below) .
A much better method for testing burnt-ore was 2)roposed by
J. Watson (J. Soc. Chem. Ind. 1888, p. 305). 2 or 3 grams of
pyrites cinders are mixed with 1 or 2 grams of sodium bicarbonate
of known titre ; the mixture is heated in a nickel, porcelain, or
platinum crucible by means of a small Bunsen flame for 5 or 10
minutes^ stirred up, heated once more for 15 minutes with a some-
what stronger flame, treated with hot water, filtered and washed.
The solution is titrated with hydrochloric acid and methyl-orange ;
the loss of titre in comparison with the original one is equal to the
sulphate formed. The escaping carbonic acid keeps the ^lass
76 RAW MATERIALS Of MANUFACTURE.
porous ; there is no spurting, and the lixiyiation of the small bulk
of the mixture is easy and expeditious.
Experiments made iu my laboratory (Zeits. f. ang. Chem.
1892, p. 447) showed that Watson's method yielded results
closely agreeing with those obtained by accurate gravimetrical
methods, whilst the method of Pelouze, even with burnt-ore,
yielded too low results, and that of Kolb was not more reliable.
But it was found that special precautions must be observed in
order to yield accurate results. 3-200 grams of the finely ground
samples are mixed with 2*000 grams sodium bicarbonate (of known
alkalimetrical titre), and heated in a nickel or iron crucible; a
platinum crucible easily leads to over-heating. The heating is
continued for 10 or 15 minutes with a very small flame, then
another 15 minutes with a strong flame, but without fusing the
mass. The crucible must be kept covered and the mass must n(»t
be stirred ; it should in the end be red-hot, and after cooling black
and porous. It is boiled in a porcelain dish with water, adding the
same volume oE neutral solution of sodium chloride. This is
an essential improvement, as without this addition some oxide
of iron invariably passes through the filter, and makes the
following titration by methyl-orange almost impossible. The
filtered solution is titrated with standard acid. The ditference
between the original titre of bicarbonate and that now found
shows the sulphur, 1 c. cm. normal acid indicating 0'5 per cent. S.
In the presence of a somewhat considerable quantity of zinc this
method does not answer.
Magnetic pyrites {pyrrhotite), FcjSg, is sometimes present,
especially in some kinds of American pyrites. As that ore yields
its sulphur very imperfectly in ordinary pyrites-kilns, its estimation
may become important. Cone (Journ. Amer. Chem. Soc. xviii.
p. 404) effects this by grinding the ore so that all passes through
a 60-mesh sieve (not more finely !), spreading the powder on a
sheet of glazed paper, applying a magnet to this, removing the
mechanically adhering pyrites by gently knocking the magnet and
separately brushing off the pyrrhotite after putting on the anchor.
This is repeated five or six times, and the sulphur estimated in the
separated portions.
Marcasite and pyrites can be distinguished by the easier de-
composition of the former when boiling with a solution of ferric
salt. This behaviour has been studied in detail by H. N. Stokes
OTHER CONSTITUEirrs OF PYRITES. 11
(Bulletin U.S. Greol. Surv. No. 186) ; but it is of no practical
importance for technical analysis.
Estimation of other constituents of Pyrites, — Usually it is
snfficient to estimate the sulphur in a pyrites whose nature is
otherwise known. If^ howeyer^ the pyrites is of unknown compo-
sition^ its value for acid-making can only be estimated by a com-
plete determination of all its constituents according to the rules
of mineral-analysis. If it contains^ for instance, calcium carbonate,
this on burning will retain its equivalent of sulphuric acid equal
to 0*32 per cent. S for each per cent. CaCOg ; if calcium sulphate
is present from the first, its sulphuric acid has to be deducted
from the whole quantity of sulphur found. If lead has been found,
an equivalent of sulphur must be considered as practically lost ;
and the same is the case with zinc, — because the sulphates of both
metals are hardly or not at all decomposed at the temperature of
a pyrites-burner. In France half of the S combined with Zn is
considered as lost, = 0*245 p. c. S per 1 p. c. Zn (for copper
they reckon 0*505 p. c. S per 1 p. c. Cu as lost). Frequently
arsenic will also have to be sought for ; and even silica may
be of importance — firstly, because in the presence of much
silica " explosive '^ properties of the pyrites must be feared (see
p. 55}^ and secondly, in the case of cupreous pyrites, because
silica lessens the value of the cinders. Even silver and ffold
are sometimes sought for (comp. Chem. News, xxvi. p. 63,
xxxiv. pp. 94, 132, 152, 172) ; but it cannot be said that the
quantities found have any influence on the commercial value of
pyrites.
It is not our object here to treat of the estimation of all
these substances, nor that of the copper which, in the majority
of cases, represents a large portion of the value of pyrites *. We
make an exception only with arsenic, because special methods for
estimating this in pyrites have been worked out which are not
found in the ordinary text-books.
The process employed at Freiberg is that used by Reich,
and is as follows : — Digest about 0'5 gram of finely pulverized
pyrites in a porcelain crucible, covered with a watch-glass, with
* For these substances, see Lunge and Hurter^s ' Alkali-Maker's Handbook ' ;
also a very extensive paper by Westmoreland, J. Soc. Cbem. Ind. 1886, p. 31,
aod criticisnis on the same, p. 277 ; and Lunge's Ghemisch-Technische Unter-
Buchungsmethoden,' i. p. 250 et seq.
I
78 HAW MATERIALS OF MANUFACTURE.
concentrated nitric acid at a gentle heat^ until the residue assume9
a lighter colour and the separated sulphur has turned a pure
yellow. After decomposition^ heat the crucible on a sand-bath to
get rid of the excess of acid^ but not to dryness. Add 4 grams
of sodium carbonate, dry completely on the sand-bath, add 4
grams of potassium nitrate, and heat the mass until the contents
of the crucible have been in quiet fusion for ten minutes. Extract
the cooled mass with hot water and filter; the filtrate contains all
the arsenic as sodium arseniate. Acidify with a little nitric acid,
keep for two hours on a sand-bath to get rid of the carbon dioxide,
add a sufficient quantity of a solution of silver nitrate, and
neutralize carefully with dilute ammonia. The reddish-brown
precipitate of silver arseniate is filtered, washed, dried, taken oflF
the filter as well as possible, the filter is incinerated in a muffle, the
precipitate put to it, a sufficient quantity of assay lead is added, and
the silver estimated by cupellation. 100 parts of silver correspond
to 23' 15 of arsenic.
Leroy M. McCay has modified and greatly simplified this method
(Chem. News, xlviii. p. 7) by estimating the excess of silver used
by Volhard's method. Later on (Amer. Chem. J. viii. no. 2) the
same author recommends as preferable another plan, namely dis-
solving the AggAsOi in dilute ammonia, and either estimating the
silver by Volhard's volumetric method (precipitation with ammo-
nium thiocyanate), or evaporating, drying, and weighing the total
in a thin platinum dish. If the arsenic is to be precipitated as
pentasulpliide, which is otherwise a tedious operation, McCay
recommends (Amer. Chem. J. ix. no. 3, and x. no. 6) to place the
solution in a flask with a well-fitting stopper, acidify with HCl,
and dilute with freshly boiled water till the flask is nearly full,
pass in H2S to saturation, insert and fasten down the stopper, and
place the whole in a water-bath for an hour. At the end of that
time all the arsenic will be precipitated as pentasulphide, As^S^,
containing no free sulphur.
Clark (J. Soc. Chem. Ind. 1887, p. 352) recommends the
following method as especially adapted for estimating very small
quantities of arsenic in pyrites rich in sulphur : — Mix 3 grams of
pyrites in a platinum crucible with four times as much of a mixture
of calcined magnesia and sodium hydrate, heat for about 10
minutes over a moderately low Bunsen flame, extract the shrunk
mass with boiling water, acidify the solution with hydrochloric
ZINC-BLENDE. 79
acid (which evolyes mach H2S)^ boil for a few minutes^ and
saturate with H2S^ when all the arsenic will be thrown down as
aulphide; wash the precipitate^ extract the sulphide of arsenic
with ammonia, evaporate the solution to d]7ness, dissolve in
strong nitric acid, and estimate the As as ammonio-arseniate of
magnesia^ or else by silver solution as above described. Or else
the calcined mixture, after neutralizing it with HCl, as above
mentioned, is reduced by cuprous chloride, and the liquid is slowly
distilled into water, repeating this operation twice with strong
HCl, which will cause all the arsenic to pass over as AsCI^; which
can be either precipitated by H2S or titrated by iodine. Clark
points out the necessity of carefully testing all the reagents
employed for arsenic, of which he had found as much as 0*02 per
cent, even in commercial caustic soda.
Nahnsen's process (Chem. Zeit. xi, p. 692 ; abstr. J. Soc. Chem.
Ind. 1887, p. 564) does not offer any special advantage. The
process described in detail by H. Fresenius (Zeitsch. f. anal.
Cliemie, 1888, p. 34) is no doubt very accurate, but lengthy and
troublesome.
Blattner and Brasseur (Z. angew. Ch. 1898, p. 262) give exact
instructions for estimating arsenic in pyrites, both in the wet and
dry methods.
3. ZiNC-BLENDE.
This mineral is now the principal zinc-ore. Previously to
reducing the zinc, the blende must be roasted in order to convert
it into ZnO, and this process was formerly carried out without
taking any care to deal with the enormous quantities of SOg
formed. Sanitary legislation at last interfered with this procedure,
which of course laid waste all the country round the zinc- works,
and compelled measures for dealing with the noxious gases. Part
of this is used for the production of liquid sulphur dioxide (comp.
Chap. IV.) ; most of it, however, serves for the manufacture of
sulphuric acid, so that blende has now become one of the more
important raw materials for this purpose. The historical develop-
ment of this industry will be related in Chap. IV.
Blende occurs in large quantities for instance in Silesia, West-
phalia, Rhineland, Saxony, Austria, Belgium, Wales, the Isle of
Man, Spain, Italy, France, the United States, in nearly all of
80 RAW MATERIALS OF MANUFACTURE.
which localities it is utilized for the manufacture of sulphuric
acid.
Pure blende^ ZnS, contains 32*9 per cent. S, and 67*1 per cent.
Zn. The commercial ore is^ of course^ always impure. We quote
the following analyses^ by Minor (Chem. Zeit. 1889, p. 1602)^ of
Rhenish blende : —
S 30-24 27-94 22-11 21-05
ZnasZnS 2273 27-17 3446 31-16
Zn in other combination .. . 503 475 5-83 6*65
Fe 15-98 1312 206 2-33
Gangue (by diff.) 21-02 27*02 35-54 88-84
Drasche analyzed a blende from Carinthia : ZnS 68*41 per cent.,
PbS 4-55 ; FeS, 205 ; ZnCOj 240 ; CaCOg 893 ; MgCOg 1062 ;
AljOa 0-63 ; gangue (principally quartz) 2-32 per cent.
Pennsylvanian blende, according to P. A. Genth : — Sulphur
32-69 to 33-06 per cent., Zinc 6647, Iron 038, Cobalt 034.
Jurisch (Schwefelsaurefabrikation, p. 61) quotes analyses of
various descriptions of blende, burnt by the Chemische Fabrik,
Bhenania, with sulphur contents varying from 18-40 to 32*20, and
zinc from 14-90 to 50*22 per cent.
Haenisch and Schroeder (Chem. Ind. 1884, p. 118) quote the
contents of Silesian blende =23 to 37 per cent. S, also an inferior
blende =8 to 21 per cent. S.
Blende frequently contains cadmium and mercury. The latter
occurs in Rhenish blende only =002 per cent., but in Spanish
(Aviles in Asturia) =0135 per cent.
According to direct information from the Rhenania Chemical
Works at Stolberg, the sulphur contents in the blende roasted
there ranges between 20 and 30 per cent. ; on the average 25 to
28 per cent. Iron does not do any harm, but lime retains its
equivalent of S as CaSO^. Lead affects the duration of the
roasting-floors ; it is partially volatilized with a little silver, and
reappears in the flue-dust, the towers, and chambers. Mercury and
fluorine are also volatilized and act very injuriously on the plati-
num vessels used for concentration. Recently the sulphuric acid
factories refuse blende containing fluorides. Arsenic generally
occurs in blende in such extremely slight quantities that the acid
made from it may be considered as technically pure.
The production of blende in Prussia in 1890 amounted to
OTHER METALLIC SULPHIDES. 81
362^464 tons; Belgium in 1889 to 12,376 tons; France and
Algiers in 1887 to 13,800 tons ; Spain in 1885 to 2,488 tons.
Analyses of Zinc-blende. — The sulphur is estimated by the wet
method as described p. 62. In the cinders from blende the same
method must be eiQployed since the dry methods, e. g. Watson-
Lunge's (p. 75), give quite wrong results. Details respecting this
and the other constituents of zinc-blende are found in Lunge
and Hurter's 'Alkali-Maker's Handbook ' and Lunge's ' Chemisch-
technische Untersuchungsmethoden.'
4. Otheb Metallic Sulphides.
Pyrites proper has hardly any other application than that for
sulphuric-acid making, and it is obtained almost exclusively for this
purpose. In the case of cupreous pyrites the sulphur constitutes
only a portion, but a very considerable one, of its value. The
working of poor copper- ores would not pay, apart from the
noxious effect of the gas produced in calcining the ore, unless the
price of the ore were very moderate ; and this is only possible by
the acid-makers paying on their part for the ore, which they can
well afford, as most kinds of cupreous pyrites behave very well
in the burners, and yield quite as much acid in proportion to
their percentage of copper as the non-cupreous ores. The case
of zinc-blende is now similar to that of cupreous pyrites.
The case is different with most other sulphuretted ores occurring
in smelting-operations, such as galena, the many mixed ores con-
taining blende and galena, besides iron- and copper-pyrites, the
richer copper-pyntes themselves, and, lastly, the intermediate pro-
ducts, " coarse metal/' " matte/' &c. These, for their metallurgical
utilization, equally require a calcination evolving sulphur dioxide ;
but the matter is very different here from what it is with a good
pyrites, whether it be pure iron -pyrites or containing a few per cent,
of copper. On the whole, all those ores and metallurgical products
are much poorer in sulphur than ordinary good pyrites ; and for
this reason they are less easily calcined in such a manner as to
allow of utilizing the gas, because the evolution of heat by the com-
bustion of their own sulphur is not sufBcient to maintain the process
energetically. An external stimulus was required before smelting-
works would seriously attempt to utilize the sulphur dioxide
contained in the gas from calcining the ores ; and this proved to be
VOL. I. O
82 RAW MATERIALS OF MANUFACTURE.
■
the damage and nuisance caused by the noxiotts vapours all round the
works. The claims for damages^ the law-suits, and the measures
taken by the authorities at last made it impossible in many places
for the works to go on in the old way ; and although it appeared at
first as if the sulphurous acid could not be condensed at all in this
case, or only at a loss, practice has now succeeded in fulfilling the
task in most (but not in all) cases, principally by the construction
of improved burners, which will be described in the fourth Chapter.
It would undoubtedly be too much to say that the task had
been solved in all its parts; the success has mostly been only
partial. In many cases where an ore could not possibly have
been calcined so as to utilize the gas, mixing of it with others
has been resorted to. Thus the Halsbriicke works, near Freiberg,
roast galena and blende along with pyrites ; and in 1870 they made
8000 tons of sulphuric acid from the gas.
In reference to the sulphur dioxide escaping as noxious vapour,
Leplay (comp. Percy, ^ Metallurgy,' 1862, i. p. 337) mentions that in
South Wales annually 46,000 tons of sulphur escaped into the air as
sulphur dioxide, along with arsenic, fluorine, lead, and zinc com-
pounds, in spite of the condensing-chambers. In fact the country
round Swansea was stripped of all vegetation. At Fi'eiberg the works
in 1864 paid upwards oi£2750 damages on account of their vapours,
in 1870, after better condensation had been eflfected, onlv .€239.
It should not be forgotten that sulphur dioxide occurs in very
large quantity, although in a much more dilute state, in all coal-
smoke, and consequently in the atmosphere and the rain-water
of all large towns, and that the most perfect *' smoke-combustion^'
cannot do away with this. Much more injurious than the vapours
escaping through high chimneys, which are soon diluted with air,
is the smoke from brick-works, coke-ovens, and other fires which
emit their smoke at a low height above the ground. Mr. Fletcher
has calculated that at St. Helens the acids escaping amounted :
From fire-gases to 800 tons per week.
„ copper- works „ 380 „ „
„ glass-works „ 180 „ „
„ alkali- works „ 25 „ „
Similar calculations have been made by Mr. Hasenclever (Chem.
Industrie, 1879, p. 225), who has given strong proof of the damao-e
done by coal-smoke alone.
ACID-GASES IN SMOKE. 83
According to the 28th Alkali Report (for the year 1891), p. 19, the
quantity of acid gas which escapes into the air at St. Helens, cal-
culating HCl as its equivalent of sulphur acids, is as follows : —
Tons sulphur
per annum.
From Copper and Lead smelting works 1 1 ,480 ^
Glass-works 7,500 V 19,313
Polishing-powder works 333 J
Coal burnt (1^ per cent, of 1,040,000) 15,600
Chance-Claus process ... 620
Sulphuric-acid chambers 173^ ^^^
„ Alkali-works 402j ^^^
Total 36,108
This is the equivalent of 72,216 tons 80^ or 110,586 H2SO4,
of which the alkali-works contribute only 1^ per cent. Since all
this is given off from an area of about 3 square miles, each square
mile at St. Helens receives the equivalent of 12,036 tons of sulphur,
against 11 tons in summer or 44 tons in winter on a square mile
in London.
In Chapter IV. we shall deal with the various attempts at
utilizing, or at least rendering innocuous, the acid gases given out
in calcining ores, and we shall here enumerate only the various
classes of ores or waste substances coming under the head of
causing " noxious vapours,^' such as might serve for the manu-
facture of sulphuric acid (apart from zinc-blende, comp. p. 79).
Copper-pyrites and mixtures of this with blende, galena, &c. are
roasted in several places in kilns so as to utilize the SO2 in acid-
chambers — at Chessy near Lyons, at Oker in the Harz, at Manz-
field, at Swansea. At the Altenau silver-works near Clausthal, in
1872, 228 tons of vitriol of 106° Tw. were made from copper-
pyrites (and 314 tons from lead-matte, Wagner's ^ Jahresb.' 1874,
p. 276). At Freiberg the Mulden and Halsbriicke works proceed
in the same way ; but they only utilize ores and products pretty
rich in sulphur for vitriol-making.
The following particulars respecting the materials burned at
the Government works at Oker in 1901 have been communicated to
me from an official source. The ores are : —
g2
84
KAW MATERIALS OF MANUFACTURE.
Lumps.
Tons.
Cop[)er-ores No. 1 1 ,205
2 4,095
3 1,417
Mixed ores 7.916
Pyritic lead-ore 805
Copper-matte 6,288
Lead-matte 3.377
*' Spurotein *' (rcguliis) ......
25,103
Total.
Tons.
Ton8.
895
2,100
88
4,183,
73
1,49a
3,770
11,686.
&4
869.
• ••
6,288
• • •
3.377-
1,620
1,620.
6,510
31,613.
Composition of these materials : —
Copper-ore.
1.
Cii 17-70
Fe 23()0
Zn I 950
Pb 3-70
S 32-00
o
9-70
475
30 40
33-50
5-80 ,
4-90
2-4(J
1-75
3600
40 50
Aiixed
ore.
•
lead-
ore.
uopi^r-
uiaite.
460
1-05
30-47
12-40
24-.50
24-40
21-50
15-.')0
8-75
1005
6-75
5-80
24-tKJ
34-00
18-70
.,* Kesulua.
matte. ®
18-20
21-70;
15 00 I
710
1700
64-,38
8-93
1-34
2-95
20-70
The production of acid at Oker is per annum 21,000 tons.
50° B., in 5 sets of chambers of 19,656 cub. mtrs. capacity, t. e.
1068*38 kg. acid of 5° B., per 1 cb. m. chamber-space. The kiln-,
gases have 4 to 5 per cent. SO2.
Traces of Hg, Tl, Cd, and Se have been found both in the ores
and in the products obtained therefrom.
Apart from pure pyrites, the " ordinary ores '' are best adapted;
for vitriol-making, because they contain their sulphur mostly
as FeS2; the "mixed ores'' are less favourable, on account of
their galena, and the rich copper-ores on account of their large
percentage of copper. Of the lead-ores only those anaply permeated,
by pyrites are fit for vitriol-making. The sulphur in the ores
worked at Oker varies from 20 to 40 per cent. ; on an average it
is 30 per cent. ; but it must be noticed that the sulphur of the
galena is altogether unavailable for vitriol-making. The case is.
not much better when copper-pyrites predominates, because this,
furnishes too poor a gas, and, moreover, decrepitatqs and fall? to.
SPENT OXIDE OF GAS-WORKS. 85
powder in roasting. If no more than 35 per cent, of copper-
pyrites is mixed with iron-pyrites, it does no harm. Blende behaves
in a similar way, but rather more favourably ; ores containing 35
per cent, blende along with 25 per cent, pyrites yield gas quite
adapted for vitriol-making. *
The first sulphuric-acid works at Oker were erected in 1841 ;'
there is now one of the largest acid works in Germany, viz. 14 sets
of chambers with a capacity of 800,000 cubic feet.
Galena is probably nowhere worked in such a way as to extract
its sulphur in the shape of sulphuric acid. The purest galena con-
tains only 13'4 per cent, of sulphur; it is transformed into lead
sulphate on roasting, and only in the strongest white heat gives off
a portion of its sulphur as SO2; moreover the metallurgical pro-
cesses to which it is subject are of such a nature that only poor
gas can be produced from it. This subject has been discussed by
Bode in his ' Beitrage zur Theorie und Praxis der Schwefelsaure-
fabrikation,' 1872, pp. 32 & 63 ; his conclusion is that even mix-
tures of galena and pyrites cannot be roasted in kilns for vitriol-
making if they contain more than 18 to 20 per cent, of galena.
^'Coarse metal '^ of copper- smelting is roasted for vitriol at
Mansfeld. A product containing 34 per cent. Cu, 28 per cent.
Fe, and 28 per cent. S, according to Bode, yields gas with 5^ per
cent, by volume of SOo, and at a sufficiently high temperature to
work with the Glover tower. In most cases, up to the present,
coarse metal cannot be roasted so as to utilize the SOo.
Lead-matte is used for vitriol-making — for instance, at Freiberg
and in the Lower Harz ; it is there roasted in large kilns of 12^
tons capacity. The matte loses half its sulphur, aud yields gas with
4 to 6 per cent, of SO2 ; the temperature, according to Bode, is
high enough for the Glover tower. In the Upper Harz the
utilization of its sulphur in metallurgy has in general not been
found practicable.
5. By-products op other Manufactures.
{Spent oxide of gas-works, soda-waste, ammonia-works, ifc.)
The^pent oxides from the purification of gas by hydrated ferric
oxide are in many places used for vitriol-making. This sulphur
originally also comes from pyrites, viz. from that contained in the
86 RAW MATERIALS OF MANUFACTURE.
coal^ which appears in the gas mostly as sulphuretted hydrogen.
Most works remove it from the gas by a mixture of hydrated iron
oxide and sawdust. In this case sulphide of iron and sulphur are
formed^ according to the equation
2Fe(bH)3 + 3H2S=2FeS + S + 6H20;
and when the mass^ having become inactive, is exposed to the air, it
again passes over into ferric hydroxide, more sulphur being preci-
pitated^ thus :
2FeS + 03 + 3H20=2Fe(OH)a + S2.
The hydroxide thus reproduced and mixed with sulphur is again
used in the purifiers, and is regenerated about 30 or 40 times
before the sulphur has accumulated therein to such an extent that
the mixture does not work any longer ; it is then replaced by fresh
oxide, and the spent ore is passed over to vitriol- makers.
Phipson states the composition of such a mass to be : —
Water 14 per cent.
Sulphur 60 „
Organic substance insoluble in alcohol. 3 „
Organic substance soluble in alcohol
(calcium ferrocyanide and sulpho-
cyanide, ammonium sulphocyanide,
ammonium chloride, hydrocarbons). 1*5 „
Clay and sand 8 „
Calcium carbonate, ferric oxide, &c.... 13*5 „
If the oxide contains considerable quantities of cyanides, it may
cause great trouble in the manufacture of sulphuric acid (31st
Alkali Report, p. 89).
Hot water extracts the ferrocyanides and sulphocyanides, along
with ammonium chloride; the solution can be evaporated to dryness,
and the residue separated by alcohol into insoluble calcium ferro-
cyanide and soluble sulphocyanide and chloride.
According to the analyses of Davis (^ Chemical News,' xxix.
1874, p. 30), three samples of spent oxides contained : —
SPENT OXIDE OF OAS-WOKKS.
87
I.
Sulphur .'..;■/.'.... &4-376
Ferric hydroxide 14*421
Insoluble ,..^»,...„ iV0i}2
Moisture 2*079
Liiue(a8Ca8) 2399
Sawdust 2470
Calcium carbonate
Ammonium sulphocjanide 2*662
Ammonium chloride 1 ,,^/.-
Ammonium cyanide J
Ammonium ferrocyanide
Prussian blue trace
II.
III.
62-3r>8
67-956
17112
15-3;i5
r»-099
8-301
5-387
3*900
1-776
1-002
5135
300(>
1-324
1102
100*064
l*66;j
0-366
100*220
trace
100-605
These samples seem to have been taken from precipitated iron
hydroxide, to judge from further analyses by Davis (Chem. News,
xxxvi. p. 189), in which also tarry substances are taken into account.
Residues proceeding from
P recipitated
Fe(0H)3.
Ferric hydroxide 17*74-19*36
Sawdust 1-98- 472
Calcium carbonate 0- - 1-04
Ammonium sulphocjanide ... 1*99- 2-74
Ammonium ferrocyanide trace
Tarry matters 0*72- 1*22
Sulphur 62-44-67-18
Insoluble in dilute HCl 3-6(>- 5-47
Prussian blue
Calcium sulphate
Ammonium sulphate
Moislu re ( by difference) , . . . . . 4-72- 5 76
Bog-iron-
ore.
15-96-26-42
1 14- 3-72
0* - 1-73
0-94- 1-93
trace- 021
0-92- 1-14
48-76-57*44
9-74-11-42
trace- 0-17
Copperas.
504- 6-84
1-(M- 3-24
0-
l-t>8- 3*41
0-27- 0 64
0-72- 118
48-76-55-74
7-82-12-68
trace- 1-74
trace- 1*43
12*78-16'72
7*22-10-82 7-98- 9-22
Bad
oxides.
8-72-20-40
2-16- 9-76
0* -10-36
1*18- 4*72
trace- 0*44
0*55- im
32*42-42-16
12*12-20-71
trace- 064
0- - 3-23
0- - 114
7*49-3;i-41
From these analyses it can be seen, first, that it is decidedly best
to extract the mass at first with water, in order to remove the am-
monia compounds, which are in themselves valuable, and which, if
they get into the chambers, destroy a good deal of nitre (their value
is certainly greatly lessened by the sulphocyanide) ; secondly, that
sometimes a considerable quantity of calcium carbonate is present,
whicji may get into it at the gas-works by lime being added, on
purpose or by mistake, and which, of course, retains an equivalent
quantity of sulphur in the shape of gypsum. In fact a sample of
the residue left after burning contained
88 RAW MATERIALS OP MANUFACTURE.
Insoluble 33-386
Ferric oxide 52-399
Calcium sulphate 13"315
Sulphur 0-200
These impurities (which cause a loss by retaining sulphuric acid)
aud the sulphates present from the first (which are not available)
must be allowed for in analyzing. This, according to Davis, was
formerly done by extracting the sulphur by means of carbon
bisulphide, evaporating the solution, and weighing off the sulphur ;
but as the presence of tarry matters causes an error, Davis now
makes the analysis by burning the sulphur in a current of air in a
combustion- tube of Bohemian glass, conducting the SO3 formed
into an absorbing-apparatus filled with iodine solution, and reti-
trating the unaffected iodine by a solution of sodium thiosulphate
(Chem. News, xxxvi. p. 190 ; comp. also Zulkowsky's process, p. 72) .
The burning of this gas-sulphur is usually done in shelf-furnaces
similar to those used for pyrites-smalls. They will be described in
detail in the fourth Chapter. Already in 1861, at Barking Creek, on
the Thames, 2180 tons of this material were used ; but much larger
quantities might have been got, since, according to A. W. Hofmann
(' Report,' 1862, p. 15), even at that time at least 10,000 tons of
sulphur were contained in the London gas. Much of the acid
made from spent oxide is sold as " brimstone acid.''
In France also, at that time, the sulphur from gas-works was
used on a large scale. The factory at Aubervilliers, belonging to
the Society of St.-Gobain, used no other ; Messrs. Seybel and Co. at
Liesing, near Vienna, and Kunheim and Co. at Berlin (Wagner's
Jahresb. 1864, p. 153 5 Hasenclever, /. c. p. 167) do the same.
The rational treatment of spent oxides for the purpose of obtain-
ing ammonium salts, ferrocyanides, and sulphocyanides is described
in Lunge's ' Coal-Tar and Ammonia,' 3rd edition, p. 716.
The sulphur contained in alkaluwaste, in the shape of calcium
sulphide, has been frequently proposed for the manufacture of
sulphuric acid, nearly always after having been first converted into
sulphuretted hydrogen. The only successful process in this line
(the Chance process of treating alkali-waste) belongs to the domain
of alkali-manufacture, and cannot be treated in this volume.
Only the contrivances for burning the hydrogen sulphide will be
described in the fourth Chapter.
Borntrager (G. P. 15,757) proposes decomposing the yellow
SULPHURETTED HYDROGEN. 89
liquors from alkali-waste by meaDS of ferric oxide (ground damp
pyrites cinders)^ to filter the sulphur and ferric sulphide which is
thus precipitated, and burn it after drying in ordinary shelf-
burners. [Oxide of iron in this state is a very inferior reagent
for removing the sulphur from yellow liquors and the like.]
An anonymous inventor has proposed to absorb the sulphuretted
hydrogen in hydrated ferric oxide suspended in water, filter, press
the residue, dry it, and bum it on shelf-burners. Wyss (Bull. Soc.
ind. Mulh. 1890, p. 281) has shown that this process is neither
novel nor in any way promising of success.
The sulphuretted hydrogen given off in the manufacture of sulphate
of ammonia (comp. Lungers 'Coal-Tar and Ammonia,' 3rd ed.
p. 857) is sometimes used for the manufacture of sulphuric acid.
Here the HoS is not merely diluted with a large quantity of inert
gases, but is also of very unequal concentration, which formerly
rendered the manufacture of sulphuric acid from this source an
unprofitable process. The same can be said of most other eases in
which HgS is given off as a by-product.
The utilization of the HjS from sulphate-of-ammonia works for
the manufacture of sulphuric acid has, however, been greatly
improved and is now no longer a rare exception, but is practised
in a good many English works. If the gas is properly introduced
into the burner (comp. Chapter IV.), the consumption of nitre is
not excessive, and it is even possible to increase the heat by this
means, if the spent oxides should not suffice for this purpose. The
action of the large quantity of carbon dioxide accompanying the
HjS in the case of sulphate-of-ammonia works would seem to
consist only in requiring a certain amount of chamber-space,
contrary to the opinion reported in Chapter VIII.
Sulphur dioxide, formed in many manufacturing operations as a
disagreeable by-product, apart from those already described, is
sometimes proposed to be converted directly or indirectly into
sulphuric acid. The special cases in which this has been attempted
will be treated in the next Chapter.
6. Nitrate op Soda, NOgNa (commonly called '' Mere '^).
i (NajO) = 31-05 36-49 per cent,
i (NA) = 5401. 63-51 „
8509 100-00
Hardness 1*5 to 2; spec. grav. 2*09 to 2*39. In the pure state,
90
RAW MATERIALS OF MANUFACTURE.
and in large crystals, it is colourless^ transparent^ and brilliant as
glass ; in small crystals it is white and opaque. The crystals are
rhombohedra with angles of 106° SO' and 73° SO'. It has a
cooling, bitter taste. Heated to a certain temperature it melts ;
at a red-heat it is decomposed into sodium nitrite and oxygen gas.
The fusing-point is 316°-319° C. (Carnelley, J. Chem. Soc. 1878,
ii. p. 277). Mixed with coal, deflagrates on heating. It attracts
moisture from the air (especially if not quite free from chlorides),
and readily dissolves in water, with a considerable lowering of the
temperature.
1 part of sodium nitrate, according to Marx, requires for solu-
tion 1-58 part of water at -6°, 1-25 at 0°, 0*46 at +119° C.
According to Kopp, 1 part of sodium nitrate at 18°*5 C. requires
1*14 of water, or 100 parts of water dissolve 87"72 parts of the
salt. In the presence of sodium chloride its solubility is consider-
ably less.
Specific Gravity of the Solutions of Sodium Nitrate at 20" C.
Parts of salt
Specific
Parts of salt
Specitic
in 100 water.
L'rftvitv.
1 V
111 100 water.
gravity.
1
10065
21
1-1498
2
10131
22
11578
3
1-0197
23
1-1659
4
102f)A
24
1-1740
o
1-0332
25
1182>
6
10399
26
1-1904
7
1-0408
27
, 1-1987
8
10537
28
1-2070
9
l-OGUO
29
1-2154
10
1 0676
30
1-2239
11
10746
31
1-2325
12
1-0817
32
1-2^12
13
10889
33
1-2500
U
1 0962
31
1-2589
15
1-1035
35
1-2679
1()
1-1109
3()
1-2770
17
11184
37
1-2863
18
11260
38
1-2958
1!)
1-1338
39
1-3055
20
1-1418
40
1-3155
OCCURRENCE OF NITRATE OF SODA.
91
Parts of salt
Specific
Parts of salt
Specific
ill 100 water.
gravity.
in 100 water.
giavity
41
1-3255
46
1-3761
42
1-3355
47
1-3861.
43
1 -3456
48
1-3968
44
1-3557
49
14^74
45
1-3659
50
1-4180
Kiti*ate of soda occurs iu many places iu small quantities; but
the only large beds which supply the world with this article are
situated in a region of the west coast of South America, formerly
belonging to Peru and now to. Chili. This occurrence, and the
industry founded thereon, have been described in various commu-
nications by Langbein (Wagner^s Jahresb. 1871, p. 300 ; 1872,
p. 290; 1879, p. 390) ; also by W. E. Billinghurst, of whose book
(written in Spanish) Darapsky gives an extract in the ' Chem.
Zeit/ xi. p. 752 (J. Soc. Chem. Ind. 1887, p. 545). Comp. also
Buchanan {ibid. 1893, p. 128) and Behrend (Zschr. deutsch. Ingen.
1899, p. 1199; Fischer's Jahresb. 1899, p. 406).
The nitre-beds are principally situated in the province ot
Tarapaca, between 68° 15' and 78° 18' longitude, and 19° 12' and
21° 18' 30", latitude ; they also occur somewhat south, especially
near Antofagasta and Taltal. They were discovered in 1821 by
Mariano de Rivero, and have been worked since 1830. The nitre
zone is situated at an altitude of 3600 feet above the sea-level.
The total area of the nitre-bearing strata is estimated by Billing-
hurst at 21,212 estacas (about 150,000 acres), and the yield
obtainable therefrom = 1980 millions of Spanish cwts. The nitre-
bearing rock, called caliche, is found in layers of from 10 inches
to 5 feet in depth, which rarely crop out at the surface. The over-
lying rock, called costra, is 18 inches to 7 feet thick, and consists
principally of a hard conglomerate of sand, felspar, phosphates,
and other minerals.
The composition of the caliche varies; it contains from 48
to 75per cent, of sodium nitrate, 20 to 43 per cent, of sodium
chloride, and varying quantities of sodium sulphate, calcium sul-
phate; potassium nitrate, potassium iodate, magnesium chloride,
also insoluble earthy portions and organic substance (guano). It
is first broken by a stone-breaking machine, and then put into the
dissolvers. These are partly open square tanks, preferably, how-
ever, closed egg-shaped boilers with two manholes—one on the
92 KAW MATGRIALS OF MANUFACTURE.
r
top for filling in the caliche, another at the bottom for emptying
the residue. The mass rests on a perforated bottom. The boilers
are filled entirely with the broken rock, and half with mother-
liquor, and were formerly always heated by direct steam injected
below the false bottom. After 1| to 2^ hours the liquid, then
sufficiently saturated with nitre, is run into settlers ; from these
it flows, after several hours, into a second settler, where, by half
an hour's rest, it allows some still-suspended common salt to sub-
side, and then runs into shallow coolers. The residue from the
dissolvers^ which still contains 15 to 35 per cent, of sodium
nitrate, is either emptied at once or boiled again with fresh water.
The crystals, separated in the coolers after draining off the mother-
liquors, are spread in layers of 12 to 18 inches thickness on a large
surface exposed to draught, and dried with frequent stirring. The
total cost of sodium nitrate, up to its reaching European ports, in
1871, amounted to £S ISs. per ton, which left a good margin
for profit at the average price of .£12 (it has reached £16 and more).
At present both the producing and selling prices are much lower.
The above-described system of dissolving by open steam was
afterwards abandoned for closed steam-coils or similarly-acting
apparatus; at the same time air heated to 120^-150° C. is forced
through the liquid by means of injectors, in order to hasten the
evaporation. This produces both stronger and purer liquors, the
quantity of sodium chloride being the same in the stronger as
in the weaker liquor.
The composition of the crude nitre-earth is shown by the fol-
lowing analyses : —
Caliche. Costra.
^ ^ ^
a. h. c. d. c.
Sodium nitrate 70-62 (50-97 51-50 49-05 18-60
Sodium iodate 1*90 0*73
Sodium iodide ... traces traces
Sodium chloride 2239 leS.) 2208 29-95 33«80
Sodium sulphate 1 80 4-5<i 899 902 16-64
Potassium chloride ... 855 4*57 2*44
Magnesium chloride ... 043 125 1-62
Magnesium bulphate 0*51 5*88
Calcium sulphate 0*87 l"3l ... •••
Calcium carbonate ... 0*12 015 0*09
Silica and ferric oxide ... 090 2-80 3'00
Insoluble 0-92 406 6-00 318 2010
Moisture 0*99 5*64 ... ... ...
10000 10000
NITRATE OF SODA AND POTASH. 9S
The analyses a and b {a white, b brown caliche) are by Machattic
(Chem. News, xxxi. p. 263) . They are somewhat suspicious, both
on account of the total absence of potassium salts and of the
extremely improbably high percentage of sodium iodate. This is
all the more noticeable as Machattie at the same time states the
average percentage of iodine in five samples of mother-liquor to be
=056, equal to 0*873 per cent, of sodium iodate, which may be
nearer the truth. The analyses c, rf, and e are by V. L^Olivier
(Compt. Rend., 26th October, 1875).
The iodine contained in the mother-liquors is now recovered to
a great extent, and forms one of the principal sources of this
article.
Another bed of nitrate of soda has been found in the South-
American State of Colombia (Journ. Soc. Chem. Ind. 1894,
p. 1001). It is about 100 kilometres distant from San Juan de
la Cienaga, and had been proved up to that time for a surface of
75 square kilometres. It has a thickness of from 30 centimetres
to 3 metres, and averages 11 to 12 per cent. NaNOy, together
with calcium carbonate, calcium sulphate, and silicates. The
recoverable nitre is estimated to exceed 7 million tons, but none
of it is as yet in the trade. Another bed has been discovered in
Texas ; this is stated to contain 98 per cent, pure nitrate (Chem.
Ind. 1902, p. 265).
A deposit of potassium nitrate has been found near Cochabamba,
in Bolivia (Sacc, Compt. Rend. xcix. p. 84). This deposit,
reported to be enormous, but not yet worked, consists of 60 "70
per cent, potassium nitrate, 30" 70 borax, a little salt and water,
8"60 organic substances. On dissolving the saline mixture in hot
water and cooling, pure potassium nitrate crystallizes out.
The consumption of nitrate of soda during the years 1898 to
1901 has been as follows : —
1898. 1899. 1900. 1901.
Tons. Tons. Tons. Tons.
Continent of Europe (chiefly
Germany) 904,500 981,000 1,026,000 l.Oyc.OOO
United Kingdom 132,500 121,000 126,000 118.000
United States 125,000 133,000 170,0(X) 192,(H)0
Other countries 10,000 25,000 28,000 18,0(H>
Total... 1,178,000 1,200,000 1,350,000 1,304,000-
Price on 31 st Dec, per cwt. ... 7^*. 0</. 8*\ tW. Jls.
94 RAW MATERIALS OF MANUFACTURE.
(Statistics of previous years are given in the first two editions of
this book, pp. 80 and 81.)
When emptying nitrate of soda from the bags a certain quan-
tity of the salt, which is always damp, remains adhering to them ;
this not only causes loss, but renders them useless, and even pro-
duces a danger of fire. It is therefore well to lixiviate the bags with
hot water and to, dry them. The solution is evaporated to a small
bulk and crystallized. The mother-liquor from this operation is
always very rich in chlorides, which seems to show that the deli-
quescence of sodium nitrate is not a property of the pure salt, but
is owing to the magnesium and calcium chloride contained in it,
since the dampest salt will adhere to the bags. The washed and
dried bags should not be stored in quantity, as they are still very
inflammable.
Composition of Commercial Nitrate of Soda. — R. Wagner
(Jahresb. 1869, p. 248) found in commercial nitrate of soda : —
Sodium nitrate 9103
„ nitrite 0*31
„ chloride 1*52
Potassium chloride 0 64
Sodium sulphate 0*92
„ iodate 0-29
Magnesium chloride 0*93
Boric acid traces
Moisture 1*36
10000
The nitrate of soda imported into England, as used by vitriol-
makers, is much purer than the above sample. The English sellers
mostly guarantee a maximum of 5 per cent. " refraction '^ (that is,
the total percentage of all foreign constituents, inclusive of water) ,
frequently, however, 4 or even 3^ per cent, refraction. English
vitriol-makers would, indeed, altogether refuse nitrate containing
upwards of 3 per cent, of chlorides, like that analyzed by Wagner,
1 per cent, being the maximum allowed. The muriatic acid gene-
rated from them, of course, gives, with nitric acid, free chlorine
and its compounds with nitrogen oxides, and causes a loss of the
latter. The average composition of nitrate for chemical works is
COMPOSITION OP COMMERCIAL NITRATE OP SODA. 95
96 sodium nitrate (including nitrate, iodate, &c.),
0'5 chlorides (calculated as NaCl),
0*75 sulphates (calculated as NaSO^),
2" 75 moisture.
Gilbert (Zsch. f. angew. Chem. 1893, p. 495) pointed out that
the Chilian nitre always contains, and always has contained,
some potassium nitrate. He states that the percentage of KNO^^
rarely exceeds 5 per cent., and the deficiency of nitrogen caused
thereby is more than compensated by the value of the potassium
for agricultural purposes. The old method of testing for
" refraction " is obstinately adhered to by the producers, and is
preferred by Gilbert to the direct guarantee of 15*57 per cent,
nitrogen demanded by the agricultural control-stations. Jones
(loc. cit. p. 698) mentions that he had met with nitre containing
much more potash ; but this nitre, which is recovered from the
bilge-water of the carrying vessels, occurs only quite exceptionally
(comp. below). Most suphuric-acid manufacturers do not share
Gilbert's opinion : see below.
An impurity formerly entirely overlooked in commercial nitrate
of soda consists in perchiorate and perhaps also chlorate of
sodium.
Beckurts (Arch. d. Pharm. ccxxiv. p. 323 ; Fischer's Jahresb.
1886, p. 305) found in all descriptions of commercial nitre small
quantities of chlorates and perchlorates, and this has been con-
firmed from all sides. A large number of methods have been
devised for estimating the perchiorate (comp. Lunge's Chem.-
techn. Uutersuchuugsmethoden, i. p. 283 et seq,), all of which
are practically founded upon converting the perchiorate into
chloride, preferably by fusing the nitrate with lime or sodium
carbonate and manganese peroxide, estimating the chloride in the
ordinary way, and deducting the chloride previously existing in
the nitrate.
In order to manufacture nitrate of soda free from perchiorate,
which does not occur in the first crystallization, but accumulates in
the mother-liquors, and after using these three or four times over
contaminates the nitrate crystals up to 1 per cent., H. Foelsch
& Co. (G. P. 125,206) cools down the impure mother-liquors to
0°C., 1 cub. metre of which then furnishes 160 kil. of a mixture
9G RAW MATERIALS OF MANUFACTURE.
of salts, containing 150 NaNOg and NaClOj. The new mother-
liquor, when employed for redissolving crude nitre, at first
furnishes nitrate free from perchlorate.
Analysis of Nitrate of Soda.
In the laboratories working for the importers of nitrate of soda
the value of nitrate is mostly only estimated indirectly, viz. the
" refraction.'^ 10 grams are well dried in a porcelain capsule,
weighed again, dissolved, the residue (if any) is estimated, the
liquid dissolved to a certain volume, and in separate portions of the
liquid the chloride and sulphate are estimated in the usual way.
The sum total of moisture, insoluble residue, sodium chloride,
and sodium sulphate is called the *' refraction,^' and it is assumed
that the remainder is real sodium nitrate. This may, however,
lead to very erroneous results, where, for instance, the nitre
contains some potassium nitrate. A case in point has been
described by me in ^ Chem. Ind.' 1883, p. 369, where an error
amounting to 2 per cent, was caused in this way. Perchlorate
causes also errors; and altogether it stands to reason that the
consumer of nitrate receives justice only by a real determination
of the nitric acid contained in (or, more properly speaking, to be
evolved from) a sample of nitrate. On the other hand, the interests
of the importers and dealers in nitrate are quite the opposite.
According to Fischer's Jahresb. 1899, p. 407, the Hamburg im-
porters insist upon the '' indirect " (f. e. altogether deceptive)
analysis, and want the perchlorate to be counted as nitrate.
The direct analysis, i. e. the estimation of the NaNO^ (or its
equivalent of KNOa), can be carried out in many ways, a complete
enumeration of which is found in my ' Chemisch-technische
Untersuchungsmethoden,' i. p. 273 et seq. Among these, those
mostly used are the Schloesing-Grandeau and the Ulsch methods
(comp. below, in the section on Nitric Acid), but at sulphuric-acid
works the method nearly always employed is the '^ nitrometer'
method/' which is carried out as follows * : —
Dissolve a good-sized sample, say 20 to 30 grams, of the nitrate
* The nitrometer in its use for the analysis of nitrous vitriol will be described
in Chap. III.
ANALYSIS or NITRATE OF SODA.
97
in twice its weight of water, employing a flask of known contents
and heating very gently. Weigh out a quantity corresponding to
about 0-4 gram NaNOg in an ordinary weighing-glass or in a tube
with stopcock, as employed for testing fuming sulphuric acid
(fig. 8). Run its contents into a nitrometer, either a "bulb-
nitrometer,'' as shown in fig. 9, or, preferably, the nou-graduated
Fifr. 8.
Fig. 9.
agitating-vessel connected with a Lunge's gas-volumeter, to be
described in the next Chapter. In the latter case no observation
of temperature and barometric pressure nor any reduction tables
are required. Do not rinse the weighing-tube (which would
dilute the liquid too much), but weigh it back as it is. Decompose
VOL. I.
98 RAW MATERIALS OF MANUFACTURE.
the solution within the nitrometer with a sufficient quantity of
strong sulphuric acid and mercury, and measure the nitric oxide
given oflE as will be described in the next Chapter, where also a
table for reducing the readings to NaNOg will be given.
7. Nitric Acid, NOgH.*
This may be called one of the raw materials of vitriol- making,
although a manufactured product itself^ in those works using it
in lieu of solid nitrate of soda.
Nitric acid proper (the monohydrate) has the equivalent 6305
(0 = 16), and may be said to contain, as formerly expressed,
85*71 nitric anhydride (NgOa) and 14'29 water. Its specific
gravity is 1*54 at 20°, or 1*55 at 15°. It is colourless if perfectly
pure; but the strongest acid of commerce is always coloured
yellow, or even red, by a partial decomposition into oxygen and
nitrogen peroxide, N2O4 (hyponitric acid) . Its boiling-point is
86° C. On boiling an acid containing a little water, at first
strong acid distils over, till the boiling-point of the remainder has
reached 126°, at which point the thermometer remains stationary,
and an acid of constant composition for any certain pressure distils
over. Exactly the same point is reached from the opposite side
by distilling more dilute acids, in which case water distils over,
and the remaining acid becomes more and more concentrated, till
the above stationary point is reached. The acid at that point has
nearly the composition 2NO3H + 3H2O (corresponding to 6ON2O5
and 4OH2O) and a specific gravity of 1'42.
The following table shows the boiling-points of nitric acid of
various strengths : —
Spec. gray.
Boiling-point.
1 Spec. grav.
Boiling-point.
115
104° C.
1--13
123° C.
1-20
108
1 1-48
115
1-30
113
1-50
99
1-35
117
! 1-52
86
1-40
119
* Some interesting notes on the early history of manufacturing nitric acid in
England are contained in Guttmann's paper, Journ. Soc. Chem. Ind. 1901, p. 7.
The cost of producing 200 lbs. of acid spec. grav. 1'375 in 1771 was £8 2s. 2d.
without labour. See also W. P. Reid's notes, ibid. p. 8.
SPECIFIC GRAVITIES OF NITAIC ACID,
99
For the percentage of nitric acid for different specific gravities,
Kolb (Ball. Soe. Ind. de Mulhouse, 1866^ p. 412) has given a
table which is now rendered obsolete by the more accurate table
derived from the experiments of Longe and Rey (Zsch. angew. Ch.
1891, p. 165) . The specific gravities are taken at 15° C, referred to
water of 4P and reduced to the vacuum. They refer to chemically
pure nitric acid; commercial acid, containing nitrous acid, &c.,
contains less real HNOg at the same specific gravity.
Specific
graTities
Degrees
100 parU by weight
contain
1
1 litre contains kilog.
, 1
at?^°
Twacld.
ii
4°-
N,0,.
HNO3.
NA-
HNO,. 1
1000
0
0-08
010
0001
0001 '
1005
1
0-85 i 1-00
0008
0010 .
1010
2
1-62 1-90
0-016
0-019
i 1-015
3
2-39
280
0-024
00-28 1
1020
4
3-17
3-70
0033
0038 '
1-025
5
3-94
4-60
0O40
0047
1030
6
4-71
5-50
0049
0057
1035
7
5-47
6-38
0057
0066
1-040
8
6-22
7-26
0064
0075
1045
9
6-97
813
0073
0085
1-050
10
7-71
8-99
0081
0094
1055
11
8-43
9-84
0-089
0104
1-060
12
9-15
10-68
0-097
0-113
i 1065
13
9-87
11-51
0105
0123
1070
14
10-57
12 3:}
0-113
0132
^ 1075
15
11-27
1315
0-121
0-141
1080
16
\\m
13-95
0-129
0151
1-085
17
12m
14-74
0-137
0160 ;
1090
18
13-31
15-53
0-145
0169 ;
' 1095
19
1399
16-32
, 0-153
0179 !
1100
20
14-67
1711
j 0161
0-188
1105
21
15-34
17-89
0170
0198
l-llO
22
16-00
1867
0-177
0-207
1115
23
16-67 19-45
0-186
0-217
1120
24
17-34
20 23
0195
0-227
1-125
25
1800
2100
0-202
0236
1-130
26
18-66 ! 21-77
0-211
0-246 \
' 1135
27
19-32 ' 22-54
0 219
0256 \
M40
28
19-98 23-31
0-228
026(5
1 1-145
29
20-(^ ' 24-08
0-237
0-276 i
1-150
30
21-29 \ 24-84
0-245
0-286
1155
31
2194 25-60
0-2r>4
0-296
1160
32
22-(M) 2rr3<)
0-262
0-306
M65
33
23-25 2712
0-271
0316
1170
34
23-90 27-88
0279
0 32r>
1175
35
24-54 28-63
0-288
0-336 i
1180
36
25-18 2938
0-297
0347
1-185
37
25-83 3013
0-306
0 357
1
E
0
100
RAW MATERIALS OF MANUFACTURE.
Table (continued).
1
Specific
giiiTities
1
Degrees
100 i>arte bj weight
contain
1
1'
1 litre contains kilog.
atl^°
Twadd.
1
1
10.
; Sfi,. HXO3.
1
1190
38
26-47 30-88
1 0-315 0-367
1195
39
2710 31-62
' 0-324 0-378
1-200
40
27-74 32-36
0-333 i 0-388
1-205
41
283(5 33-09
0-342 0-399
1-210
42
28-99 ; 33-82
! 0-351 0-409
1-215
43
29-61 , 34-55
0-3(>0 0-4-20
1-2:^0
44
30-24 35-28
0369 0-430
1225
45
3088 3(v03
0378 0-441
1-230
46
31-53 36-78
0-387 0-452
1-235
47
3217 37-53
0-397 0-463
1-240
48
1 32-82 38-29
0-407 0475
1-245
49
33-47 3905
0417 0486
1-250
60
3413 39-82
0-427 0-498
1 1-255
51
34-78 40-58
0-437 0-509
' 1 2()0
52
35-44 41-34
0-447 0-521 ;
1-2(55
53
3609 4210
0-457 0-533
1-270
54
3(;-75 4287
1 0-467 0-544
1-275
55
37-41 43-64
0-477 0-566 ■
1-280
56
3807 44-41
0-487 0-508
1-285
57
38-73 4518
0-498 0-581
1-290
58
39-39 45-95
0-508 ' 0593
1-295
59
4005 4(»-72
0-519 0(505
1-300
00
4071 47-49
0-529 0617 1
1305
01
41-37 ; 48-26
0-540 0-630 '
1-310
62
4206
4907 '
0-551 ! 0-643 1
1-315
63
4276
49-89 1
0562 ' 0-656 '
1-320
64
43-47
50-71
0-573 0(569
1-325
65
44-17
51-53
0-585 0083
l-33(>
66
44-89
52 37
0-597 0-697
1-335
67
45-62
53-22
0-(MX) 0-710
1-340
68
4r.-35
54-07
0621
0-725
1-345
(i9
47 08
54-93
0(53;5
0-739
1-350
70
47-82
55-79
! 0-(>45
0-753 1
1-355
71
48-57
56-66
0-658
0-768 !
1-3G0 .
72
49-35
57-57
0671 \ 0-783
1-305
73
5013
58-48
0(584 0-798
1-370
74
50-91
59-39
0-698 0-814
1-375
75
51 (59 60 30
0-71 1 0 829
l-3^<0
76 1
52-52 61-27
0-725 084(5
1-385 ;
77
53-35 ! 62-24
0-739 , 0-8(52
1-390
78
54-20 63-23
0-753 0879
1-395
79
55-07 &4 25 J
0-7(58
0-896
1-400
80
5597 65-30
o-7as
0-914
1-405
81
5()92 66-40
0-8(X>
09;53
1-410
82
5786 ' 67-50 ,
0-816
0-952
1-415
83
58-83 j 68-63
0-832
0-9J1
1-420
84
59-83 ' 69-80
0-849
0-991
1425
85
60-84 ; 70-98
0-8(57
1011
1-430
m
61-86 72-17 '
0-885
1-032
1-435
87
62-91 73-39
0-903
1053 !
1-440 ,
88
6401 74^
1
0-921
1075 1
SPECIFIC GRAVITIES OF NITRIC ACID.
101
Table (continued).
Specific
grarities
Degrees
100 parts by weight,
contain
1
1 litre contains kilog.
at ^P-
Twadd.
40-
N,0,.
HNOj.
NjO,. 1 HNO3.
1-445
89
65-13
75-98
0941 , 1098
1450
90
66-24 1 77-28
0-961 1 1-121
i 1 455
91
67-^8 78-60
0-981 1144
1-460
92
(i8-56 79-JI8
1001 1 168
1-465
93
(;9-79 , 81-42
1023 1-193
' 1-470
94
7106 82-90
1-(H5 1-219
1-475
95
72-39 84-45
1-068
1-246
1-480
96
73-76 86a')
1092
1274
1-485
97
7518 87 70
1116 1-302
1-490
98
7680 89-60
1-144 1-335
1-495
99
78-52 9160
1174 1-369
1-500
100
80-65 94-09
1-210 1-411
1501
—
8109 94-60
1-217
1-420
1-502
—
81-50 9508
l-:>>4
1-428
1-503
—
81-91 ; 95-.55 i
1-231 1-436
1-504
—
82-29 96-00
1-238
1-444
1-505
101
82-63 96-39
1-244
1-451
1-506
—
82-94 9676 |
1249
1-457
1-507
83-26 97-13
1-255
1-464
1-508
~^
83-58 97-50
i-2<;o
1-470
1509
83-87 97-84
1-265
1-476
1-510
102
8409 9810
1-270
1-481
1-511
84-28 98-32 i
1-274
1486
1-512
84-46 98-53
1-277
1-490
1-513
84-63 98-73
1-280
1-494
' 1-514
—
84-78 98-90
1-283
1-497
1-515
103
84-92 J»9 07
1-287
1-501
1-516
—
ar04 99-21
1-1^89
1-504
1-517
85-15 99-34
1-292
1-507
1-518
—
8^3-26 99-46
1-294
1-510
1-519
__
85-35 99-57
1-296
1-512
1-520
104
85-44
9967 1
1-299
1
1-515
Correction of the observed specific gravities for temperatures
a Uttle above or below 15° C.
With spec. grav. between 1*000 — 1
1-021—1
1041—1
1071—1
1-101-1
1131—1
9}
)*
if
)f
7f
i>
'>
»)
}>
)i
Add for -1°C.
Deduct for +1°C.
•020 00001
•040 00002
•070 0-0003
•100 00004
•130 0-0005
•160 0-0006
102
RAW MATERIALS OF MANUFACTURE.
Correction of the observed specific gravities (continued).
With spec. grav.
between
1-161— 1-200
0-0007
1-201— 1-245
00008
1-246-1-280
00009
1-281 1-310
00010
1-311—1-350
0-0011
1-351 1-365
00012
1-366 1-400
00013
1-401—1-435
0 0014
1-436— 1-490
0 0015
1-491— 1-500
0-0016
1-501 1-520
0-0017
Loring Jackson and Wing, and a little later on R. Hirsch
(Chem. Zeit. 1888, p. 911), have shown that the presence of
lower oxides of nitrogen in nitric acid has a considerable influence
on its specific gravity. Thus the first runnings from a distillation
possessed a specific gravity =1*62, but contained 12 percent,
by weight of HNO2. Hirsch assumes (without strict proof) that
each per cent, of HNO2 raises the specific gravity by 0 01. If
this is correct, an acid of sp. grav. 1*44, but containing 1 per
cent. HNO2, really contains only 99 per cent, of HNOj, of spec,
grav, 1*43. Now 100 grams of pure acid of 1*44 are = 74'4
grams HNO3, but 99 grams of 1"43 only =710; hence the 1 per
cent, of HNO2 present makes a difterence of 3*4 per cent, of HNOg
in the real strength^ compared with the apparent strength as taken
from the specific-gravity tables.
This subject has been more accurately investigated by Lunge
and Marchlewski. From their paper (Zsch. f. angew. Ch. 1892,
p. 10) I give the following table, showing this influence in the
150
case of nitric acid of spec. grav. 1*4960 (at —75-) : —
4^
PROPERTIES OF NITRIC ACID.
103
If A 1
per cent.
Alteration of
per (jent.
1
Alteration of ,
spec. grav. by
N,0,
spec. grav. by
I 0-25
0-00050
6*55
004475
0-50
0-00075 ,
70{)
0-04650
075 !
000150 1
7-25
004720
I-OO
0-00300
750
005000 '
i 1-25
0-00475
775
005165
1 1-50
000675
8-00
005325 '
t 175
000775
8-25
005500
( 200 ;
0-01050
8-50
0-05660
2-25
001250
8-75
0058-25
2-50
0-01425
900
006'-)00
2-75
001625
9-25
006160
300
0-01800
9-50
0-06325
325
001985
9-75
0 06500
3oO •
002165
10-00
006ii00
3-75
002350
10-25
006815
4-00
002525
10-50
006976
4-25
002690
10 75
0-07135
' 4-50
002875
1100
007300
4*75
0 03050
11-25
007450
500
0-03225
11-50
007600
1 5-25
003365
1175
0-07750
' 5-50
003600
12-00
0078nO
575
003775
1225
008050
; 600
003950
12-50
0-08200
6-25
0-04175
12-75
0-083r>()
6-50
0(>i300
Saposehnikoff (Chem. Centralbl. 190Q, ii. p. 708, and 1901, ii.
p. 1330) has studied the conditions of equilibrium between HNOj^,
HNO2, and NO, i. e. the formation of HNO., and NO from HNO2,
and the reduction of HNO« by NO to HNOo.
The oxidizing properties of nitric acid are well known and can-
not be described in detail here ; but it should be mentioned that
an acid containing the lower oxides of nitrogen, such as the '^ red
fuming nitric acid/^ has even more strongljr oxidizing properties
than the pure acid, and this helps to explain some points in the
theory of the formation of sulphuric acid, as we shall see in a
subsequent chapter.
Manufacture of Nitric Acid. — This acid has been known since
the times of Geber, in the eighth century; and Raymundus
Lullus in 1225 taught how to prepare it by distilling a mixture of
clay and saltpetre. Now-a-days it is always made by distilling
nitrate of soda with sulphuric acid, an excess of the latter acid
beyond the theoretical quantity being used in practice. By the
104 RAW MATERIALS OF MANUFACTURE.
equation : 2NaN0jj + H2SO4 = 2HN08 + Na2S04, in theory 85 parts
of NaNOg require 49 parts of SO4H2, and yield BSNOaH along with
71Na2S04 ; this comes to the same as 57*6 parts of SO4HS, or,
say, 60 parts of ordinary strong oil of vitriol (with 95 per cent, of
SO4H2) to 100 parts of 95-per-cent. nitrate. If these proportions
are used, a portion of the nitric acid is always decomposed inio O
and N2O4, and red fuming acid is obtained. This arises from the
fact that the above equation is only realized at a high temperature,
whilst at less elevated temperatures sodium bisulphate is formed :
NaNOj + H2SO4 = HNO3 = NaH SO4. Moreover, the presence of
strong sulphuric acid, owing to its attraction for the elements of
water, has a tendency to split up some HNO3 into H2O and N2O5,
the latter compound being at once decomposed into N2O4 and O.
To avoid the loss involved in this operation, generally more dilute
nitric is produced by employing weaker sulphuric acid, say of
14<y*-148° Tw., and more than the theoretical quantity of it, gene-
rally from 20 to 30 per cent, in excess of the simple equivalent.
In this case the admixture of a certain quantity of sodium bi-
sulphate makes the residue of distillation much more easily fusible
and facilitates its removal from the retorts. When the acid is made
at factories where salt is decomposed, even more sulphuric acid
than the above is generally employed, as the excess is not lost, the
residual " cylinder-cake " or '* nitre-cake " being regularly mixed
with the salt to be decomposed in the sulphate-pans ; in this case
as much sulphuric acid is saved as corresponds to the bisulphate
contained in the cylinder-cake.
Nitric acid was formerly made in glass retorts, which are prac-
tically obsolete now, cast-iron retorts being universally employed.
The retorts belong mainly to two different types — horizontal
cylinders, charged sideways, and pots or stills, charged from
the top.
The ordinary type of French cylinder apparatus, used in many
places outside of France as well, is represented in figs. 10 and 11
on a scale of 1 : 25.
The ends of the cylinders are here exposed to the air and con-
sist of cast-iron disks, 1^ inch thick, cemented into the rebates
cast on to the ends of the cylinder by the usual rust cement (100
iron filings, 5 flowers of sulphur, 5 sal-ammoniac), or by a mixture
of this with ground fire-bricks and the like. At all events the
back end is fixed in this way, and is provided with a pipe for
BAW MATERIALS OF MANUFACTURE.
Fig. 11.
taking away tbe vapours ; the other end is made to take off and'
serves for charging the nitre and discharging the residue. A hole
and S-shaped funnel in the man-lid allow of running in the sul-
phuric acid. These cylinder-ends cause a good deal of cooling,and
consequently a loss of fuel, which can be avoided to a great
extent if they are not made of iron but of a single stone flag each ;
the charging end also remains fixed in this case, the nitre being
charged through a small general opening in it, and the residue
being run off in a liquid state through a pipe generally closed by a
ground-in iron stopper. A little more sulphuric acid is employed'
in this case, so as to obtain a very fluid mass at the end. If
sulphuric acid of 144° Tw, is employed, the receivers will contain'
nitric acid of 77° to 82° Tw.; if weaker acid is desired, a little
water is put into the receivers. The strongest acid, of 100° Tw,„
can only be made from strong I'itriol and dried nitrate.
CYLINDERS FOR NITRIC ACID. 107
At first the cjliaders are fired rather strongly; but as soon as
the first receivers get warm the fire is slackened, and during 18
hours is kept so that of eight receivers only the firsffive are warm
to the touch. If the heat gets up too high, the coutents of the
retort may boil over, and far more ruddy vapours will be formed.
The end of the reaction is known by the cooling of the receivers ;
then the fire is increased again for a little time, and at last is
allowed to go down.
In regular work, there are some red vapours at first, but much
less with rather weaker acid (say 135° Tw.) than with stronger
acid. At the end the red or yellow vapours appear again and last
np to the finish.
The cylinders are sometimes cast so that their upper half can be
protected against the attack of the acid by lining it with acid-proof
bricks (see fig. 12) ; but some believe that this does more harm
Fig. 1-2.
than good, as the upper part of the metal cylinder is all the less
acted upon by the nitric acid the hotter it becomes.
A more perfect form of cylinders is that employed at the
Griesheim works, aud shown in full detail iu figs. 13 to 16 with
all the measurements marked in centimetres *.
These retorts work off 8 cwt. of nitre per 24 hours, with an
expenditure of 2i cwt. of coal or very little more, inclusive of the
time of filling and emptying. In case of need, 25 per cent, more
can be charged, without fear of boiling over, with careful work.
• These fipires are taken from HaussermaQn's article " Nitric Acid, ' in
Huepratt-Bunte'B Enc, d. Teclin. Chemie, 4tU ed. vii, p. 051 et »eq. Some notes
from thiB article are also given in tbe following piiges.
RAW UATERIALB OF UANTTPACTITHE.
GRIKSHEIM RSTORT8 POK NITRIC ACID.
109
III
I ii
III
III
III
o
o
>
•*^
G
O
P
o
o
c
.2
'■+3
u
RAW MATERIALS OF MANUFACTURE.
Fig. 10.
tjectional plan.
Such boiliug over occurs more easily with stroug sulpliuric acid
(of 1'84 spec, grav.) than with weaker acid (140° Tw.). The setting
is so arranged that the tire-gases entirely surround the cylindrical
part, and the ends are made of sandstone slabs, cemented and
kept in their places by iron bars pressing against them on the
flutside. In this kind of apparatus no corrosion takes place,
wherefore the arrangement shown in fig, 12 (p. 107) is quite
unnecessary. The exit opening continues into a glass tube,
forming connection with the condensing-apparatus. The arrange-
ment for running off the nitre-cake is made quite clear in fig. 13.
The objection sometimes made to cylindrical retorts, that they
do not easily permit the manufacture of the strongest nitric aciil,
is refuted by the experience of many years at the various works
belonging to the Grieaheim Company.
Vol. I. Sidiihuric Aad.
STILLS FOR NITRIC ACID. Ill
The other class of retorts, taking the shape of pots or stills, is
either entirely surrounded by the fire^ or with its top exposed to
the air. The former kiiid, as employed in France for very strong
nitric acid, is shown in fig. 17. It consists of cast-iron pots of
from 4 to 5 feet diameter and equal height, and a metal thickness
of 1 J to 2 inches (rather thicker at the bottom). There is a wide
neck a at the top, closed by a lid, fastened on with a suitable
cement, e.ff. a mixture of clay and gypsum. There is a tube b
for carrying away the gas, either coming out perpendicularly at
the top, as shown here, or bending away horizontally, as in a
laboratory retort ; this tube should be lined with an earthenware
tube as far as it is at all liable to cool down below the point
where the metal can be acted upon by nitric acid. Another
tube c serves for introducing the sulphuric acid ; this is preferable
to running it in through the neck a, after charging the nitre.
Sometimes there is no pipe provided for running out the liquid
residue, but it is better to have one, as shown at d, and to protect
this tube against the direct action of the fire by a fire-proof
covering. Where this is not provided, the liquid residue must
be ladled out through a at the close of the operation, which is a
disagreeable proceeding. The pot is set in a furnace so that
it is altogether surrounded by the fire, even at the top, to which
access is afforded for charging by lifting off a fire-clay slab, or
a metal plate filled with ashes, as shown at e. By this means a
saving of fuel is effected, and the equal heating of the retort
causes it to stand the action of the nitric acid very well. A
pot 5 feet wide and 5 feet high takes a charge of 9 cwt. of nitre,
and requires from 16 to 18 hours to work this off, including
the time for charging and emptying. Of this time, one hour
may be reckoned for charging and making the joints, six hours
for the first stage, where red fumes appear, about as much for
the second stage, where the vapours are almost colourless, and
four or five hours for the last stage, where the temperature has to
be raised much higher, the yellow and red vapours appear again
and more water comes over with the acid.
That kind of pot where the top (with the small hole) is exposed
to the air is shown on a scale of 1 : 60 in fig. 18. It forms part
of the Valentiner vacuum system to be described below. The pot,
for a charge of a ton of nitre, is composed of two parts. The
outlet-pipe for the nitre-cake runs almost horizontally, so that it
112 KAW UlTEKIALS OF MANUFACTURE.
is easily cleaned. The gas-pipe first eaters a amall receiver,
destined to keep back any froth caiTied over. Such a vacuum
retort is finished in 12 hours, with about 4 cwt. of coal per tou
of nitre.
4 rOU MTKIC ACID.
114 RAW HATEBIALS OP UANUPACTUBE.
Figs. 19, 20, and 21, from drawings which I owe to the
kindness of Mr. H. H. Niedenfiihr, show the kind of still
employed by Mr. O. Guttmaun in couDexioii with his coudensiiig-
plant, but of course applicable in every other case. The drawings
are to scale and are clear enough without further explanations.
The brick setting, as arranged by Mr. Niedenfuhr, has had an
excellent effect, and admits of huishiug a charge of 13 cwt. in
24 boura with only 17 per cent, of coal. The internal flanges
have the effect of diverting the contents of the pot, when
frothing, towards the centre, and of preventing any acid con-
densing from running down the sides. These retorts have been
found specially useful for making a maximum of strong acid.
Usually it is preferred to cast such conical pots in one piece
(like caustic-soila pots), and to set them so that the cover projects
over the brickwork. In order to kee|) the cover sufficiently hot, it is
KETOKTS FOB XlTKIt ACIU. 115
covered by a layer of ashes or by bricks ; the drying of the nitre
Id this case is equally performed by the waste heat of the fire, but
in a separate place at the end of the retort-bench.
At some works they use large semicylindrical trougha of cast
metal with broad flanges aad a vertical rim all round, in order to
cover them by a brick arch or a stone slab. At others they employ
FiK- 21.
lai^c pots with rounded bottom, lying on their side : the open
end, which forms part of the front of the furnaces, being closed
by a stone slab. Neither of these forms has found very much
favour elsewhere.
Prentice (Engi. pat. 6960, 1893) carries on the process in a
continuous manner. The nitre is mixed with sulphuric acid
outside the retort, preferably in a kind of mortar-mill (according
to patent No. 8902, 1893, a lai'ge excess of sulphuric acid should
be employed and the residue subsequently used for the manu-
116 KAW MATERIALS OF MANUFACTURE.
facture ot superphosphate) . The mud thus produced is conveyed
iuto a heated chamber in which the nitrate dissolves in the sul-
phuric acid^ but no nitric acid is as yet split off [?] ; here chlorine
and nitrogen tetroxide are liberated and escape into a condensing-
apparatus. Owing to this the nitric acid is later on free from
volatile impurities [?]. The mixture is now charged into the
retort^ which has an oblong section and is divided into a number
of separate chambers by partitions stai*ting from the cover and
not reaching entirely to the bottom. Above each chamber the
cover is provided with a head and vapour-pipe. The retort is
heated from below, and the fire-gas subsequently travels round the
sides. The mixture coming from the heating-chamber enters the
first chamber of the retort, where it begins to boil and gives off a
large quantity of strong nitric acid mixed with a little nitrogen
tetroxide. The mixture now, without ceasing to boil, flows on from
chamber to chamber, giving oft' pure nitric acid, which gets weaker
and weaker in the following chambers. In the last chamber, placed
immediately above the fire, the temperature is highest ; here the
last nitric acid is driven off, together with much water and a little
sulphuric acid, and the nitre-cake is drawn off, free from the nitrate.
By this method § of the fuel and f of the condensing-plant [?J
can be saved. A still producing 4 tons per week weighs less than
two tons. — A communication by Prentice in the ^ Journal of the
Society of Chemical Industry,^ 1894, p. 323, gives nothing new.
In discussing it doubts were expressed as to whether that process
is applicable at such works where the residue cannot be utilized
for the manufacture of superphosphate, as it contains a large
excess of acid. Indeed, this extra quantity of acid required for
working that otherwise very ingenious process has made it an
economical failure.
Uebel (patents of the Chemische Fabrik Ehenania, E. P. 27240,
1 898, and 3305, 1901) proceeds in a novel way for the manufacture
of nitric acid. The nitre-cake, running out at a high temperature
from the retorts, is mixed with sulphuric acid of about 1*71 spec,
grav, in such proportions that a '^ polysulphate '* is formed, the
water being evaporated by the heat remaining in the nitre-cake.
This polysulphate, of a composition =NaH3(S04) 2, serves instead
of fresh sulphuric acid for manufacturing the next batch of nitric
acid, in which case the latter is obtained of the highest strength,
just as if the strongest sulphuric acid had been employed.
uebel's process for nitric acid, 117
Of course the fresh nitre-cake formed from the second operation
amounts to much more than that from the first ; therefore a portion
of it is set aside in the ordinary manner^ and the remainder is again
employed for making nitric acid. The practical work carried out
with this process has led to the construction of a new and original
style of retorts, which may be very usefully adopted even when
nitrate of soda is to be decomposed in the ordinary way by
sulphuric acid, not by " polysulphate.''
The principal advantages of the new process are : that the
alternate heating and cooling of the apparatus is avoided, which
produces a saving in fuel and prolonged duration of the plant ;
that there is not the same inequality of the nitric acid as is pro-
duced during the period of working oflF a retort on the old system ;
that a much smaller and consequently much cheaper plant and
only half of the ordinary ground-space are required for the same
output, both for distillation and condensation (the Uebel plant
may be combined with any of the systems of condensation to be
described below) ; that there is less labour -required ; and that the
strongest acid can be made by means of ordinary sulphuric acid
140° Tw., as its water is evaporated free of expense in the formation
of the polysulphate — ^' waste-acid '^ from nitrating processes being
equally fit for employment.
The Uebel process is illustrated in figs. 22 to 24. Two parallel
retorts A, A' (consisting of a cast-iron bottom piece and a stone or
stoneware cover) are alternately charged with nitrate of soda and
the requisite quantity of liquid '^ polysulphate'' (or else, if there
is a sufficient quantity of strong acid at disposal, e.g. where
manufacturing by contact processes, directly with such acid). The
heating is performed by the waste fire-gases of the lower retort B,
a temperature of 170° to 200° C. being thus attained in A and A'.
When most of the nitric acid has beeu expelled, the contents of
A and A' are let down into retort B, where the last portion
of nitric acid is liberated, at a temperature of about 300° C. The
contents of B have, of course, been previously run out, but never
completely, leaving always a stock of hot liquid " bisulphate " in
the retort. The bisulphate (nitre-cake) is run from B into the
cast-iron pan C, where it is mixed with previously heated sulphuric
acid of about 140° Tw. (or with waste acid from nitrating pro-
cesses of equal strength) in the proportion of forming a tetrasul-
phate, H2SO4, NaHSOi. In consequence of the high temperature
KfiVi MATERIALS OF MANDFACTfKK.
Fig. -22.
UEBEL^S PROCESS FOR XITRIC ACID.
119
of the liquid nitre-cake, the water present in the fresh sulphuric
acid is evaporated (together with any nitric acid present in waste
acids, if such are employed) and is carried away by a pipe, or a
vapour hood, not shown in the drawing. Half of the '' poly-
sulphate " formed in C is employed for a fresh operation in A, A',
%m..
wp..
"t^-^t-i—
Scale 1 : 8(i.
\ 0,5 0
I i 1 1 I I ' ' ■ ' t
ftm
the other half beiug at disposal for any outside utilization in lieu
of highly concentrated acid. If such utilization is not possible,
half of the nitre-cake is run out from C before adding fresh
sulphuric acid and only the other half is employed for making
^' polysulphate," the unemployed half being utilized in the manner
followed in other cases.
(An English patent by Claes, No. 1072, 1900, describes the
manufacture of such ^^ polysulphates '^ from nitre-cake and sul-
phuric acid as a commercial article.)
The flame of the hearth F first heats retort B, which is pro-
tected by an arch, and then, according to the position of the
damper, either A or A', v^v^ are valves for discharging the
contents of A, A' into B ; v'^ another valve for discharging B into
C, placed high enough to leave a stock of liquid nitre-cake in B.
After heating-up, A is charged with 8 cwt. nitrate of soda,
previously dried on the shallow basins T, T' on the top of the
furnace. The charging is performed through manholes H, H^
which are afterwards closed by covers and kept tight by nitrate
120 RAW MATERIALS OF MANUPACTURB.
of soda heaped upon them. Then fused polysulphate is gradually
run in through a swan-neck pipe from the tilting-box K, pre-
viously lifted to the top of the furnace. This polysulphate will
be at a temperature of about 120° or 130°, and must therefore
be gradually introduced within from half an hour to an hour, so
as to avoid a tumultuous evolution of nitric acid. Of course, in
lieu of this, fresh acid can be employed, as stated above. During
the introduction of the hot polysulphate damper z is closed and
z' is opened ; when finished, z is opened so that the fire-gases play
upon A. When the temperature has risen to about 170°, nearly
all HNO3 has been driven out; the thin, liquid melt contains a
little nitrate and all the water introduced with the nitrate and the
polysulphate or fresh acid. It is now slowly run through valve v
into retort B, where it meets with a stock of hot bisulphate.
Here, without the frothing taking place in the old, directly fired,-
retorts, the last nitric acid and the water is drawn off within a
very short time. The contents of B remain there until the con-
tents of A' (which has been charged and worked in the mean-
time) are ready to be run oflF, which takes from 3^ to 4 hours.
Then B is discharged into C, where fresh acid is run in as before
described, so that room is made in B for receiving the contents of
A'. A in the meantime has been left empty for an hour, damper
z being closed, in order to cool down. Now the polysulphate is
run out of C into the tilting-box K, which is hoisted up to serve
as before. Thus a charge of 8 cwt. of nitrate can be worked oft*
in each of the retorts A and A' every four hours.
A good cement for permanent joints against nitric acid in the
retorts consists of 10 parts powdered volvic lava, 7 iron filings, 7
powdered brimstone, 7 fire-clay, 10 ground fire-bricks, moistened
with as little water as possible. If carefully stemmed into the
joint, it becomes very hard.
Cement for the earthenware pipes, &c., in the condensing-
apparatus can be made from 5 parts hot linseed-oil, 2 parts
brimstone, 2 parts india-rubber scraps, and enough sulphate of
baryta to produce a thick paste which is employed in the hot state.
The most universally employed cement for nitric acid (and other
acids) is asbestos cement, Guttmann gives the following pre-
scription for preparing it: — 1 part silicate of soda is dissolved in
19 parts water [or, which is preferable, the usual 33 percent,
solution is diluted with 8 times its volume of water] and asbestos
liowder is kneaded with it in small quantities at a time, until a
Vi^POURS FROM NITRK-ACID RETORTS. 121
toagh paste is obtained. Should a cement be required whidi will
become very hard^ a little finely powdered barium sulphate is mixed
with it ; but this is not desirable where pipes have to be changed^
as the cement cannot be removed without risk of damaging the
pottery. If too little water is used from the first, the water
absorbed by the cement will cause it to swell and to burst the
sockets. The cement is applied in small bits, and tightly pressed in
by means of a piece of wood with a blunt-ended tool. The sur-
face is flattened and moistened with some silicate of soda solution.
If the putty in the sockets shrinks, about § incli of it is raked out
and fresh cement put in.
Asbestos cement must never be treated with pure water to
begin with, but with acid or acid fumes, in order to ensure its
setting by the separation of silica.
The gases and vapours evolved during the working of the retorts,
apart from the vapour of nitric acid itself, consist of aqueous
vapour, nitrogen peroxide, hydrogen chloride (which is, however,
almost entirely converted into the following gases), chlorine,
nitrosyl chloride, and a little iodine. HCl and the gases derived
from it (NOCl and CI) are principally formed at the commencement,
through the decomposition of sodium chloride ; but towards the
end they appear again owing to the decomposition of perchlorate.
All of these impurities are dissolved by cold nitric acid, and hence
occur in ordinary, " unbleached " nitric acid.
Volney (Journ. Amer. Chem. Soc. 1891, xiii. p. 246) showed
that the frothing in the ordinary nitric-acid process takes place
only in the last stage, when the strong acid has passed over and
when the last portion of the nitre is suddenly acted upon, with
formation of weaker acid. At this period practically only
NaHSOl is ^^ the retort besides NaNOg.
Later on (ibid. 1901, p. 489) Volney returned to this subject,
with the following results. During the first period of the process,
when the temperature of the retort is not above 100°, the com-
pound NaHa (804)3 is formed, which Volney calls '^ trisulphate ^^
(it is really a tetrasulphate and identical with UebeFs ^' poly-
sulphate,'' p. 117), by the reaction :
NaNO3"f2H2SOj=NaH3(SO.02 + HNO3.
The second period sets in at 100° and ends at 121° ; here the
principal reaction is :
NaN03 + NaH3(S04)2=2NaHS04+ HNO,.
122 RAW MATERIALS OF MANUFACTURE.
During the first period pure nitric acid^ boiliug between 81^ and
88°, distils over. During the second period the HNOs begins to
decompose into lower oxides and HgO, so that more dilute acid
must now be formed.
Later again (Journ. Soc. Chem. Ind. 1891, p. 544; comp. also
Journ. Amer. Chem. Soc. 1902, p. 226), Volney states that the
first phase, where acid of 77°-90° boiling-point comes over, takes
place at a temperature of 97^-122° ; the second phase, where acid
ofspec. grav. 1*505, boiling at 94°-100°, comes over, requires a
temperature in the retort of 130°-165°. Above this the decom-
position of HNO3 sets in and yellow acid comes over. All this
refers to working under ordinary pressures. When, however,
working at reduced pressure on the Valentiner system (originally
proposed by me, comp. infra), say at 300 mm., the first stage
runs from 55^-100°, the boiling-point of the strongest acid at that
pressure being only 45°-75° ; the second stage, where the poly-
sulphate acts upon more nitrate, begins at 100^ and is finished
at 120°, instead of at 165° at ordinary pressure.
[In considering Volney's results, we must not overlook that
they were obtained in glass vessels on a small scale. On the
manufacturing scale, and working in iron vessels, higher tempera-
tures cannot be avoided and the formation o£ nitrous vapours is
proportionately increased. The frothing at the last stage of the
distillation is generally ascribed to the partial decomposition of
sodium bisulphate into pyrosulphate and water: 2NaHS04=
NagSA + HjO.]
In a subsequent paper (Journ. Soc. Chem. Ind. 1901, p. 1189)
Volney gives the results of working at a much stronger vacuum,
viz., 110 mm. absolute pressure. During the first phase, at about
74°, free sulphuric acid acted on the nitrate \ during the second
phase, at about 85°, the polysulphate did the principal action.
The outside temperature rose to 140°. The distilling nitric
monohydrate boiled at 30°. With concentrated sulphuric acid
there is a great amount of frothing, which does not take place with
acid of 60° B. When working with such acid at ordinary
pressure, the greatest portion of the nitric acid distils at about
118° without frothing; at a pressure of 110 mm. at 74°, with
slight frothing. In both cases the remaining salt-cake consisted
of sodium bisulphate and water of crystallization. The nitric acid
produced showed in both cases 1*38 spec. grav. at 21° C. ; to
CONDENSATION OF NITRIC ACI1>. * 123
produce stronger acid^ it must be redistilled with concentrated
sulphuric acid.
The condensation of nitric acid is^ properly speakings only a
cooling-process, but it is preferably carried on in such manner
that^ in the first instance, stronger and weaker acid, coming over
at different stages, are separately collected, and that, if possible,
the above-mentioned impurities (p. 121) are kept out of the acid
and are separately treated. For many years, which now appears
strange to us, in this distilling and condensing process only air-
cooling was employed, and it is only during the last ten or fifteen
years that the much more efficient system of water-cooling, which
is so universal in other distilling processes, has been applied for
nitric acid, and that with complete success.
The old condensing-plant for nitric acid consisted entirely of a
series of earthenware receivers (Woulfe's bottles), combined in
sets of six to twenty and more, according to the size of the retort.
Sometimes two or even more of these jars are superposed over one
another, in order to increase the cooling-surface. According to
the strength of acid intended to be made, they are either left
empty or charged with a little water, as already mentioned.
Sometimes they are cooled on the outside with water ; but generally
this is not the case. Each of the jars is provided with a bottom
tap for running off the condensed acid; that from the first two
receivers is more impure than that of the others, as it contains a
little sulphuric acid and iron carried over from the retort. The
acid becomes weaker the further the receivers are from the retort.
Since the receivers now and then crack with the heat, it is
advisable to put them on stoneware saucers provided with a spout,
for collecting the acid running out.
The cement used for joining the receivers with the pipes, &c.,
has been described on p. 120.
This style of condensing-plant is still found at many, especially
smaller, works ; but it must be considered quite out of date, as its
cooling action is too imperfect, as the great number of joints is
troublesome to keep tight, as the receivers, on breaking, cause great
loss and danger to health and life, and as there is always some loss
of uncondensed vapours.
A minor improvement, which can be made, consists in inter-
posing between the retorts and the receivers a somewhat long
glass tube, or rather a number of air-cooled glass pipes of slightly
124
RAW MATERIALS OP MANUFACTURE.
conical shape, as shown in fig. 25, with sufficient fall for the
condensed acids not to stop in the tubes and run out of the joints.
Fig. 26.
With a length of from 10 to 13 feet (the longer the better) the
cooling by air is already very efficient. The single tubes are put
loosely together, without any cement.
Gobel proposed (Dingl. Journ. ccxx. p. 241), to my knowledge
for the first time, the system of cooling such pipes by water,
employing a long glass tube placed in a trough. This plan must
have led to frequent breakages and was probably soon abandoned
for that reason, but in principle it was perfectly correct.
Much more durable, as well as efficient, are water-cooled
stoneware cooling-worms, which do away with the necessity of
employing a large number of receivers.
These coils or worms were first manu-
factured by Messrs. Doulton & Watts, of
Lambeth, as shown in fig. 26, but are now
made by many other English and German
firms, in various shapes, and are compara-
tively very durable; good worms stand 300
operations and upwards. They admit of
fractionating the products, but are mostly
employed when acid of medium strength
(up to 82° T\v. ) is to be made.
It is nowadays considered indispensable to employ some
contrivance for depriving the gas issuing from the condensers of
not condensable oxides (NOo and NO) . Where the acid is made
at a sulphuricacid factory, we may employ a small Gay-Lussac
absorbing-tower, consisting of a stack of stoneware pipes filled
with coke and continually fed with sulphuric acid of 144° to 152°
Tw. The gas enters at the bottom and issues at the top, and
thereby gives up all its nitrous and hyponitric acid to the vitriol,
which arrives at the bottom as a more or less rich " nitrous
vitriol,^' and can be employed in the manufacture of sulphuric
acid. For a ton of nitre from 3i to 4 cwt. of vitriol are required ;
and nitrogen acids corresponding to 3 to 5 per cent, of nitric acid
a
9
m
t
r
8
B
i
r
I.
r
a
r.
t
a
r
h
e
e
(I
e
0
<
<
I. Stdphurie Acid.
o
f
e
h
f.
s
c
1
fi
1
n
fi
ti
o
fi
ei
{'
C(
11 <
a1
al
w
T
tli
vi
ac
ai
f
^-^^'^•■^'^-■^
kV\ .\ CV ,\ ,\s\ v\Us ^^^\^\^
CONDKNSATION OF MTKIC ACID. 12'J
of 1"33 spec. grav. are absorbed by the same^ more or less,
according to the percentage of chlorides in the nitre.
For such manufactures of nitric acid as are not in connexion
with sulphuric-acid works, the coke-tower can be fed with water.
If care is taken that an excess of air is always present, not
merely the last remnant of nitric acid is condensed, but the lower
nitrogen oxides are likewise converted into nitric acid and thus
saved.
Farpreferable to coke-towers for both the just-mentioned purposes
are the " plate-towers " constructed by the Author, in conjunction
with L. Rohrmann, and frequently called " Lunge towers.*' These
will be described in detail further on (in Chapter YI.; ; in this
place we will only point out that the use of coke is irrational,
because it destroys nitric acid, converting it into nitrous acids
and lower oxides. The plate-towers, of which hundi*eds are in use
at nitric-acid works, explosive works, &c., for the above-mentioned
purpose, are made of the best fire- and acid-resisting stoneware,
and, as will be seen hereafter, they are so constructed as to ofler
the most intimate possible contact between the gases and liquids.
Fig. 27 shows the arrangement as suj^plied by the Krauschwitzer
Thonwaarenfabrik near Muskau, Germany, for the condensation
of the last fumes, or the recovery of nitric acid from lower oxides.
A B C is the stoneware shell of the tower, D the cover with
distributing arrangement, E the plates, F the anmiUir bearers,
G drawing-off tap with hydraulic lute, H exit-tube, J sight for
observing the condensation, K recfeiver, M injector for drawing-
air through the neck N, in connexion with the last receiver O.
Figs. 28 & 29 show how a cooling- worm can be combined with
other pieces to form a complete condensing-apparutus. The
vapours, coming from two retorts charged with 6 cwt. of nitrate
of soda each, first pass through receivers a a for impure acid
carried over, which is run oft* into the carboy b, then through the
cooling- worms c c, discharging their contents of j>ui'c acid into
carboys d rf, then into a collecting-receiver e and into a '^ plate-
tower ^* / for recovering nitric acid from N2O4, and finally into a
last receiver g and injector h. The pots i i, equally discharging
their vapours into e, serve for refining (bleaching) the yellow
nitric acid (see below).
Such apparatus is especially needed for recoveiing huge
quantities of nitrogen oxides, given oft* in the treatment of wunIc
KAW MATKHI.^I^ or MANrFACTURE.
[
receiver^ is run into V3 or V4, according to its strength, that
condensing in receiver ff always into V^.
The refining or bleaching of nitric acid consists in drivin*^ off
the lower oxides, so that the acid becomes colourless. This can
be done by long heating in a water-bath, which is a tedious opera-
tion—much more quickly if a current of air is blown throu^^h the
gently heated acid ; the air along with the gas contained in it is
conducted through a small coke-tower, or, preferably, a small
''plate-tower^' (see above), fed with water, where dilute nitric
To fa<x p.^127.] Pol. I. S„lp\wie Add.
CONDENSATION OF NITRIC ACID. 127
uitratiiig acids from the manufacture of nitroglycerine^ nitro-
cellulose^ aniline^ &c.y and reconverting them into nitric acid.
For such purposes it is sometimes necessary to combine several
plate-towers in sets, one tower being placed higher than the others
and delivering its weak acid to those lower ones, to be got up to
strength (Rohrmann & Niedenfuhr, E. P. 29746, 1897).
The weak acid can also be pumped back to be used over and
over again, till it has got up to the maximum strength (1*38 or
1'40 spec. grav.). Figs. 30 aod 31 show a complete nitric-acid
regeneration plant, as employed after long experience by H. H.
Niedenfuhr. Pipe a (6 inches wide) comes from the denitrator ;
it is preferably rather long, say 40 feet, so as to cool the gases
which pass through receiver b into the first tower Ri ; glass tube
c admits of watching the process. R^ is filled with 34 LuQge
plates. The second tower Rg contains 30 Lunge plates and above
them 2 feet of stoneware balls. The connecting-pipe between the
towers contains a " KirchhoflF pipe '^ d, which admits of mixing air
with the gases in case of need. Pipe e takes the gases away from
R2, through the regulating '* sight ''/ into receiver^, and through
h into the chimney. V^, V2 are cisterns for feeding the towers ;
Vj in the beginning receives a little water, V^ is entirely filled
with water. As soon as the denitrator has been started, water is
run down from Vi into R^ drop by drop, and from Vj into R2 ^^ &
strong jet, sufficient for absorbing all nitrous products. The liquid
from R2 is collected in V^, and is pumped back by the "Plath^'
pump P (to be described later on) into Vi and Vj. With this weak
acid Ri is later on fed to such an extent that the acid recovered
here comes out with the desired strength and runs into. V3. On
the other hand, R3 is always fed up to the extent of completely
recovering the last traces of nitric acid. If the weak acid from
V4 does not suffice for that purpose, it is made up with water,
preferably at a temperature of 50° C* The acid condensing in
receiver b is run into V3 or V4, according to its strength, that
condensing in receiver g always into V^.
The refining or bleaching of nitric acid consists in driving off
the lower oxides, so that the acid becomes colourless. This can
be done by long heating in a water-bath, which is a tedious opera-
tion— much more quickly if a current of air is blown through the
gently heated acid ; the air along with the gas contained in it is
conducted through a small coke-tower, or, preferably, a small
'* plate-tower '^ (see above), fed with water, where dilute nitric
OKIESBEIM PROCESS FOR NITRIC ACID. 1^9
acid is contained. This refining is, of course, unnecessary for
nitric acid used in the manufacture of sulphuric acid.
Hirsch (G. P. 46,096) runs the impure acid through a stoneware
worm, placed in water of 80° C. Air is blown in at the bottom,
and the feed of nitric acid is so regulated that it runs out at the
bottom at a temperature of 60^ C. and sufficiently bleached. It
runs through a second worm placed in cold water, and can then
be put into the carboys. The nitrous gases escaping at the top
are treated in the usual manner. One worm can purify several
tons of acid per diem. The same process may also serve for
treating the waste acid of nitroglycerine and nitrobenzene works ;
in this case air heated to 150° C. or steam is blown in at the
bottom, and the feed of acid is so regulated that it issues, at a tem-
perature of 140° C, as comparatively pure sulphuric acid.
The French Government gunpowder works have employed such
worms for a long time for this purpose. An ordinary size of worm
furnishes 2 cwt. acid per hour, with less than 0*3 per cent. N2O4
and practically free from chlorine.
The Griesheim process does away with the necessity of bleaching
the nitric acid, as it is produced at once in a pure state. (A
similar result is aimed at by the apparatus of Guttmann in its
recent form and by that of Valentiner, to be described later on.)
The Chemische Fabrik Griesheim (Germ. pat. 59,099) places
behind the retort a reflux-cooler, consisting of aRohrmann stone-
ware worm contained in a water-tub kept at about 60° C. by the
heat of the operation itself. The acid vapours ascending in this
cooler are partially condensed there ; in consequence of the high
temperature the lower nitrogen oxides (together with the chlorine)
escape in the state of vapour, and are condensed by air and water
in a " Lunge tower '^ (plate-column) to weak nitric acid. The acid
condensing in the worm flows into a receiver, kept at 80° C, and
is therefore perfectly pure, no '* bleaching '^ being required. Air
is advantageously introduced into the receiver.
This system is illustrated in fig. 32. We notice that the
receiver, into which the acid flows back from the reflux-cooler, is
provided with a glass pipe reaching down to the bottom, through
which air is introduced, which aids in driving ofi^ N3O4 and
chlorine, &c., and also somewhat cools the acid. The strong and
pure acid from this receiver is from time to time run off into
a lower receiver^ not shown in the figure, where it cools down
VOL. I. K
180 BAW MATERIALS OP MANUFACTURE.
sufficiently for being withdrawn into carboya; the vapours rising
here are also paBse<l into the condensing-apparatus. The vapours
issuing from the top of the worm are passed through a few
receivers and ultimately iiito the plate-tower.
In practice the temperatiiVe of the water in the worm-tub rises
to 40° or 50° C. in the case of slow distillation, and 60° in that of
rapid distillation ; there is always some fresh water run iu, care
being taken not to cool too much.
Tip. n-J.
This apparatus is generally combined with the retorts shown on
p. 107. Each chaise of 400 kil. nitrate of soda (undried) with
450 kil. sulphuric acid (95 per cent. HJSO4) yields first two litres
of impure acid, which is run off separately, then 240 or 250 kit.
acid of 48° Baume =91 per cent. HNOa, with less than 1 per
cent. NjOj, usually only 0"5 per cent. N^Oi, and no chlorine at
all ; then 60 or 65 kil. acid of 43° to 44° Baume, perfectly clear,
aud 8 kils. of 25° to 30° Baume impure acid from the tower.
guttmann's condensing-affaratus. 131
When working with dried nitrate^ the first 250 kil. of acid show
49°B.=93 per cent. NHO3. The weaker acid (42° B.) can be
put back into the retort and recovered as strong acid in the next
distillation. The nitre-cake tests aboat 30 per cent. '^ iree'^ SO].
The Griesheim system has been working for many yeaiB at a
number of factories with perfect satisfaction.
The same principle as that followed in the Griesheim condensing-
plant is the foundation of the patents of Skoglund (G. P. 104^357
and 105,704), where first strong acid is condensed in a stoneware
apparatus and the weak, impure acid in a lead condenser [?].
Piatt (E. P. 9133, 1901) describes an improvement of the same
system.
G. Guttmann (E. P. 8915, 1890) has constructed a nitric-acid
condensing-apparatus on the principle of building it up entirely
of perpendicular pipes^ so arranged that the gases travel upwards
and downwards and the acid is run off at the bottom, as it
is liquefied, by means of hydraulically sealed branch-pipes, into
a common reservoir. The pipes are 8 feet long and have thin
walls {^ inch thick) ; they perform the condensation incom-
parably better than ordinary receivers, even when merely cooled
by air, but in this case there must be a considerable number of
such 8-feet pipes, say 15 or 20, to finish off a charge of 12 cwt.
in a shift of 10 or 11 hours. The last pipe is connected with a
*' Lunge tower " (plate-tower, comp. p. 125), fed with water, in
order to retain the uncoiidensed oxides, in the shape of weak
nitric acid, amounting to about 5 to 7 per cent, of the total acid.
The Guttmann system is especially suitable for the manufacture
of strong nitric acid, of course by means of dried nitrate and strong
sulphuric acid.
The former shape of Guttmann^s system is shown in our second
edition, pp. 890 and 891. Many plants have been erected on that
principle with great success. Later on, Guttmann improved his
system in various ways. He fixed an injector, fed with com-
pressed air, immediately behind the exit-tube from the still, where
the whole of the vapours are still uncondensed. They are thus
rapidly drawn away from the retort and mixed with hot air. It is
contended that in this way nitric acid with not more than 0*7 per
cent. NcOj and at a concentration of 97 per cent, can be made
during the whole of the process. The acid is as nearly colouriess
as can be desired. [If all the acid is 97 per cent., what becomes
k2
132 RAW MATERIALS OF MANUFACTURE.
of the water contained in the sulphuric acid^ unless this is of
equal strength? Comp. also Volney^ p. 121.]
Another improvement of Guttmann's is the substitution of water-
cooling for air-coolings which admits of a considerable saving in
pipes. There are only six of these, one exposed to the air and the
other five contained in a water-tank^ through the bottom of which
they pass water-tight by means of rubber rings. Guttmann states
that here all the acid can be obtained with 96 per cent. HNOs,
and slightly over 1 per cent. N^O^ ; this increase of NjO^ is caused
by the greater rapidity of the cooling, which does not leave so much
time for the oxidation of NOg by the air. Otherwise the con-
densing battery is similar to the first, inclusive of the plate-tower.
A recent patent of Guttmann's is E. P. 18,189, 1897. In his
English patent No. 13,694, of 1901, he describes an arrangement
for condensing nitric acid at a comparatively high temperature,
when it is less liable to absorb nitrous gases.
Figs. 33, 34, 35 show a Guttmann plant with the most recent
improvements, as designed by H. H. Niedenfuhr and supplied by
the Krauschwitz pottery. The drawings are to scale and explain
themselves. Each retort, with its set of cooling-pipes &c., decom-
poses from 12 to 14 cwt. of nitrate of soda in a working day of
12 or 14 hours ; with somewhat larger retorts, the cooling.battery
can be driven up to 20 cwt. nitrate per day.
Hart (Engl. pat. 17,289, 1894) uses an apparatus consisting of
a series of superposed glass tubes, slightly inclined to the hori.
zontal, which starts from an upright stand-pipe and ends in another
upright pipe. The pipes are cooled by squirting water upon them
or otherwise. The vapours pass simultaneously through all the
inclined pipes. As the water squirted on to the glass tubes
evaporates, its cooling-action is very strong. Hart asserts that
by his method the distillation is efiected in half the usual time,
with very little fuel and slight formation of N2O4. There are
15 tubes to each condenser, about an inch wide and six feet long^
for a retort working off 1000 lbs. of sodium nitrate in 8 hours.
This system has been introduced with great success in a number
of American and English works.
Dieterle and Rohrmann (G. P. 85,240) promote the evolution
of gases in the nitric-acid retort by the introduction of some inert
gas, which must certainly be au obstacle to condensation. Their
object is more rationally attained by the application of a vacuum, as
proposed by Valentiuer (see also Guttmann's injectors, p. 131).
GUTTMANN 8 C0NDENSIX6-APPARATUS.
133
134
Raw materials of xianufacture.
Valentiner (Eng. pat. 610, 1892, and 19,192, 1895; figs. 36 and
37) manufactures nitric acid in a vacuum. The retort in which
the sodium nitrate is decomposed with sulphuric acid is connected
with a cooling-worm, and this is connected with a receiver, from
which, with the interposition of a Woulfe^s bottle, the air is
aspirated by an air-pump. In this way the most highly con.
centrated pure nitric acid can be obtained. Perfectly pure nitric
monohydrate, produced by this process, is now found in commerce.
Fig. :34.
Section A-B.
[Before Valentiner (whose patent dates Sept. 8, 1891), I had shown
that this can be done by distilling in vacuo, Zsch. f. angew. Ch.
1891, p. 167, published March 15, 1891.]
Hallwell (Chem. Zeit. 1895, p. 118) gives some details of the
practical application of Valentiner^s process. The cast-iron retort
holds 16 cwt. of nitrate of soda and is nearly globular in form
(it is shown in fig. 18, p. 112). It is not heated directly by the
JjtTHQS : ^
valestixer's system for nitric acid.
135
CD
CO
SI£
Ix
tu
136 RAW MATERIALS OF MANUFACTURE.
flame^ but is surrounded by hot gases. At the top there are necks
for the acid vapour, for charging the nitrate, for letting in air, for a
thermometer (in a pipe closed at the bottom), and for introducing
sulphuric acid. The 8-inch wide neck which carries away the
acid vapours is continued into a glass cylinder, through which
the inside can be observed, then into an earthenware bend, con-
nected with an earthenware worm of 2.J-inch bore and 50 square
feet cooling-surface^ ending in a three-way cock. Then follow
two earthenware receivers of 60 gallons capacity each, with outlet-
cocks at the bottom, a receiver of 18 gallons, a smaller earthenware
worm (2^ -inch bore and 25 square feet cooling-surface), a 60-gallon
receiver and five 18-gallon receivers, all of which are provided with
outlet-cocks at the bottom and air-cocks at the top. The second
small receiver behind the second worm is charged with water, the
fourth with sulphuric acid; in these the inlet-pipes are deep enough
to dip into the liquid. The last receiver is connected with the
air-pump. The two large receivers behind the first worm take
most of the condensed acid and are alternately put into series by
means of the three-way cock ; thus the acid can be drawn off
without interrupting the work. The joints are all made by mclans
of flanges provided with rills and a cement made of silicate of soda
and asbestos (comp. p. 120). The vacuum helps to keep the joints
tight, and they stand very M'ell.
When the nitrate has been charged, all necks are closed, the
air-pump is started^ and by opening a tap in the connecting-pipe
sulphuric acid is drawn in from a store-tank. Much gas is given
off at once ; first of all nitrosvl chloride, which mostlv travels as
far as the receivers behind the second worm. When the vacuum
has gone up to 500 millimetres of mercury, the fire is started and
the thermometer rises to 80°, which tem|)erature is kept up during
the principal phase. The vacuum is kept at 650 or 670 millim.
In the end the temperature must be raised to 120^^ or at most
130°. When no more acid distils over, the air-pump is stopped,
and the temperature is raised to 170° or 175°, in order to render
the nitre-cake more liquid.
Through this low temperature the decomposition of the nitric
acid and the reducing-action of the iron are brought to a minimum.
Therefore the yield is almost equal to theory, and that mostly in
the shape of strong acid. With undried nitre and sulphuric acid
of 142° Tw. the yield, according to Hallwell, is 95*7 per cent, of
valentiner's process for nitric acid. 137
the theoretical, in the shape of acid of 90J°Tw. (.78 per cent. HNO3) ,
and 3*8 per cent, impure acid of 21^° Tw., together 99*5 per cent.
With undried nitrate and sulphuric acid of 160° Tw. the yield is
99 per cent, nitric acid of 93^° Tw. (81 per cent. HNOg), and
08 per cent, as dilute acid, together 99*8 per cent. The weak
acid is left in the receivers till it has risen to 66° Tw. With
dry nitrate and sulphuric acid 168° Tw., real nitric monohydrate
can be obtained. The usual strength, 100° Tw., contains pnly
O'OJ? per cent. NgOg and no chlorine at all, against 0*95 to 192
per cent. N2O8 in the ordinary 100° acid.
Hall well states that while in other processes charges of 6 to
8cwt. nitrate require from 15 to 20 hours to be worked off, here a
16 cwt. charge takes only 7 or 8 hours, and two charges are easily
made in 24 hours, which means four or five times the usual quantity.
The consumption of coal is 8 or 9 parts for firing and 6 or 7 parts
for the vacuum, together 14 or 16 parts to 103 nitrate, against 32 to
35 parts in the old process. There is also no steam or compressed
air needed for refining the acid. The durability of the retorts is
greater than in the old process, and the earthenware vessels do not
suffer at all. The temperature of the first receivers is only 35° to
42°, the back receivers are cold ; they never crack, nor do they
collapse through the atmospheric pressure, as they are made rather
thick- walled and nearly globular in shape. Any cracks in the
pipes or bad places in the receivers are cured by putting on asbestos
paper soaked in silicate of soda solution. If too much frothing
is observed through the glass cylinder, this is at once remedied
by opening the air-cock. The total length of the apparatus is
only 40 feet, the width of the furnace 17 feet, that of the con-
densing.plant 5 feet. Large apparatus holding from 50 to 60 cwt.
is to be constructed. The sanitary drawbacks of the ordinary
nitric acid manufacture are absent when working with a vacuum.
The action of the acid gases escaping from the condensation
(especially NOCl) upon the air-pump must be prevented by
charging the last receiver with milk of lime, which should, of
course, be renewed from time to time.
Further communications on the Valentiner process have been
made by Franke (Chem. Zeit. 1897, p. 511). For each ton of
nitrate decomposed there should be about 87 cubic feet condensing-
space. The retorts should bs entirely surrounded by the fire, or
their upper part and the branch-pipe should be protected by an
138 HAW MATERIALS OF MANUFACTURE.
acid-proof liDing. With sufficient condensing-space and good
cooling, and if the formation of NoOj is prevented, the yield should
be almost the theoretical, since only part of the chlorides can
escape. The nitre-cake is free from nitrogen compounds, and the
first receiver, which is filled with water, is not changed even after
several operations. The temperature during the distillation itself
is hardly 100° C, but later on it must rise to 175°, to make the
nitre-cake sufficiently liquid. The hot vapours then formed may
decompose some of the condensed acid, unless they are very well
cooled immediately behind the retort. Theoretically 1000 kil.
96 per cent, nitrate of soda, with 2 per cent, water, treated with
1000 kil. 94 per cent, sulphuric acid, should furnish 796'5 kil.
nitric acid of spec. grav. 1-486 (89"3 per cent. HNOa) ; from 1000
kil. 96 per cent, nitrate, previously dried, 771*5 kil. nitric acid
spec. grav. 1*497 (92*2 per cent. HNO3) should be obtained, and
practically the yield need not be much less. But with con-
centrated acid there is much frothing, through the production of
N2O4, because the acid does not at once penetrate through the
nitrate; this is avoided by employing more sulphuric acid, or
by introducing liquid nitric acid into the retort. Valentiiier
has recently recommended the preparation of stronger nitric acid
by redistilling the weak acid in the vacuum with concentrated
sulphuric acid ; but this double distillation costs decidedly more
acid, coals, and wages in comparison with the direct distillation.
Comp. also the same author in Zsch. angew. Ch. 1899, p. 269.
Bergmann (ibid. 1900, p. 1003) also re^xirts favourably on this
process, but finds it necessary to pass the gases through milk of
lime before they reach the pump, in order to retain the nitrosyl
chloride. He consumes 5 cwt. coals for a charge of 16 cwt.
nitrate.
The manufacture of nitric acid bv means of a vacuum has also
been patented by Dreyfus (E. P. 13,826, 1895), who mentions a
temperature of 170°-190'^.
In spite of the disadvantage that a double distillation is required
in order to obtain very strong nitric acid, the Valentiner system
has been introduced in many factories. Owing to the low tempe-
rature, the material of the apparatus is more durable, the product
is pure, and the yield almost equal to theory. In the latter
respect, as Guttmann remarks, the results in all well-conducted
factories are about the same. 100 parts real NaMOs ought to
yield 74*13 parts real HNO3; but of course any NO2 contained in
VARIOUS SYSTEMS FOR NITRIC ACID. 139
the acid must be deducted from tbis^ and at least 5 per cent, is
necessarily obtained as weak acid (tower acid) .
The Valentiuer system has been represented as decidedly
inferior to Guttmann's by some observations made in a scientific
publication in 1901, but has been quite as strongly defended from
the other side. We can here only give the respective references :
Zsch. fur angew. Chemie, 1901, pp. 413, 495, 571, 658, and 731.
Compare also Volney's results on the smaller scale when dis-
tilling in a vacuum, p. 121,
The principal choice of a condensing system nowadays seems
to lie between the systems of Griesheim, Guttmann, and Valen-
tiner. Their results (allowance being made for the exaggerations
of interested parties) seem to be so nearly alike, that it becomes
very much a question of first cost of the apparatus, which we can-
not treat of here, since patent licences play a considerable part in
.this case.
Concentration of Nitric Acid, — (Comp. Valentiner's process,
p. 138). Colin (French pat. 211,045) prepares fuming nitric acid
of spec. grav. 1*5 by distilling nitric acid of 1*4 with sulphuric acid
of 1*84 in enamelled cast-iron retorts, and employing a glass three-
way cock for separating the distillates. [This apparatus docs not
appear to be practicable.]
Erouard (Germ. pat. 62,714) also concentrates dilute nitric
acid, or waste acids from nitrating processes, by adding strong
sulphuric acid, or a solution of CaClj or MgCl2 [!], and distilling
the mixture in a vessel in which it travels in a zigzag direction.
H. A. Frasch (Germ. pat. 82,578) prepares highly concentrated
nitric acid by passing the vapours from the nitric-acid retort
through a tower heated above the boiling-point of the acid, in
which hot concentrated sulphuric acid is descending, or in >vhicli
other dehydrating substances, such as anhydrous sulphate of soda
or burnt pi aster-of -Paris, act upon the mixed vapours.
The Verein Chemischer Fabriken in Mannheim (G. P. 85,042)
places between the still and the condenser a dephlegmator, from
which the dilute acid condensed there runs back into the still.
This dephlegmator is kept at a temperature of 85*^ C, so that the
concentrated acid can pass on.
The fixed residue from the manufacture of nitric acid (called
''nitre-cake/' or, in the workmen's language, '^ sally nixon,'' a
corruption of " sal enixum ") is practically a mixture of neutral
arid acid sodium sulphate. It generally contains from 25 to 30
140 RAW MATERIALS OF MANUFACT^JRE.
per cent, '^free aeid^' (i, e. bisulphate acid) calculated as SOs, J^nd
only traces of nitrate. We shall treat of it in detail in Vol. II.,
and here only remark that most of it is worked up with com-
mon salt into ordinary salt-cake and hydrochloric acid ; part of
it is also used up directly for glass-making, but no doubt not to
great advantage. Kirkman (£. P. 5703, 1889) employs it as an
absorbent for ammonia, in which case a profitable utilization of
NagSOj will be very difficult.
Giles, Roberts, and Boake (Engl. pat. 11,979, 1890) convert
ordinary nitre-cake by addition of sulphuric acid into '^penta-
sulphate,'' Na^O, SSOg, SHgO, which can be packed in iron
drums [?] or ordinary casks and usefully employed for certain
purposes (comp. Uebel's *^ polysulphate,*' p. 117).
The cost of manufacturing nitric acid at a Continental factory
(some years ago) was as follows, and can be easily reduced to
current English prices : —
1. For acid of 3&^ Baume (=sp. gr. l-334or SOper cent. NO3H).
£ s. d.
4 charges of 4 cwt. each nitrate of soda, at 16^.... 12 16 O
16 cwt. sulphuric acid 144^ Tw., at 3^ 2 8 O
10 cwt. lignite (very inferior quality), at 9rf 0 7 6
2 men, at 3^ * 0 6 0
Interest and writing off the plant 0 8 0
General expenses 0 16
Packages, &c 1 16 0
Yield : 21 cwt. acid at 36^ Baume 19 3 0
Cost of 1 cwt. acid at 36° Baume 0 18 3
2. For acid ofoO"" Baume {^sp. gr. 1-532=93 per cent. NOgH).
£ s. d.
6 cwt. of nitrate of soda (dry), at 19^ 5 14 0
6 cwt. sulphuric acid 168° Tw., at 5^. 6r/ 113 0
6 cwt. lignite, at 9rf. 0 4 6
Wages — 2 men 0 6 0
Sundries and general expenses 0 12 0
Smallstores 0 11 0
Yield : 3 cwt. 2 qrs. 10 lb. acid 9 0 6
Deduct value of 5 cwt. nitre-cake 0 15 0
8 5 6
Cost of 1 cwt. of acid 50'' Baume 2 6 0
WASTE ACIDS FROM NITRATING PROCESSES. 141
The cost price at a French works is stated by Sorel, for 100 kil.
of acid of 36° Baume^ made in large cylinders : —
francs.
Supervision (^ of the foreman^s wages) 0*59
Wages of workmen 113
76-33 kil. nitre at 27 frs 20*69
83-95 kil. acid 60° B. at 2-25 frs 1*89
36 kil. coal at 2-15 frs 077
Lighting 0*27
Kepairs 2*55
General expenses 0*65
28-54
Deduct 88-54 kil. nitre-cake at 2*25 frs. ... 1-99
Cost of 100 kil. nitric acid 36° B 26-55
Utilization of waste acids from nitrating processes. — ^Enormous
quantities of nitric acid, always mixed with strong sulphuric acid,
are consumed in the manufacture of nitrobenzene, nitrotoluene,
and other aromatic compounds required in the manufacture of
colouring-matters, as well as in that of nitroglycerine, nitrocellulose,
and other substances serving as explosives. The waste acids formed
in the first class of processes contain but little nitric acid, generally
only about 1 per cent., with about \ per cent, of nitrobenzene &c.
They may be used directly in the Glover tower, where the nitric
acid is utilized as well as the sulphuric acid. Special processes for
denitrating these acids are rarely employed, least of all with a view
of recovering the nitric acid, which does not pay for the trouble and
expense.
The case is entirely different with the waste acids from the
manufacture of explosives. These contain much more nitric acid
than those from the manufacture of nitrobenzene. Nitroglycerine
waste acid contains about 10 per cent. HNO3, 70 H3SO4, 20 HgO;
gun-cotton acid 11 or 12 HNO3, 80 H2SO4, 8 HjO. These acids,
if it is not possible to consume them in a Glover tower, can be
used for replacing part of the sulphuric acid in the nitric-acid
manufacture; or else they are denitrated by steam, producing
dilute sulphuric acid (which is concentrated by evaporation and
used over again) and nitric acid, together with lower nitrogen
142 RAW MATERIALS OF MANUFACTUREt
oxides. By passing the vapours mixed with air through some
receivers and then through a '' Lunge tower/' the nitrogen oxides
are also converted into nitric acid^ the total being recovered as
nitric acid of 70** or even 80° Tw. (comp. pp. 125 & 127).
A detailed description of the denitrating process is found in
O. Guttmann^s 'Manufacture of Explosives' (London^ 1895),
vol. ii. p. 177.
The denitration is altogether similar in this case to the process
to be described in Chapter VI. in connection with the recovery of
nitre in the lead-chamber process. The best kind of apparatus is
a column of Volvic lava^ made in one piece and packed with bits
of flint or quartz. Steam is injected to such an extent that the
outflowing denitrated sulphuric acid has a specific gravity of about
1'635; it is often strongly coloured. The vapours are passed,
together with air, injected or aspirated by suction, through a
number of receivers, say 6 or 8, or else a small Guttmann battery,
and then into a Lunge tower fed with a very thin stream of water,
followed again by a few receivers.
"Where very large quantities of nitrous vapours have to be re-
generated, whether it be from waste acids of the just-mentioned
kind or from other chemical processes, it is best to combine several
plate-towers, one of which may be placed above the others so as
to feed them with the weak acid produced therein. Thus all
the recovered nitric acid can be brought up to spec. grav. 1'38.
The Krauschwitzer Thonwaarenfabrik, Muskau (Silesia), furnishes
this kind of plant. A very complete apparatus for this purpose
has been arranged by Niedenfuhr, as described and illustrated
supra^ p. 127.
Where sulphuric anhydride is made, these waste acids can b^
brought up to strength by means of SO3 and used over again.
Various Processes for the Manvfacture of Nitric Add.
Glock (G. P. 110,254) heats nitrate in a pan provided with
stirring-gear to 120°-150° and runs in the equivalent quantity of
sulphuric acid, previously heated to 100^-130°, in a thin jet. The
end of the decomposition is facilitated by steam or a thin jet of
water. At last air is blown through, whereupon the solid, pul-
verulent neutral sodium sulphate is ladled out. Or else the nitrate
VAIIIOUS PROCESSES FOR NITRIC ACID. 143
is from the first heated to 250° C. [This process looks extremely
impracticable.]
Manufacture of Nitric Acid ivithout Sulphuric Acid, — Campbell
.& Walker (E. P. 9782, 1894) grind nitre-cake with nitrate of soda
and charge the mixture into retorts provided with a mechanical
agitator.
Garroway (E. P. 6777, 1899; Journ. Soc. Chem. Ind. 1901^
p. 1191) mixes ordinary acid uitre-cake with sodium nitrate, heats
the mixture in a retort, and blows a spray of weak nitric acid
by means of compressed air over the mixture. The nitrogen
oxides^ mixed with steam and air, are regenerated by condensing-^
tubes and towers into nitric.acid (see pp. 125 & 127). ' It is alleged
that ultimately a 96-per-cent. acid can be produced [? this cer-
tainly cannot be done by any ordinary regenerating process] . The
residue is neutral salt-cake, testing 9836 per cent. This process
is stated to have been worked for three years at Glasgow.
Garroway (E. P. 2466, 1895) also prepares nitric acid by heat-^
ing a mixture of nitrate of soda and ferrous sulphate or alum,
obtaining sulphate of soda and oxide of iron or alumina.
He also patents the decomposition of sodium nitrate by silica
(E. P. 2489, 1896).
Very interesting are the processes intended for decomposing
nitrate of soda in such manner that, besides nitric acid, caustic
soda is formed . Theoretically these processes are enormously supe-
rior to the ordinary process ; there is no waste of sulphuric acid and
no production of nitre-cake, which is a product of very low value,
but the soda is brought into its most valuable form, as caustic or
carbonate. But not one of these processes has come into regular use.
They all suffer from the drawback that the temperature of decom-
position is too high, which necessitates special shapes of retorts
and great wear and tear of these, and that some of the nitric acid
(often a very considerable proportion) is reduced to lower oxides.
These can be reconverted into nitric acid by means of intimate
contact with air or water, in *' plate-columns ^' (Lunge towers) or
otherwise ; but this cannot be done without a perceptible loss, and
it never leads to the production of very strong nitric acid, such as
is required for the manufacture of explosives and many other
purposes.
We therefore only refer to the enumeration of these processes
144 »AW MATERIALS OF MANUFACTURE.
in vol. iii. (2nd edition) p. 254 et seq., and briefly qaote the more
recent additions to the subject.
A special retort for the Lunge and Lyte process (decomposition
of NaNOs by Fe20g) has been constructed and patented (G. P.
90^654). It consists of a revolving cylinder placed in a sloping
position, with inner projections and contrivances for feeding and
exhausting without stopping the process. A full description of
the whole process has been given by J. L. F. Vogel in the Eng. &
Min. Journ. 1900, p. 408,
Main, Stevenston, and McDonald (E. P. 23,819, 1895) heat
sodium nitrate with manganese oxides. Garroway (G. P. 79,699)
employs lime and superheated steam. Vogt (E. P. 22,018, 1891)
heats with lime, ferric or manganese oxides in a current of super-
heated steam and carbon dioxide.
Darling and Forrest (E. P. 5808, 1894) propose obtaining nitric
acid by the electrolysis of fused alkaline nitrates, together with
metallic potassium or sodium. Patent no. 13,171, 1895, extends
this process to the production of nitrous fumes (together with
sodium oxide) for the manufacture of sulphuric acid.
Siemens and Halske (Oerm. pat. 15,103) suggest utilizing
the well-known formation of nitric acid by the direct union of
oxygen and nitrogen under the influence of the silent electrical
discharge. This reaction, which under ordinary circumstances is
much too slow, is to be promoted by adding ammonia-gas, which
causes an abundant formation and separation of solid ammonium
nitrate.
McDougall also employs electricity (E. P. 4643, 1899).
Crookes (Elect. World, xxxiii. p. 319; Fischer's Jahresb. 1898,
p. 307) shows that by means of induction- sparks 74 grams NaNO^
are formed by one kilowatt-hour, or 1 ton of nitrate by 14,000 kilo-
watt-hours. This would at present cost £25 with the best steam-
engines and dynamos working day and night, and is therefore out
of the question, but it might be done by very cheap water-power.
W. Ostwald (B. P. no. 898, 1902) improves the well-known
formation of nitric acid from ammonia, under the influence of
platinum, by passing that gas, with more or less its volume of
atmospheric air, at a red-heat over smooth platinum, coated with
& layer of spongy or black platinum. The smooth platinum causes
the ammonia to be burnt into nitric acid, with practically no for-
mation of free nitrogen. The finely divided platinum accelerates
TRANSPORTATION AND PUMPING OP NITRIC ACID. 145
both reactions^ the second one more than the first. By moderate
use of the finely divided platinum with the smooth platinum, the
operation can be so performed that the reaction takes place rapidly,
but without any great formation of free nitrogen. The same
effect is produced by iridium, rhodium, palladium, the peroxides
of lead and manganese, the oxides of silver, copper, iron
chromium, nickel, and cobalt.
Statistics concerning nitric acid are very difficult to obtain. For
the year 1 894 the well-known firm Gehe & Co., of Dresden, reckon
that the productign of nitric acid in Germany was 54,000 tons, for
which 40,000 tons nitrate of soda were consumed. In the United
States, in 190i), 27,890 tons nitric acid of various strengths were
made (U.S. Census Bulletin, No. 210).
Transportation of Nitric Acid. — Nitric acid is usually sent out in
glass carboys or earthenware jars. Strong acid, when kept in
tightly closed vessels, may give off a dangerous amount of vapours,
especially in the sunshine or under similar conditions. Guttmann
recommends covering the carboys loosely with earthenware or
glass cups fitting over the necks, also making the straw in which
the carboys are packed incombustible by dipping it in a solution
of zinc chloride or nitre-cake [which will very quickly corrode it;
neutral sodium sulphate is decidedly preferable I], and to lime-
wash the top of the carboys exposed to sunlight ; also to store the
carboys in an excavated chamber, where a breakage will not cause
much damage, 21s the place can be swamped with water. Some-
times fires are caused by nitric acid from a broken carboy inflam-
ing the straw ; in which case the poisonous vapours are exceedingly
dangerous to inhale (comp. Chap. VI.) .
Strong nitric acid can be kept and treated in cast-iron or
wrought-iron or leaden vessels, but, of course, any dilution caused
by the moisture of air or otherwise will cause a violent action.
Aluminium resists nitric acid even in a somewhat more dilute state.
Protection against accidents caused by fumes of Nitric and
Nitrous Acid. — In the sixth Chapter we shall treat of this
subject in connection with Gay-Lussac and Glover towers.
I will here only allude to the official rules published by the
" Berufsgenossenschaft fiir chemische Industrie,'' Fischer's
Jahresb. L899, p. 411, and the report oy Duisberg, Zsch. {. angew.
Ch. 1897, p. 492.
Pumping of Nitric Acid. — For this purpose Paul Kestner, of
VOL. I.
s
RAW MATERIALS OF UANSFAC^URE.
Fig. 88.
ANALYSIS OF NITRIC ACID. 147
Lille^ has constructed special ''pulsometers^' entirely of stoneware,
one form of which is shown in fig. 38 — exhibiting the cylinder A,
the float B, the air-valve C^ and the delivery-pipe D. All
flanges are made tight by thin sheet asbestos^ the faces being ground
and polished. The working of this ^' pulsometer ^' is exactly like
that of the cast-iron pulsometer described in Chap. YI.
Another automatic and continuous pulsometer has been con-
structed by Plath^ and is sold by the Vereinigte Thonwaarenwerke,
Charlottenburg (fig. 39) . It also consists entirely of stoneware^
including the valves. The acid enters through valve a, the hollow
ball of which is weighted so that it just sinks in the liquid^
and is therefore easily raised by the inflowing liquid. When
the vessel is full^ the acid gets up to the other valve, the ball of
which, by is so light that it even floats upon water. It is, of course,
raised by the acid and shuts ofF the upper air-way c. The com-
pressed air now enters through d and forces the acid up into the
rising-main e, and thus into the upper store<tank. As soon as the
vessel is empty, the ball b descends, being aided by the short
column of liquid in c. The compressed air cannot now enter the
vessel, and, as there is now no counter pressure, the ball a is again
raised and the acid flows in from the lower tank, the air escaping
through e. This action continues so long as there is acid in the
lower store-tank and compressed air in d. Balls a and b are
accurately ground as usual for pumps &c. A new model made by
the same firm prevents any escape of compressed air during the
charging.
Schartler (Zsch. angew. Ch. 1901, p. 729) describes apparatus
constructed by O. Guttmann from stoneware for the purpose of
raising acids, viz., first an injector for working with steam or com-
pressed air, and, secondly, a constantly acting acid-egg.
Analysis of Nitric Acid, — Nitric acid is frequently merely tested
by the hydrometer, but this is quite illusory, owing to the influence
of the nitrogen peroxide (p. 102). When tested by titration it
should be noticed that methyl-orange is destroyed by nitrous acid ;
but this can be overcome by adding the indicator only towards
the end of the titration (comp. Chap. III., where the indicators
are more specially treated). When titrating with caustic soda,
this indicates ail acids : HNOs, NjO^ (which reacts = HNOj +
HNO2), HsS04, &c. At least a permanganate titration should be
made as well, in order to estimate Ns04«
l2
148 BAW MATERIALS OF MANUFACTURE.
The impurities contained in commercial nitric acid are as
follows — chhrincy sulphuric acid, fixed residue, iron : all found and
estimated by well-known methods. Nitrous acid or nitrogen
tetroxide is best estimated by means of potassium permanganate^
running the acid from a burette into the warm diluted solution of
permanganate, as will be described in the next Chapter. (The
influence of NO^H on the specific gravity of nitric acid has been
noticed above.) Iodine is recognized by boiling 1 c.c. in order to
remove the lower nitrogen oxides and to oxidize all iodine into
iodic acid, diluting with 5 c.c. previously boiled water, and adding
a few drops of a solution of potassium iodide and starch, made
with water free from air. A blue colour shows the presence of
iodine in the original acid, according to the reaction
HI08 + 5KI + 2H,0= 5KOH-f-6I.
This test, according to Beckurts (Fischer's Jahresb, 1886, p. 305),
is much more delicate than the ordinary one of reducing the iodate
by zinc and extracting the iodine set free by carbon bisulphide ;
but Beckurts^s test would, of course^ lead to serious errors unless a
check test was made with the iodide of potassium employed, which
might itself contain some iodate.
SULPHUR DIOXIDE. 149
CHAPTER III.
THE PROPEllTIES ANI) ANALYSIS OF THE TECHNICALLY
EMPLOYED OXIDES AND ACIDS OF SULPHUR.
Sulphur Dioxide, SOg. (Sulphurous Anhydride; erroneously
designated Sulphurous Acid.)
Sulphur dioxide is at the ordinary temperature and pressure a
colourless gas of suffocating smell, neither supporting combustion
nor combustible itself directly. Even when greatly diluted with
air it has a very injurious action upon plants and animals (comp.
further on, p. 154).
Sulphur dioxide contains 50 per cent, by weight of sulphur and
60 per cent, oxygen. Its specific gravity has been found by various
observers =2*222 to 2*247 (air = l); calculated from the mole-
cular weight =2*216. A litre of the gas at 0° C. and 760 millim.
pressure weighs 2*8608 grams. Its coefiScient of dilatation is not
exactly equal to that of air, but rather larger, especially at lower
temperatures, namely for each 1° C, according to Amagat : — -
Between 0° and 10^ 0004233
10° „ 20° 0004005
At 50° 0*003846
100° 0*003757
if
»
150° 0*003718
200° 0-003695
250° 0*003685
Its specific heat, compared with its equal weight of water is
=0*1544; compared with its equal weight of air =0*3414. Mathias
(Compt. Rend. cxix.p.404) gives it as between —20° and + 130° C:
0*31 712 + 0*0003507/ + 0000006762 /«.
The density of saturated vapour of SO2 (that is, in contact with
150 PROPKRTIES OF OXIDES AND ACIDS OF SULPHUK.
liquid SOj) at various temperatures (water of 0°=1) is, according
to Cailletet and Matbias (Compt. Rend. civ. p. 1536) : —
At 7-3 C.
... 0-0()621
„ 16-5
... 000858
„ 24-7
... 00112
„ 37-5
... 0-01fi9
„ 45-4
. 0-0218
„ 58-2
... 00310
,. 78-7
... 0-0464
.. 910
... 00626
At
100-6 C.
... 0-0786
123
... 0-1840
130
... 0-1607
135
.. 0-1888
144
... 0-2195
152-5
... 0-3426
154-9
... 0-1017
„ 156 critical point.
The heat of formation of one molecule of SO3 ( = 64) from
ordinary (rhombical) sulphur is =71,080 cal. (Thomsen), or 69,260
cal. (Berthelot).
By moderate cooling sulphur dioxide can be condensed to a
liquid, even without application of pressure. Liquid SO2 is a
colourless mobile fluid, of about the same refractive power as
water, boiling at —10° C. ; but on drawing it ofl^ at the ordinary
temperature from a reservoir it remains liquid for some time, the
evaporation cooling it down below its boiling-point. Its latent
heat at 0° is 912, at 10° 88-7, at 20° 84-7, at 30° 805.
Its vapour-tension is : —
At 0 C. = 0'53 atmosphere overpres?ure.
10° = 1-26
„ 20° = 2-24 atmospheres „
„ 30- = 3-51
„ 40° =5-15
}9 »
The specific gravity of liquid sulphur dioxide at various tem-
peratures has been accurately determined by A. Lange (Zsch.
angew. Ch. 1899, p. 275) as follows : —
Tempeniliire.
Specific pravity.
Temperature.
Specific gi'Hvity.
-20'^ C.
1-4846
+ 20° C.
1-3831
-10°
1-4601
25°
1-3695
- 5°
1-4476
30°
1-3556
0°
1-4350
35°
1-3441
+ 5°
1-4223
40°
1-3264
10°
1-4095
50°
1-2957
15°
1-3964
60°
1-2633
SULPHUR DIOXIDE. 151
He found that absolutely anhydrous liquid SO3 does not aot
upon iron up to 100° C. Technical sulphur dioxide has a slight
action^ owing to the presence of a little water. The temperature
at which this takes place increases with the purity of the acid,
e.g. it is 70° with ^cid containing 07 per cent. H2O. Since liquid
SO2 cannot dissolve more than 1 per cent, water, even the most
impure product cannot act on the iron vessels in which it is
transported at ordinary temperatures. The mixture of ferrous
sulphite and thiosulphate formed acts as a protecting crust. lu
ice-machines where SO2 is the active agent, and where the tem-
perature in the pumps may rise considerably, only absolutely
anhydrous SOg should be employed. (Comp., on this subject, also
Zsch. angew. Ch. 1899, pp. 300 & 595.)
Sulphur dioxide is produced by burning brimstone, and by
heating (roasting) many metallic sulphides, in the presence of air;
by the action of strong mineral acids, both on its own salts, the
sulphites, and on the thiosulphates and all polythiouic acids ; by
heating sulphuric anhydride with sulphur, or by heating oil
of vitriol with brimstone, coal, organic substances, or several
metals; by strongly heating the vapour of sulphuric anhydride,
or sulphuric acid, with simultaneous formation of oxygen and
water respectively ; and by igniting many sulphates, whereby the
sulphuric anhydride first liberated at once splits up into sulphur
dioxide and oxygen.
Thus sulphur dioxide is produced from sulphuric acid or
anhydride in many ways by reductive processes. On the other
hand, the sulphur dioxide passes over, even more easily, into sul-
phuric acid by oxidation processes; and it is accordingly one of the
most frequent and potent reducing agents. Under certain condi-
tions, by the action of light, of the electric current, or of a very
high temperature and pressure combined, the sulphur dioxide splits
up into sulphur and sulphuric anhydride. In the presence of oxygen
(for instance, that of atmospheric air), or of bodies easily parting
with their oxygen (such as the higher oxides of nitrogen, of man-
ganese, of lead), sulphuric acid or its salts are formed. A very im-
portant reaction is that between SO2 and sulphuretted hydrogen,
HjS. When completely dry the two gases do not seem to act
upon each other. Even in the presence of moisture no action
takes place if the temperature is above 400° C. (E. Mulder). At
the ordinary temperature water and sulphur are produced, but at
152 PROPERTIES OP OXIDES AND ACIDS OF SULPHUR.
the same time also pentathionie acid^ according to the equation
5 SO2 + 5 HjS = S5O6H2 H- 4 H2O + 5 S.
This action occurs simultaneously with the simple reaction,
«
S02 + 2H2S = 2H30 + S3,
one or the other of these prevailing, according to the proportion of
the two gases in the mixture.
With water, sulphur dioxide does not form sulphurous acid proper,
SOgHg, but only, under certain conditions, a solid compound with
much more water (9, 11, or 15 HgO to SOj), which has not yet
been definitively investigated. Sulphur dioxide dissolves pretty
freely in water; and this solution behaves in every way as if it
contained the real acid SO3H2 ; but constantly, even at the ordinary
temperature, the dioxide (SO2) evaporates from it. One volume
of water absorbs, under 760 millim. pressure and at 0°, nearly 80
volumes SO2. The coefficient of absorption, according to Bunsen
and Schonfeld, at temperatures ranging between 0° and 20°, is
79-789 -2-6077/ + 0-029349/';
at temperatures between 21° and 40°,
75182-2-1716/ + 001903/\
The saturated acid contains at 0^68*861 volumes of gaseous SO3,
and has a specific gravity of 1*06091 ; at 10° it contains 51*383
volumes gaseous SO2, and has the spec. grav. 1*05472 ; at 20°,
36*206 volumes SO2, spec. grav. 1*02386. The absorbed gas does
not escape on fieezing — and on boiling, only completely after a
long time. Alcohol absorbs a much larger volume of sulphur
dioxide (at 0° and 0*76 metre pressure, 338*62 volumes SO2) .
A table, not very much deviating from the above statements, of
the solubility of sulphur dioxide in water at 0*76 metre mercurial
pressure at difierent temperatures, is given in Kopp and Will's
* Jahresberichte' for 1861, p. 54.
Giles and Shearer (J. Soc. Chem. Ind. 1885, p. 305) give the
following table of the precentage of SO2 in solutions of various
specific gravities : —
SOLUTIONS OP SDLFHCKOU8 ACID.
153
Temp.
15°-50.
t»
I*
Spec. grav.
Per cent,
SO.,
'a*
10051
10102
10148
10204
10252
10297
1-0353
099
205
2-87
404
4-99
5-89
701
Temp.
1
Spec. grav.
Per cent.
SO,.
1 15°-5C.
10399
808
1
ft
1-0438
8-68
1
1-0492
9-80
ft
1^541
10-75
1205
10597
ii-a5
11°0
106(»8
13-09
1
Another table^ for the temperature 15°C., is given by Scott
(Pharm. Soc. J. & Trans, xi. p. 217) : —
Per cent. SOj.
Specific gravity.
Per cent SOj.
Specific gniyity
0-5
1C028
5-5
10302
1-0
10056
60
1-0328
1-5
1-0085
6-5
1-0353
2-0
10113
7-0
10377
2-5
10141
7-5
10401
3-0
1-0168
80
1-0426
3-5
1-0194
: 8-5
10450
4-0
10221
90
10474
4*5
1-0248
9-5
1-0497
50
10275
100
10520
Much higher are the figures given by Pellet (Journ. Soc. Chem.
Ind. 1902, p. 171) :—
P. c. SO2 in 100 H,0 12 3 4 5 6
Sp. gr. at 15°-17°C. 10025 1-015 10225 1030 10375 1045.
An apparatus for the production of aqueous solutions of sul-
phurous acid has been described by Holzhausel (G. P. no. 49194).
Solutions of sulphurous acid in the presence of oxygen are
partly converted into sulphuric acid.
According to Scott, when sulphur dioxide (mixed with COj)
is to be made by the process mostly used, viz. that of heating
sulphuric acid with charcoal, it is best to employ acid of 74 per
cent. 808=165° Tw. If stronger acid be used, a portion of it is
reduced to sulphur, which may give iron sulphide with the iron of
the apparatus ; with weaker acid sulphuretted hydrogen is formed.
In order to obtain the gas as pure as possible, the washing- water
should be mixed with lead sulphate or coarsely powdered charcoal.
154 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
Sulphurous acid forms two series of salts (sulphites) — ^saturated
or normal ones, SO3M2 ; and acid ones, SO^MH, isomorphous with
the corresponding carhonates.
Sulphur dioxide is absorbed by anhydrous barium oxide at 200°,
better at 230° ; by strontium oxide at 230°, better at 290°; in both
cases the normal sulphite is formed. Calcium oxide forms at 400^
a basic sulphite, CagSsOie, which at 500° splits up into sulphate
and sulphide. Magnesia absorbs SO2 very slowly at 326°^ and
slightly above this the sulphate is formed (Birnbaura & Wittich,
Ber. d. d. chem. Ges. 1880, p. 651).
The reactions taking place between sulphur dioxide and the
oxides and acids of nitrogen will be described in a later part of
this Chapter.
Injurious action of Sulphurous Acid {Sulphur Dioxide). — The
quantity of SO2 which may be present in the air without being
injurious to health has been stated by Hirt (^Gewerbekrankheiten/
p. 15) =1 to 3 per cent. This is an obvious error. Ogata (Archiv
f. Hygiene, 1884, p. 223) found that 0'04} per cent. SO2 causes
difficulty of breathiDg after a few hours ; he could not take a single
full breath in air containing 0*05 per cent. SO2. It is an acute
blood-poison.
Lehmann (Zsch. f. angew. Ch. 1893, p. 612) shows that persons
not habituated to sulphurous acid are very little affected by
0*012 per mille, but perceptibly so by 001 5 per mille SO2. The
presence of 0*030 per mille after a few minutes causes strong
irritation of the nasal membranes^ sneezings and slight coughing,
which symptoms decrease after 10 minutes. The employes and
workmen accustomed to it are but little affected by 0*037 per mille ;
the sensibility to SO2 seems to be lessened by habitually respiring
air containing it.
Sulphur dioxide is also very injurious to vegetation, and is one
of the chief constituents of the '^ noxious vapours '' so much com-
plained of in most manufacturing districts. It is true that these
vapours contain other injurious constituents, chiefly of an acid
character, viz. sulphuric anhydride, hydrogen chloride, and some-
times even the acids of nitrogen. Disregarding the latter, and
even HCl, which will be treated of in the Chapter devoted to that
subject, we shall now enter upon a description of the effects
produced by the ordinary '^ acid smoke *' of metallurgical and
NOXIOUS VAPOURS. 155
similar works^ where SO^^ and generally also SO,i, are the principal
acids concerned.
A detailed investigation of the influence of the noxious vapours
at Freiberg, where verr large and numerous smelting- works are
situated^ on vegetation and on the health of domestic animals
has been made by Freytag (abstracted in Wagner's Jahresb. 1873,
p. 180). The acid, arsenic, and zinc vapours of the Freiberg
smelting-works under favourable circumstances, even with the
present condensing arrangements, may injure the vegetation of
the neighbourhood in the following way : at a sufficient concen-
tration they are taken up by the leaves when covered with dew ;
on the evaporation of the water the organs affected are corroded
and reduced to the same state as that which they assume when
vegetation ceases. This injury can always be proved both by the
eye and by chemical analysis. A " poisoning '' of the soil or of
the whole plant is out of the question. The assumption of an
invisible injury done to the vegetation by the smelting-works'
vapours and the awarding of damages founded thereon are un-
warranted ; they contradict the fundamental principles of all exact
investigation and foster the desire of the unreasoning multitude
to incessantly raise fresh claims for damage alleged to have been
done by the works. A decrease of the nutritive value of food-
plants, in cases of visible injury done to the leaves, can only occur
in consequence of the loss of these leaves and the lessened ability
of the plants to decompose carbonic acid and produce organic matter
therefrom. Any metallic oxides or salts adhering to the leaves of
food-plants may become dangerous to the animal organization by
causing inflammation of the mucous membranes, and, under very
unfavourable circumstances, may produce death ; but this fact can
always be established with certainty by post-mortem examination
and by chemical analysis. The supposition that the '^acid disease''
and tuberculosis occurring in a particular neighbourhood among
the cattle are produced by the noxious vapours from smelting-
works is utterly unfounded and must be most emphatically con-
tradicted. Freytag considers that air containing more than 0*003
volume per cent, of SO2 will do injury to vegetation.
Schroeder (Wagner's Jahresber. 1874, p. 277) made extensive
experiments on the influence especially of sulphurous acid on vege-
tation, with the following principal results: — From air containing
156 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
as little as tjj/od of its volume of SO2 this gas is absorbed by the
leaves of "leafy ^' (deciduous) trees and conifers; these retain it
mostly, a smaller portion penetrating into the wood^ the bark, and
the leaf-stalks, either as such or after oxidation to sulphuric acid.
Conifer-leaves absorb less sulphurous acid from the air for an equal
surface of leaves than deciduous trees ; the absorption takes place
equally over the whole surface of the leaf, not by the stomata^
and therefore has no relation to the number of the latter. A
principal efiect of the injurious action of sulphurous acid is its
causing a check to the normal evaporation of water, the disturbance
being in a direct ratio to the quantity of SO2 ; the evaporation is
mostly affected by absorption of SO3 in sunlight, at a high tempe-
rature and in dry air. The transpiration of conifers is not visibly
lowered by the same quantity of sulphurous acid as that which affects
other trees (deciduous ones). The injury done by sulphurous acid
is greater if the absorption takes place at the lower than if at the
upper side of the leaf.
Other communications on this subject, partly contradicting those
of Schroeder, have been made by Stockhart (Wagner's Jahresb.
1874, p. 2528). According to his observations at Zwickau, a distance
of 630 metres (=690 yards) protects even the most sensitive vege-
tation against the effect of large volumes of vapours, if they escape
through chimneys not less than 82 feet high. Conifers are much
more sensitive than deciduous trees ; the decreasing series of sen-
sibility is — pine, pitch-pine, Scotch fir, larch, hawthorn, white
beech, birch, fruit-trees, hazel-nut, horse-chestnut, oak, red beech,
ash, linden, maple, poplar, alder, mountain-ash. In the parts of
plants corroded by sulphurous acid, not this acid, but sulphuric
acid can be found, and that to a larger extent than in the same
parts of plants collected at the same time in districts free from
smoke.
Schroeder and Schertel (Wagner's Jahresb. 1879, p. 234) found
in healthy fir-leaves 0*162 to 0*237 per cent. SOg; damage was
only done when the percentage rose above 0*250; the highest
found was 0*592 near Freiberg, 1-33 in the Oberharz.
Other figures given by Fricke (Chem. Ind. 1887, p. 492) state the
difference in the amount of SO2 found in healthy and damaged
plants as follows : —
NOXIOUS VAPOURS. 157
Healthy. Damaged.
Beans 6119 6*551
Buckwheat 5-110 5*880
Grass 7-105 8-336
Rye 3-684 5-610
Wheat 2-179 4412
Cabbage 27290 30-843
Oats 2-926 6788
Potatoes 13-000 17500
In most cases the differences are too slight to base any trust-
worthy conclusions on them. Oats, wheats and potatoes stand the
acid gases better than young meadow -plants.
Just and Heine (Chem. Industrie^ 1889^ p. 252) also found very
varying percentages of sulphuric acid in plants alleged to be
damaged by SO2, so that this means of tracing such injury is very
unreliable.
F. Fischer, in the 230th volume of Dingler^s Journali has given
a short synopsis of the researches made in this direction up to 1878.
A special treatise (in German) has been published on acid smoke
by Bering (Cotta, 1888).
Morren (Chem. Trade Journ. ii. p. 18S) shows that leaves are
more sensitive than flowers to sulphurous acid. When present in
a proportion of 1-80^000 in the air, the leaves of fruit-trees are
visibly affected in three to five hours, and this effect seems to
spread after direct action of the gas ceases. Adult leaves are
usually more sensitive than young leaves. The nerves are least
affected and usually'remain green. Sulphurous acid dissolved in
water is almost without effect on the upper surface, whilst on the
lower surface each little drop causes the formation of a spot
visible on both surfaces. This solution is not quickly changed
into sulphuric acid ; the effect of the latter is quite different from
that of sulphurous acid.
Koenig (Dingler's Journal, ccxxix. p. 299) describes the appear-
ance of trees destroyed by the vapours from roasting blende.
Hasenclever (Chem. Ind. 1879, p. 22 •) gives coloured and
photolithographic illustrations of the ravages caused by acid
vapours and metallic sulphates upon the leaves of plants aud
plantations of trees, side by side with those caused by frost,,
autumnal decay, fungi, drought, overgrowth of other trees, &C.,.
158 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
which closely resemble the phenomena produced by tlie acid
vapours from chemical works, and hence are frequently wrongly
attributed to the latter cause. Neither is the estimation of sul-
phates and chlorides in the damaged leaves &c. at all a safe guide
to the detection of the real cause^ looking at the enormous
quantity of acids sent into the air wherever coal is consumed on a
large scale. At Stolberg, near Aachen, on a superficial area of
1600 acres, 220 chimneys daily emit 34.} tons of sulphur dioxide
firom coal^ and nearly 51 tons of SO2 &om zinc- works, glass-works,
&c., the alkali- works adding only ^ ton (more correctly 480 kilog.)
of SOs^ and | ton of HCl. Hence alkali-makers ought not to be
saddled with the whole, or even the chief part, of the damage
observed in the neighbourhood.
An important paper on the subject in question has been pub-
lished by Hamburger (Journ. Soc. Chem. Ind. 1884, p. 202). His
conclusions, founded upon a large number of analyses of damaged
leaves, are practically the same as Hasenclever^s, namely, that
undoubtedly injury is done to vegetation by the acids in the
smoke ; but much difficulty exists as to proving this with certainty
in special cases^ and at all events the SO2 contained in ordinary coal-
smoke contributes very largely to the injurious action popularly
attributed to the emanations from chemical works.
A professional forester, Reuss, partly by himself and partly
together with Schroeder, has embodied the results of laborious
work on this subject in several German publications. Their con-
clusions have been attacked by another forester, Borggreve, but
Hasenclever (Chem. Ind. 1895, p. 496) has shown many mistakes in
that criticism. They all agree in the conclusion that the growth
of trees is only impeded by acid vapours if visible damage is
done to the leaves. When the leaves or needles remain green,
chemical analysis may prove an action of acid gases, but no real
damage.
Winkler (Zsch. angew. Ch. 1896, p. 371) ascribes the effect of
noxious vapours from brick-kilns &c. principally to their contain-
ing aqueous vapour, which, on cooling, causes the condensation
of sulphurous and hydrochloric acid.
Further papers on noxious vapours, with details as to their
action, are those by Hagen (Chem. Zeit. 1896, p. 238), Ost (ibid.
1896, p. 165), Nissenson and Neumann (Berg- u. Hiittenm. Zeit.
1896, p. 145), Schroder & SchmitE-Durbont (Dingler's Joum. ccc.
DETECTION OV SULPHUROUS ACID. 159
p. 65), Ost and Wehraer (Chem. Ind. 1899, p. 233), Seydler
(Fischer's Jahresb. 1899, p. 358), Ramann and Sorauer (ibid. 1900,
p. 832), Wislicenus (Zsch. ang. Ch. 1901, p. 689).
Detection and Estimation of Sulphurous Acid and Sulphur Dioxide^
Qualitative reactions of Sulphurous Acid. — The sense of smell is
a very good means for detecting the presence of SO3, when other
odorous acids are absent. Gaseous mixtures containing SO2 together
with such acids are best passed through an absorbent, e. ff, sodium
carbonate, with which afterwards the ordinary reactions for SO2 are
made. When passing such gaseous mixtures through a solution
of potassium permanganate, or of iodine in potassium iodide, these
liquids are decolorized, the iodine being reduced to HI. This last
reaction may also be utilized on test-paper. On the other hand,
a test-paper, soaked in a solution prepared by boiling 2 grams
wheat-starch with 100 c.c. of water, and adding 0'2 gram of
potassium iodate dissolved in 5 c.c. water, is turned blue by SO^,
by the formation of free iodine. These reactions may also be
utilized for recognizing the presence of SO2 when set free from its
salts by the action of sulphuric acid. One of the best reactions,
and one specially adapted for discovering SO2 in sulphuric acid
itself, is its conversion into H2S by means of pure zinc, or,
preferably, aluminium in an acid solution. The H2S is then
"recognized by its reaction on lead paper, or by the purple colour
produced in an ammoniacal solution of sodium nitroprusside.
A solution of a sulphite, either neutral or with addition of some
sodium bicarbonate (just acidulated with acetic acid), when poured
into a solution of zinc sulphate containing a little sodium nitro-
prusside, produces a red colour or precipitate, either at once or, if
very little SO2 is present, after adding some potassium ferricyanide.
This reaction is not given by thiosulphates, which are, moreover,
distinguished from sulphites by their giving (generally only after
a little time) a precipitate of sulphur on being treated with a
stronger acid. According to Reinscli, SO2 can be detected by
boiling the acid solutfon with a strip of clean copper, which is
thereby blackened. This is caused by the formation of cupric
sulphide, and the colour is not changed by heating the strip in a
glass tube ; but when the colour is produced by arsenic, there is a
sublimate of white arsenious acid formed in the tube.
According to Schiitzenberger, sulphurous acid contained in a
160 PROPERTIES OF OXIDES AND ACIDS OP SULPHUR.
solution can be recognized by adding a very little indigo solution
and agitating with a zinc rod ; owing to the formation of hypo-
sulphurous acid (Schiitzenberger's ^^ hydrosnlphuric *' acid); HSOj,
the blue colour will be destroyed^ but will quickly reappear in
contact with the air.
The quantitative estimation of sulphurous acid in the free state can
take place either as will be described in the case of sulphites^ or by
titration with standard alkali. In the latter case^ however, it must
be noted that the point of neutrality is reached with phenolphthalein
when the normal salt, NasSOg, has been formed, so that each c.c.
of normal alkali (containing 0031 NaOH) indicates 0*032 SO,.
Of course, as is always the case with phenolphthalein^ the standard
alkali must be soda or potash, ammouia being useless for this pur-
pose. But when employing methyl-orauge as indicator, the point
of neutrality is reached exactly at the formation of NaHSOa, so
that each c.c. of normal alkali indicates 0*064 SO3. Litmus
gives somewhat uncertain results, and is therefore useless as an
indicator. It is thus possible to estimate free SOs in the presence
of stronger free acids in this way : one portion of the liquid is
titrated with methyl-orange, and another with phenolphthalein as
indicator ; in the latter case more alkali will be used, and the
difference of c.c. of normal alkali, multiplied by 0*064, shows the
quantity of free SO2 present (LuDge, J. Soc. Chem. Ind. 1883,
p. 513 ; Thomson, Chem. News, xlvii. p. 136 ; Blarez, in Compt.
Bend. ciii. p. 69, adduces nothing new).
The acid sulphites are neutral to methyl-orange, which con.
sequently allows us to estimate any SO3 present over and above
NaHSOg. The SOj present in the NaHSO« itself can be titrated
with normal soda and phenolphthalein, each c.c. of normal alkali
indicating 0*064 SO3. Normal sulphites, as NajSO^, can be titrated
by means of methyl-orange and standard hydrochloric or sulphuric
acid, the red colour appearing when NaHSOg has been formed, so
that each c.c. of standard acid indicates 0064 SO^.
Other methods of estimating SOj either in the free state or in
its salts are based on its reducing properties. The reagents
serving for this purpose are either a standard solution of iodine or
one of potassium permanganate, both of which are well known and
require no description here. A dccinormal solution of either
indicates per cubic centimetre 00032 gram SO^. The method
to be recommended for testing gaseous SO^ in burner-gas will be
APPLICATION OF SULPHUROUS ACID. 161
described whea treating of that gas. Special attention must be
drawn to the necessity of employing water free from air in esti-
mating SO2. This is not necessary if the solution of the sulphite
or sulphurous acid is run into the solution of iodine (Giles and
Shearer, J. Soc. Chem. Ind. 188*, p. 197, and 1885, p. 303).
In many cases the quantitative estimation of sulphurous acid
can take place by converting it into sulphuric acid by means of
oxidizinj^ agents: chlorine, bromine, iodine, hydrogen peroxide, &c.
The sulphuric acid is then estimated in the usual way.
Sulphur dioxide in the presence of hydrogen sulphide, which
gases may exist together in a state of great dilution by inert gases
(as in the exit-gases from '^Claus kihis'"*), can be estimated by
passing the gases through a solution of 1 in KI, followed by one
of caustic soda or, preferably, sodium thiosulphatc. The iodine
oxidizes H.,S into H.,0 + S, and S0-» into H.,SO. : hence the
acidity of the solution is not affected by II3S, merely by SO2.
On the other hand, each c.c. of dccinormal iodine indicates
0'003.2 gram of sulphur in either case, so that the difference
between the iodometrical and the alkalimetrical test gives the
H^S present. The addition of a tube with sodium thiosulphatc
solution is necessary, because the gaseous current carries away
some iodine which is retained in that solution ; the latter, before
titrating the iodine solution back, is added to it (details in my
paper, J. Soc. Chem. Ind., Nov. 1890).
Applications of Sulphurous Acid {Sulphur dioxide).
The greatest quantity of SOo is produced for the manufacture of
sulphuric acid. Next to this in importance comes its use for the
manufacture of wood-pulp, mostly in the state of calcium bisul-
phite (or a solution of CaSOa in an excess of sulphurous acid).
One of the oldest uses of sulphur dioxide, in the shape of burning
sulphur, is as a disinfecting and antiseptic agent. For the former
purpose it is not so much valued now as formerly, since it has been
shown that many of the disease-germs resist the action of SOo for
a long time. The antiseptic function of SO2 comes into play in
the fumigation of wine-casks, in the arresting of the fermentation
of wort, in the manufacture of glue (where it acts also as a
bleaching agent), and in many other cases.
In the textile industries sulphurous acid is largely used as a
VOL. I. M
162 PROPERTIES OP OXIDES AND ACIDS OF SULPHUR.
bleaching agent, especially for wool, silk, straw, &c. It is not
quite certain in which way it acts in this case, possibly by forming
a compound with the colouring-matters contained in the fibres.
Formerly it was generally assumed that the SO2 in bleaching
acted as a reducing agent, which indeed must be true in some
cases, although probably noc in all. The reducing functions of
SO3 are utilized in chemical and metallurgical operations in too
many cases to be enumerated here.
Sulphuric Anhydride, S0;{
(Sulphur Trioxide),
consists of 40 per cent, by weight of sulphur, and 60 per cent, of
oxygen. According to Marignac (Arch. Sci. Phys. Nat. xxii.
p. 225, 1853; Hi. p. 236, 1875; Iviii. p. 228, 1877) and Schultz-
Sellack(Berl.Ber.iii.p.215),it exists in two different modifications,
a liquid and a solid. The liquid, a-anhydride, melts at + 16° C,
and begins to boil at +35° (according to Schultz-Sellack, at 46°).
Spec. grav. at 13°= 1-9516, at +20° (melted) =1-97. In the
melted state it is less oily than oil of vitriol, and , if pure, colourless,
but usually coloured brown by dust. When kept for some time
at the ordinary temperature (below 25°) it is changed into the
solid, y9-anhydride, whose melting-point is stated very differently,
from 50^ to 100° C. Probably it begins to melt' at 50°, and
gradually passes over into the ot-modification ; it slowly evaporates,
even at the ordinary temperature. It forms fine, feathery,
asbestos-like, white needles. The /3-anhydride is probably a
polymer of the a-modification. Buff (Ann. Chem. Pharm., Suppl.
iv. p. 151) confirms this. According to R. Weber, however (Pog-
gendorff's Ann. clix. p. 313 ; Berl. Ber. xix. p. 3187), the sulphur
trioxide, obtained absolutely pure and free from water by his
method, is at the summer temperature a very mobile, colourless
liquid, which, on gradually cooling, solidifies to long, transparent,
prismatic crystals -similar to nitrate of potash, quite different from
the white, opaque crystals of the ordinary anhydride containing a
little water. These crystals melt at 14°'8 C, and boil at 46°*2.
Under certain conditions the anhydride can, like many other
bodies, be cooled much below its proper melting-point without
solidifying, but then solidifies suddenly. After a twelvemonth it
SULPHURIC ANHYDAIDE. 163
Still shows the same composition and the same melting-point as
it had when freshly prepared. Weber accordingly rejected the
assumption of two different modifications^ and ascribes the
phenomena of this kind observed by others, especially the for-
mation of the modification resembling asbestos, to a minute
residue of water. So much seems to be correct in Weber's con-
clusions that the transition of the first to the second modification
is promoted by a minute quantity of water.
Oddo, in 1901 (Rend. Ace. Lincei, [5] x. p. 207; Chem.
Centralbl. 1901, i. p. 969), definitely proved the existence of two
modifications by cryoscopic estimation of the molecular weights.
The compound melting at 13°*8 is the real sulphur trioxide, SO3 ;
the fibrous compound, which does not melt unchanged, but at 50°
slowly, and at 100° quickly changes into SO3, is disulphuric
anhydride, SgOc. SOy instantly burns organic tissues and causes
deep wounds ; S2O6 is much less active and can be touched
with the hand. SO3 dissolves at once in H2S0i ; S2O0 but
slowly. Oddo gives the structural formula of SO3 : p.ZL^'^^f
ot S^Oqi p. S S ^.
Schenck (Lieb. Ann. cccxvi. p. 1) regards the liquid modification
as a solution of the asbestos-like polymer in real SO3 in a state
of unstable equilibrium.
The heat of formation of one molecule of SO- (=80 parts by
weight) from S and O-, is = 103,230 cals. (Thomsen); from SO3 + O
= 34',400 cals. in the solid state, or =22,600 cals. in the gaseous
state (Berthelot). The heat of vaporizing 1 mol. SO3 is =11,800
cals. ; that produced by dissolving 1 mol. SO3 in a large quantity
of water =39,170 cals. (Thomsen).
In moist air sulphuric anhydride at once forms dense white
fumes; with water it hisses like red-hot iron. Many organic
substances are at once charred by it. In the complete absence
of water it does not redden litmus. It gives several compounds
with sulphur, whose colour, with the quantity of sulphur decreas-
ing, changes from brown to green and blue. In the blue modi-
fication Weber has proved the presence of the sesquioxide, SgO^.
With sulphur dioxide there seems to exist a distinct compound,
SO2 + 2 SOa. With water SOy at once combines to form sulphuric
m2
164 PROF£KTI£S OF OXIDES AND ACIDS OF SULPHUR.
acid (SO4H2) and its different hydrates. It is, however, not easy
to completely condense the sulphuric anhydride often produced in
considerable quantity in technical processes, even with a large
quantity of water and manifold contact, and special precautions
have to be taken for this purpose.
The anhydride SO3, when conducted through a red-hot tube,
splits up into SO2 and O, but is reformed from these gases at
a somewhat lower temperature, especially in the presence of
platinum and several metallic oxides. The technical application
of this reaction is described in Chapter XI.
For scientific purposes sulphuric anhydride is made by gently
heating fuming oil of vitriol, or by ignitinjr sodium pyrosulphatc
(Na2S207) . Its production in a perfectly pure state is described
by Weber (/. c). Formerly it was not used for technical purposes,
owing partly to the costliness of its production, partly to the
supposed difficulty of handling and keeping it. Recently, however,
its production has been made so much cheaper that certain
branches of manufacture already employ it advantageously. Its
a])plication has turned out to be a very simple affair, as it can be
sent out in drums made of tinned iron. Its handling is certainly
unpleasant, since the contact of the skin with liquid anhydride,
or even just liquefying by absorbing moisture, causes very malig-
nant and slow-curing burns. Its production on a manufacturing
scale will be described in Chapter XI.
Pyrosulphuric Acid, SgOjHj
(Structural formula, SO.— OH
SO2— OH),
contains the elements of 89*89 parts of sulphuric anhydride and
10- 11 of water, or equal molecules of hydrate and anhydride.
A transparent crystalline mass, melting at 35° C, it decom-
poses at moderate heat into anhydride (SO3) and oil of vitriol
(SO.Ho).
Pyrosulphuric acid is contained in the Nordhausen fuming acid
of trade, which often consists altogether of it, and then bears the
trade name " solid oleum/' Pyrosulphuric acid can also be easily
obtained from the ordinary liquid fuming Nordhausen acid by
FUMING OIL OF VITRIOL. 1()5
cooling below 0°. Lastly^ it can be made by carefully mixing
anhydride with a small quantity of oil of vitriol. Weber (/. c.)
obtained an intermediate hydrate, H2SO4, 3 SOg, corresponding to
94-69 per cent, of SOg.
Pyrosulphuric acid forms salts, of which those of the alkaline
metals are the best known and most important. This sodinm
pyrosalphate (S^O/Nas) is formed by fusing acid sodium sulphate
(SO^NaH) at incipient red-heat. At a full red-heat it splits up
further into neutral sulphate (SO^Naj) and sulphuric anhydride
(SOg) ; this reaction is sometimes utilized for producing the
latter compound. In contact with water, the pyrosulphates are
gradually retransformed mto acid sulphates.
A compound with 14*41 per cent. H2O can also be obtained,
which crystallizes in thin transparent prisms^ fumes in the air, and
melts at 26° C. Formula—
SO2-OH
3 HA 4 SOg, or >0 + 2 SO^/^ ^
SO2— OH ^^"
The Nordhausen or fuming oil of vitriol, the manufacture of
which will be described in the 11th Chapter, is a viscous oil,
representing a mixture of pyrosulphuric acid or sulphur trioxide
with sulphuric hydrate in varying proportions, and therefore
solidifying at very different temperatures. It fumes in the air,
and gives out vapours of anhydride, whilst monohydrate remains
behind. Water transforms it at once into ordinary sulphuric
acid, with strong evolution of heat. It is often coloured brown
by organic snbstances, and, according to its mode of preparation,
contains many other impurities, such as iron, sodium, calcium,
aluminium, &c. (as sulphates), sulphurous acid, selenium, organic
matters, &c. When the receivers used in its preparation are
charged with ordinary strong acid, the impurities of the latter will
likewise pass into the fuming acid.
The subjoined tables refer to the mixtures of SO3 and H^SOg,
comprised under the designation of fuming oil of vitriol (abridged
O.V.), even when they consist mostly of SOg and are solid at the
ordinary temperatures.
166
PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
Melting-points of fuming 0,V,
Knietsch (Ber. 1901, p. 4100) gives the following table of
the fusing.points of fuming sulphuric acid (comp. also under
" Sulphuric Acid '' below) : —
Per cent.
Melting-point
Per cent.
MpI ting-point
SO3.
SO3.
°C.
0
+ 100
55
+ 18-4
5
+ 3-5
(K)
+ 0-7 ;
10
- 4-8
65
+ 0-8 '
15
-11-2
70
+ 90
20
-110
75
+ 17-2
25
- 0-6
' 80
+220
30
+ 15-2
85
+33-0 (270)»
35
■4-20 0
90
' +34-0 (27-7)
40
+338
95
: +360 (260) !
46
+34 8
100
+40-0 (17-7)
1 50
1
+28-5
The boiling-points of fuming O.V. are stated by Knietsch
(/. c. p. 4110) as follows : —
1
SO3 total
SOafree
Boiling-point
Barometric
per cent.
per cent.
°0.
pressure, mm.
823
3&1
212
759
83-4
9-63
170
759
8(>-45
20-23
125
759
89-5
42-84
92
759
93-24
6:5-20
60
759
99-5
97-2
43
759
The vapour-pressures of various descriptions of fuming O.V. are
given in the same place; comp. also the curves, infra, p. 174.
Specific gravities of fuming O. V,
CI. Winkler gives the following table of the specific gravities of
fuming sulphuric acid at 20° C; but it should be remarked that he
worked only with '' commercial acid/^ made by the old process,
and that consequently all the densities found are sensibly higher
than those belonging to pure acids ; nor is it a matter of surprise
that the values found by Messel with another description of
♦ The numbers in brackets denote the fusing.points of fresh, not yet polymerized
acids.
SPECIFIC GRAVITIES OF I'UMING O.V,
167
'* commercial acid'' (see below) do not entirely agree with
Winkler's table.
Specific
graritj
at20°C.
! 1
i 1
i 1
1
1
1
1
1
1
1
8(J0
8<>:)
870
875
880
8a'i
890
895
900
9a>
1-910
915
920
925
930
935
9-10
945
950
9J)0
965
970
Percentage of
I H,0.
SO3.
81-84
8212
82-41
82-63
82-81
82-97
8313
83-43
83-48
83-57
83-73
^08
84 56
85-06
a")-57
8<>23
86-78
8713
87-41
87 65
88-22
88-9:3
89-83
18-16
17-88
1759
17-37
1719
1703
1687
16-66
1652
16-43
16-27
15 92
1544
1494
14-43
13 77
13-22
12-87
12-59
12-35
11-78
11-08
1017
Percentage of
Q^ i acid of
"'^3- I 66° B.
26-45
27-57
28-76
29-95
30-38
3103
31-67
3252
33-09
33-46
34-10
;i3-52
37 27
39-49
4156
44-23
^6-46
47-88
49-01
49-98
5229
55-13
58-81
73-5;')
72-43
71-24
70a5
69-62
68-97
(«-23
67M8
66-91
66-54
65-91
64-48
6273
60-51
58-44
55-77
53-54
52-12
50 99
5002
47-71
44-87
41-19
Percentage of
free SO,,.' H.,SOj.
1-54
266
4-28
514
6-42
7-29
816
9-34
10 07
10r)6
11-43
1333
15-95
J 8-67
21-34
25-65
2803
29-94
31-46
32-77
39-68
44 64
98 46
97-34
9576
94-:i6
93-58
9271
91-84
90 66
89-93
89-44
88-57
8667
84-a5
8i-;«
78-66
74-a5
71-97
70-06
68-54
67-t^3
64-13
60-32
55-.'36
Messel (J. Soc. Chem. Ind. 1885, p. 573) gives the following
specific gravities of commercial Nordhausen acids, both at 2G°*6 C,
as determined by himself, and calculated for 15°*5 C. : —
Specimens.
' Percentage
of SO3.
Liquid
»» •
Cr^'stalline wnss, resembling nitre..
f» H 1»
It »• II
>» • »l »»
Liquid
f »
t»
Crystallized
II
II
8-3
3^)0
400
44-5
46-2
59-4
60-8
650
69-4
72-8
80-0
82-0
Specific gravities
at 80° F.
(-=26°-6C.)
calculated
for 60° F.
(=15°-5C.)
1-842
1-852
1-930
1-940
1956
1-970
1-961
1 -975
1-963
1-977
1-980
1-994
1-992
2006
1-992
2-006
2 002
2016
1-984
1-988
1959
1-973
1-953
1%7
168
FROFEIITIES OF OXIDES AND ACIDS OF SULPHUR.
Knietsch (Ber. 1901, p. 4101) gives the following tables for com-
mercial fuming O.V., made by the contact process. The weighings
were made at 15°, referred to water of 15°, with brass weights
without reduction to a vacuum.* The temperatures to which they
refer are 35° and 45° for fuming O.V.^ and 15° for this and the
strongest ordinary acids.
The curves (p. 174) show a maximum for ordinary acid slightly
below the percentage of monohydrate (HgSOJ ; for fuming acids
the maximum is at 60 per cent, free SOg at 15° C, at 56 per
cent. SOj at 35° C, at 50 per cent. SOg at 45° C.
Specific gravities at 15° and 45° C.
H,SO,
Total SO3
FreaSO, 1
Spec. grar.
Spec. grav.
per cent.
per cent.
per cent.
1
at 15° C.
nt 45° C.
95-98
1 78-35 ,
1-8418
96-68
7892
1-8429
96-99
1 79-18 I
1-8431
1
• • • • ■
97 66
79-72
18434 Max.
•••••• 1
98-6.5
80-53
1-8403
99-40
8114
1 a388 Min.
1
«• ^ . . . ■ 1
99-76
81-44
1-8418
1
1
1 10000
81-63
o-b
1-8.500
1 -822 1
1
83-46
100
1-888
: 1-858 '
85-30
200
1-920
1-887 i
8714
300
1-957
1-920
88-97
400
1-979
; 1-945
• • • • •
90-81
50O
2-009
1-964 Max.
92-65
60-0
1, 2020 Max.
1-959
94 48
700
' 2-018
1942
96-32
800
1 2-008
1890
9816
900
1-990
1-8(H
30000
1000
1
' 1-984
1-814
1
* The values for acids of 100 per cent. H^SO^ and below do not quite agree
with those found by Lunge and Naef (comp. later on), which could not be other-
wise, as the latter worked with pure acids and referred their figures to water of
4° C. and to the vacuum. The only essential de^nation is that Knietsch does not,
like the authors mentioned, as well as Kohlrausch and Schertel, find the minimum
spec, gravity at 100 per cent. IlaSO^, but at 99-40 per cent#
FUMING SULPHURIC ACID.
169
Specific gravities of fuming O.V. at 35° C.
Total SO,
1
Free SO- 1
a
i Total SO,
Free SO.,
d
per cent.
per cent.
1
Spec. gray.
1-8J86
1 per cent.
1
per cfcnl.
Spec. grav.
81-63
0
9118
52
1 -9749
81 •«9
: 2
1-8270
91-55
54
1-9760
8236
4
1-8360
91-91
56
1-9772
8273
6
1-84-25
92-28
58
1-9754
8309
8
1-8498
92tW)
60
1-9738
83-46
1 10
1-8565
93-02
62
1-9709
83-82
12
1-8627
93-38 I
64
1-9672
84-20
14
1-8692
9375
66
l-9<)36
84-66
16
1-8756
9411 '
68
l-9<-»00
84-92
1« 1
1-88:30
94-48
70
19564
8530
20
1-8919
94-85
72
l-950i
85-66
22
1-9020
95-21
74
19442
8603
24
1-9092
95-58
76
19379
86-40
26
1-9158
95-95
78
1-9315
86-76
28
1-9220
1 9632
80
1-9-251
8714
30
1-9280
96'69
82
19183
87-60
32
1-9338
' 97-05
84
1-9115
87-87
34
1-9405
97-45
86
1-9046
88-24
36
19174
97-78
88
1-8980
88-60
38
1-95:34
9816
90
1-8888
88-97
40
1-9584
9853
92
1-8800
89-33
42
19612
98 90
94
1-8712
89-70
44
1-9643
99-26
i ^
1-8605
9007
46
1-9672
99 63
98
1-8488
90-44
48
19702
10000
100
1-8370
90-81
60
1-9733
1
t
1
1
1
1
The specific heats were found by Knietsch as follows (those
marked * were directly observed) : —
Spec. beat.
Total SO3
per ceDt.
Free SO3
per cent.
Spec. Iieat.
76-8
0 3691 *
78-4
0-3574 *
80
0-850
800
0-3574 *
81-5
0 3478 *
82
2-0
0-34o
83-46
100
03417*
84
12-89
0-340
a5-48
20-95
0-3391 •
86
23-78
0-340
8713
29-74
0-3:392 *
88
34-67
0-350
88-75
38-75
0-3498*
90
45-56
0-360
90-1
4(Vl
0-3599 *
90-73
49-4
1
0-3660*
Total SO.,
FreeSOg
per cent.
per cent.
1
91
510
92
5645
93
61-89
93-3
635
94
67-34
94-64
70-6
95
72-78
90
78-23
9652
811)
97
83 67
97-92
88-6
99
89-12
99
94-56
lK)-8
98-9
100
1000
0-370
0-400
0-425
0-43-25 *
0-455
0-4730 *
0-495
0-535
0-5598 *
0-5W)
06526*
0-650
0710
0-7413 *
0-770
170
PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
The heats of solution were observed by Knietsch both by means
of a calorimeter and on a large scale. Those referring to
fuming O.V. are given here, those of ordinary sulphuric acid under
that heading.
SO3 total
SOgfree
Onlories.
Heat of solu-
tion of solid
per cent.
1
per cent
1
O.V.
82
20
199
83
7-5
210
84
12-9
223-5
80
18-3
237-5
1
1
86
23-5
250
1
87
202
265
88
34-7
278
89
40-1
292
90
45-6
308
"286
91
ra-o
325
304
92
56-4
344
322
93
(Jl-9
363
340
M
073
i«l
360
95
72-8
1 401
:«0
i 96
78-3
421
402 !
97
83-7
442
423 1
98
891
465
442
99
94-6
490
463
UK)
1000
515
486
Knietsch (Joe, cit.) gives also tables of the electrical resistance,
the velocitf/ of outflow (viscosity), and the capillary rising of sul-
phuric acids and O.V. from 0 per cent, to 100 per cent. SOg ; the
results are exhibited in the curves, infra, p. 174.
The action of fuming O.V. on iron (cast iron, wrought iron, and
steel) will be mentioned later on, together with that of ordinary acids.
The analysis of fuming sulphuric acid is described after that of
ordinary acid.
Sulphuric Acid, HgSOi.
Natural Occurrence,
Free sulphuric acid is found very exceptionally in nature, whilst
some sulphates, especially that of calcium, occur in enormous
quantities.
In the free state sulphuric acid has been found especially in a
few springs of volcanic origin, and in the rivers fed by those
springs. One of the best known cases of this kind is the Rio
Vinagrein Mexico, which contains 0111 per cent, free sulphuric
acid (calculated as SO3), and 0"091 per cent, free HCl ; it daily
MONOHYDRATED SULPHURIC ACID. 171
carries into the sea 38 tons, according to others even 69 tons of
both acids. Many other similar instances have been discovered in
various parts of the T,vorld (comp. the first edition of this work,
i. p. 17). Other cases again occur from the oxidation of sulphur
ores, the acid being expelled by heat from the sulphates generated
atfirst.
Miners are only too familiar with the occurrence of free acid in
pit-watera from similar causes, by the corrosion of metal pumps
and steam-boilers ; even the leather of the valves thereby becomes
brittle and the wooden parts are charred. In the case of vol-
canoes, sulphuric acid is formed by the oxidation of the sul-
phuretted hydrogen and the sulphur dioxide from the fumarolcs
and solfataras.
Even in the animal kingdom free sulphuric acid has been found,
viz. in the salivary glands of several mollusks, especially of Dolium
^alea, which contain 2'47 per cent, free sulphuric acid and 0*4
per cent, free hydrochloric acid (Boedeker and Troschel ;
De Luca and Panceri).
Monohydrated Sulphuric Acid, H2SO4.
The proper sulphuric hydrate, commonly known as mono-
hydrated sulphuric acid, has the rational formula : —
SO -OH
and may be considered as containing 81*63 per cent. SO3 and 18'37
per cent, water. It is a limpid, colourless, oily liquid. Its
specific gravity at (f is 1*853 ; at 15° (compared with water of 4°) :
1-8384 (Lunge and Naet), 1*8378 (Schertel), 1*8372 (Marignac,
F. Kohirausch, MendelejefF) *. The specific gravity changes to
the extent of +0001 for each degree Centigrade. Both the
addition of very little SOj, and that of very little water raises the
specific gravity (see below) . The pure monohydrate solidifies at
about QP and forms large, plate-shaped crystals which melt at
-f 10*^*5 ; they remain liquid a good deal below that temperature,
but solidify on agitation, or even better when a fragment of the
solid hydrate is introduced. The acid begins to boil at 290°, but
the boiling-point rises up to 338° (Marignac). This shows that
it does not distil unchanged : in fact a mixture of hydrated acid,
* Comp. Berl. Ber. 1884, pp. 1718, 2536, 2711.
172 PROPERTIES OP OXIDES AND ACIDS OF SULPHUR.
anhydride, and water passes over (see below). This dissociation
begins much earlier; the pure monohydrate fumes^ that is gives
oft' S0|^ very slightly even at the ordinary summer temperature^
distinctly at 30° or 40°. Accordingly it cannot be obtained by
boiling down or distilling dilute acid^ but by adding an exactly
sufiicient quantity of anhydride to the strongest acid obtainable
by concentration, or by cooling such acid below QP and several
times recrystallizing the monohydrate in the same way. The
latter process has been made commercially available by the Author
for the manufacture of monohydratic sulphuric acid, which conse-
quently is no longer a laboratory product, but has become an
important article of commerce (comp. Chap. IX.).
The vapour of sulphuric acid consists for the most part, or even
entirely (according to the temperature), not of molecules of
SO4H2, but of isolated molecules of SO3 and H2O ; theory would
give to SO4H2 (2 vols.) a vapour-density of 3'862^ for separated
molecules of SO3 and HgO (4 vols.) a vapour-density of 1*6931^
whilst Deville and Troost at 440° found it actually =1-74. The
dissociation is therefore as good as complete in the state of vapour ;
and this assumption agrees very well witli our present notions
concerning the state of vapour (Dittmar, Chem. News, xx. p. 258).
Formation of Sulphuric Acid, — It has been asserted that sul-
phuric acid is formed in damp flowers of sulphur, even at the
ordinary temperature ; this is certainly the case on heating sulphur
with water at 200° C, or by applying the electric current. Sulphur
is easily oxidized to sulphuric acid by chlorine, hypochlorous acid,
nitric acid, aqua regia, &c. It is produced, along with sulphurous
acid and sulphur, from tri-, tetra-, and pentathionic acids — from
the former by merely heating, from all three by the action of
chlorine or bromine, or even on the prolonged action of stronger
acids, which set the thionic acids free ; also the thiosulphates yield
sulphuric acid under the action of chlorine. All these decomposi-
tions have to be kept in view in alkali-making.
Mostly sulphuric acid is formed from sulphur dioxide. The
aqueous solution of the latter is gradually transformed into sul-
phuric acid by the action of the air alone, and it is so transformed
at once by chlorine, bromine, iodine, hypochlorous acid, nitric
acid, and several metallic salts, such as manganic sulphate,
mercurous nitrate, &c. Sulphur dioxide and oxygen conducted
through a red-hot tube containing platinum, platinized asbestos^
SULPHURIC ACID CONTAINING WATER. 173
ferric oxide^ and a number of other substances^ yield sulphuric
anhydride, or in the presence of water sulphuric acid. This
reaction, which was formerly only of scientific interest, has
become of the greatest technical importance and is, according to
some opinions, destined to supersede the formerly universal, and,
up to this, most general process of making sulphuric acid from
sulphur dioxide, air, and water by means of nitrogen oxides as
oxygen carriers. All this will be explained in detail later on.
The heat of formation of 98 parts SO4H3 is : —
Liquid. In diluted solutions.
From SO2, 0, HgO 54,400 72,000 calories.
„ S,08,H20 124,000 141,000 „
„ S,04,H2 193,000 210,000 „
The heat of neutralization of 1 mol. (98 parts) HjSO^ by 2 mols.
(80 parts) NaOH in the presence of 400 mols. of water is given by
Thomsen =31,380 cals. Pickering (Journ. Chem. Soc. 1889,
p. 323) states it only =28,197 cals.
Sulphuric add containing water, — The strongest oil of vitriol
obtainable by boiling-down ordinary pure sulphuric acid contains
a quantity of water which is not stated alike by different observers
(Marignac, Pfaundler, Roscoe, Dittmar, Lunge and Naef, &c.).
The statements differ from 97*86 to 98'99 per cent, of SO4H2; it
is at all events very nearly 98*3 per cent. This distilled sulphuric
acid solidifies a little below 0° ; but it also shows the phenomenon
of superfusion in a very high degree. It boils at 338° (Marignac),
or 315° to 317° C. (Pfaundler and Polt). Usually Marignac's
statement is considered the most reliable ; the acid of Pfaundler
and Polt probably contained a little more water. The boiling
takes place quietly under a stronger pressure than the ordinary
one, but at a lower pressure with violent bumping, which can be
avoided by putting in platinum wire or scraps, according to
Dittmar even better by conducting a slow current of air through
it during the boiling (see Chapter VIII., purification of sulphuric
acid).
This acid of 983 per cent. H3SO4, distilling unchanged, possesses
a number of other peculiarities, marking it out as representing a
certain equilibrium, or so-called *^ critical concentration." This
comes out very well in a series of curves, given by Knietsch (Ber.
1901, p. 4089), fig. 39, p. 174. The boiling-points at that point
174
PJIOPERTIGS OF. OXIDES AND ACIDS OF SULPHUR.
bo
FROPEHTiES OF SULFHURIC ACID. 175
(330°) show a sharp apex ; below this, water or dilute sulphuric
acid, above this, sulphuric anhydride is volatilized until in either
case the constantly boiling acid of 98*3 per cent, is reached. The
vapour-tension at that critical concentration is = zero, measured
at 100° in a vacuum ; the specific gravity of hydrated acid here
reaches its maximum, from which it descends in both directions ;
the electrical resistance at this point begins to increase suddenly
towards a maximum reached at nearly 100 per cent. H2SO1 ; in con-
nection with this the action upon iron decreases, which is of great
importance for the durability of apparatus (comp. below).
In fig. 39, curve 1, marked , shows the melting-points ;
curve 2, the specific gravities at 15°, . at
35°; curve 3, , the specific heats; curve 1, , the
heat of solution; curve 5, o — o — o, the electric resistance at
25°; curve 6, + 1 , the boiling-points ; curve 7, • -|- 1 ,
the vapour-tensions at 100° ; curve 8, o o . — , the
viscosities (times of outflow) ; curve 9, -•-•-•-•^ the capillarity ;
curve 10, -1-1-1-1-1-1-1-1-, the action upon iron.
The following fact concerning a property of sulphuric acid con-
taining about 98 per cent. H2SO4 is of great importance in the
manufacture of sulphuric acid from SOa by the contact process.
Knietsch found that the task of converting the SO3 into hydrated
acid cannot be accomplished by absorbing it in a series of vessels
filled with water or dilute acid, although the heat of dissolution
in this case is at a maximum; but acid of 97 or 98 per cent.
HjSO^ at once and completely absorbs the SO3, so that only one
vessel is required, in which the proper concentration is maintained
by continuously running in water or dilute acids and running
off concentrated acid. Sackur (Zscli. f. Elektroch. 1902, p. 81)
explains this by saying that at ordinary temperatures the 100
per cent, monohydrate, H2SO4, is slightly dissociated into
H2SO4, H2O + SO:, (comp. p. 172), but in the presence of very
little water (equal to 98 per cent. H2SO4) the partial pressure of
SO3 is at a minimum, and hence this acid has the maximum
absorbing power for SO3. Up to this point no free H2O is
present, but with greater dilution it is found. That these
dilute acids are inferior solvents of SOs is explained by the fact
observed by Oddo (comp. p. 1G3) that the true SO3, melting at
14°, is easily dissolved in H2SO1, but the polymer, SoOg, but
slowly. Now the latter is formed from SO3 by the influence of
176 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
minute quantities of water, and therefore also when SO3 is passed
into acids below 98 per cent. H2SO4.
The ordinary ^^ rectified oil of vitriol" always contains more
water than that obtainable by the highest degree of concentration
or distillation. Exceptionally 98 per cent, acid is made for com-
mercial purposes; the usual rectified O.V., formerly called 170°
Twaddel), now more accurately 168° Twaddell, varies in strength
from 96 down to 93 or even 92 per cent, of real monohydrate.
This variation is partly caused by the fact that at the higher degree
of concentration a slight difference in specific gravity corresponds
to a great difference in percentage ; partly by the fact that the
specific gravity of commercial acids, owing to the presence of
impurities, is always higher than that of the pure acid; but, apart
from this, the correctness of ordinary hydrometers is rarely of a
very high order. Still, it must be conceded that in England at
least there is a possibility of making the hydrometers all alike,
the basis of TwaddelPs system being a plain and unmistakable
one, as every degree is equal to a difference of 0005. But
matters are far worse on the Continent and in America, where
Baume^s hydrometer is almost universally used; unfortunately
the degrees of this instrument, as stated by various authorities,
answer to very different specific gravities, and those of the
instruments found in trade often show even far greater devi-
ations. The only rational hydrometer on Baume's system which
rests on a mathematical basis, and which should therefore l>e
always obtainable with a uniform scale, is that graduated according
to the formula
141-3--W'
where d signifies any special density (specific gravity), and n the
degree of the scale corresponding to it. (The mathematical de-
duction of this formula is j^iven in the first edition of this work,
pp. 20 and 21.) This scale is also generally accepted in Germany
and France now. It is the only one in which the degree 66,
which is that everywhere accepted for rectified O.V., comes near
the real specific gravity of pure sulphuric acid of 90 per cent., or
commercial acid of 94 to 95 per cent., namely 1*840. Unfortu-
nately, apart from various other scales, Gerlach's scale is also
sometimes used, although this is far lower than the '^rational''
scale, and, to make the confusion still greater, the American
SPECIFIC GRAVITIES.
177
Comparison of Baumi's Hydrometers with the Specific Gravities.
"
'
1
•
a
Rutiontvl
Hydrometer,
144*3
Baum^'ci
Hydrometer
(Qerlach).
American
Hydro-
a-
Rational
Hydrometer,
144-3
Baum^'s
Hydrometer
1
American
Hydro-
Q
J XXX %J
144-3 -»•
meter.
36
- X J^ M. \J
'^""144-3-»-
(Gtorlach).
meter.
1
1
1-007
1-0068
1-006
1-332
1-3250
1-334
2
1014
10138
1-011
37
1-345
1-3370
1-342
3
1022
10208
1-023
38
1-357
1-3494
1-359
: 4
1-029
1-0280
1-029
39
1-370
1-3619
1-368
6
1037
1-0353
1036
40
1
1-383
1-3746
1-386
6
1-045
1-0426
1-043
41
1-397
1-3876
1-395
7
1052
1-0501
1050
42
1-410
1-4009
1-413
8
1060
1-0576
1-057
43
1-424
1-4143
1-422
9
1-067
1-0653
1-064
44
1-438
1-4281
1-441
10
1-075
1-0731
1-071
,46
1-463
1-4421
1-461
11
1083
1-0810
1-086
46
1-468
1-4564
1-470
12
1091
1-0890
1093
47
1-483
1-4710
1-480
13
1-100
1-0972
1-100
48
1-498
1-4860
1-600
14
1108
1-1054
1107
49
1-514
1-5012
1-510
15
1-116
11138
1-114
50
1-530
1-5167
1-531
16
1125
1-1224
1122
:51
1-540
1-6325
1-541
17
1134
1-1310
1136
, 52
1-563
1-5487
1-661 '
18
1142
1-1398
1143
53
1-680
1-6662
1-573 ;
19
1-152
1-1487
1-150
54
1-697
1-6820
1-694 ,
20
1162
1-1578
1158
55
1-615
1-5993
1-616
21
1171
1-1670
1172
56
1-634
1-6169
1-627
22
1180
1-1763
1-179
1 67
1-652
1-6349
1-660
' 23
1190
1-1858
1-186
58
1-671
1-6533
1-6(U
1 24
1-200
1-1955
1-201
59
1-691
1-6721
1-683
25
1
1-210
1-2053
1-208
60
1-711
1-6914
1-705
26
1-220
1-2163
1-216
61
1-732
1-7111
1-727
27
1-231
1-2254
1-231
62
1-753
1-7313
1-747
28
1-241
1-2357
1-238
63
1-774
1-7620
1-767
29
1-252
1-2462
1-254
64
1-796
1-7731
1-793
30
1-263
1-2569
1-262
65
1-819
1-7948
1-814
31
1-274
1-2677
1-269
66
1-842
1-8171
1-836
32
1-285
1-2788
1-285
33
1-297
1-2901
1-293
34
1-308
1-3015
1-309
35
1-320
1-3131
1-317
VOL. I.
N
178
PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
Comparison between the Degrees of TwaddelVs Hydrometer and
Specific Gravities,
Degrees,
Tw.
1
2
3
4
5
k\
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Specific
Gravity.
005
010
015
020
025
030
035
040
045
050
055
060
005
070
075
080
085
090
096
100
105
110
115
120
125
130
135
140
145
150
155
160
165
170
175
180
185
190
195
200
205
210
215
Degrees,
Tw.
Specific Degrees,
Gravity. , Tw.
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
1-220
1-225
1-230
1-235
1-240
1-245
1-250
1-255
1-260
1-265
1-270
1-275
1-280
1-285
1-290
1-295
1-300
1-305
1-310
1-315
1-320
1-32.5
1-330
1-335
1-340
1-345
1-350
1-355
1-360
1-365
1-370
1-375
1-380
1-385
1-390
1-395
1-400
1-405
1-410
1-415
1-420
1-425
1-430
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
Specific I Degrees,
Gravity. Tw.
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
435
440
445
450
460
465
470
475
480
485
490
495
500
505
510
515
520
525
530
535
540
545
550
555
560
565
570
575
580
585
590
600
605
610
615
620
625
630
635
640
645
Specific
Gravity.
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
ir)6
157
158
159
160
161
162
163
164
165
166
167
168
169
170
1-650
1-656
1-660
1-665
1670
1-675
1-680
1-685
1-690
1-695
1-700
1-705
1710
1-715
1-720
1-725
1-730
1-735
1-740
1-745
1-750
1-755
1-760
1-765
1-770
1-775
1-780
1-785
1-790
1-795
1-800
1-805
1-810
1-815
1-820
1-826
1-830
1-835
1-840
1-845
1-850
SPECIFIC GKAVITIES.
179
manufacturers have adopted again another scale^ based on the
formala
«=145 —
145
In order to clear the way as far as possible, we give herewith tables
(pp. 177, 178) showing the value of a degree Ban me according to the
rational scale, to Gerlach's^ and to the American scale (the last is
copied from A. H. Elliott, 'Chem. Trade Journal,' vol. ii. p. 183).
The percentage of mixtures of sulphuric acid and water is in the
great majority of cases tested by the hydrometer only, and many
tables have been constructed for this purpose. It would be very
desirable, as Hasenclever points out ( Hofmann's Report, i. p. 181), if
all sulphuric-acid makers used the same reduction-tables for their
calculations ; for in the statements on the yield of acid, and in
many other cases^ frequently different tables are used ; so that the
working results of different factories are not always comparable
with each other. This very clearly appears from the following
comparative Table : —
Degrees
10
1>0
30
M)
oO
60
Spec.
Gray.
(Kolb).
Percentnge of SO^Hj in the Vitriol according to
Vaiique- D'^^eet. ! Tables of different works.
liii. I
1075
11G2
1-263
1-^383
1-530
1-711
1-812
11-73
24-01
3i5-5-i
50-41
66-54
l(X)-00
6<V45
82-34
100 00
11-5
2.'V3
36-0
51 -C
l>6-9
83-3
11-40
23-46
3<»-(30
51-49
mn
82-80
(>3-8
70-4
10-98
21-97
35-93
49-94
63 92
79-90
Bineau.
110
22-4
34-9
38-4
62-7
780
100-0 100 00 910 97-87 1000
Kolb.
10-8
222
34-7
48-3
62-5
78-1
lOO-O
The totally incorrect tables of Vauquelin and D'Arcct are used,
up to the present time, exclusively in the south of France.
We shall here not take any notice of those old tables, nor
of those of Ure, Dalton, &c., and we refer to our first edition
«s to the more modern and reliable tables of Bineau and Kolb.
In this place we give only the most modern and accurate results,
viz., those of Lunge and Isler (Zsch. f. angew. Ch. 1890, p. 129)
for the strengths up to 142° Tw., and also of Lunge and Naef
(Chem. Ind. 1883, p. 37) *.
* A very extended study of the speciiic gravities of sulphuric-acid solutions
has been published by Pickering (Journ. of the Chem. Soc. vol. Ivii. pp. 04 H
seqq.). The reasons why I do not see any occa.<«ion for accepting Pickering's
figures in lieu of my own are stated in the * Journal of the Society of Chemical
Industry/ 1890, p. 1017.
n2
180
PROPERTIES OP OXIDES AND ACIDS OF SULPHUR.
•ft.
/^
c
• mm
s
c
o
o
'd
•s
£ • •
S!
aS
.a
«*-
o
1
u.
a.
8
OH
•T3 o
OH
O
GO
c«
1
I '^ Vrf ^^ _ _
«»r;t>-o
"ir T^ T^ T^ T^ 'sr "y* w* <—''"•'"•'"' ""^ '"• ^^ •"• w"^ w>i w>t .^ w>i «M ■«'< VI wj WW « .
ooooooooooooooooooooooo6c©o
ooooo6oooo6od>ooooooooooo©oo
o
oooooo6ocooc©666e6ooo66o©co
acid of
106° Tw.
OH
1— t
o
00
c«
o
IS
tic's
^H
11
,_pHi-(,-ir-ii-rr-<f-Hi-(fl^(N6l<MC^<?IC'l^??l
'^J -Ti I?! & "
• •
OOi-i<M(NC0'«*«Tti»r:;0C0t^XX05OO»^r^(Nc53cb'^'^Ocbcb
i
X
=-1
OkQC
8
'COOOOO»-^i-ii— »i-Hi— if-ti— I
*■ ■•■ ■••••■■A
SPECIFIC ORATITIE8.
181
oooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooo
t-
t*
«©
'^C^I'©QC»^cDoco2JlMooagQt--«p»0'4co(^l1-HO®xso»Q'tf'^(^^(^^^OQOC5g
T1G^Cl0^c<l(^l?^c<^<f^5^«cococcc^5«oo55co«co'^^^T^T^T^
O666c>66e©66^6666666o6±)6666o66666c6oo
^ ^ ^ ^ 5) C^l <M i>l C'l r) ?! C>1 ?
oooooocoooooc
■ " I ca o O ^ -M
0 300000000 00000000000 00
»-^^SS55c'l5'l?5?Ic^l¥lC'lS^SlS8Swo3o300COo8«S«r5S««r3C03^
C5 CaO
^•QQPO'-«cqcQ^ii5tP|;»gD©Q«-*gcO'^iO«traQ^<5'^<NoO'^i-'^^b-oQq5P
»7*t*cpX'?OQpooqaTt<pij:9>»p©»o©»pp'^o:'^giTHcpcpi^
^^f»^i.'5©^fc^t-«aDCiCi©©»-jj-Hg2aSSSrc
t
»c
UV ^^^ l^. Wl *>W ^Sl'
OC C5 CS Q © -—
<M 5i 5i « CO CO
■ • •
182
PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
^ ^
n
°H
e
'«o
§8
a
r— •
8
I
3
. 1
*«s
5 O
« rf
P<«
r-«
>»!
1 1
• ■
.S o
o
s^
CO
a>*i<
«^
o
b
•
<»-i
rr^
O
a;
f.
1^
*j
•
m»
• mm
n
0
^^
O
'-M
wH
QQ
c
1
(co
•
00
^
£
'!§
c)
^^
uri
•
of Sulph
1 J.
OH
Is
:2o)
1
^ G.
«c
*3
•
.Si
-S)-!
•«
••"A
• pSi ei
QQ
••m
« ^
g
^§
};.»
fe**^
1
1
C5
.a
1
^
A.
s.
•
^
^
g
O
►^
1— 1
^
1
w oa p -M e5 lO
X 00 ^ C^ S^ 0)
666006660
OaO^iO
C -^ '"' •"
•^1— C5«COOl-»'f^t>''^»7iOOiC^QO»-'^<MaO»i:t5'IX^O«SC<C-»^
'f:co«'?f»h<;b6i^aocic666.^(NCi«"^-^»b6«i*-qC^
Oiioo®cit«c^a)cp*'^cb'^o»po»p^^^:c^»7^ic:p»i:w
6 '^ c^i (i^i »? cJi rh rH ic ib cp «i b» (» cC' cs 6 6 6 »— "^ (f^
—« 00 iC p 1-- — Tt* or (?i ic Ci fM »o X ^1 >r. op — « cc r- 1'- cp CO »*': "^ C5 ci
w«'iH3^ij-^^cbt^r-i^xxd>6i660'^'^'fi6i'fic?«T^
<^^c5<:oco?oi-t-i'-t«i.'-b-.i>i>i>-i-.o5xxxxxaDXxx^
^OO'^X'MOO^XC^lCOO'^'X'-iir. X'MCCOCCl^^rt<X'-<'<^X
<m • W'» y» V, c, •"* <• V vj .^. v» rj v5 ^» • . ^^ Ty Ty ■»7" ■^T" ^rp ^p "j< Ty »p ^ tJ" ^y T
OS 2^
I
CO
03
ja t
.S 5
»ftc»^j;oL':oiCpi^p»<;o^Q»fiO^^P»^ooo^Qi^O'^0
SPECIFIC ORAVITIES.
183
M t>.
m
"^ ^ '^
■ • • •
^ "t l^ -H O » N
O ^ I'" Ci O — ^ 00
5* W 30 CC -J^ ^ ^
CO to s« 00
^ "^ ^ -f o
OOOOO
g5P «0 50 CO CO t— l- r» t>- 5P « lO »it '5 '^ Tf "^ 4?: »C ^ t^ 00 X Ci Q o O Q
X0S0i05CiCi0iCsS5Ci06OOOOOOOOOO»-<'^^^-^'~''^'~*'~t
ft* go t^ « « »o «s
oooOQOCJoaoaoooxxdaoa _ _______
oooooooooooooooooooo
Oat^lOCO--3£^»r500^5STQO"*t<CO-^C:i^lJ5CO-^CS
» cc CD « CO t^ t>- 1- 1-* t- r> i-«- 1-* I* t'- 1" i>- X ® Qo 00
0000003000000000000000
0000000000000000
^*5P3!05C0»OXiJ5'MCS'rCi'*Qi3QOJCO<Ml>"00X"^XC0i-<"«^X'MkO«(M^XCS^X
OCkJtriOl>-Oyi2it»u'^O^lOt*»^^OQO"T-^00'*<'-Ht»"^pt*"^»-IXCeWOXiO--Xii5rH
C^ClrH-H—* oxt-x
'"h — < 'n ri « Tt< Th »b ih
c^Wb-t«t* t»t*t«-l>
r- «^Ht-H^H^H^Hi-Hi— ti-H^^i-Hi-Hi-H^-^^^— i-Hi-Hp-^^^i-h»-^i— l»-«i-H-H.— »H^H
184
PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
o
us
s
8
t I
^n m m
^^
S O
•g
o
£
:5o
^ 9=> ?> «?^ 9* 9* P P 9 9 9 T* T" T" T^ T" T^ ?^ <?* <?• <?< ^ ^ « w ^ ^
<M — '^ ^
__ OQ.
CD
94
S
»«
CO
»H2^w<^loo^co^>«oa'^clco:*?p^-oa•^©^■3♦«Dt*oi^co»oo■*
•• • • •• • •• •• • • ■• ••••• • •• ••• 9^
s
1
■gs
o
OQ
OH
I?,
O
00
oooo
,V .>. 2-k J Jl ^ ^ .i^ i. ^ i. i. • . • . J^ ^, X, ^ ^ J^
t* t^ t<» t* i> t
II
f-«»-4— ii-Hi— i^i^^-,— ir-Hi— li-Hi— i^Hi— (— 4»-Hi— ii-H^Hi— Ir-ii— it-^f-«r-ni-HrH»-H
^a»f7H'^t*ci<?iTt<i^p(N»r:t-9'nTt<?ooa^"^coaorHccioooo©i
o o
B
a,
oc
•S|
S ci5 CO
COCO'
^pcp^-t*l>t•l>,^-^-^-^.;,^.^-^,lJ,^>ls,,>.ifi
SPECIFIC GRAVITIES.
185
1— I W T
Oei»^t---*»^0iG0i-Hi000C^«pO'^QQ'>1wQ't<Q0«Q«C^0i«D«0"'X)C>?CC0^^l'-O
cOTfQc^cooDXX23'i:X2P^2S'2^^SQ25SOgocaQO«>i^Q^C'i02J(N(NC^c<i<N^
pao«3w65»7H»;7«OTOqfc*7i«®p©lie5i'-pw25(»(N
o o o >n o to
- - 1 -^ t«- Ci
OOQQQC»OiQpQQOp»OQOQOQ<NO»3C9QpQpQ'OQOOOC
^l>ooo5p«o©o5^paDp7>l25^-o5-H55lO^*OTh«'?^25^
iOib'i?bt*«6sppppp'-H.lH»l^i^flie35idio»M««-^
*<* w ^ c»
^^p6
t*®05p^'MeQ'-r
tOiQiOdcOOCOCO
• lA • • • • CO • • •
• Ia • • • • 3t • • •
• ^0 • ■ • • ^0 • ■ •
&
iS
■ • • •
lOt^pd-ffCCX© ^
Seo -^ <^ -^ "^ "^ lO . lO
0» CO "«f< lO CO
* CC "CO 'CD * 5D
X 0»
■ • • • •
iJJ P iTi o ic p ^c
8
opiApiop»cpi-<©icO'fir5cot*gQgap'-tSi^3^»2S?tr!52S2!2!:2:i2SSs
t«i>t«aDxxxxxooxxxxxxxxa5aoxxooxxxxxxxxxx
186 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
The specific gravities were in each case estimated at exactly
15^ C. and compared with water of 4P, the weighings being reduced
to the vacuum. Special notice should be taken that all older
tables (including those of Bineau, Otto^ and Kolb) are entirely
wrong in the case of the acids of highest strength^ as the maximum
of specific gravity does not (as it is made to do in those tables)
coincide with the greatest strength^ that is, pure monohydrated
sulphuric acid, H2SO4. The maximum it at about 98*5 per cent.,
and from this point the specific gravities decline to 100 per cent.
H2SO4; above this point, that is when SO3 is present, they
instantly rise again. (Comp. Kohlrausch, Pogg. Annal. Erganz-
ungsband viii. p. 675 ; Schertel, Journ. f. prakt. Chemie, [2] xxvi.
p. 246; Lunge and Naef, Chem. Ind. 1883, p. 37; and others.)
Special attention should be drawn to the point that all tables
indicate the specific gravities only for chemically pure acids ; those
for commercial impure acids are always higher ; we shall return
to tliis subject further on. The accuracy of the above-given
tables may be taken as +005, so that the first decimal is right,
but the second serves only for determining the first.
A paper by MendelejeflF (Zeitsch. f. physik. Cb. i. p. 273) on
the specific gravities of mixtures of sulphuric acid and water has
only theoretical interest.
A correction for any deviation of the temperature from 15° must
be made, whenever the acid tested by the hydrometer is above or
below that temperature. Bineau has given a small table for this
purpose, which, however, is wrong for the more dilute acids.
From a very large number of observations made in the Author's
laboratory, a table has been constructed showing the alterations
to be made in the specific gravities observed for all strengths of
acid, aud for all temperatures from 0^ to 100°, in order to reduce
them to 15° C. This table is found in Lunge and Hurter's
* Alkali- Maker's Pocketbook ' ; in this place we give only the
average figures. For each degree above or below 15° you should
add to or deduct from the specific gravity observed : —
00006 with acids up to l" 170
0-0007 with acids from 1170 to 1-450
00008 ,. „ 1-450 ,. 1-580
0-0009 „ ., 1-580 „ 1-750
0-0010 „ ,, 1-750 „ 1-840
HYDRATES OP SULPHURIC ACID, 187
All observers agree that the solutions of sulphuric acid contain
different hydrates. The literature of this subject is very large,
and we shall here quote only the most important facts (comp. also
Mendelejeffy supra, p. 186, and Pickering, p. 179). A sesqui-
hydrate of sulphuric acid cannot be established as a certain
chemical compound^ but the double hydrate, SO4H2 f HgO, is
known with certainty. It crystallizes from rather more dilute
acid — for instance, acid of 144° Tw. in the cold. It contains
84-48 monohydrate +15-52 water (or 68-97 anhydride -f-3103
water), melts at +8° C, but, owing to superfusion, generally only
solidifies below this temperature (for instance, in the deptii of
winter) ; at 205° to 210° C. it already loses 1 mol. H^O and
leaves ordinary oil of vitriol behind. The crystals form large,
clear, hexagonal columns with six-pointed end-faces. Spec. grav.
1*78 to 1*79. By the crystallization of this hydrate carboys are
often cracked in winter ; acid of 144"" Tw. and the like ought
therefore only to be warehoused in places where the temperature
will not sink too low, for instance below the acid-chambers.
Stronger or weaker aoid can 1)e exposed to the cold of winter
without any danger.
A third hydrate, SO4H2+2H2O, is assumed, because, on
diluting strong vitriol with water down to this point (that is,
corresponding to 73*13 per cent, monohydrate, or 59 70 per cent,
anhydride), the largest contraction, viz. from 100 volumes to
92-14 volumes, takes place. Bourgoin (Bull. Soc. Chem. [2] xii.
p. 433) infers the same from observations on electrolyzing dilute
vitriol. The density of this mixture is variously stated by
different observers : — by Graham at 1*6321 ; by Bineau, 1*665 ;
by Kolb, 1-652 ; by Jacquelain, 1-6746. According to Liebig it
boils at 163° to 170° ; between 193° and 199° it loses one molecule
of water, and is changed into SO2H2 + H2O (Graham).
Pickering (Chem. News, Ix. p. 68) has obtained a hydrate of
the formula HjSO^, 4H2O, containing 57*66 per cent, real
sulphuric acid. It fuses at —25°. By adding a little water or
sulphuric acid the fusing-point is at once lowered to —70°.
The specific gravities given in all the tables refer only to pure
acid, and cannot be accepted as quite correct for the ordinary acid
of trade, which always contains impurities. Kolb has examined
into this matter, and has determined the influence of the common
impurities upon the density of sulphuric acid, viz. that of lead
188 PROPERTIES OP OXIDES AND ACIDS OP SULPHUR.
sulphate/ of the oxygen compounds^ of nitrogen, and of sulphurous
acid. Arsenic, and perhaps iron, usually occur in too small a
quantity in sulphuric acid to influence its density ; but certainly
there may be cases, not mentioned by Kolb, in which sulphuric acid
is strongly contaminated with salts of iron, aluminium, sodium, &c.
The iron, for instance, may come from pyrites-dust ; aluminium
from the packing of the Glover tower, or from the fire-clay
frequently employed for stopping leaks ; sodium from solutions of
nitrate or sulphate of soda, which sometimes inadvertently get into
the chambers.
For saturated solutions of sulphurous acid in sulphuric acid of
varying density, Kolb (Bull. Soc. Ind. de Mulhouse, 1872, p. 224)*
gave a table which has been proved to be incorrect by J. T. Dunn
(Chem. News, xliii. p. 121, and xlv. p. 270). The latter has also
shown that KolVs figures are too low. By passing a current of
pure dry SO2 through sulphuric acid of spec. grav. 1*841 he found
that this acid dissolves : —
Tenii:eratiire.
At 11°-1 C.
Volume at 700
pressure.
33-78 vols.
luillim.
SO,
Spec gray, of
solution at temp.
of experiment.
1-823
„ 16 1
28-86
«l
,. 17-1
28-14
a
„ 26 -9
19-27
>>
1-822
„ 42-0
12-82
if
1-821
„ 50 -9
9-47
it
1-818
,. 62 -3
7-21
99
1-816
., 84-2
4-54
J »
1-809
Dilute acids dissolve the following quantities, at temperatures
varying from 15° to 16°, reduced to 760 millim. pressure : —
Spec. grav. of Absorbs vols,
sulphuric acid. SO^.
1-753 2083
1-626 2517
1-456 5^987
1-257 30-52
1-151 31-82
1-067 3408
• Given in our 2nd edition, p. 127.
SPECIFIC GRATITIES OF IMPURE SULPHURIC ACID.
189
In sach quantities (up to saturation) sulphurous acid certainly
never occurs in commercial vitriol ; and it is very rarely that more
than traces of it are found therein^ since it does not agree with
the nitrogen oxides which are most frequently found in com-
mercial vitriol. Nitric acid is^ if at all^ only present in extremely
small quantities in the sulphuric acid of trade^ and therefore does
not modify its density to a sensible extent ; especially it will not
be found in sulphuric acid of more than 144^ Tw.^ except perhaps
in the nitrous vitriol from the Gay-Lussac towers ; but even this^
according to my analyses (see below) ^ under normal conditions
contains mere traces of NO^H. Nitric oxide can also be neglected;
neither concentrated nor diluted sulphuric acid dissolves more than
mere traces of it. Nitrom acid certainly has a very marked
effect on the apparent percentage of a sulphuric acid^ according
to the hydrometrical test, although only in " nitrous vitriol '' such
large propo rtions of nitrous acid occur as to influence the specific
gi*avity of the sulphuric acid.
Kolb {he. cit. and our 2nd edition, p. 129) has given a table for
the specific gravities of solutions of N^Oa (or rather of SO5NH,
comp. below) in sulphuric acid which is useless, as it is founded
upon erroneous assumptions.
R. Kisling (Chem. Ind. 1886, p. 137) has examined the effect
of a percentage of arsenic on the specific gravity of sulphuric acid.
The specific gravities of two commercial acids, A and B, were
observed at 15° C. and calculated for water of 4** C. and the
vacuum, in order to be comparable with Lunge and Naef's figures
for pure acid (supra, p. 119).
A.
B.
Spec. gray. SO^H.^ , Aafi^ Spec. gray. SOJIj
at 15®. per cent, pet cent. | at 15°. percent.
1-8377
I'8ii87
ivsa^.w
18401)
1-8412
1-8413
1-8414
1-8^15
92-87
93-28
94-25
93-60
93-93
93-77
0137
0-137
0-192
0-258
0 219
0-254
0-231
0-231
1-8367
1-8372
18373
1-8384
1-8386
1-8388
93 82
93-67
9312
93-72
93-96
9404
ASA
per cent.
0024
0-035
0028
0037
0037
0039
190 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
When comparing these results with those of Lunge and Naef^
the considerable influence of the arsenious acid on the specific
gravity of sulphuric acid is very apparent.
With respect to lead sulphate, Kolb found that, at the ordinary
temperature, there were dissolved in
100 parts vitriol of 1*841 spec. grav. 0039 part.
100 „ „ 1-793 „ 0011 ,,
100 „ „ 1-540 ,, 0-003 ;,
In more dilute acids the lead can hardly be estimated. Nitric
acid, which anyhow occurs in very small quantities, does not
strongly influence the solubility of lead sulphate in sulphuric acid,
nitrous acid not at all. The effect of lead sulphate on the
density of vitriol can accordingly be neglected for the ordinary
temperature; at most it w^ould influence the fourth place of
decimals.
Although, as we see, the impurities of ordinary sulphuric acid,
leaving aside " nitrous vitriol,^' have very little effect on its
density, still the latter, at the highest degrees of concentration, is
no trustworthy means of estimating the percentage of real SO4H2
in the acid, even when the correction for temperature mentioned
on p. 186 is applied, because at this concentration a small difference
in density corresponds to a very large difference in percentage.
Many factories have special hydrometers, in which the last few
degrees are spread over a large area and are further subdivided ;
but in fact the density ought to be estimated by more accurate
methods, for the hydrometers are frequently not reliable, and
certainly not so unless the normal temperature for which they
have been made be exactly observed. But any determination of
density for estimating the percentage of the very strongest acids
must be rejected, after what we have sieen on p. 186. The acids
from 96 per cent, upwards ought therefore always to be estimated
alkalimetrically.
The following table of Anthonys will be of practical value. It
shows in column a how many parts of oil of vitriol of 168^ Tw.
must be mixed with 100 parts water at 15^ or 20° in order to
obtain an acid of the specific gravity b.
HELTINQ-POINTS OF SULPHURIC ACIJ).
191
a. ;
1
h.
a.
h.
a.
h.
1 :
1-009
130
1-456
370
1-723 1
2 1
1015
140
1-473
380
1-727
5
1035
150
1-490
390
1-730
10 i
1-060
160
1-510
400
1-733
15 '
1-090
170
1-530
410
1737
20 i
1-113
180
1-543
420
1-740
25 1
1140
190
1-556
1 430
1-743
: so ;
1165
200
1-568
440
1-746
35 ;
1-187
210
1-580
450
1-750
40 ■
1-210
220
1-593
460
1-764
45 !
1-229
230
1-606
1 470
1-757
50 1
1-248
240
1-620
480
1-760
55
1-265
' 250
1-630
490
1-763
60 ;
1-280
260
I 1-640
1 500
1-766
65 :
1-297
270
1 1-648
510
1-768
70 1
1-312
280
1 1-654
520
1-770
75
1-326
290
1-667
630
1-772
80 ;
1-340
300
1-678
540
1-774
a5 '
1-357
310
1-689
1 550
1776
90 ;
1-372
320
, 1-700
1 560
1-777
1)5 '
1-386
330
1-705
580
1-778
100 ;
1-398
340
1-710
590
1-780
1 no ;
1-420
350
1-714
600
1-782
120 i
1-438
360
1-719
1
The melting-points of sulphuric acids of diflferent degrees of
concentration are given by Payen in a table quoted in our 2iid
edition, p. 133.
A new determination by myself (Berl. Ber. 1881, p. 2649) gave
the following results : —
Melting-point.
liquid,
do.
do.
-8-5
+ 4'5
+ 6-5
'+80
Spec. grav. of
acid at 15° 0.
Freezing-point.
1-671 . .
. liquid at -20° C
1-691 . .
do.
1-712 . . .
do.
1 727 . .
- 7-5°
1-782 . .
-8-5
1-749 . . .
-0-2
1-767 . . . .
-I-1-6
1.-790 . .
+ 4-5
1-807 . . .
-90
1-822 . .
. liquid at -20° C
1-840 . .
do.
-6-8
liquid.
do.
192
PROPERTIES OF OXIDES ikND ACIDS OF SULPHUR.
Pickering (J. Chem. Soc. 1890^ Ivii. pp. 331 et seq.) has pub-
lished an extensive memoir on the freezing-points of sulphuric
add and its solutions. Thilo (Journ. Soc. Chem. Ind. 1893,
p. 827) gives very extended tables as to the results obtained by
him in B. Pictet's laboratory. Pictet, himself, subsequently
(Compt. Rend. cxix. p. 642) supplies the following table : —
Pormuln.
SOjH.,
Spec.
1
Freezing-
Formula.
SO,H,
Spec.
Freesing-
per cent.
10000
griiv.
point.
+10°
per cent.
grav.
1196
point.
H»SO,
1-842
H,JS04+ I5U3O
26-63
-340
T +H,0
2HaO
84-48
1-777
+ 3
16 „
25-39
1-187
- 26-5
73-08
1-650
-70 j
18 .,
23-22
1-170
. - 19
4 „
57-65
1-476
-40
20 ..
21-40
1-157
-17
6 ..
47-57
1-376
-50 1
25 ,
17-88
1129
- 8-5
8 ,.
40-50
1-311
-65 ;
50 ..
9-82
1-067
- 3-5
., 10 .
36-25
1-268
-88
75 .
6-77
1045
0
.. 11 .,
3311
1-249
-75
100 ..
516
1-032
+ 2-5
,. 12 ,.
31-21
1-233
-55
300 ,.
1-78
1-007
+ 4-5
.. 13 .,
29-52
1-219
-45
„ 1000 .,
0-54
1-001
+ 0-5
» 14 „
28-00
1
1-207
-40 i
j
1
Knietsch (Ber. 1901, p. 4100) gives the following table of tlie
melting-points of sulphuric acid, ordinary and fuming, from 1 per
cent, to 100 per cent. SO3, which I have supplemented by adding
the corresponding percentages of HjSO^. By '' melting-point *'
he understands the temperature at which the cooled acids in which
crystals had commenced to form remained constant when the
vessel was taken out of the cooling medium during the process of
solidification. In a set of curves appended to the original (com p.
fig. 39, p. 174) he also shows the temperatures at which the
first crystals form and those at which, on cautiously heating,
the last crystals were liquefied. The curves show a decided
maximum near the point H2S04,H20, a minimum at the point
2H2S04,H80, a maximum (nearly coinciding with the first)
at the point H3SO4, a minimum at 4HjS04,S03, a strong
maximum at HsS04jS03 (=pyrosulphuric acid), a minimum at
H2S04,2S03, and the highest maximum for SOs in the polymerized
state.
BOILING-POINTS OF SULPHURIC ACID.
193
S03
H2SO,
Melting-
SO3
1
1 H.^S04 Melting-
SO3
Melting-
p. cent. p. cent..
1
point °C.
- 0-6
- 10
p. cent.
j p. cent. point °0.
p. cent.
point °0.
1
2
]-22 '
2-45;
• • •
••■ 1 below -40°
•• •
82
83
-f 8-2
- 0-8
3 ' 3(57j
- 1-7
61
74-72 -40
84
- 9-2
4 1 4-90
- 20
62
75-95 -20
85
-110
0
612
— 2*7
63
7717 -11-5
86
- 2-2
6
7-35
- 3-6
64
78-40 - 4-8
87
+ 13-5
7
8-57
- 4-4
65
79 02 - 42
88
+260
8
9-80,
- 53
66
80-85 , +1-2
89
+ 34-2
9
1102
- 6-0
67
82-07 + 8-0
90
+34-2
10
1225
- 67
68
83-39 1 + 80
91
+ 25-8
11
13-47
- 7-2
69
84-52 i H- 70
92
+14-2
12
1470
- 7-9
70
85-75
+ 40
93
+ 0-8
13
15-92
- 8-2
71
86-97
- 10
94
+ 4-5
14
1715
~ 90
72
88-20
- 7 2
95
+ 14.8
15
18-37;
- 9-3 ,
73
89-42
- 16-2
96
+20-3
Ifi
19-60
- 9-8 ,
74
90 65
-250
97
+29-2
17
20-82
-11-4
: 75
91-87
-34 0
98
+338
18
2205
-13-2
1 76
93101 ^ -320
99
+36-0
19
23-27
-15-2
1 77
94-83 1
7 -^32
100
+400
20
24-50
-171
' 78
9505
S -16-5
21
25 72 1
-2-2-5
1 79
96-77 i - 5-2
22 26-95 1
-310
! 80
98-00
+ 30
23 ; 2S17 !
-401
81
99-25
H- 7-0
1
1
81-63
100-00
+100
On hoiling dilute sulphuric acid^ at first nothing but aqueous
vapour escapes ; according to Graham^ acid vapour is mixed
with the steam ouly when no more than 2 molecules of water
are present to 1 of SO3 — that is, with a percentage of 84s'48
SO4H2 or a specific gravity of 1*78. After several discussions
about the loss of sulphuric acid in concentrating it, by myself.
Bode, Walter, &c., it may be assumed that in manufacturing
practice no sensible loss of acid takes place by real volatilizalion up
to a strength of 144!° or even of 152° Tw. ; but from violently
boiling acid there is always a little acid carried away mechanically
in the shape of small drops, especially in pans fired from the top
and also in the Glover tower, or in a '' vesicular state." When
the evaporation up to that point takes place quietly at a moderate
heat, there is probably no loss of acid at all.
The boiling-point of sulphuric monohydrateis stated by Marignac
=338^ by Pfaundler =317°. The boiling-points of sulphuric acid
containing water were examined by Dalton in the beginning of
this century. His table, which was obviously wroiig, has been
replaced by another, founded upon the author's investigations
(Berl. Ber. xi. p. 370).
VOL. I. O
194
PROPERTIES OF OXIDES AND ACID8 OF SULPHUR.
Table T.
Specific
Teojpe-
. Spec. grav.
Percent-
BoilJDg-
Barometer
rature.
reduced to
j »ge of
' point.
reduced to (»-:
gravity.
°C.
15° C.
i s6,H,.
' ^C.
raillims.
1-8380
17
1-844K)
95-3
297
718-8
1-8325
16-5
1-8334
92-8
280
72;3-9
1-8240
15-5
1-8242
90-4
264
720-6
1-8130
16
1-8140
88-7
257
7260
1-7985
15-5
1-7990
86-6
241-5
7201
1-7800
15
1-7800
843
228
720-5
1-7545
16
1-7554
81-8
218
726-0
1-7400
15
1-7400
80-6
209
720-6
1-7185
17
1-7203
78-9
203-5
725-9
1-7010
18
1-7037
77-5
m
725-2
1-6760
19
1-6786
75-3
183-5
725-2
1-6690
16
1-6599
73-9
180
725-2
1-6310
17
1-6328
71-5
173
7252
1-6055
17
1-6072
69-5
169
7301
1-5825
15
1-5825
67-2
160
728-8
1-5600
17
1-5617
654
158-5
730-1
1-5420
17
1-5437
64-3
151-5
7301
1-4935
18
l-49(iO
59-4
143
7301
1-4620
17
1-4635
56-4
133
730-1
1-4000
17
1-4015
50-3
124
730-1
1-3540
17
1-3554
45-3 ,
118-5
730-1
1-3180
17
1-3194
41-5 I
115
730-1
1-2620
17
1-2633
34-7
110
7329
1-2030
17
1-2042
27-6
107
732-9
1-1120
17
1-1128
15-8 ,
103-5
732-9
10575
17
1-Oo80
8-5 1
101-5
735()
Table II.
(Calculated by graphical interpolation.)
Percent.
Boiling-
1
Percent.
Boiling-
Percent.
Boiling-
. Percent.
Boiling-
SO^H,.
6
point.
SO,H,.
point.
o
1 118-5
70
point.
1 o
170
; so,H,.
point.
o
101
45
1,
'' 86
238-5
10
102
50
1 124
72
1745
, 88
251-5
! 15
1035
53
1285
74
180-6
'i 90
262-5
20
105
5(;
133
76
189
91
268
25
106-6
60
141-5
78
199
92
274-5
30
108
62-5
147
80
2()7
93
281-5
35
110
. 65
153-5
82
218-5
14
288-5
40
114
67-5
r
161
84
<»7
95
21»5
AQUEOUS-VAPOUR TENSION OF SULPHURIC ACID.
195
The tension of aqueoiM vapour in sulphuric-acid solutions of
various strengths was determined by Regnault in 1845 (Ann.
de Chim. [3] xv. p. 179) for temperatures from 5° to 35° C. We
here give his table (for every other degree), adding to the hydrates
quoted by him the percentage composition and specific gravities.
We also subjoin SorePs table (p. 196), computed for a wider
interval of temperatures, better suited for the wants of sulphuric-
acid manufacture. The tensions are stated in millimetres of
mercurial pressure.
Reguault^s Table of the Aqueous- Vapour Tensions of Dilute
Sulphuric Acid.
H,80,
+5h,o
H,SO,
1
1 H,SO,
«c.
4-3H.,0
' +4H,6
84-5o/o.
731%.
64-5o/„.
57-6" o.
5
1-780.
1-654.
1-554.
1-477.
0105
0-388
0-861
1-2W
7
0108
0-480
0-985
1-510
9
0112
0-476
1125
1-753
11
0118
0-527
1-280
2025
13
0-124
0-586
1-454
2-331
15
0131
0-651
1-648
2674
17
0139
0-725
1-865
3059
19
0149
0-808
2108
3-492
21
0159
0-901
2-380
3-977
23
0171
1-006
2-684
4-523
25
0184
1125
3024
5-135
27
0199
1-258
3-405
5-822
29
0216
1-408
3-83()
6-594
31
0-235
1-557
4-305
7-459
a3
0-256
1-767
4-838
8-432
35
1
1
0-280
1-981
5-432
9524
H.,s04 n.,sOi H2SO4
-|-5H,0 +7H,0 +mfi
521 %. 43-7 -0
1-420 1-340.
'0»
2137
2-464
2-829
3240
3-699
4-215
4-793
5-440
6-366
6-979
7-892
8-914
10-060
11-345
12-785
14-400
3168
3643
4176
4-773
5443
^•194
7036
7-980
9039
10-226
11-557
13-050
14-723
16-600
18-704
21-0(>3
37-7
1-287.
4-120
4-728
6-408
6-166
7013
7-958
9014
10191
11-506
12-974
14-613
16-443
18-485
20-765
23-311
26-152
H3SO, ' H,SO,
-l-llH.,0 +I7H2O
33- 1 %. 24-3 °/o.
1-247. 1176.
4-428
5164
5-980
6-883
7-885
8-995
10-222
11-583
13090
14-760
16-610
18-659
20-920
23-443
26-228
29-314
5-478
6-300
7-216
8237
9-374
10-641
l2-a>4
13-628
15-383
17-338
19-516
21-944
24-650
27-66(i
31025
34-770
Knietsch (/. c. p. 4111) has also determined the vapour-tensions
of sulphuric acid, both ordinary and fuming, at various tempera-
tures up to 100° C. As may be imagined, the aqueous-vapour
pressures decline rapidly with the concentration of the ordinary
acid, and for acids from 90 to 98*6 per cent, they are=0 even at
100°. From this point the vapour-tension, now of course pro-
duced by SOs, rises very rapidly, as is shown by the following
table (p. 197), where the pressure of | vol. fuming acid + i vol.
air is expressed in atmospheres.
o 2
•
o
o
3
1
1
• •
fee*.
o
• •
c :
a ;
9»
183-5 2220
1600 1950
138-5 169-5
118-7 ' 146-0
100-7 125-0
83-7 1060
700 880
9
s
44-4 , 570
33-7 , 43-4
•
I-H
CO
lo
9
fi
00
lb
T-l
9
CO
r-H
•
f-H
CO
00
cb
1
o
8
o
a :
•
•
•
•
•
•
- :
•
f-H
9
in
• •
S i
E :
•
•
•
•
•
•
s
us 00 (?l o p
s ? s s as
IN
00 iO
CO (N
CO ifi
lO
CI
9 Cl
I-H
o
»b
in
1-4
•
i>
•
•
o
o
• •
S :
S :
o
CO
l-H
CO
C& Cq CO
S S g
1— 1
9 9
5? §
CO
u-S P
3 s
f-H
00
r-l
•
•
do
Cl
■
CO
o:>
CO
1
o
• •
s i
■ •
g !
E :
1-H
Si
.-1 »o
3 S
CO
00
■
CO Th i>
S CO S
O •««< ^
g g S
So
"^ CO
Tjl
CI
oa »o
-H i-H
^H
Cl
f-H
9
lb
9
c^
<b
md Water (
o
00
CO CJ
O Cl
1-^ .-J
1-H
•
Cl
•
S :
a :
o
•
I-
a
&
G>
2 "^ ^
3 Th CO
o
<7> T
15 2
lb
«-H
CO
CI
uO
I-H »0
Cl Cd
1-H
lO uO
• •
•p
9
9
00
•
9
9
uo
CO
o
CO
CO
cb
Cl
Cl
CO
Cl
9
Cl
•
1-H
' Sulphuric Acid c
o
o
§
•
S i
a :
• •
S :
ib
I'*
©
•
CO
o
a
•
CO
00
CO O l»
^ s s
l> CI
do »b
1-4 »-^
"*
^
<=> r- <P
1-H CO 1-H
CO G^ C^l
m
1— 1
9 »?
■»ii ^
1— J F-1
»;s p
t-» CO
■
o
• CO
EdD
1-H
•
o
Od
^
»o
•
CO
• • •
^ O CO
3^1 C^ fH
O)
CO
1— 1
O 0)
6 do
CI
l>
Ca QO
■ •
9
CO
Cl
m
f-H
»
rH
o
CO W3 Tt* l-H
§3 S S ^
eo Gi yp Cd
S S^ 55 2
CO
•
1— 1
cb
1— «
>P 30 O
do »b ^
rH i-H ^H
T' ? 9
•^ <M O
»— 1 »-l r-«
lO
•
o
1-1
Cl o
do cb
•
^ CD
00 Cl
do w
f-H
Cl
f-H
Cl
CD
•
rH
•
^H
i
ii
do
CO o
?,
f-H
Cl
•
F-1
CO
•
1-H
•
o .
queotu Vapour in
o
1— t
do
1-1
1— <
1-H
9 r- »P
l-H C» l-*
1-^
cb
ip IP
00
CO
9
00
CO 00
do Cl
•P 9
Cl Cl
Cl
1-H
f-H
■
1-H
I-H
•
1-H
t '
1
1
o
»0
00
o
•
o
\«5
I-H !?1 fH
do t^ cb
o
»b
O U3
«
I-H
00
o
6
1
2-5
s2
»o
•
o
1— t
00
do
m
9
O I-! CO
CO « Tfi
•
CO
p O
CO CI
f-H
CI
op Th
^H ^H
Cl
•
1-H
o
•
t-H
00
■
o
CO
•
o
•
o
^
V
o
?,
o
f-1
•
•
CO
00
CO LC O
^ CO CO
^
w
CI 00
c^ ^
n
CO p
I-H 1— 1
CD
■
■
o
6
00
6
1
o
1
1-^
Ecb
S
•
o
lO
■
o
CO
cb
1-1 CD i-H
w 6i <N
00
•
r-1
CO Tti
Cl
•
o 00
I-H O
CO
■
o
o
CO
•
o
Cl
•
o
I-H
o
1
t^
1^
o
•
CO
CO
CO
•M Ci CO
• • >
CI ^ T-l
«
CI ^
• ■
■
o
00 t>
o o
uo
•
o
•
o
CO
o
Cl
•
o
■ 1
o
-J
Mi4
^
s
s
7^
If:
3
5fl S 8
g
s s
s
s
00
s
Cl
X
1
£1]
1
5-3
CI
^
w
8
§8
8 8 8 8 8 2;
T-H 1-^ 1— 1 rH
00
I-H
FH
a V;
I-H I-H
— 1«
CO
I-H
I-H
Cl
-.it*
r-l
CO
rH
f-H
• •
CO to
•
1— 1
•
■
f-H
X
Ttl
•
1— 1
f
■
C^ p CI
O oD O
-^ TT O
• • •
1-H ^H ,— 1
CI
*
f— I
• •
Cl
Ci
lO
•
• •
»-H I-H
Cl
•
I-H
■
I-H
O
1-H
•
I-H
1-H
SPECIFIC HEAT OF SUIiPHt'RIC ACID.
197
Temp.
°0.
SO3
30 p. c.
SO3
40 p. c.
SO3
50 p. c.
SO3
60 p. c.
SO3
70 p. c.
SO3
80 p. 0.
1
1 SO3
100 p. c.
atm.
atm.
atm.
atm.
atm.
atm.
atm. '
35
• ■ •
• • •
• »•
1
• • •
• • •
0-160
1 0-400
40
• • •
0-075
• • •
0-225^
0-375
0-500
" 0650
45
0050
0125
• • •
0-360 ;
0-576
0-650
0-875
50
0100
0175
0-350
0-525
0-775
0-875
1-200
55
0140
0-225
0-450
0-675
1025
1-2C0
1-600
60
0-200
0-275
ObLO
0-825 '
1-400
1-500
1-850
65
0-225
0-350
0-700
1025
1-650
1-900
2-250
70
0-275
0*400
0-825
1-275
2050
2-300
' 2-725
75
oaio
0-475
1000
1-570
2-525
2-800
3-300
80
0-400
0-575
1150
1-850
3-100
3-500
4000
85
0-450
0-675
1-400
2-150
3-700
• 4175
4-900
90
0530
0-825
1-700
, 2-575
4-400
i 5-050
6-900
95
0-625
0950
2-050
i 3150
5-200
• 6-000
• t*
100
0-730
1-OOU
2-400
1
1 3-700
6000
• ■ ■
1 • • •
The specific heat of monohydrated sulphuric acid at 16^-20° C.
is 0*3315 (waters: 1). Marig^ac states the specific heat of acid
diluted with water as follows : —
HjSO^
+ 5 aq.
=0-5764
10 „
0-7212
15 „
0-7919
25 „
0-8537
50 „
0-9155
100 „
0-9545
200,,
0-9747
400 „
0-9878
The following table, by Bode, is more convenient for practical
use (Zeitschr. f. angew. Chem. 1889, p. 244) : —
Spec, gi-av.
1-842
1774
1-711
1-615
1-530
1-442
1-383
Spec. heat.
0-3315
0-38
0-41
0-45
0-49
0-55
0-60
Spec. grav.
1-020
1-263
1-210
1-162
1-116
1-075
1037
Spec. heat.
0-67
0-73
0-78
0-82
0-87
0-90
0-95
Comp. also Pickering (J. Chem. Soc. Ivii. p. 90).
198
PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
Kuietsch (Berl. Ber. 1901^ p. 4102) gives the speeific heats of
acids of higher concentratiou and fuming acids^ up to 100 per cent,
anhydride. The curve (fig. 39, p. 174)- falls continuously with
the concentration, till 20 per cent. SO3 is reached ; it then rises
again, and at 100 per cent. SO3 reaches the considerable amount
of 0*77. The principal points are : —
Total SO,.
Free SO3.
Sijec. lieat.
1
Total SO3.
Free SO,.
S[)ec. heat.
80
•••
0-3574
1 92
.>t)-45
0-400
82
20
0345
1 94
07-34 .
0-465
84
12-89
0-340
i 90
78-23
0-535
m
23-78
0-340
98
8912
0-050
S8
34-r)7
0-350
la)
100
0-770
90
45-56
0-300
■
Chemical behaviour of Sulphuric Acid.
On mixing oil of vitriol with water a considerable rise of tempe-
rature takes place, water being chemically fixed by the formation
of certain hydrates, as described above. Besides, on mixing
concentrated acid with water, as already mentioned, a not in-
considerable contraction takes place, which must equally lead to an
evolution of heat. But on mixing strong sulphuric acid with snow
excessive cold is produced by the heat becoming latent on the
liquefaction of snow, which considerably exceeds that becoming
free in consequence of the chemical combination. This cold,
however, is only produced when the proportion between acid and
ice does not exceed certain limits : for 1 part of sulphuric acid
there must be 1^ part of snow present ; with less snow there is a
rise of temperature.
Even when more strongly diluted much heat is liberated. Many
observers have worked upon this subject ; but we quote here only
a few. Thomsen (Deutsch. chem. Ges. Ber. iii. p. 496) states
that one gram-molecule (that is 98 grams) H2SO4 gives the
following amounts of heat when combining with x molecules of
water : —
HEAT OF SOLUTION OF SULPHURIC ACID,
199
X,
1 .
6272 metrical heat-units.
9364
}>
w
3 .
. 11108
:j
a
5 . .
, 13082
sy
3}
9 . .
, 14940
jf
yi
19 . .
, 16248
}i
n
49 . .
. 16676
•}
;»
99 .
. 16850
j»
;»
199 . .
. 17056
>*
j»
499 . .
. 17304
a
j»
799 .
. 17632
i)
;>
1599 . .
, 17848
3i
3>
Somewhat higher results were obtained by Pickering (J. Chem.
Soc. Ivii. p. 94).
Knietsch (Berl. Ber. 1901, p. 4103) gives the following figures
of the heat of dissolution, found both in the laboratory and by
large scale experiments with 400 kgs, water (for the values of
fuming acid, see p. 170) : —
SO,
H2S0,
per cent.
per cent.
50
61-25
51
62-48
52
63-70
53
64-93
54
66-15
r^
67-38
5<?
r>8-t50
57
69-83
58
7105
59
72-28
m
73-.7O
61
74-73
02
75-95
63
77-18
(*>4
78-40
♦>5
79-63
m
80-85
Calories.
SO3
per cent.
ir,so,
per cent.
Caloriee.
39
{ 67
82-08
93
41
' 6S
83-30
98
44
m
84-53
103
46-5
70
85-75
108
49
71
86-98
113
51-5
' 72
88-20
119
54
73
8943
126
57
74
90-65
133
59-5
75
91-83
139
62
76
9310
146
6^3
77
94-33
152
m
78
95-55
160
72
79
96-78
168
75
80
9800
178
79
81
99-23
188
Qiyry
8H53 ,
100-00
193
ss
1
The curve both for ordinary and fuming acid runs on quite
steadily, without any breaks, so that the formation of the different
hydrates evidently does not cause any special heat-action.
On account of this considerable evolution of heat, concentrated
200 PROPERTIES or OXIDES AND ACIDS OF SULPHUR.
sulphuric acid and water must always be mixed with care : the
water ought never to be poured into the acid, but the acid run in
a thin jet into the water with constant stirring. On a sudden
mixture, so much heat is liberated at once that the acid may be
raised to the boiling-point and splash about ; and glass vessels are
easily cracked thereby.
The affinity of sulphuric acid for water is also proved by its
great hygroscopicity. Concentrated sulphuric acid is one of the
best means for drying gases ; and it is not only used in this way
for innumerable scientific, but also for some technical purposes —
for instance, in the manufacture of chlorine by Deacon^s process,
where the gaseous mixture, having been deprived of its hydro-
chloric acid by water, is passed through a coke-tower fed with
sulphuric acid, in order to be deprived of its moisture. In the
same way, chlorine gas is dried for the process of utilizing tinned
scrap-iron by treatment with chlorine, which in the dry state does
not act upon iron, but gives with tin anhydrous tin tetrachloride.
Concentrated sulphuric acid acts also upon liquid and solid
bodies by depriving them of water or even splitting off the elements
of the same if the substance only contains the latter, but no readily
formed water. Upon this action, too, a host of scientific and
technical applications of sulphuric acid are founded. In this case
frequently the Bo-called sulphonic acids are formed, generally com-
pounds easily lending themselves to further reactions. Instances
of this are: — the formation of ether by the splitting-up of sulphu-
ric acid, with the intermediate formation of sulphovinic acid ; that
of ethylene on the further splitting-off of water; the preparation
of nitrobenzene, picric acid, nitronaphthalene ; the manufacture of
resorcine and alizarine by the alkaline fusion of the sulphuric acids
of benzene and anthraquinone ; and many other cases.
The charring of many organic substances, such as wood, sugar,
&c., by contact with strong sulphuric acid, proceeds from the
same source. Necessarily this acid, in its concentrated form, must
have an extremely prejudicial effect on the animal body. The
remedy usually applied, viz. burnt magnesia, cannot do much good
when the epithelium of the oesophagus and the stomach are once
destroyed.
The affinity of concentrated sulphuric acid for water is also
shown by the fact that it easily runs over, when kept in open
vessels, by attracting moisture from the air — a fact to be
remembered in the case of balances &c.
DECOMPOSITION OP SULPHURIC ACID. 201
Decompositions of Sulphuric Add. — Some of these have been
roentioned already — for instance, that into anhydride and water by
evaporation. The mixed vapour, on account of the unequal velocity
of diffusion of the two vapours, can be separated to a great extent
into its two constituents, so that at 520° C, in an hour, a residue of
60 per cent, monohydrate and 40 per cent, anhydride, at 445° C.
75 percent, monohydrate and 25 per cent, anhydride was obtained
(Wanklyn and Robinson, Proc. Roy. Soc. xii. p. 507). Perhaps a
process for preparing fuming acid can be founded upon this fact.
Even far below the boiling-point the dissociation begins in the
liquid acid. It has been pointed out that already at 30° to 40° C.
the hydrate begins to give off vapours of anhydride (Marignac) ,
which fact has been confirmed by the exact researches of Dittmar
(Chem. Soc. Joum. [2] vii. p. 446) and Pfaundler & Polt (Zeitschr.
f. Chemie, xiii. p. ^6).
A more thorough decomposition into sulphur dioxide, oxygen, and
water takes place on conducting the vapour of sulphuric acid
through porcelain or platinum tubes filled with bits of porcelain
and heated to a bright red heat. This mode of decomposition has
been recommended by Deville and Debray as a *' cheap '^ plan for
making oxygen ; but it does not seem to have answered, owing to
the insufficient sale of sulphurous acid or its salts ; it was expected
to play a great part in the manufacture of anhydride by Winkler's
process, but even for this purpose it has not been found economical.
On heating with charcoal to 100° or up to 150° C, sulphuric acid
yields carbon dioxide and sulphur dioxide; on boiling with phos-
phorus, sulphur ; on boiling with sulphur, sulphur dioxide ; by the
action of the electric current, hydrogen, oxygen, sulphur, &c. (in
dilute sulphuric acid the electric current merely causes the decom-
position of water).
Sulphuric acid at temperatures below its boiling-point behaves
as the strongest of all acids, and expels all other acids from their
salts when the solubilities &c. allow this; but, inversely, sodium
sulphate is also decomposed by hydrochloric acid. In fact, the
*' avidity " of hydrochloric and nitric acid at ordinary temperatures
much exceeds that of sulphuric acid. Boussingault (Ann. Chim.
Phys. [5] ii. p. 120) has shown that dry hydrochloric-acid gas at a
red heat decomposes the sulphates of sodium, potassium, barium,
strontium, and calcium (see further on). More refractory acids
expel sulphuric acid at higher temperatures — for instance, boric
acid, silica, and phosphoric acid.
202 FROPERTIKS OF OXIDES AND ACIDS OF SULPHUR.
With the bases sulphuric acid forms two principal series of
salts, viz. acid ones, of the formula S02QvYr> and neutral ones, of
the formula SOa<^z:^J. Very frequently it also occurs in basic
salts, rarely in hyperacid salts.
The technical applications of sulphuric acid to a great extent
rest on its great affinity to all bases. Of its salts the acid and
neutral ones are soluble in water, excepting the neutral salts of
barium, strontium, lead, silver, and mercury (in the state of
protoxide), which are little or not at all soluble in water and dilute
acids. Calcium sulphate is sparingly soluble in water. Most
sulphates are insoluble in alcohol. The basic sulphates are mostly
insoluble in water, but soluble in acids. The sulphates incline
very much towards the formation of double salts, of which those
are called alums which contain a combination of univalent and
trivalent (corresponding to a double atom of quadrivalent) metals.
The neutral salts of the alkali-metals, of calcium^ magnesium,
silver, manganese, and ferrosum, the latter only if entirely free
from acid and peroxide (a condition very rarely realized), do not
redden blue litmus-paper, whilst all other soluble sulphates do
this.
On heating to a red heat, only the neutral sulphates of the
alkalies, of the alkaline earths, and of lead remain unchanged. At
a still higher temperature (that of melting iron) the two latter
classes are also decomposed, but the alkaline sulphates are vola-
tilized without a change. Even zinc sulphate and manganous
sulphate are not easily decom])osed. This explains the difficulty
of completely converting blende into oxide of zinc.
On roasting, the decomposable sulphates yield metallic oxides,
sulphur dioxide, and oxygen. They are much more easily split
up on heating by certain additions, such^as coal, iron, &c.
On the metals sulphuric acid acts in a very different way. The
water-decomposing metals in the cold yield nothing but hydrogen
with it ; at high temperatures even zinc and iron already yield
sulphurous acid ; and zinc, if certain conditions are obsen'ed, can
even yield sulphuretted hydrogen (Fordos and Gelis) .
Most of the heavy metals do not act upon sulphuric acid in the
cold and in a dilute state (they yield sulphur dioxide only on
being heated with concentrated acid), such as arsenic, antimony.
BEHAVIOUR OP CAST-IRON TOWARDS SULPHURIC ACID. 203
I)i8muth^ tin, lead, copper, mercury, silver (sulphates being formed
at the same time) ; gold, platinum, iridium, abd rhodium do not
act on sulphuric acid at any temperature.
While pure sulphuric acid has hardly any action whatever on
platinumy even at its boiling-point, some action is exercised by the
impurities never absent from commercial acid. Impure platinum
may also be acted upon to some extent, but the action on gold is niL
We shall go into this subject when treating of the concentration
of sulphuric acid in platinum vessels.
The behaviour of cast-iron towards sulphuric acid is of great
technical importance. It has been known for many years past
that concentrated oil of vitriol acts very little indeed upon cast-
iron, whether hot or cold, provided the access of air is excluded,
the moisture from which would dilute the acid and cause it to act
more strongly. It was, however, at first considered an extremely
bold step when Lancashire alkali-makers began to decompose
common salt with sulphuric acid in cast-iron pots, as is explained
in the Chapter treating of the manufacture ofsaltcake.
Since that period manufacturers have become much bolder, and
for many years past cast-iron vessels have been employed for
*' parting ^' alloys of silver and gold by boiling sulphuric acid, for
making nitrobenzene and analogous products by a mixture of
strong sulphuric and nitric acid, and innumerable other purposes
where strong acids Ifove to be manipulated either hot or cold,
even for the lasit concentration of sulphuric acid itself. Dilute
sulphuric acid, if the dilution be not too strong, acts very little
on cast-iron in ^he cold or at a gentle heat, if air be excluded ; it
can hence be employed for ''acid eggs,'' in which chamber-acid is
forced up, in lieu of pumps, and similar purposes.
It is usually assumed that some descriptions of cast-iron resist
the attar*.k of acids more than others. This point, together with
some others of importance, has been the subject of experiments
made in my laboratory (Chem. Industrie, 1886, p. 74) *.
These experiments lead to the following conclusions : —
(1) At the ordinary temperature all acids down to 106^ Tw.
act very little on all descriptions of cast-iron.
(2) At 100° C. the action is much stronger. It is least in the
* A full report is also given in the second edition of tliis book, p. 141 to
p. 143.
204,
PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
case of acid of 168° Tw., IJ times stronger with acid 142° Tw.,
and 3 times stronger with acid 106° Tw.
(3) At the boiling-point the differences are far more pronounced.
Acid of 168° Tw. acts very little even then, both in the pure state,
or as commercial acid (containing a little ^2^*)> ^^ when
containing a little SO2. But acid of 142° Tw. acts 14 times
stronger at 200° than the same acid at 100° C, and 20 times
stronger than acid of 168° Tw. At 285° C. commercial acid of
142° Tw. does not act very diflFerently from pure acid of the same
strength. Acid of 106° Tw. at its boiling-point ( = 147°C.) acts
less than acid of 142° Tw. at 200° C, but still four times as much
as acid of 168° Tw. at 295° C. There is no diflference between
pure and commercial acid in this case.
(4) The differences between various mixtures of cast-iron are
of no importance against acid of 168° Tw. in all cases, and against
the weaker acids at 20° and 100° C. But the latter acids at their
boiling-point act decidedly less on charcoal-pig and on chilled
cast-iron than on all other kinds. A difference between hot and
cold casting could not be found. The strongest attack was made
on Scotch pig and on mixtures containing such.
In another series of tests we examined the action of mono^
hydrated sulphuric acid on various metals, both on standing 6 days
at 20° and on heating 2 hours to 100° C, always with exclusion
of air. The loss of weight was : —
Loss per cent.
6 days at 2 hours at
20^ 0. 100° C.
Cast-iron 0041 0071
Wrought-iron... 0*175 0313
Copper 2*630 excessive
Lead 3*480 3650
Lo&s in fptiins per square
centimetre of surface.
6 days at
2 hours at
20=0.
100° C.
0062
0*015
0-056
0095
1*115
excessive
1-790
1*847
Wrought-iron is much more acted upon than cast-iron by weaker
acids, but at the ordinary temperature it resists the action of strong
sulphuric acid down to 140^ Tw., and probably even a little below
that strength. Where, however, through the action of moist
atmospheric air, the acid gets more diluted, a very strong action
sets in. Hence the wrought-iron vessels in which sulphuric acid
ACTION OF IRON ON SULPHURIC ACID.
205
is now very generally carried must be protected inside against
any access of air when empty; and at the manholes &c., where
temporary access of air is unavoidable, they are best protected by
a sheet of lead.
Knietsch (Ber. 1901, p. 4109) gives the following table respect-
ing the action of sulphuric acid (ordinary and fuming) of various
strengths on cast-iron, mild steel ('^Flusseisen^'), and puddled iron.
The cast-iron contained 2*787 percent, graphite and 3*55 per cent,
total carbon, the mild steel 0*115 per cent, carbon and the puddled
iron 0'076 per cent. The figures denote the loss of the metal per
superficial metre an hour in grams, after treating with acid for
72 hours at 18°-20° C, with exclusion of air.
H.,80, 1
SO3
Cast-iron.
Mild steel.
Puddled
•
percent
1
per cent.
1
(
iron.
48-8
39-9
0-2177
• •• I
I
■ • «
61-2 ,
50-0
0-1510 '
1
03032
67-7 '
55-3
0-0847
00789
73-4
59-9
00662
• t ■
0 0623
79-7
650
0-1560
• ■ ■
0-1159
83-7
<^-4
01388
• ■ •
01052 1
851
09-5
0-1306
• ■ •
0-1034 '
88-2
720
01636
• • •
0-1417
iK)-6
73-9
01760
• • ■
0-1339
920
752
00933
• • •
0-1040 '
930
76-9
00736
00987
00855 i
943
770
00723
00987
00708
95-4
77-9 ;
01274
00933
0-1209 !
96-8
79-0 ,
01013
00815
00988 1
98-4
; 80-3
00681
00533
0-0655 ;
98-7
' 80-6
00589
00509
00570 i
99-2
1 81-0
00568
00418
00504
9930
; 81-07
0057
0042
0050
1 99-50
! 81-25
0060
0-033
0049
99-77
81-45
0-066
0042
0049 j
100-00
! 81-63
0-087
1 0088
0-076 1
total SO,
free SO,
81-8
0-91
0-201
0393
0-323
8202
200
0-190
0-285
0-514
8-2-28
364
0132
0-441
0-687
82-54
; 4-73
0154
0-956
1-075
1 82-80
1 745
0-151
0-566
1-321 i
1 83-60
i 1017
0079
0-758
1-540
84-20
12-89
0-270
1024
0-892 ;
1 84-62
16-16
0-271
1-400
0-758
8505
18-34
0076
1-988
1-530 ;
86-00
, 23-78
0070
0-245
0-471
88-24
' 34-67
0-043
0033
0-053 :
9007
4556
1
i 0040
0018
0-019
206 PROP£KTI£S OF OXIDES AND ACIDS OF SULPHUR.
Knietsch (/. c. p. 4090) makes the following further remarks on
this subject. Whereas cast-iron vessels are very suitable for the
manufacture of hydrated sulphuric acid^ this is not the case for
fuming acids/ for these, although they corrode the metal but
slightly^ cause it to cracky sometimes quite suddenly, with a
loud report. Evidently the fuming acid penetrates into the pores
of the metal, where it is reduced to SO2 and HjS, with formation
of COo from the carbon of the cast-iron — all of them gases with
somewhat low critical temperatures which produce high tensions in
the interior of the metal.
WroughUiron (and zinc) is somewhat strongly acted upon by
fuming acids up to 27 per cent. SO^ ; this is explained by the
increase of electric conductivity, apparent from the curve No. 5,
fig. 39, p. 174. We notice that the minimum lies at 100 per cent.
H2SO4 : then the conductivity rises sharply, the maximum being
at 10 to 15 per cent, free SOg, whereupon ii sinks just as rapidly
to a minimum. At 27 per cent, free SO3, wrought-iron is again
entirely passive, and apparatus made of it may serve for many years
for the manufacture of high-strength fuming acids.
The action of sulphuric add on lead has been the subject of
many experiments. Calvert and Johnson (Compt. Rend. Ivi.
p. 140) came to the conclusion that lead is all the more acted upon
the purer it is, and that an energetic action only takes place above
the specific gravity of 140° Tw. Similar results were obtained
by Mallard (Bull. Soc. Chem. 1874, ii. p. 114) and Hasenclever.
Bauer (Berl. Ber. 1875, p. 210) found that 50 grams of
strongest oil of vitriol (168° Tw.) with 0*2 gram of pure lead
produced a sensible evolution of gas at 175°, stronger at 196® ; at
230°-240° suddenly all the lead is changed into sulphate, with
formation of SOg, H, and S. Lead containing varying quantities
of bismuth (0*71 or 4 or 10 per cent.) was even more strongly
acted upon, whilst small quantities of antimony and copper make
the lead more resisting j platinum acts in a similar way. Lead
containing 10 per cent, tin behaves like pure lead.
According to experiments made by J. Glover (Chem. News, xlv.
p. 105), pure lead is less acted upon when suspended in a vitriol
chamber than such containing 0*1 to 0*75 per cent, copper, or 0*1
to 0*5 per cent, antimony. N. Cookson (ibid. p. 106) found that
strong acids at a high temperature act more upon lead containing
antimony than upon pure lead, whilst weaker acid at a lower
ACTION or LEAD ON SULPHURIC ACID. 207
temperature acts the other way. Comp. also Mactear (Chem.
News, xli. p. 236).
In the North of England those rolling-mills which roll the sheet-
lead from the vitriol-works supply a special kind of '^chemical lead *'
which is made from the melted-up old chamher-lead, pipes, &c.;
in this case many impurities, especially antimony, from '^ regulus '^^
valves, &c., get into the lead, which are supposed to improve its
quality for acid-chambers.
Napier and Tatlock (Chem. News, xlii. p. 314) found that the
action of sulphuric acid on lead at the ordinary temperature is
accompanied by an evolution of hydrogen, which may cause bulging
out of the closed lead vessels in which the acid is sometimes sent
out for sale.
The experiments of Veley (J. Soc. Chem. Ind. 1891, p. 211),
according to which mixtures of nitrous and nitric acid have a
stronger action on lead than either acid by itself, have no practical
bearing on the behaviour of the acid in vitriol chambers, since if
nitric acid occurs there it is always accompanied by nitrous acid,
and, what is far more important, the immense excess of sulphuric
acid greatly modifies all conditions of the case.
An extensive investigation made by myself together with E.
Schmid has been published in Zsch. f. angew. Chem. 1892, p. 642>
also partly in Journ. Soc. Chem, Ind. 1891, p. 146. I here give
a very brief summary of our results, some of which are of great
practical importance.
1. At higher temperatures the purest lead iu all cases resists
both pure and nitrous sulphuric acid, with or without access of
air, much better than '' regulus metal '^ (82 Pb, 18 Sb) or '' hard
lead'^ (1*8 per cent. Sb), or even soft lead with only 0*2 per cent.
Sb. In the cold, lead with 0*2 per cent. Sb is very slightly superior
to the purest lead ; regulus metal behaves much worse, and hard
lead worst of all.
2. Concentrated nitrous vitriol is always more active than pure
acid. In the case of somewhat dilute acid (spec. grav. 1*72 to
1' 76), nitrous vitriol acts less than pure acid on soft lead and hard
lead, owing to a protective coating of lead sulphate being formed.
If more dilute, the action is again stronger (comp. below). In all
cases nitrous vitriol acts more in the presence than in the absence
of air.
3. Comparing two samples of soft lead, the purer sample was
208 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR,
•
found decidedly better ; even a very slight proportion of bismuth
(0044 per cent.) acts injuriously.
4. It is altogether inadmissible to judge of the resistance of lead
to sulphuric acid from the quantity of the ^a* (hydrogen) evolved.
Soft lead gives off at the ordinary temperature^ after a week's
contact^ much gas ; hard lead^ although losing much more weighty
very little gas (.^^ of the theoretical quantity), probably owing to
galvanic action. But^ on this account, if sulphuric acid is to be
sent out in tightly-closed or soldered-up leaden boxes, they should
not be made of soft, but of hard lead, since otherwise the hydrogen
may bulge up or burst the vessels.
5 (a). Lead containing up to 0*2 per cent, copper (alloys con-
tainiug more copper cannot be homogeneously rolled) is in the cold
acted upon by strong sulphuric acid more than pure lead ; with
nitrous acid there is not much difference. At 100*^ C. all kinds of
acid act in the same way on pure lead and on lead containing copper ;
oonoentrated pure acid rather less than concentrated nitrous vitrol^
but more than nitrous vitriol of spec. grav. 1-72 (comp. No. 2).
At 200^ C, concentrated acid acts alike on pure lead and on lead
containing 0'02 per cent. Cu ; lead containing more copper is
slightly less acted upon by pure acid, but rather more by nitrous
vitriol.
(*) Above 200° (225^ to 25&°) lead alloyed with 1 per cent,
antimony is far more strongly acted upon than pure lead (in the
proportion of 265 to 1 at 225°) j but lead containing 0*2 per cent,
copper resists the acid at 235° much better than pure lead in the
proportion of 1 : 17, and at 255° in the proportion of 1 : 26*5.
6 (a) . Pure soft lead gives no visible evolution of gas with pure
concentrated sulphuric acid up to 220°. From this point more gas-
bubbles are continually given off, and at 260° the lead is momentarily
•dissolved with strong frothing, smell of SO2, and precipitation of
sulphur, the temperature rising to 275°.
(b) The same lead, alloyed with 0*2 per cent, of copper, shows a
-visible evolution of gas only at 260°, regularly increasing up to the
boiling-point (310°), at which the lead is very gradually dissolved.
(c) Soft lead, alloyed with 1 per cent. Sb, gives with sulphuric
acid the first visible gas at 175°, more strongly at 225°, and
between 275° and 280° there is the same turbulent, sudden
solution as in the case of pure, soft lead. [Bauer, Berl. Ber.
1875, p. 210, found similar results; according to him 0*73 per
ACTION OF SULPHURIC ACID ON LEAD. 209
cent, bismvth lowers the temperature of sudden decomposition
from 240° to 160°.]
Hence the purest lead is subject to instantaneous solution by
sulphuric acid at 260°. An addition of 1 per cent. Sb. raises this
temperature only about 20% but 02 per cent. Cu completely
destroys this liability to sudden decomposition.
7. The percentage of oxygen in lead is very slight even in
extreme cases, and does not seem to have any connection with
its liability to be acted upon by acid. But the latter, as may be
imagined, is less when the density of the surface is mechanically
increased.
8. The final considerations in selecting the kind of lead best
suited for constructing apparatus for the manufacture of sulphuric
acid are as follows : —
For vitriol-chambers^ towers^ tanks, pipes, and all other instances
where the temperature can rise but moderately, and certainly never
up to 200°C., the purest soft lead is preferable to every other rfc-
scription of lead, being least acted upon by hot acid, whether dilute
or concentrated, pure or nitrous.
Any sensible proportion of antimony is in nearly all cases
injurious; copper causes at least no improvement. This, of course,
does not apply to those cases where the lead requires an addition
to its tensile strength, nor to that mentioned sub No. 4 of
packages for acid to be closed air-tight. Hence an addition of
about 0*2 per cent, antimony may be useful in the case of apparatus
which is only in contact with cold acid ; but with warm acid even
this percentage is to be avoided.
For very high temperatures, e. g,, the hottest boiling-down pans,
which ought not to be heated above 200° C, but may sometimes
be raised to that point, an addition of 01 to 0*2 per cent, copper is
advantageous, while antimony should be avoided here under all
circumstances (comp. no. 6 b). That percentage of copper has no
action at 200°, but only above 220°; and in the presence of
bismuth it protects the lead from the sudden destruction sometimes
observed.
9. Technical "sulphuric monohydrate^' at 50° C. acts far more
strongly on lead than concentrated sulphuric acid, [The " mono-
hydrate'' employed in our experiments had attracted a little water
and tested only 98'85 per cent. H2SO4 : its action upon lead was
13^ times that of ordinary concentrated acid of 96*5 per cent
VOL. I. P
210 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
H2SO4. Fresh monohydrate of 99*5 or 99*75 per cent, would, no
doubt, have shown even more action.]
10. Nordhausen fuming oil of vitriol acts upon lead much more
strongly than ordinary concentrated acid. Wlien it contains
20 per cent, free SO3 it has 32 times the effect of ordinary acid ;
stronger Nordhausen acid has rather less effect than the 20 per cent,
acid, because a protective coat of lead sulphate is formed. At all
events lead must not be brought into contact with Nordhausen acid.
11. Nitric acid of spec. grav. 1*37 to 142 may be brought into
contact with lead at the ordinary temperature, but no acid of less
strength. Stronger acid than the above acts more powerfully upon
lead^ but no more than concentrated sulphuric acid. Mixtures of
concentrated sulphuric acid and strong nitric acid act very little
indeed upon lead, much less than either strong sulphuric acid or
strong nitric acid by themselves : such' mixtures can be treated in
lead vessels without any hesitation. (Later observations have
shown that this applies only to ordinary temperatures and when
no moisture can be attracted from the air. Hot mixed acids act
strongly on lead.)
12. Mixtures of sulphuric acid and nitroso-sulphuric acid,
partly also containing nitric acid, all of them originally containing
0*1 per cent. N, but by heating to 65° C. brought to the state in
which they can really exist in vitriol-chambers, give the fol-
lowing results : — If a little nitric acid is added to dilute sulphuric
acid, and the mixture is heated, a little HNO3 is volatilized, but
no nitrosyl-sulphuric acid is formed until the concentration has
reached spec grav. 1*5. From this point oxygen escapes, and at
spec grav. 1*768 the whole of the HNOj has vanished, SOgNH
appearing in its stead. Inversely, nitric acid is formed from
nitrous sulphuric acid on diluting it; in the case of prolonged
heating, this evidently takes place not by splitting up into HNOs
and NO, but by absorption of oxygen from the air.
The action of the acid on lead is least just about the lowest
point where the nitrosyl-sulphuric acid is stiU capable of existing.
It increases with its dilution, and in proportion to this, evidently
through the formation of nitric acid, equally with its concen-
tration, and later on rapidly so, the action of stronger sulphuric
acid combining with that of nitrosyl-sulphuric acid and nitric acid.
The minimum action is between spec. grav. 1-6 and 1*6 — that is, just
at that concentration above or below which the acid ought not to be
ACTION OF SULPHUKIC ACID ON LEAD AND ZINC. 211
kqi)t in vitriol-chambers. This proves that it is not rational to
keep the acid in the first chamber too strong (comp. Chap. VII.).
Bis muih is generally (as shoAvn above) considered very injurious
to the resis ting-quality of lead for sulphuric acid. H. O. Hofmann
(Min. Ind. v. p. 398) certainly states that bismuth up to 2 per cent.,
or up to the amount usually found in commercial lead^ does
not affect its resistance in vitriol-chambers^ and that it is far more
important that the surface should be clean and smooth^ to prevent
droplets of condensed acid adhering to the lead. While the
second part must be accepted without contradiction^ the first
(concerning the bismuth) is contrary to every other experience.
In a special case which has come under my notice, two Glover towers
behaved quite differently, one of them going without lead repairs
for 13 years, the other one being damaged before two years were
over. Analysis proved the lead to be of almost identical compo-
sition in both cases ; but the first contained O'OOl, the second
0'012 per cent, bismuth.
According to Junge (Sachs. Jahrb. f. Berg- u. Hiittenwesen,
1895) some acid-makers prefer lead desilverized by the Pattinson
process to that treated by the Parkes process, because the latter
is supposed to contain more zinc ; but Hofmann proves this to be
contrary to facts ; Parkes lead contains less zinc, but more
bismuth, than Pattinson lead. It is not denied that in concen-
trating-pans Pattinson lead stands better than Parkes lead. This
would be explained by the fact that Pattinson lead contains more
copper and less bismuth than Parkes lead (comp. supra, p. 209).
Fluorine sometimes occurs in blende, and the fluorides may
cause trouble, as they are converted into HF in the Glover tower,
and this acid contaminates the sulphuric acid. Trost (Chem. Zeit.
1902, p. 12) asserts that even a very slight quantity of HP in
sulphuric acid causes great wear and tear of the lead of Glover
towers and chambers, not only directly, but by facilitating the
corrosion by sulphuric acid and the nitrogen acids. Most descrip-
tions of blende contain only traces of fluorine, but once he found
7 per cent. (The experiments quoted in the original do not show
any essential difi^erence between pure acids and those containing up
to 2 per cent. HP in their action on lead. Gaseous HF seems to
act much more than that contained in sulphuric acid.)
About the behaviour of zinc towards sulphuric acid, I will only
quote the fact that concentrated acid yields hydrogen, together
p2
212 JPROPERTIBS OF OXIDES AND ACIDS OF SULPHUR.
M'ith hydrogen sulpliide, down to acid of the formula SO4H3,
5 H2O. Acid of the formula SO4H2, 6 HgO yields pure hydrogen
(Muir and Robb^ Chem. News, xlv. p. 70).
nn is not acted upon by acid of the formula SO4HS, 3 H^O.
Behaviour of Sulphurous and Sulphuric- Acid towards the Oxides
of Nitrogen. — The reactions between the oxides and acids of sul-
phur and nitrogen are of extreme importance for the theory of
the sulphuric acid process in general, and for the recovery of
the nitrogen compounds in particular.
The older researches in this field, of Clement-Desormes, Dalton,
Davy, Berzelius, Gay-Lussac, W. Henry, Gaultier de Claubry, De
la Provostaye, A. Rose, Koene, Weltzien, Rebling, and Miiller,
have now merely an historical interest. The modern literature of
this subject begins with the labours of B. Weber, during the years
1862 to 1867, published in the Journ. f. prakt. Chem. Ixxxv.
p. 423 and c. p. 37; Poggendorflfs Annalen, cxxiii. p. 34-1, cxxvii.
p. 543, cxxx. p. 277 ; and partly in Dingler's Polyt. Journal, clxvii.
p. 453. Other very important papers have been published by CI.
A. Winkler (' Researches on the Chemical Processes going on in
the Gay-Lussac Towers,' Freiberg, 1867), by Rammelsberg (Ber.
d. deutsch. chem. Gesellsch. 1872, p. 310), by Michaelis and
Schumann (ib. 1874, p. 1075).
My own researches referring to this subject are found in the
following publications : — 1877 : Berl. Ber. x. pp. 1073 & 1432.
1878 : Berl. Ber. xi. pp. 434 & 1229 \ Dingler's Journal, ccxxviii.
pp. 70, 548, & 553. 1879 : Dingler's Journal, ccxxxiii. p. 63 ;
Berl. Ber. xii. pp. 357 & 1058. 1881 : Berl. Ber. xiv. pp. 2188
6 2196. 1882 : Berl. Ber. xv. pp. 488 & 495. 1884 : Chem.
Ind. 1884, p. 5. 1885 : J. Soc. Chem. Ind. pp. 31 & 447 ; Berl.
Ber. xviii. pp. 1376, 1384, 1391 ; J. Chem. Soc. xlvii. pp. 457 &
405. 1888 : Berl. Ber. xxi. pp. 67 & 3223. 1889 : Zeitsch. f.
angew. Chemie, pp. 265 & 385. 1890 : ibid. p. 195. 1899 : ibid,
p. 393.
Of the different oxides of nitrogen, nitrous oxide, N2O, need
not detain us here. It is very slightly soluble in sulphuric acid,
much less so than in pure water, as, once formed, it is not oxidized
again into NO or higher nitrogen oxides ; it is altogether lost for
the manufacture of sulphuric acid, nor does it form any chemical
compound with that acid.
Nitric oxide, NO, was said by Henry and Plisson to be absorbed
BEHAVIOUR OF NITRIC OXIDE TO SULPHURIC ACID. 213
by oil of vitriol^ if left a loDg time in contact with it, with forma-
tion of nitrous p}'rosu]phuric anhydride (see below) ; but Berzelius^
Gay-Lussac, and many others have long ago refuted this state-
ment^ more especially Winkler (/. c. p. 58). In fact the sulphuric
acid of the absorbing-apparatus cannot retain that portion of the
nitrogen oxides which have been reduced to the state of nitric
oxide ; and from this follows the necessity of an excess of oxygen
in the gas issuing from the chambers^ since only this prevents the
existence of nitric oxide in the same. Small quantities of nitric
oxide may, however, escape oxidation even in the presence of
oxygen, and are hence found in the chamber exit-gases.
The solubility of NO in sulphuric acid, although not nil, is
extremely slight. CI. Winkler already in 1867 showed that it is not
absorbed by strong vitriol. Kolb also made experiments with acids
of varying concentration (Bull. Soc. Indust. Muhl. 1872, p. 225),
and has found that acid of 1*841 does not absorb even traces of
NO; acid of 1*749 to 1*621 merely traces (2 to 6 milligrams to
100 grms. acid) ; acid of r4?26 absorbs 0*017 grm. NO ; acid of
1*327, 0*020 grm. NO to 100 grms. My own experiments (Journ.
Soc. Chem. Ind. 1885, p. 447, and 1886, p. 82 ; also Berl. Ber.
1885, p. 1391, and 1886, p. Ill) show that concentrated O.V.
absorbs per c. c. only 00000593 grm. = 0035 c. c. NO, and acid
of sp. grav. 1*500 only half that quantity.
In the presence of oxygen nitric oxide is absorbed by sulphuric
acid (Bussy, Winkler) ; but then it is really nitrous acid which is
absorbed ; and Winkler was the first to prove that it is precisely the
presence of sulphuric acid which causes the oxidation not to pro-
ceed beyond the formation of nitrous acid, the latter combining
afterwards with the sulphuric acid to form nitroso-sulphuric acid
and water : —
2 S03(OH)2 + N303=2 S02(OH)(ONO) + HjO.
In several of the above-quoted papers I have given clear proofs
of the same fact, viz., that on passing nitric oxide, together with
a very large excess of free oxygen, through strong sulphuric acid,
nothingbut nitroso-sulphuric acid is formed, which means that 2N0
lake up only 1 0; but once out of the range of the acid, imme-
diately above it, the reaction 2N0-f 02=N204 sets in, and this
compound, when afterwards meeting sulphuric acid, yields equal
molecules of nitric acid and of nitroso-sulphuric acid (vide infra).
214 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
Nitrous acid, — Real nitrous acid, HNOj, is not known in the
pure state, only in that of dilute solutions. When dissolving
nitrous anhydride^ NjOg, in water, some nitrous acid is formed
and remains dissolved in the excess of water, but much splits up
according to the formula :
3N02H = 2NO + N03H + H20.
The anhydride, N2O3, also called nitrogen tetroxide, is known
as a dark blue liquid which boils below 0°C. The vapours imme>
diately dissociate for the most part into NO and NO2 (with more
or less N3O4, according to temperature), but a small quantity of
N2O3 evidently exists even in the state of vapour. We shall not
here enter upon the much debated question concerning the exist-
ence of N203in the gaseous state, but merely quote the literature:
Luck (Berl. Ber. 1878, pp. 1232 & 16i3),Vitt (ibid. p. 2188),
Ramsay & Cundall (Journ. Chem. Soc. xlvii. pp. 187, 672), Lunge
(ibid. p. 457 ; Zsch. anorg. Chem. vii. p. 209), and Dixon &
Peterkin (Proc. Chem. Soc, May 1899, p. 115).
Although there is no doubt about the fact that most of the
" nitrous vapours," formerly considered as N2O3 in the state of gas
or vapour, is in reality principally a mixture of nitrogen oxide (NO)
and peroxide (which, for the sake of simplicity, we shall in this case
consider as NO2), with very little N2O8, we must bear in mind the
equally undoubted fact that a mixture of equal molecules of NO
and NO2 behaves in all its reactions towards other chemical com-
pounds exactly as if it were N20g. When passed into an alkaline
solution, it quantitatively yields a nitrite :
2NaOH + NO + N02=2NaN02 + H20.
When brought into contact with concentrated sulphuric acid it is
quantitatively converted into nitroso-sulphuric acid :
2H2SO4 + NO + NO2 = 2SO5NH + H20.
We have, therefore, the right to assume that such a mixture of
equal molecules of NO and NO2 chemically comes to the same thing
as gaseous N2O3; and we shall throughout this book simplify many
of our explanations and discussions by speaking of the above
mixture as N20g, although physically it is only a mixture of NO
and NO2. We are all the more entitled to do this, since there is
no doubt that some N2O3 exists in the gaseous state as such, and
BEHAVIOUR OP NITROUS ANHYDRIDE TO SULPHURIC ACID. 215
since^ according to the law of mass action^ this N2O3 must be
constantly reformed when taken away by some chemical reaction.
I have all the more right to take this line, since everybody
speaks of distilling sulphuric acid or subliming ammonium
chloride, although we know that on distillation nearly all sulphuric
acid is split up into SOg and H2O, which recombine on conden-
sation (comp. p. 172), and ammonium chloride in the state of
vapour is mostly=NH3 + HCl.
Nitrous anhydride {nitrogen trioxide), N2O3, dissolves in sul-
phuric acid, all the more easily when the latter is concentrated,
but also, as we shall see, when it contains a certain amount of
water. The reaction taking place is this :
2 H2SO4 + N203=2 SO5NH -f H2O.
Rammelsberg {/. c.) asserts that when nitrous anhydride is em-
ployed in excess, nitric acid and nitric oxide are also formed :
H2SO4 + 2 N2O3 = SO5N H + HNO3 -I- 2 NO ;
but this reaction only takes place in presence of water, and the
formation of nitric acid and nitric oxide must be regarded as
a secondary reaction, owing to the well-known action of free
nitrous acid on an excess of water. Where there is enough sul-
phuric acid and no excess of water, Rammelsberg's reaction does
not come into play.
The compound formed by the action of nitrous acid on sulphuric
acid of the empirical formula SO5NH has long been known,
both in the solid state, as '' chamber-crystals,'' and dissolved in
an excess of sulphuric acid, as '* nitrous vitriol " ; but its true
composition and nature have only comparatively recently been
elucidated.
The easiest way of preparing the chamber -crystals in a state of
purity is by conducting sulphur dioxide into well-cooled fuming
nitric acid until the whole mass has been converted into a magma,
but not until the nitric acid has been entirely decomposed, and
drying the crystallized mass on a porous slab under a bell-jar over
some oil of vitriol. Obtained in this way, or collected in the con-
necting-pipes of vitriol-chambers or other places where there is a
deficiency of steam, they consist of four-sided prisms or ortho-
rhombic crystals; but generally, when prepared on the small scale,
they appear as a scaly, feather-like, or granular mass, colourless
216 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
and transparent* Their fusing-point is stated by Weltzien=73°,
by Gaultier de Claubry=12(r to 130°; but they are partly
decomposed before fusing^ with evolution of red fumes.
The composition of chamber-crystals was formerly uncertain ;
the question was, in the language of the older chemists^ whether
they were a compound of sulphuric acid with nitrogen peroxide or
with nitrous a^id (nitrogen trioxide). Miiller (Ann. Chem. Pharm.
cxxii. p. 1) still pronounced for the former; but R. Weber proved
in 1862^ and more rigorously in the following yeaf, by estimating
all their constituents according to unexceptionable methods^ that
their formula must be constructed on the second supposition.
His results were as follows : —
Molecular weight. Calculated. Found.
2SO3 160 62-39 6400
N3O3 76 29-92 27-96
H2O 18 7-69 10-50
SOg, N3O3, H2O 254 100-00 102-46
This formula has also been confirmed by Michaelis and Schumann
(Ber. 1874, p. 1075), who at the same time maintained, from the
products of decomposition by phosphorus perchloride, that the com-
pound whose molecular weight has to be halved must be regarded as
OH
nitromlphonic acid, S02<(>tq — that is, as sulphuric acid, one of
whose hydroxyls is replaced by the nitro group, or as nitric acid,
for whose hydroxyl is substituted the sulpho group. This mode
of explaining the constitution of that substance cannot, however,
be strictly maintained. Both from its formation and the decom-
position, it is certain that it does not contain the nitro group
NO2, but the nitroso group NO. Its formula must therefore be
written :
and it must be called nitrosyl sulphate, or, more correctly, nitroso-
sulphuric acid, which means sulphuric acid, one of whose hydrogen
atoms is replaced by the nitroso group, that is the radical of
nitrous acid, NO (OH). It is a mixed anhydride of sulphuric and
NITROSO-SULPHURIC ACID. 217
nitrous acid^ as is proved both by its formation and its decom-
position by water.
There exists also a complete anhydride of nitroso-sulphuric
acid^ of the empirical formula^ NjOs^ 2 SOg, which is rationally
written :
NO\Q
bat this is only formed by mixing liquid sulphur dioxide and nitro-
gen tetroxide in the cold under pressure (Provostaye), or sulphuric
anhydride with dry nitric axide (H. Rose)^or by heating sulphuric
anhydride with nitrogen tetroxide (Weber), none of which reac-
tions are possible in the manufacture of sulphuric acid.
Neither is this the case with the compound produced by R.
Weber (Poggend. Annalen^ clxii. p. 602) by conducting sulphuric
anhydride into the most highly concentrated nitric acid, which
has the empirical formula N20fi, 4 SO3, 3 HgO.
Nitroso-sulphuric acid is formed not merely as above indicated,
but in many other ways. We have already mentioned its
formation from sulphur dioxide and nitric acid :
S02 4- NOsHrr S02(0H) (ONO).
It is also' formed when a mixture of strong sulphuric and nitric
acids is heated, oxygen being evolved (A. Rose) :
H2SO4 + HNO3 = SOa (OH) (ONO) -I- HjO + O.
On the other hand, sulphur dioxide can form that compound even
with the lower oxides of nitrogen, if there is water and (except
with Nj04, where this is unnecessary) oxygen present. In the
perfectly dry state sulphur dioxide does not act on the dry nitro-
gen oxides ; but in presence of the smallest quantity of water
'^ chamber-crystals" are formed, if SOj meets with N2O4, or with
NO or N2O3 and oxygen. Winkler has shown that, in an atmo-
sphere of moist carbon dioxide^ nitrogen trioxide does not form
chamber- crystals with sulphur dioxide, but nitrogen peroxide does
so, and he distinguishes N2O3 and N2O4 in this manner. The
fumes of N2O3, with an excess of SO2 and H2O, but in the absence
of oxygen, give no chamber-crystals at all; they are decolorized,
nitric oxide and sulphuric acid being formed. If oxygen or air
218 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
is admitted^ chamber-crystals instantly appear^ and this is also
the case when nitrogen peroxide meets sulphurous acid in the
presence of water. These observations of Winkler's have been
repeatedly confirmed; but we must now add that what he
called " fumes of N2O3 " is in reality mostly a mixture of equal
molecules of NO and N02^ behaving chemically like N2O3.
As some points had not been entirely cleared up by pre-
vious investigators^ and there were partial contradictions among:
their results^ I undertook a new investigation on the interaction of
sulphur dioxide and nitric oxide, with or without the presence
of water (Berl. Ber. xiv. p. 2196), which led to the following
results : —
1 . DiT NO and SO2 have no action upon one another, be it at the
ordinary temperature, or at 50° or at 100°, provided that air and
moisture are rigorously excluded.
2. NO, SO2, and water act in such a way that as much N2O is
formed as corresponds to the quantity of SO2 originally present.
A reduction down to N could not be established.
3. If NO and SO2 meet in the presence of dilute sulphuric acid,
of spec. grav. 1*455, no reduction of NO to N2O takes place, even
when there is a very large excess of SO2 present, whether the
digestion be carried on for many hours at ordinary temperature or
at 60°. Even with acid of spec. grav. 1*32 no reduction of NO by
SO2 could be established.
4. If NO, SO2, and oxygen meet in the presence of water, a
slight but distinct reduction down to N2O takes place. If, how-
ever, in lieu of water, dilute acid of spec. grav. 1*32 is employed,
no such reduction can be observed.
The bearing of these results on the theory of the chamber-
process will be discussed later on (Chapter VII.) .
A further investigation by myself (Berl. Ber. xviii. p. 1384;
Journ. Chem. Soc. xlvii. p. 465) confirmed the above results. It
was shown that in the dry state nitric oxide combines with
an excess of oxygen to form N2O4 exclusively, or nearly so ;
dry nitric oxide in excess with oxygen yields N2O8 along with
N2O4; in the presence of water, NO with an excess of oxygen is
altogether converted into nitric acid. If, however, NO meets O
in immediate contact with concentrated sulphuric acid, there is
neither N2O4 nor HNO3 formed, even with the greatest excess of
oxygen; oxidation proceeds only to the stage of N2O3, which.
ACTION OF WATER ON CHAMBER-CRYSTALS. 219
however, is not formed in the free state^ but passes into nitroso-
salphuric acid :
2 S04H2+2NO-f 0=2S02(OH)(ONO) + H2O.
Outside the immediate contact with the acid the reaction is
again as before with dry gases, viz. 2N04-02 = N204; that is,
here NO is oxidized to a higher state than within the sulphuric
acid.
A very elaborate investigation of the interaction between
nitrous and sulphurous acid was published by Raschig (Lieb.
Ann. ccxli. pp. 161 et seq.). He found a number of new com-
pounds^ and rectified some of the statements of Fremy and Claus
concerning compounds formerly described by them. He also
discovered a very convenient method of preparing hydroxylamine.
But as nearly all Raschig's experiments were made with alkaline
solutions^ and those which were performed with acid solutions were
made under conditions utterly different from those of a lead
chamber, namely at the freezing-point, we cannot stop to give any
details of his results. Under the just-mentioned circumstances,
apart from NjO and NO, amidosulphonic acid, hydroxylamine,
and ammonia are observed, but only in small quantities; and
above the>low temperatures employed by Raschig the occurrence
of those substances is altogether too uncertain and minimal to be
taken into consideration for our purposes.
Action of water on chamber^cry stats. — ^These crystals are very
deliquescent; they absorb water rapidly from ordinary air. In con-
tact with a little more water, they dissolve quickly with evolution
of heat, much nitric oxide being given off. When introduced
into a large quantity of water, they dissolve without visible evolu-
tion of gas ; but in point of fact nitric oxide is formed as well,
also nitric acid, together with nitrous acid. This has led to many
attempts at explanations, and Rammelsberg and Philipp have
asserted that exactly \ of the nitrogen appears as NO, § as nitrous
acid, and ^ as nitric acid :
16 SO,(OH)(ONO) +9H80=16 SO4HS + 4NO
-f2N02(OH)+5N20a.
But this complicated and very unlikely reaction need not be
assumed at all. Every fact observed in this c<»nnection can be
220 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
quite Simply explained by the following reaction :
S02(0H) (ONO) + Hfi = SO4HJ + NO(OH) ;
that is, nitroso-sulpuric acid takes up the elements of water, to
form sulphuric acid and nitrous acid ; the latter^ as is well known,
is unstable in .the presence of an excess of water, and hence
partly splits up into nitric oxide and nitric acid :
In the presence of less water, nitrous anhydride can be formed
from chamber-crystals, and escapes in the shape of brown fumes
(of course mostly dissociated into nitric oxide and peroxide) :
2S02(OH)(ONO) 4-H20=2S04H2 + N208-
For nearly every purpose nitroso-sulphuric acid or its solution
in sulphuric acid may be regarded as a solution of nitrous acid in
sulphuric acid.
The behaviovr of nitroso-sulphuric add towards sulphuric acid of
various concentrations is of great interest for our purposes. In
concentrated oil of vitriol the crystals dissolve easily and without
decomposition. This solution is stable enough to be distilled
without losing any nitrous acid, whilst the isolated crystals are
decomposed on being gently heated. I have shown (Zeitsch. f.
angew. Chemie, 1888, p. 661, and 1890, p. 447) that on distilling
such a solution for four hours, when 40 per cent, of the sulphuric
acid had passed over, the distillate contained only 5 per cent., the
residue 95 per cent, of the nitrous acid, none of it having been
destroyed. It is possible to obtain solutions of 1*9 sp. gr. ; they
evolve with water nitric oxide, inflame phosphorus at 62° C, oxidize
sulphur and many metals on distillation with evolution of NO ;
heated with ammonium sulphate to 160° they evolve nitrogen gas.
Sulphur dioxide evolves nitric oxide; but a solution of nitroso-
sulphuric acid in strong oil of vitriol (of 170° Tw.), even ou long-
continued treatment with dry sulphur dioxide, is only incompletely
decomposed, and on addition of water still shows the presence of
nitious acid by the evolution of brown vapours. This explains
the fact (well known to manufacturers) that concentrated sulphuric
acid contaminated by nitrous acid is only with difficulty purified
by sulphur dioxide. At a higher temperature sulphur dioxide
decomposes chamber-crystals with evolution of nitrous oxide
(Fremy). Further statements respecting the behaviour of sulphur
ACTION OF NITROGEN PEROXIDE ON SULPHURIC ACID. 221
dioxide towards the solution of chamber-crystals in sulphuric acid^
the so-called '^ nitrous vitriol/^ will be made when examining the
process going on within the Glover tower. It is remarkable^ and o£
great importance for the practice of sulphuric-acid making, that
even dilute acids of 170 down to 1*55 sp. gr. dissolve the crystals
in the cold without decomposition; the decomposition only
commences when the specific gravity of the dilute acid has fallen
below 1*55 — that is, below the density of ordinary chamber-acid.
Nitrogen peroxide^, whether in the state of a liquid or a gas,
strongly acts on sulphuric acid. If, according to Weber, nitrogen
peroxide, made by gently heating fuming nitric acid [and there-
fore not quite free from nitric acid] , be added to sulphuric acid of
different degrees of concentration, the following is observed: —
The strongest oil of vitriol, down to 1*7 sp. gr., absorbs the
nitrogen peroxide without coloration. Acid of 1*55 turns yellow;
here the nitrogen peroxide is probably absorbed to a large extent
similarly as by nitric acid, and no decomposition, as represented
by the equation on page 223, has taken place, whilst this has to
be assumed in the case of the stronger acids. Acid of 1*49 turns
greenish yellow; of 1*41 intensely green; acid of 1*31 turns blue
and evolves nitric oxide, which on applying a gentle heat escapes
with violent eflfervescence. Weak acids are only coloured for a
short time. Prom this Weber inferred : — that acids of 1*8 to 1*7
combine with nitrogen peroxide with formation of nitroso-sulphuric
acid ; weaker acids simply absorb it ; and the more dilute acids
decompose it with formation of nitric oxide, nitrous acid, and nitric
acid. The action of sulphurous acid on these mixtures is different
according to their concentration. As mentioned above, the solu-
tion of chamber-crystals in concentrated sulphuric acid is but
incompletely decomposed even by a prolonged action of sulphurous^
* We shall generally give this name to the compound formerly called
" hyponitric acid " and now sometimes " nitrogen tetroxide " or " nitrogen
dioxide. '^ It is well known that in the gaseous state it consists of a mixture of
molecules of NO, and N^O^, of varying proportions, according to the tempera-
tui-e. At a low temperature, especially in the liquid state, it is SN2O4; above
140^'= NO^. The following are the intermediate proportions : —
At 2(3-7 '^ C 80 per cent. N^O^ 20 per cent. NO,.
„ 39*8 „ 71 „ „ 20 „ „
„ 60-2 „ 47 „ „ 63 „
„ 80-6 „ 23 „ „ 77 „
V
222 PROPERTIES OP OXIDES AND ACIDS OP SULPHUR.
acid ; but the yellow mixture of 1*55 sp. gr. and the coloured, more
dilute, acids are decomposed with strong effervescence of nitric
oxide. It will be shown afterwards what part all these reactions
play in the recovery of the nitrous gas in the manufacture, where
the object is first to absorb the gas in sulphuric acid of 1*7, and
then again to liberate it from that solution.
Winkler gave a different account of the behaviour of liquid
nitrogen peroxide from that of Weber. He stated that it may
be mixed with sulphuric acids down to 142° Tw., but that it
yields a solution totally different from that of chamber-crystals in
sulphuric acid, viz., one of a yellow colour and constantly evolving
red fumes. On heating, it effervesces and gives off streams of
gaseous nitrogen peroxide ; if the mixture is made with sulphuric
acid of 142° Tw., the NO2 completely volatilizes far below the
boiling-point of sulphuric acid, so that the residue on dilution
with water does not decolorize potassium permanganate. If, how-
ever, acid of 170° Tw. has been employed, the liquid on heating
certainly yields up the larger portion of its NOj ; but the residue
behaves like a solution of chamber-crystals in sulphuric acid, and
on being mixed with water it evolves red fumes which can be
proved to be NjOs, not NO2, by their not forming any chamber-
erystals with moist SO2.
There are some essential differences between the statements of
Weber and those of Wiiikler, more especially so far as the beha-
viour of nitrogen peroxide is concerned, which were cleared up by
my own researches (see below) .
If concentrated sulphuric acid is mixed with a little concentrated
nitric acid, and sulphur dioxide is passed into the mixture, the
nitric acid in the cold is only reduced to nitrous acid, which
remains combined with the sulphuric acid : this compound resists
the further action of the sulphur dioxide, similar to the solution
prepared from concentrated oil of vitriol and chamber-crystals.
On the other hand, more dilute mixtures of sulphuric and nitric
acid, below 1*7 spec, grav., are more or less easily decomposed by
SO2, in the ratio of their dilution.
Since the labours of Weber and Winkler did not in all points
agree with one another, and the subject seemed to call for
another investigation, I undertook a long research (Dingler's
Journal, ccxxxiii. p. 63), the conclusions of which (also published
in the Berl. Her. xii. p. 1058) are as follows : —
BEHAVIOUR OF NITROGEN OXIDES TO SULPHURIC ACID. 223
1. Nitrogen peroxide, under ordinary circumstances, cannot
exist in contact with sulphuric acid, but at once splits up into
nitrous acid, which, with a portion of the sulphuric acid, yields
nitroso-sulphuric acid and nitric acid (dissolving as such), thus : —
NA + S0,(0H), = S02(0H)(0N0) +N08H.
2. Nitroso-sulphuric acid, on dissolving in an excess of sulphuric
acid, forms a colourless liquid, but only up to a certain limit of
saturation, which is all the higher the more concentrated
the sulphuric acid. This limit for acid of sp. gr. 1*84 is not yet
reached at 55-34 milligrams N208=185 millig. S02(0H)(0N0)
in 1 cubic centim. of acid.
3. Beyond that limit at first a yellowish tint appears, of course
with stronger acids only when more nitroso-sulphuric acid is pre-
sent than with weaker acids. This took place with a mixture of
sp. gr. 1*887 (made from pure sulphuric acid of sp. gr. 1'84),
containing in 1 cub. centim. 147 milligrams Nj03=372 milligr.
S02(0H)(0N0), and also with acid of sp. gr. 1*706, containing in
1 cub. centim. only 56*7 milligr. N2O3=190 milligr. SOsCOH)
(ONO). Since these acids also are rendered colourless by pro-
longed boiling, the excess of nitroso-sulphuric acid seems to be
rather loosely held; but the temperature of the water-bath is not
sufScient to affect it.
4. The phenomenon observed by Winkler, a mixture of strong
vitriol and nitrogen peroxide showing an orange-colour even when
cold, emitting red vapours, and exhibiting a tempestuous evolution
of nitrogen peroxide on being gently heated (which proves the
existence of unchanged nitrogen peroxide), can evidently take
place only when the mixture contains far more N^O^ than the
strongest mentioned above, or the strongest ever occurring in
vitriol-works under any circumstances. Many experiments of
heating in the water-bath for a prolonged period demonstrate the
absence of free Ns04 in all cases observed. Still less can the
presence of nitrogen peroxide be assumed in more dilute acids ; it
is therefore inadmissible to cite it as such in analyses.
5. All nitrous vitriols, t. e, solutions of nitroso-sulphuric acid
in sulphuric acid, whether they contain nitric acid at the same
time or not, on being heated far below their boiling-point assume
a golden-yellow or even darker yellow colour, but entirely lose it
again on cooling. This change of colours may be repeated any
224 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
number of times. It hardly indicates a loosening of the com-
bination^ since this proves to be very stable even at much higher
temperatures ; but it may rather be compared to the deeper colour
which ferric-chloride solutions assume on being heated.
6. The stability of nitroso-sulphuric acid in its solution in sul-
phuric acid is very great^ even at the boiling-point^ providing
the specific gravity is not below 1'70. It is true that on boiling
it some nitrogen is always lost^ and all the more the less con-
centrated the acid is ; but if the boiling takes place so that the
vapour cannot condense and flow back^ there is some nitroso-
sidphuric acid found in the residue, even from acid of sp. gr. 1'65
(comp. p. 221) . But if the vapour is condensed and the condensing
liquid (which, in the case of vitriol of sp. gr. 1*80 or below, consists
of very dilute acid or almost pure water) is allowed to flow back,
a considerable loss is caused by denitration.
7. Down to a concentration of sp. gr. 1*65 the affinity of sul-
phuric acid for nitrous acid, t. e. the tendency to the formation of
nitroso-sulphuric acid, is so great that any nitric acid present at
the same time, whether added as such or formed by the decompo-
sition of nitrogen peroxide, is reduced with loss of oxygen, and
employed to form nitroso-sulphuric acid. In the case of acid of
sp. gr. 1*71 and upwards, this transformation takes place almost
completely after a brief boiling, but at sp. gr. 1*65 only incom-
pletely. This is a further argument against the existence of N2O4
in the solution.
8. Below sp. gr. 1*65 the nitroso-sulphuric acid possesses so
little stability that, for instance, from acid of sp. gr. 1*60 some
nitrogen oxides (but only a very small percentage) are expelled in
the water-bath, and nearly all of them by boiling for a short time.
In the case of acid of sp. gr. 1*5, it is evident that, even without
heating, the nitrous acid added is partly decomposed into nitric acid
and nitric oxide ; but after heating for an hour in the water-bath
a considerable quantity of nitroso-sulphuric acid remains undecom-
posed, whilst another portion has been converted into sulphuric
acid. In the case of still weaker acids, of course these phenomena
occur even to a greater extent ; but it is Yery probable that even
very dilute sulphuric acid may contain, while cold, a little nitroso-
sulphuric acid if reducing-agents are absent.
9. Most of the nitric acid present along with nitroso-sulphuric
acid in dilute acids (of sp. gr. 1*5 and under) remains behind in
BEHAVIOUR OF NITROGEN PEROXIDE TO SULPHURIC ACID. 225
the liquid even after prolonged boiling. If, therefore^ the nitrous
yitriol of acid-works, in consequence of a faulty process, contains
nitric along with nitrous acid, it cannot possibly be completely
denitrated by hot water or steam, in which case a less strength
than sp. gr. 1*5 is never reached ; the denitration can be only
effected by reducing-agents, such as sulphur dioxide in the Glover
tower or mercury in the nitrometer. In the latter it can be very
clearly seen with how much more difficulty and slowness the deni-
tration goes on in the presence of nitric acid.
10. The tendency to form nitroso-sulphuric acid is so strong
that even if a large quantity of air (oxygen) is passed through
sulphuric acid along with nitrons acid no oxidation to N3O4 or
N3O0 takes place, just as in the case of oxygen and NO.
11. Nitrous acid cannot be absorbed by caustic-soda solution
without loss, because a portion of it is decomposed into nitric acid
and nitric oxide.
12. The purple colour which is developed in nitrous vitriol by
the action of redacing-agents is caused by a solution of nitric
oxide in such acids, and is possibly produced by a very unstable
compound of nitrogen and oxygen, midway between NO and
N,0«.
Although my experiments had decidedly proved (comp. No. 1
and 4 of the just-quoted conclusions) that nitrogen peroxide does
not dissolve as such in sulphuric acid, with formation of an
unstable solution from which the N2O4 can be drawn off by heat-
ing, the former erroneous assertion of Winkler (since that time
recognized as such by himself) did not vanish from chemical
literature, and, for instance, gave rise to a decidedly erroneous
explanation of the process of Lasne and Benker for treating the
absorption in the Gay-Lussac tower. This caused me to investi-
gate the subject once more (Berl. Ber. xv. p. 488). I pointed out
that mixtures of pure nitrogen peroxide with even somewhat dilute
sulphuric acid, down to spec. grav. 1*65, behave quantitatively as
mixed solutions of equal molecules of nitroso-sulphuric acid and
nitric acids ; that on prolonged heating in a watei'-bath such solu-
tions in acid of spec. grav. 1*75 do not lose any, and in acid of
spec. grav. 1*65 only very little, of their nitrogen compounds. On
prolonged boiling part of the latter escapes, but a large quantity
of nitroso-sulphuhc acid remains behind, more than that originally
present, part of the nitric acid having passed into it with loss of
VOL. I. Q
226 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
oxygen. The idea of a " loose '^ union between NgO^ and
sulphuric acid must therefore be entirely abandoned ; and from
this follows the fallacy of the idea held by some manufacturers
that N2O4 is less easily absorbed by sulphuric acid than N2O3,
and that therefore much N2O4 is lost in the Gay-Lussac tower.
I directly disproved this idea by showing that vapours of nitrogen
peroxide are most easily, quickly, and completely absorbed by
sulphuric acid of spec. grav. 1-71, such as is used in the Gay-
Lussac tower, and that this (colourless) solution is not changed
either by long heating to 100° or by passing a current of air for a
long time through it.
A concluding investigation on the behaviour of nitrogen per-
oxide towards sulphuric acid has been made by myself together
with Weintraub (Zsch. angew. Ch. 1899, p. 393), of which I here
give merely a summary of the results.
1. The reaction between sulphuric acid and nitrogen peroxide
is reversible, since the nitric acid formed has some action on
nitroso-sulphuric acid, forming sulphuric acid and nitrogen
peroxide :
H2SO4 + NA Z SO5NH + HNO3.
In mixtures of sulphuric acid and nitrogen peroxide an equili-
brium is formed, all four substances — sulphuric acid, nitrogen
peroxide, nitroso-sulphuric acid, and nitric acid — being present at
the same time. 2. In contact with concentrated sulphuric acid
(95 per cent. H2SO4) nearly all the nitrogen peroxide is converted
into nitroso-sulphuric and nitric acid. The inverse reaction sets
in to a sensible extent only when very little sulphuric acid is present
in comparison with nitric acid, 3. The affinity of sulphuric acid
for nitrogen peroxide quickly decreases with the increase of water,
so that in the case of sulphuric acid of spec. grav. 1*65 the action
of nitric acid on nitroso-sulphuric acid begins to prevail ; there-
fore very much of the nitrogen peroxide added remains in the free
state, although the quantity of HNO3, which is only formed by
the reaction itself, is but small. 4. In the practice of sulphuric-
acid manufacture, the quantity of sulphuric acid in the con-
centrated state so largely prevails over that of the nitric acid, that
all nitrogen peroxide may be practically regarded as quantita-
tively changed into SO3NH and HNO3. This, of course, also
holds good when absorbing nitrous gases in concentrated sulphuric
acid for analytical purposes. Therefore the conclusions No. 1
REDUCTION OP NITROSULPHURIC ACID. 227
and No. 4 (p. 223), although not mathematically exact, are to all
intents and purposes valid.
77i€ tension of nitrous acid in presence of dilute sulphuric acid
at different temperatures is a matter of great importance for the
theory of the formation of sulphuric acid in the lead-chambers.
The first observations on this point were published by Sorel
(Zsch. angew. Ch. 1889, p. 272) ; but these have become obsolete
by the far more extended observations published by myself in the
Zeitsch. f. angew. Chemie, 1891, pp. 37 et seq. The tables con-
structed from these indicate the loss of N2O3 suflfered by acids
of four different concentrations, containing quantities of N2O3
varying from 1 gram per litre upwards, in a current of air" at
temperatures from 50° to 90° C. (pp. 228-231) .
The behaviour of nitroso-sulphuric acid towards reducing
reagents is of the greatest importance, both for the chamber-
process in general and for the recovery of nitrogen compounds.
The most important of the reagents in question is sulphur dioxide,
which acts as follows : —
2SO2(0H)(ONO)+SO2+2H2O = 3SOJl3 + 2NO;
that is, it forms with nitroso-sulphuric acid both sulphuric acid
and nitric oxide. This is the leading reaction of the Glover tower,
as we shall see hereafter; and it must also occur within the
chambers, more especially in the first part of the set.
Sorel (Zeitsch. f. angew. Ch. 1888, p. 273) has shown that if a
mixture of SO2 and O is made to react upon nitrous sulphuric acid
and NO, there may be either a reduction of nitrous acid to NO
or an oxidation of NO to N0O3 (i^ the shape of SO5NH), accord-
ing to variations in the following conditions : temperature, dilution
of the acid, proportion between SOj and O, percentage of NO.
The extreme cases were well known before : a reduction takes
place at high temperatures, with scarcity of oxygen and excess of
water ; an oxidation with excess of oxygen, concentrated acid, and
low temperatures. For the intermediate cases Sorel made some
special experiments, from which it followed that in identical
mixtures an increase of the temperature from 70° C. to 80° C.
was suflScient to change the oxidation into reduction. At equal
temperatures a reduction took place when the gaseous mixture
contained 31 per cent. SO2, 10 per cent. O, 59 per cent. N, but
an oxidation with a mixture of 21 SO,, 12*1 O, 66*9 N, &c. The
q2
228
PROPERTIES OF OXIDES AND ACIDS OP SULPHUR.
1. Nitrous vitriol of spec, grav. 1*720 (say, 78 per cent. H2SO4).
1
N^O 3 originally '
Loss of N2O3 in grams per litre at
present.
Grams in 1 litre.
1
50°.
60°.
70°.
80°.
W. 1
1
■ ■ ■
• ■ •
...
0006
2
• • •
■ • ■
• • •
0018
3
■ • •
• • •
• • •
0025 '
4
• ■ •
■ • •
• • ■
0031
r>
• * •
■ ■ •
• • •
0087
6
• ■ •
» • »
• « ■
0-0i3 ;
7
■ • •
• ••
0006
0-056
8
• • •
• ■ •
0010
0-068
9
• • •
• • m
0012
0081
10
• ■ ■
• • ■
0018
0-093
11
• ■ •
• • •
0-025
0112
12
• ■ •
» • •
0030
0125 1
13
...
• ■ •
0-031
0-143
14
• • •
• ■ •
0043
0168 '
15
• • •
0006
0056
0193
16
• « •
0-010
0068
0-218
17
• ■ ■
0-006
0087
0250
18
• • •
0012
0106
0-281 i
19
• • •
0-025
0125
0-318 1
'JO
« • ■
0-031
0150
0-356 .
21
• • •
0043
0175
0-400 1
n
• ■ •
0-064
0-200
0-450
23
• • •
o-6o6
0081
0-237
0-500 1
24
0006
0018
0-100
0-275
0-550 1
25
0012
0-031
0125
0-312
0-600 ,
26
0018
0-043
0150
0-3.56
0-662 .
27
0031
0-062
0-181
0-400
0-725 I
28
0-043
0081
0-212
0-450
0-800
29
0062
0-100
0-25(3
0-500
0-850
30
0081
0-125
0-293
0-550
0-956
31
0093
0162
0-337
0-612
1-043
32
0112 '
0200
0-387
0-641
1-125
33
0125
0-237
0-391
0-743
1-206 i
34
0-143
0-275
0-476
0-806
1-287 1
35
0156
0-312
0-525
0-868
1-375 1
36
0175
0-350
0-575
0-931
1-456 1
37
0193
0381
0-618
1-000
1543 '
38
0-206
0-418
0-662
1062
1-625 •
39
0-237
0-456
0-718
1-125
1-712 ,
40
0'2(i8
0-500
0-775
M93
1-800 '
41
0-293
0-543
0-831
1*256
1-890
42
0-325
0-587
0-887
1-331
1-975 .
43
0-350
0-631
0-937
1-400
2062
44
0-376
0-675
0-993
1-468
2-150
45
0-406
0-712
1-050
1-537
2-237 '
46
0437
0-756
1106
l-60(i
2-3-25
47
0-462
0-800
1-162
1-675
2-392
48
0-493
0-837
1-218
1-743
2-500
49
0518
0-881
1-268
1806
2-587
50
0-550
0-931
1-325
1
1-875
2-675
1
TENSION OF NITHOUS ACID IN SULPHURIC ACID.
229
2. Acid of spec. grav. 1-686 (say, 76 per cent. H2SO4).
N2O3 originallj
Lots of NjOs in grams
per litre at
1 present.
G^faiiiB in 1 litre.
50^
60
mm ^m ^w —
' 70".
' 80°.
90**.
8
• • m
1
• a
■ ■ •
« ■ •
0025
9
m m
• ■
■ a ■
• • ■
0036
10
» •
• ■
• • •
0-012
0050
11
^ -
• •
• • •
0018
0062
12
■ •
• •
• • •
0-025
0086
13
• ■
* ■
0-010
0050
0125
14
• ■
1
1
0012
0-075
0-162
15
• 1
• • «
0025
0100
0-225
16
• •
• ■ i
0050
0-150
0-286
17
• • 4
0010
0062
0-200
0-350
18
• m M
0-012
0-100
0-262
0-436
19
■ • •
0025
0-150
0-350
0-525
20
• • ■
0-060
0186
0425
0-625
21
• • ■
0075
0-250
0-525
0-750
22
■ ■ •
0112
0-300
0-650
0-975
23
• * •
0136
0-350
0-775
1-200
24
• • ■
0175
0-400
0-900
1-436
20
O-OIO
0-200
0-462
1-025
1-662
26
0-012
0-236
0-512
1175
1-900
27
0025
0-262
C-562
1-300
2-125
28
0036
0-800
0-612
1-436
2-350
29
0-050
0-336
0-675
1-575
2-600
30
0-062
0-362
0-750
1-700
2-812
31 ;
0-100
0-412
0-850
1-812
3086
:f2
0136
0-462
0-950
1-975
3-350
33
0-186
0-512
1050
2100
3-625
34
0-225
0575
1150
2-236
3-900
;i6
0-275
0-625
1-250
2-362
4175
36
0-312
0-675
1-336
2-500
4-450
37
0-350
0-725
1-436
2-625
4-736
;«
0-400
0525
1-536
2762
5000
39
0-436
0-836
1-636
! 2-900
5-275
40
0-486
0-886
1-736
3025
5-550
41
0-550
0-950
1-850
3-160
5-850
42
0-612
1050*
1-986
3-275 ,
6125
43
0-686
1125
2112
3-412
6-400
44
0-750
1-212
2-250
3-525
0-700
45
0-825
1-300
2-325
3-736
6975
46
0-886
1-386
2-500
3-825
7-250
47
0-962
1-475
2-636
3-962
7-536
48
1025
1-512
2-762
4-100
7-825
49
1100
1-650
I
2-886
4-236
8-100
230
PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
3. Acid of spec. grav. 1*633 (say, 71*5 per cent. H2SO4).
NjOj originally
present.
Grams in 1 litre.
1
2
3
4
f)
6
7
8
9
10
11
12
13
14
15
Ifi
17
18
19
20
21
22
23
24
25
26
27
28
29
80
31
32
50°.
0012
0-050
0-100
01H2
0-212
0-300
0-400
0-500
0-600
0-700
0-800
0-871
0-986
1-086
1-186
1-275
1-375
1-475
1-562
1-662
1-812
1-975
2-18(5
2-436
2-700
3-000
3-312
3-6(>2
4-025
4-412
4-800
5-23C>
Lobs of N.^Og in grams per litre at
60°.
0-025
0-075
0150
0-212
0-300
0-425
0-562
0-712
0-835
0-936
1125
1-262
1-400
1-536
1-675
1-800
1-936
2-036
2-250
2-412
2-612
2-812
3062
3-336
3-(i36
3-950
4-300
4-636
5-012
5-412
6-836
6-325
70°.
0036
0-08(»
0186
0-300
0-436
0-636
0-836
i-a%
123(>
1-436
1-636
1-825
2-03(i
2-225
2-412
2-612
2-786
2-975
3-18(i
3-400
3-650
3-912
4-250
4-612
5-000
5-412
5-850
6-325
6-812
7-350
7-950
8-575
80°.
0100
0-225
0-350
0-536
0-736
1000
1-275
1-550
1-812
2-086
2-350
2-636
2-900
3175
3-450
3725
4-000
4-262
4-550
4-85(»
5-162
6-512
6912
6-350
6-800
7-300
7-812
8-400
8-025
9-675
10-386
11175
90°.
0150
0-300
0-436
0-612
0-825
1-112
1-425
1-750
2-062
2-375
2-712
3075
3436
3-800
4-162
4-512
4-886
5-236
6-600
6-986
6-400
6-862
7-425
803(1
8-662
9-362
10-150
11-000
11-986
13126
14-500
16-362
reaction of SO2 on nitrous sulphuric acid is nothing like so simple
as previously assumed. If the acid exceeds the strength of 1 '630,
the SO2 does not reduce the NjOs to NO, but forms with it and
sulphuric acid nitroso-sulphuric acid, so long as there is oxygen
in excess and the atmosphere contains more N2O3 than corresponds
to the tension of the acid in question at that special temperature
(comp. above). Otherwise reduction to NO takes place. Acids
below spec. grav. 1'600 are able to fix N2O3 under the same con-
ditions, but only at comparatively low temperatures; at higher
temperatures there is reduction even in the presence of an excess
of O and N2O3.
REDUCTION OF NITROSO-SULPHURIC ACID.
231
4. Acid of spec. grav. 1*60 (say, 69 per cent. H2SO4).
N.jOa originally
Loss of NgOg
in grams per litre at
present.
Grams in 1 litre.
50°.
' 60°. '
70°.
80°.
90°.
1
0050
0086
0175
0336
0-412
0
0100
0-236
0-436
0-725
0-912
a
0-325
0-525
0-775
1-150
1-500
4
()o62
0-836
1-250
1-910
2-100
T)
0-812
1-I.tO
1-500
2120
2-700
6
1050
1-450
1-900
2-500
3-350
7
l-28(i
1-800
2-350
3100
4-112
s
1-512
2-150
2-800
3-725
4-900
{)
1-750
2-425
3-250
4-336
5-686
10
1-975
2-812
3-712
4-900
6-475
11
2-250
3-162
4-436
5-265
7-300
ll>
2-512
3-536
4-675
6-325
8-125
la
2-786
3-936
5150
7012
8-962
14
3065
4-2.M)
5-650
7-700
9-750
1.3
3-3^)
4-612
6-125
8-400
10-625
16
3-600
4-975
6-612
9125
11-462
17
3-862
5-350
7-100
9-525
12-250
18
4150
5-712
7600
10-462
13136
19
4-425
6-075
8086
11-350
13-975
2()
4-700
6-425
8-5()2
11-850
14-800
Another reducing agent whose action had formerly been over-
looked is carbon, in the shape of the coke employed for packing
the Gay-Lussac tower. I have shown (J. Soc. Chem. Ind. 1885,
p. 31) that coke has a very strong reducing-action.on nitric acid
dissolved in sulphuric acid, which goes far towards explaining
the fact that the " nitrous vitriol " from the Gay-Lussac towers
never, except under altogether exceptional circumstances, contains
any nitric acid, even when considerable quantities of N2O4 had
been present in the exit gases. But the reduction goes further;
some N2O3 itself, in the shape of nitroso-sulphuric acid, is by the
coke reduced to lower nitrogen oxides and is thus lost. This has
been proved by myself in my laboratory (Zeitsch. f. angew. Chem.
1890, p. 195) ; and as it is a matter of importance, we give the
results obtained in the following table (p. 232).
It will be seen that at 40° C. two hours' contact reduced the
percentage of NgOg by 2"4 to 4'5 per cent. ; at 7QP the reduction
sometimes went as far as 28 per cent. The latter temperature
ought never to occur in a Gay-Lussac tower, but it does occur
regularly in Glover towers up to the top. The conclusion is that
232
PROPERTIES OF OXIDES AND ACIDS OF SVI-PHUR.
1
1
Percentage decrease
Material used.
Tempe-
! rature.
«>C.
1
; Time
(hours).
1
Original
percentage
grams. 1
in
grms.
HA-
■
of the N„0
originallj
present.
1.
2Jitrous Vitriol of spec. grar. r8375
GsB Coke in lumps ..
.... 16
24 '
18-93
0-330
1-71
>t ft **
.... 14
2
18-92
0-539
2-86
II II ••
.... 70
2
19-30
0742
3-84
Oven Coke in lumps ..
.... 15
24
19-30
0-285
' 1-48
ft ft '•
.... 40
2
18-92 ,
0-362
1-91
II If • •
.... 70
2
19-30
0-452
2-34
Ghis Coke in powder ..
.... 15
24
19-30
0-790
4-09
tf tf "
.... 40
2
18-92
0-858
4-54
>f t* • •
.... 70
2
lH-22 >
0-903
5-57
tt tt •'
.... 100
2
16-22
4-611
i 28-43
Oven Coke in powder
15
24
19-30 ;
0-379
1 1-96
'1 tf
... 40
2
1892
0-451
2-38
1* 1*
... 70
2
16-22
0-527
1 3-25
... 100
2
16-22
1
2-770
1708
1
2. Nitrous Vitriol of spec. grav. 1-725.
Gas Coke in powder . .
..... 15
' 24
19o0 1
0-333
1 1-98
»• i»
...J 40
; 2
19-50 1
0-574
; 2-^
M tt
70
2
19-50
0-891
4-57
ft ■» • ■
....; 100
2
1 1
19-50 .
1
3-410
17-49
coke packing should be entirely avoided in Glover towers, and
that it is not advisable even for Gay-Lussac towers (comp.
Chap. VI.).
Analysis of Sulphuric Acid.
Qualitatively sulphuric acid is always recognized best by the
white precipitate of barium sulphide which it gives with barium
chloride, both in the free state and in the solutions of its salts,
even when very much diluted. This precipitate mostly settles down
as a heavy powder, but in extremely dilute liquids occasionally
appears only after some little time as a white cloud. Barium
sulphate is as good as insoluble in water, solutions of salts, and
free dilute acids ; in concentrated acids it is a little soluble,
especially on heating, also in concentrated sulphuric acid itself
and in solutions of ferric chloride. On the other hand, in a very
concentrated liquid free from sulphuric acid, but containing
much hydrochloric or, especially, nitric acid, the addition of
£STIMATION OF SULPHURIC ACID. 233
barium chloride may cause a precipitate of barium chloride itself
or of barium nitrate^ which, however, is distinguished from
barium sulphate by its crystalline appearance, and even more by
vanishing on dilution of the liquid ; barium seleniate is distin-
guished from barium sulphate by its solubility bn boiling with
concentrated hydrochloric acid, and by its behaviour with the blow-
pipe. The reaction proves the presence of sulphuric acid either
in its free state or in its salts. In order to find sulphuric acid in
the free state in the presence of sulphates of acid reaction, either
the alcoholic solution of the substances can be tried with barium
chloride (free acid being soluble, but all sulphates insoluble in
absolute alcohol) , or the charring properties of concentrated oil
of vitriol are made use of by evaporating the solution mixed with
a little cane-sugar in a small porcelain capsule on the water-bath,
and observing whether a blackening of the sugar takes place.
This reaction, however, also takes place with the sulphates of
very weak bases, such as alumina or ferric oxide ; nor can sul-
phuric acid be distinguished with certainty in this way from
hydrochloric or nitric acid ; but in phosphoric, acetic, tartaric acid,
&c. a very small proportion of sulphuric acid can be proved by this
reaction. Another reaction for free sulphuric acid, as well as for
any strong free acid, is that with methyl-orange : the latter does
not change colour by adding metallic salts, but is changed by the
smallest quantity of free sulphuric acid.
In insoluble sulphates the acid is recognized by fusing them with
alkaline carbonates, or by boiling with concentrated solutions of
the same and filtering the solution of the alkaline sulphate formed
thereby from the insoluble carbonates, or with the blowpipe, on
charcoal, by the formation of sodium sulphate, according to well-
known methods.
The quantitative estimation of free sulphuric acid for technical
purposes is almost exclusively effected by volumetric methods or
by the hydrometer. In both cases, of course, impurities will have
a disturbing action ; but for technical purposes they may nearly
always be neglected (compare p. 187 et seq,). The hydrometric
estimation of sulphuric acid has been already described in detail;
and we shall here only point out again that the temperature must
not be neglected in this case.
The volumeirical estimation of free acid generally takes place
by means of a standard solution of potash, soda, or ammonia.
234 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
According to the accuracy required, either a normal solution is
used (that is, one containing per litre an equivalent expressed in
grams), or a semi- or decinormal solution, &c.
Formerly tincture of litmus was most frequently used as indi-
cator. Litmus is not well fidapted for working in artificial light :
the red appears almost as clear as water, the blue like a dark
violet; but the transition from bright red into purple, &c., cannot
be seen with certainty. This can be remedied by monochromatic
light, if the artificial light be coloured yellow by common salt :
the red appears clear as water, the blue like deep black ; and the
transition is even sharper than in daylight.
Litmus has, moreover, the disadvantage that it is sensitive to
all weak acids as well, and that it is destroyed by sulphuretted
hydrogen. If, therefore, carbonates are to be tested with it, thia
must be done at a boiling heat, and the boiling must be prolonged
for some time. If any sulphides are present, an excess of acid
must be added, and all the H2S expelled by prolonged boiling ;
only then should the litmus be added, and the analysis finished
by re-titrating. This makes application of litmus very troublesome
in alkalimetry ; in fact a real error is introduced by the necessity
of long boiling, if this is done in glass vessels which yield up some
alkali thereby. In acidimetry this drawback is less felt, but only
when the standard alkali is kept entirely free from carbonates,
which is very difficult in the daily practice of alkali-works.
Fhenolphthalein has in many cases taken the place of litmus. It
is one of the most sensitive indicators known, and the change from
no colour in acid solutions to a decided pink with the faintest trace
of free alkali is easily noticed even in artificial light. But this
indicator has two drawbacks : it is too sensitive even towards the
weakest acids [e,g, CO2), and it fails in the presence of ammonia.
The former circumstance entails exactly the same difficulties as in
the case of litmus. Hence, while phenolphthalein is the best of all
indicators for titrating weak acids, it is decidedly inferior to methyl-
orange for the titration of alkalies containing carbonates or
sulphides, and in the acidimetry of strong acids.
The indicator which in alkali and acid works is now universally
employed is methyUorange *. This is sulphobenzene-azodimethyl-
* I have proposed this name for the indicator introduced by me, in lieu of
the commercial names of Poimer's Orang^ No. III. or helianthin (Chem. News,
xliv. p. 288) ; and it has been generally adopted.
METHYL-ORANGE AS INDICATOR. 235
aniline^ or the sodium salt of this compound :
(SOgNa) CeH4-N2-CeH4N(CH3)„
%vhich is employed in an aqueous solution of 1 in 2000 water, or
even more dilute^ and a very small quantity of the solution is
used for each test. It is best to keep it in a bottle with a
perforated cork^ a glass tube drawn out to a point and inserted in
the cork serving as a pipette for regulating the supply. Methyl-
orange is orange in neutral solutions or in the presence of free
alkali, but is faintly yellow in very dilute solutions, and no more
ought to be added to the liquid to be tested than suffices to colour
it just perceptibly yellow. In this case a single drop of fifth-
normal sulphuric or hydrochloric acid will cause a transition into
red. But when too much of the indicator has been added, so
that the colour of the solution is orange, the transition into red
(or, rather, in this case into pink) is only gradual, and therefore
useless. Warm solutions behave in a similar manner. It is
therefore a distinct rule to be observed with methyl-orange, to
employ as little as possible of it, and to work always at the
ordinary temperature. This is made possible by the fact that
methyl-orange is not acted upon by weak acids, such as COj, HjS,
acetic acid, &c. ; and this is undoubtedly one of its most valuable
properties, since the trouble and loss of time in boiling the liquids,
and the error introduced in the case of glass vessels, are thereby
avoided. Both NajCOg and NaHCOg can be titrated directly in
the cold just as well as NaOH^ the total available soda being
always indicated. Sulphurous acid behaves in the manner explained,
sui/ra (p. 160), that is, the compound NajSOs is alkaline, NaHSOg
neutral, to methyl-orange. Oxalic acid, as well as other strong
organic acids, come in between sulphurous acid and the strong
mineral acids ; no sharp results can be obtained with them,
and hence oxalic acid cannot serve as standard acid with methyl -
orange. On the other hand, ammonia, which cannot be titrated
with phenolphthalein, behaves quite normally towards methyl-
orange, just like potash and soda. The normal sulphates of
peroxide of iron, alumina, &c., which give an acid reaction with
litmus, are neutral towards methyl-orange, so that any free acid
present with them can be estimated by means of this indicator.
Methyl-orange is destroyed by nitrous acid. Nevertheless it
can be easily employed in titrating sulphuric or nitric acid con-
236 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
tainiDg nitrous acid in two ways : either by adding the indicator
shortly before the saturation is completed and quickly finishing
the titration^ or by supersaturating the acid with caustic soda and
retitrating with standard acid.
Nitrous acid acts upon methyl-orange like a strong mineral
acid, and is therefore completely saturated before the pink colour
has changed to yellow,.if there is not time for the colouring-matter
to be destroyed.
The tropseolins, formerly recommended as indicators^ are
nothing like so sensitive as methyl-orange, and are best not used
at all as indicators^ especially since several totally distinct com-
pounds are comprised under this name, and the dealers do not
always supply that which is really wanted.
Free sulphuric acid (including that contained in commercial
ferric or aluminium sulphate, or any other sulphate) is estimated
by adding a drop of methyl-orange solution, which causes a pink
colour, and then adding a standard solution of alkali, till the pink
tint has changed into pure light yellow. It is best to check this
by reproducing the faint pink shade by means of a drop of standard
acid.
The normal alkaline solution itself is best standardized by
means of a normal acid, be it sulphuric or hydrochloric acid ;
and this on its part is best standardized by pure ignited
sodium carbonate, which is easily obtained or prepared —
for instance, by washing and igniting sodium bicarbonate. If
sodium carbonate, bought as chemically pure, dissolves in water
without any residue, and shows by the ordinary reagents no
chloride or sulphate, or only unweighable traces of these, it can
be used at once for standardizing normal acids after moderately
igniting. If methyl-orange be used as indicator, this roundabout
way need not be taken, especially at works* laboratories, but pure
sodium carbonate itself can be used as acidimetrical liquid, either
in a normal solution containing 53 grams per litre or in more
dilute solutions. The latter are more to be recommended, since
the really normal solution causes efBorescences of sodium carbonate
at the lower ends of the burettes, &c., which does not happen with
semi-normal or weaker solutions, at least not for some time.
Although it is more important in alkalimetry than in acidimetry,
we will here treat of the standard acid itself. As such many
factory-chemists use sulphuric acid, but we recommend as more
STANDARD ACID. 237
suitable hydrochloric acid, both because it can be used for esti-
mating alkaline earths us well^ and because it admits of a twofold
way of checking the standard^ either volumetrically by pure sodium
carbonate, or gravimetrically by argentic nitrate. The gravimetric
estimation of sulphuric acid by barium chloride is nothing like so
accurate as the estimation of HCI in the shape of AgCl. Oxalic
acid, most strongly recommended by Mohr, and formerly used
by very many chemists, has great drawbacks. It is extremely
difficult to prepare in the perfectly pure and dry state, without
losing some of the water of crystallization ; it does not keep in
weak solutions, and it cannot be employed with methyl-orange.
For standardizing normal acids, sometimes a solution of pure
sodium carbonate is made, of which portions are taken out with a
pipette. For the most accurate estimations it is, however, always
preferable to weigh each portion of sodium carbonate, directly after
igniting and cooling, into the beaker, since it is never possible to
measure as accurately as to weigh, because, among other reasons,
the volumetrical apparatus very rarely agree quite accurately one
with another. In spite of the trouble, it should most certainly not
be neglected to compare, in the first instance, the pipettes with all
the measuring-flasks, in order to see whether the former fill the
latter precisely ; secondly,* to calibrate the burettes accurately, in
which case it will often be found that the upper parts diflTer by
several per cent, from the middle and lower parts, and cause a
corresponding error. Of course the burettes must again be com-
pared with the other measuring-apparatus.
The standard acid is made to represent equivannts, not molecules ;
that is, if sulphuric or oxalic acid, it will contain one half of the
molecular weight in grams, viz. 49 or 63 grams, because these
acids are bivalent ; but if it is the univalent hydrochloric or nitric
acid, it will contain the total molecular weight, viz. 36'46 grams
HCI, or 65 grams NO3H. First of all, the acid is diluted a little
less than necessary, and it is found out how many cub. centims.
of it are required for a certain quantity of sodium carbonate.
From this the quantity of water is computed which is required
for obtaining an exactly normal acid ; and after mixing this with
the acid the accuracy of the standard is ascertained by repeated
titrating with sodium carbonate. Not less than 2 to 3 grams of
the latter should be taken for each test. If litmus is to be the
indicator, to the alkaline solution drops of tincture of litmus are
238 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
added till it becomes very markedly blue, then acid till strong
effervescence sets in ; and the liquid is now made to boil ; then to
the hot liquid gradually more and more acid is added, till the blue
colour has passed through the purple and reddish purple of the
CO2 reaction to the bright red of the SO4H2 reaction. The liquid
cooled by the addition of acid must be constantly heated again.
Often, after several minutes^ boiling, the apparently red liquid
again turns purple and then blue. When working with boiling
liquids there is never any doubt, to a single drop, respecting the
point where the pure red sets in. The test must be made in a
porcelain capsule, not in a glass beaker. Precisely the same
troublesome method must be employed with phenolphthalein. All
this trouble is saved by using methyl-orange as indicator and
working in the cold.
When a perfectly accurate normal acid has been obtained, the
normal alkali, whether ammonia, soda, or potash, is most easily
made from it ; and this is now used for the acidimetric test of
sulphuric acid. Concentrated sulphuric acid must, of course, first
be diluted in the usual manner.
Analysis of Fuming Oil of VitnoL
Several communications have been made on this subject, as by
Ftirstenau (Chemiker-Zeitung, 1880, p. 18), Moller (ibid. p. 569),
Becker (ibid. p. 600), W^inkler (Chem. Ind. 1880, p. 194), Clar and
Gaier (ibid. 1881, n. 251). We shall, in the first instance, prin-
cipally describe tW methods contained in Lunge and Hurter's
Alkali-Maker's Hand-book,' as derived from practical information,
with a few improvements.
In the present case even the taking of t/w sample is not quite a
simple task. Measuring it in a pipette is out of the question, it
must be weighed. But even for this purpose the article, if solid,
must be first liquefied. This is comparatively easy with partly
crystallized acid or with solid pyrosulphuric acid; these can be
liquefied without any danger in a closed vessel by gently heating
to 30° in a sand-bath. Soldered-up tins are generally placed in a
suitably heated stove. There is no sensible loss of strength if the
aperture for this purpose is previously opened and at once covered
with a watch-glass. This prevents any pressure within the vessel
during the heating, which must otherwise be guarded against in
ANALYSIS OP PUMINQ OIL OP VITRIOL. 239
Opening it. The case is different with products containing a larger
percentage o£ anhydride. These do not liquefy completely, a por-
tion always remaining in the state of a gelatinous residue. This
residue is, however, composed exactly like the liquid portion, so
that the sample may be taken out of the latter without any danger
of making a mistake.
The sample is weighed either in glass bulbs or in a glass tap-
tnbe. The former are very thin bulbs of
about J inch diameter, ending each way in Fig. 40.
a capillary tube (fig. 40). The liquefied
acid (2 or 3 grams) is sucked into the bulb,
without danger to the operator, by means of
a bottle closed with an india-rubber cork,
through which passes a tightly- fitting glass
tap, connected at its free end with an elastic
tube. Suction is applied to the latter, the
tap is closed, the elastic tube is drawn over
one of the capillary ends of the weighing-
bulb, and by opening the tap a sufficient
quantity of acid is admitted into the bulb.
The capillary tube ia cleaned and one of
the two ends is sealed at the lamp. The
other end can be left open without fear of
any loss of SO3 or attraction of moistiu-e during wcigliin<r. The
weighing is best done on a small platinum crucible with two nicks,
on which the ends of the bulb can rest. If the latter should be
accidentally broken, the acid runs into the crucible, not on the
balance. Then the bulb, open end downwards, is put into a small
Erlenmeyer flask, into the neck of which it should fit exactly
(fig. 40), find which contains so much water that the capillary
tube dips pretty far into it to prevent any loss of SO^ on mixing
the acid with water. Now break off the other point, allow the
acid to run out, squirt a few drops of water into the upper capillary,
and ultimately rinse the whole bulb-tube by repeated aspiration of
water. Dilute the liquid to 500 c.c. and take 50 e.c. for each test.
The testing is done with fifth-normal solution of soda (1 c.c.=
0008 grm. SO3) and litmus or methyl-orange aa indicator. The
acidity found is diminished by that proceeding from SO., found
by titrating another sample with iodine.
In lieu of the bulb-tube (first proposed by Clar and Gaier) we
240
PROPEKTIES OF OXIDES AND ACIDS OP SULPHUR.
prefer the glass tap-tnbe, as shown in fig. 41. The tap should be
tight without greasing, and the tube below it should taper gradu-
allr. It is charged by suction, in the same way as described above,
with about 0*5 grm. of Nordhausen acid, no more, in order to be
able to titrate it directly, without taking an aliquot portion. After
the proper quantity of acid has been introduced, the tap is closed.
Fig. 41.
Fig. 42.
the tube is cleaned outside with filtering-paper, aud it may be
weighed at once, without any fear of a change of weight during the
operation. It is, however, preferable to employ a tube (as shown
in fig. 42) ground into an outer glass case, which is, of course,
tared together with the empty tube.
After weighing, place the tube point downwards in water, or, in
the case of nearly pure anhydride or the strongest Nordbausen
ANALYSIS OF FUMING SULPHURIC ACID. 241
acids, in a layer of crystallized, coarsely powdered Glauber's salt,
and slowly run out the contents. Then squirt a drop of water from
above into the tube, allow it to stand for a moment, and rinse
thoroughly with water. Anhydride once melted for the purpose
of filling the tube remains liquid enough to complete the weighing
and running out without requiring to be heated again.
The most convenient apparatus for weighing oflF fuming sul-
phuric acid (as well as other fuming acids or substances evolving
vapours) is the ''bulb-tap pipette,'^ proposed by Lunge & Rey
(Zeitsch. f. angew. Chem. 1891, p. 165), and shown in fig. 42,
in which both the filling as well as the weighing and discharging
are performed without any loss of vapours and without requiring
any special aspirating-apparatus. Above tap a there is a bulb b,
rather less than an inch in diameter, and above this a second tap c.
The lower portion of the pipette is ground into a glass tube (/, closed
at the bottom. In the conical part of the pipette there is a groove &,
reaching halfway down, the corresponding half of the groove /
being in the outer tube. By turning the pipette in the latter, the
tube d can be made to communicate with the outer air, or the
reverse. When the pipette is to be used, close tap a, suck at the
top with the mouth, and before leaving off shut tap c, so that the
bulb b contains a partial vacuum. Dip the point of the pipette
into the acid and open a ; the acid will rise up, but a is shut
before it gets so far, or even sooner, when enough has got in.
Clean the pipette outside, put it into dy and weigh. In most ordi-
nary cases (with other acids, ammonia, &c.) the grooves e and /are
made to communicate ; water is squirted through c into b and then
run through a, with the contents of the pipette, into d, the air
escaping through e and/. The dilute acid is run into a beaker and
titrated. In the case of Nordhausen acid it is preferable to take
the pipette out of d, rinse the latter into a beaker, run the contents
of the pipette, by opening a, into water or Glauber's salt contained
in the same beaker, then squirt water through c (during which
time a may be closed), and rinse the pipette into the same beaker.
The sucking at c is quite sufficient to produce the necessary
rarefaction of air in 6, and no vapours are lost, as is inevitable
with any other kind of aspiration.
Very strong oleum (70 per cent. SO3 and upwards) is best
-weighed in glass bulbs (p. 239) sealed at both ends ; these are
put into a bottle containing sufficient water, which is closed by a
VOL. I. u
242 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
ground-in stopper ; the bulb is smashed by shaking the bottle, and
the titration is made.
The sampling of solid sulphuric anhydride is not an easy matter^
This substance, which is now a regular article of trade and is sent
out in iron bottles, is much too compact and tough to enable a
sample to be taken out by means of an auger. Before using it, it
is always heated in a stove till it has completely liquefied ; but in
this state the bottle on opening emits such a dense cloud of fumes
that any sampling is out of the question. The way out of the
diflBculty is this : — In a stoppered bottle some lumps of the solid
anhydride are weighed off on a large balance and are theYi mixed
with a sufficient quantity of accurately analyzed monohydrated
sulphuric acid, to form an acid of 70 per cent. SOg, which is
liquid at ordinary temperatures. The solution is promoted by
gently heating the bottles, say to 30° or 40° C, with the stopper
loosely put on. At last a sample is taken out by means of the
pipette described above (p. 240) and the analysis performed in the
usual \rayj taking account of the slight proportion of water present
in the *'monohydrate^' employed.
Rosenlecher (Zeitsch. anal. Ch. xxxvii. p. 209) describes the
method employed at Freiberg. A number of bulbs are made from a
glass tube, 6 or 8 mm. wide, of the form shown in fig. 43, and keeping
Fig. 43.
them exactly to the dimensions indicated. The capillary ends are
contracted before a spirit-lamp to ^ mm. bore, in the case of very
strong anhydride to ^ mm. The bulbs are filled by aspiration by
means of a capillary rubber tube drawn over the shorter end, in
case of need interposing a test-tube filled with soda crystals. The
suction is continued until the acid arrives in the bulbs before the
heavy fumes enter the shorter capillary. All the bulbs are turned
with the points of the capillaries upwards, cleaned outside, and
placed in a paste-board box provided with nicks. The weighing is
performed in a platinum crucible (p. 239) or on a specially made
wire stand. No attraction of moisture need be feared during the
ANALYSIS OP FUMING SULPHURIC ACID.
243
weighing, but the bulbs must not be heated by touching them
with the fingers. They are then placed in bottles, charged with
20 or 30 c.c. water of ordinary temperature and the indicator^ in
such manner that the acid does not flow out. The wetted stopper
is put tightly in^ the bottles are placed sideways (up to this
time the colour of the indicator should not have changed)^ the
bulbs broken by shakings and then, after the white fumes have
vanished, the titration is performed in the bottle itself.
The results of titration are first calculated for the total (com-
bined and uncombined with water) SOs, each c. c. of normal soda
solution indicating 0*040 grm. SOg, and the proportion of free
SO3 and HgSO^ present is then read off by means of the following
formula : S03=S— 4*444 (100— S), in which SO3 denotes the free
sulphur trioxide, and S the total SOg as found by titration.
This calculation is saved by the following table, computed by
Knietsch (Ber. 1901, p. 4114) :—
S03
SO3
SO3
SO3
SO3
SO3
SO3
Total. Free.
81-63 00
Total.
Jrr€e.
Total. Free.
1
Total.
Free.
Total
92-2
Free.
Total.
Free.
71-7
Total. Free.
84-3 14-6
87 0 29-2
89-6
43-4
57-5
94-8
97-4
85-8
81-7 '
0-4
84-4
151
87-1
29-8
89-7
43-9
92-3
581
94-9
72-2
97-5 86-4
81-8
0-9
84-6
15-6
87-2
303
89-8
44-5
92-4
58-6
95-0
72-8
97-6 ; 86-9
81-9
1-5
84-6
16-2
873 30-9
89-9
450
92-5
59-2
951
73-3
07-7
87-5
820
20
84-7
16-7
87-4 31-4
900
45-6
92-6
59-7
95-2
73-9
97-8
88-0
82-1
26
84-8
17-2
87-5
31-9
901
461
92-7
60-3
95-3
74-4
97-9
88-6
82-2
31
84-9 17-8
87-6
32-5
902
46-6
92-8
60-8
95-4
750
980 : 89-1
82-3
3-6
85-0
18-3
87-7
330
90-3 ; 47-2
92-9
61-3
95-5
75-5
98-1 89-7
82-4
4-2
851
18-9
87-8 33-6
90-4 1 47*7
930
61-9
95-6
761
98-2 1 90-2
82-5
4-7
85-2
19-4
87-9 341
90-5 ' 48-3
931
62-4
95-7
76-6
98-3
90-7
82-6
5-3
85-3
200
88 0 34-7
90 6 48-8
93-2
630
95-8
771
98-4
91-3
82-7
5-8
85-4
20-5
88-1 35-2
90 7 i49-4
93-3
63-5
95-9
77-7
98-5
91-8
'82-8
6-4
85-5
210
882 35-8
90-8 J49-9
93-4
641
960
78 3
98-6 i 92-4
82-9
69
85-6
21-6
88-3 36-3
90-9
50-5
93-5
64-6
961
78-8
98-7 92-9
830
7*5
85-7
22-2
88-4 36-8
91-0
510
93-6
65-2
96-2
793
98-8
93-5
83-1
8-0
8o-8
22-7
88-5 37-4
911
51-6
93-7
65-7
96-3
79-9
98-9
940
83-2
8-5
85-9
23-2
886 37-9
91-2 ' 521
93-8
66-2
96-4
80-4
990
94-6
83-3
91
860
23-8
88-7 38-5
91-3 , 52-6
93-9
66-8
96-5
81-0
991 951
83-4
96
861
24-3
88-8 39-0
91-4 ' 53-2
940
67-3
96-6
81-5
99-2 95-6
83-5
10-2
8<5-2
249
88-9 39-6
91-5
537
94-1 67-9 1
96-7
820
99-3 96-2
83-6
10-7
863
2o-4
890 401
91-6
54-3
942
68-4
9('v8
82-6
994 96-7
83-7
11-3
86-4
260
89-1 40-6
91-7
54-8
94-3
690
969
831
99-5 ; 97-3
83-8
11-8
860
26-5
89-2 :41-2
91-8 55-4 1
94-4
69-5
970
83-7
99-6 1 97-8
839
12-3
86-6
27 0
89-3 41-7
91-9
55-9
94-5
701
97-1
84-2
99-7 98-4
84 0
12-9
86-7 ,27-6
89-4 42-3
920
56-4
94-6
70-6
97-2
848
99 8 98-9 '
841
13-4
86-8 28-1
89-5 i 42-8
921
570
94-7
71-2
97-3
85-3
99-9 99-5
84-2
140
860
28-7
..-)
k:^
242 PROPERTIES OP OXIDES AND ACIDS OP SULPHUR.
ground-in stopper; the bulb is smashed by shaking the bottle, and
the titration is made.
The sampling of solid sulphuric anhydride is not an easy matter^
This substance^ which is now a regular article of trade and is sent
out in iron bottles^ is much too compact and tough to enable a
sample to be taken out by means of an auger. Before using it^ it
is always heated in a stove till it has completely liquefied; but in
this state the bottle on opening emits such a dense cloud of fumea
that any sampling is out of the question. The way out of the
diflSculty is this : — In a stoppered bottle some lumps of the solid
anhydride are weighed off on a large balance and are theti mixed
with a sufficient quantity of accurately analyzed monohydrated
sulphuric acid, to form an acid of 70 per cent. SOg, which ia
liquid at ordinary temperatures. The solution is promoted by
gently heating the bottles, say to 30° or 40° C, with the stopper
loosely put on. At last a sample is taken out by means of the
pipette described above (p. 240) and the analysis performed in the
usual way, taking account of the slight proportion of water present
in the ''monohydrate^' employed.
Rosenlecher (Zeitsch. anal. Ch. xxxvii. p. 209) describes the
method employed at Freiberg. A number of bulbs are made from a
glass tube, 6 or 8 mm. wide, of the form shown in fig. 43, and keeping
Fig. 43.
them exactly to the dimensions indicated. The capillary ends are
contracted before a spirit-lamp to ^ mm. bore, in the case of very
strong anhydride to i mm. The bulbs are filled by aspiration by
means of a capillary rubber tube drawn over the shorter end, in
case of need interposing a test-tube filled with soda crystals. The
suction is continued until the acid arrives in the bulbs before the
heavy fumes enter the shorter capillary. All the bulbs are turned
with the points of the capillaries upwards, cleaned outside, and
placed in a paste-board box provided with nicks. The weighing i&
performed in a platinum crucible (p. 239) or on a specially made
wire stand. No attraction of moisture need be feared during the
ANALYSIS OF FUMING STTLPHURIC ACID.
243
weighing, but the bulbs must not be heated by touching them
with the fingers. They are then placed in bottles, charged with
20 or 30 c.c. water of ordinary temperature and the indicator^ in
such manner that the acid does not flow out. The wetted stopper
is put tightly in, the bottles are placed sideways (up to this
time the colour of the indicator should not have changed)^ the
bulbs broken by shaking, and then, after the white fumes have
vanished, the titration is performed in the bottle itself.
The results of titration are first calculated for the total (com-
bined and uncombined with water) SOs, each c. c. of normal soda
solution indicating 0*040 grm. SOa, and the proportion of free
SO3 and H3SO4 present is then read ofE by means of the following
formula : S03=S— 4-444 (100— S), in which SOa denotes the free
sulphur trioxide, and S the total SO^ as found by titration.
This calculation is saved by the following table, computed by
Knietsch (Ber. 1901, p. 4114) :—
S03
SO3
SO3
SO3
SO3
SO3
SO3
Total.
81-63
Free.
00
ToUl. Free.
Total.
Free.
Total.
Free.
43-4
Total
92-2
Free.
Total.
Free.
71-7
Total. Free-
84-3 14-6
87 0
29-2
89-6
57-6
94-8
97-4
85-8
81-7
0-4
84-4
151
87-1
29-8
89-7
43-9
92-3
581
94-9
72-2
97-5 86-4
81-8
0-9
84-5
15-6
87-2
303
89-8
44-5
92-4
58-6
95-0
72-8
97-6 1 86-9
81-9
1-5
84-6
16-2
87-3
30-9
89-9
450
92-5
59-2
951
733
07-7 1 87-5
82-0
20
84-7
16-7
87-4
31-4
900
45-6
92-6
59-7
95-2
73-9
97-8 1 88-0
821
2-6
84-8
17-2
87-5
31-9
901
461
92-7
60-3
95-3
74-4
97-9 ! 88-6
82-2
31
84-9 17-8
87-6
32-5
90-2
46-6
92-8
60-8
95-4
750
980 ' 89-1
82-3
36
850
18-3
87-7
330
90-3
47-2
92-9
61-3
95-5
75-5
98-1 89-7
82-4
4-2
851
18-9
87-8
33-6
90-4
47-7
930
61-9
95-6
761
98-2
90-2
82-5
4-7
85-2
19-4
87-9
341
90-5
483
931
62-4
95-7
76-6
98-3
90-7
82*6
5-3
85-3
200
880
34-7
906
48-8
93-2
630
96-8
771
98-4
91-3
82-7
5-8
85-4
20-5
88-1
35-2
907
49-4
93-3
63-5
95-9
77-7
98-5 91-8
82-8
6-4
85-5
210
88-2
35-8
90-8
49-9
93-4
641
960
78 3
98-6 92-4
82-9
69
850
21-6
88-3
36-3
90-9
50-5
93-5
64-6
961
78-8
98-7 92-9
830
7-5
85-7
22-2
88-4
36-8
91-0
510
93-6
65-2
96-2
793
98-8
93-5
83-1
8-0
85-8
22-7
88-5
37-4
911
51-6
93-7
65-7
96-3
79-9
98-9
940
83-2
8-5
85-9
23-2
88-6
37-9
91-2
521
93-8
66-2
96-4
80-4
99 0 : 94-6
83-3
91
86-0
23-8
88-7
38-5
91-3
52-6
93-9
66-8
96-5
810
991 951
83-4
96
861
24-3
88-8
39-0
91-4
53-2
940
67-3
96-6
81-5
99-2 95-6
83-5
10-2
86-2
24-9
88-9
39-6
91-5
53 7
94-1
67-9
96-7
820
99-3 96-2
83-6
10-7
86-3
25-4
890
401
91-6
54-3
94-2
68-4
9^-8
82-6
994 96-7
83-7
11-3
86-4
260
89-1
40-6
91-7
54-8
94-3
690
96-9
831
99-5 97-3
83-8
11-8
86-5
26-5
89-2
41-2
91-8
a.)*4
944
69-5
970
83-7
99-6 97-8
839
12-3
86-6
27-0
89-3
41-7
91-9
55-9
94-5
701
97-1
84-2
99-7 98-4
840
12-9
86-7
27-6
89-4
42-3
92-0
56-4
94-6
70-6
97-2
848
998 98-9 '
841
13-4
86-8
281
89-5
42-8
921
570
94-7
71-2
97-3
85-3
99-9 1 99-5
84-2
140
1
860
28-7
k2
242 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
ground-in stopper ; the bulb is smashed by shaking the bottle, and
the titration is made.
The sampling of solid sulphuric anhydride is not an easy matter^
This substance^ which is now a regular article of trade and is sent
out in iron bottles^ is much too compact and tough to enable a
sample to be taken out by means of an auger. Before using it^ it
is always heated in a stove till it has completely liquefied; but in
this state the bottle on opening emits such a dense cloud of fumes
that any sampling is out of the question. The way out of the
diflSculty is this : — In a stoppered bottle some lumps of the solid
anhydride are weighed off on a large balance and are theti mixed
with a sufficient quantity of accurately analyzed monohydrated
sulphuric acid, to form an acid of 70 per cent. SOg, which is
liquid at ordinary temperatures. The solution is promoted by
gently heating the bottles, say to 30° or 40° C, with the stopper
loosely put on. At last a sample is taken out by means of tho
pipette described above (p. 240) and the analysis performed in the
usual way, taking account of the slight proportion of water present
in the "monohydrate^' employed.
Rosenlecher (Zeitsch. anal. Ch. xxxvii. p. 209) describes the
method employed at Freiberg. A number of bulbs are made from a
glass tube, 6 or 8 mm. wide, of the form shown in fig. 43, and keeping:
Fig. 43.
them exactly to the dimensions indicated. The capillary ends are-
contracted before a spirit-lamp to ^ mm. bore, in the case of very
strong anhydride to J mm. The bulbs are filled by aspiration by
means of a capillary rubber tube drawn over the shorter end, in
case of need interposing a test-tube filled with soda crystals. The
suction is continued until the acid arrives in the bulbs before the
heavy fumes enter the shorter capillary. All the bulbs are turned
with the points of the capillaries upwards, cleaned outside, and
placed in a paste-board box provided with nicks. The weighing i&
perfoimed in a platinum crucible (p. 239) or on a specially made
wire stand. No attraction of moisture need be feared during the
ANALYSIS OF FUMING SULPHURIC ACID.
243
weighing, but the bulbs must not be heated by touching them
with the fingers. They are then placed in bottles, charged with
20 or 30 c.c. water of ordinary temperature and the indicator^ in
such manner that the acid does not flow out. The wetted stopper
is put tightly in, the bottles are placed sideways (up to this
time the colour of the indicator should not have changed), the
bulbs broken by shaking, and then, after the white fumes have
vanished, the titration is performed in the bottle itself.
The results of titration are first calculated for the total (com-
bined and uncombined with water) SOg, each c. c. of normal soda
solution indicating 0*040 grm. SOb, and the proportion of free
SO3 and H2SO4 present is then read ofE by means of the following
formula : S03=S— 4*444 (100— S), in which SOa denotes the free
sulphur trioxide, and S the total SOg as found by titration.
This calculation is saved by the following table, computed by
Knietsch (Ber. 1901, p. 4114) :—
S03
SO3
SO3
SO3
SO3
SO3
SO3
Total.
81-63
Free.
0-0
ToUl. Free.
1
Total. Free.
1
Total.
Free.
43-4
TotaL
92-2
Free.
Total.
Free.
Total.' Free.
1
1
1
97-4 i 85-8
84-3
14-5
87 0
29-2
89-6
57-5
94-8
71-7
81-7
0-4
84-4
151
87-1 :29-8
89-7
43-9
92-3
58-1
94-9
72-2
97-5 86-4
81-8
0-9
84-5
15-6
87-2 30 3
89-8
44-5
92-4
58-6
95-0
72-8
97-6 86-9
81-9
1-5
84-6
16-2
873 30-9
89-9
450
92-5
59-2
951 ,73-3
07-7 87-5
82-0
20
84-7
16-7
87-4 , 31-4
900
45-6
92-6
59-7
95-2 73-9
97-8 ; 88-0
82-1
2-6
84-8
17-2
87-5
31-9
901
461
92-7
60-3
95-3
74-4
97-9
88-6
82-2
31
84-9
17-8
87-6
32-5
902 46-6
92-8
60-8
95-4
750
98-0
89-1
82-3
36
850
18-3
87-7
330
90-3 ! 47-2
92-9
61-3
95-5 75-5 1
981
89-7
1
82-4
82-5
4-2
851
18-9
87-8 33-6
90-4
47-7
930
61-9
95-6
761
98-2
90-2
4-7
85-2
19-4
87-9 ' 341
90-5
48-3
931
62-4
95-7
766
98-3 90-7
82-6
5-3
85-3
200
88 0 ' 34-7
90-6
48-8
93-2
630
95-8
771
98-4 91-3
82-7
5-8
85-4
20-6
88-1 35-2
907
49-4
93-3
63-5
95-9
77-7
98-5 91-8
82-8
6-4
85-5
210
88-2 35-8
90-8
49-9
93-4
641
960
783
98-6 |92-4
82-9
69
85-6
21-6
88-3 36-3
90-9 50-5
93-5
64-6
961
78-8
98-7 92-9
830
7-5
85-7
22-2
88-4 36-8
910 510
93-6
65-2
96-2
79-3
98-8 93-5
831
8-0
85-8
22-7
88-5 37-4
911 51-6
93-7
65-7
96-3
79-9
98-9 940
83-2
8-5
85-9
23-2
88-6 37-9
91-2 521
93-8
66-2
96-4
80-4
99 0 94-6
83-3
91
800
23-8
88-7 38-5
91-3 52-6
93-9
66-8
96-5 810
991 951
83-4
96
861
24-3
88-8 39-0
91-4 53-2
940
67-3
96-6 !81-5
99-2 95-6
83-5
10-2
86-2
24-9
88-9 39-6
91-5
537
94-1
67-9
96-7 820
99-3 9tJ-2
83-6
10-7
86-3
25-4
890 401
91-6
54-3
94-2
68-4
96-8 82-6
994 96-7
83-7
11-3
86-4
260
89-1 I40-6
91-7
54-8
94-3
690
969 831
99-5 97-3
83-8
1 11-8
86-6
26-5
89-2 '41-2
91-8
55-4
94-4
69-5
970 83-7
99-6 97-8
839
12-3
86-6
27-0
89-3 41-7
91-9
55-9
94-5
701
97-1 184-2
99-7 98-4
84 0
12-9
86-7
27-6
89-4 42-3
92*0
56-4
94-6
70-6
97-2
848
998 98-9
841
13-4
86-8
28-1
89-5 ;42-8
921
57-0
94-7
71-2
97*3 ! 85-3
99-9 i 99-5
84-2
140
1
86-9
28-7
1
1
K^
244 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
This table serves also for the frequently performed operation of
mixing Nordhausen acid of a certain percentage of SO3 with
concentrated sulphuric acid in order to produce an acid with a
smaller percentage of SOg, This can be done by means of a
formula given by Gerster (Chem. Zeit. 1887, p. 3),
a?=100
a
where x represents the quantity of sulphuric acid which must be
added to 100 parts of the Nordhausen acid ; a the total sulphur
trioxide in 100 parts of the acid desired ; b the total SO3 in 100
parts of the Nordhausen acid to be diluted ; c the total SO3 in 100
parts of the ordinary acid to be used for diluting. The values of
a and b are taken from the preceding table ; c is easily calculated
80
by multiplying the percentage of HgSO^ with — or 0'816. An
example will make this clearer. Supposing there is a Nordhausen
acid of 25'5 per cent. SOs in stock, as well as sulphuric acid of
98*2 per cent. H2SO4, and an acid of 19 per cent. SO3 is required,
we have then :
a=85-l ; 6=86-3; c=98x 0-816=80-1,
a?=100 . =100 --- — — — = -,^-=24.
a—c 851 — 80-1 o
That is : by mixing 100 parts of acid of 25-5 per cent. SOs with
24 parts of sulphuric acid of 98*2 per cent. H2SO4, Nordhausen
acid is obtained containing 19 per cent. SOg. [In reality the
strength of the mixed product will be slightly below that calculated,
as a certain loss of SO3 is hardly avoidable in the manipulation.]
In Zsch. f. angew. Chem. 1895, p. 221, I have drawn attention
to the fact that in allowing for the SO3 sometimes a serious mis-
take is committed. The SO2 is always tested for by iodine solution,
and is then subtracted from the total acidity. Here we must
consider that the neutrality-point in the case of phenolphthalein
is reached when 1 SOj has been combined with 2 NaOH, but in
the case of methyl-orange only 1 NaOH is consumed for 1 SOj.
Litmus cannot be used at all, as it gives uncertain results between
ANALYSIS OF FUMING SULPHURIC ACID. 245
these two limits. With methyl-orange 1 c.c. normal soda
solution indicates 0-040 gram SOs, but 0064 gram SOj. Hence
for each c.c. of decinormal iodine solution only 0*05 c.c. of
normal or 0*1 c.c. of semi-normal solution of NaOH must be
deducted. If this is overlooked, a very serious mistake is com-
mitted ; for since everything which is not present as SO3 or SO2
is assumed to be water, the incorrect allowance for SOj will cause
not merely a deficiency of SOg, but a surplus of HjO ; and as this
must be represented as combined with 4*444 its weight of water,
far too little free SO3 is found.
A practical instance will illustrate this. 3*5662 grams fuming
acid were diluted to 500 c.c, and 100 c.c. (=0*71124 gram)
employed for each test. This consumed 5*40 c.c. iodine solu«
tion, =5*40x0-0032=001728 gram SO2 or 2*43 per cent. SOg.
On titrating with seminormal soda solution and methyl-orange,
34*40 c.c. was used. By erroneously deducting 0*2 X5'40= 1*08
c.c, there remained 3332 cc.=s0*6664 gram SOg or 93*70 per
cent. The fuming acid therefore would have contained 93*70 per
cent. SOg, 2*43 SO2, 3*87 HgO. The 387 H^O is = 17*20 SOg,
and the free SOg would be =93-70— 17*20 =76*50 per cent.
In reality the 5*40 c.c. decinormal iodine corresponds to only
0*54 c.c seminormal soda, leaving 32*86 c.c. =0*6772 gram
808 = 95*21 per cent. Composition of the acid : 95*21 per cent.
SOg, 2*43 SO3, 2*36 H2O. The 236 HgO is =10*49 SOg, leaving
95*21 - 10*49= 84*72 per cent, free SOg. Hence by that erroneous
calculation the factory committed an error of 8*25 per cent, to its
own disadvantage I
Exactly the same result ia obtained when performing the calcu-
lation by means of the above table.
We have, as already mentioned, assumed everything as HgO
which has not been found to be present as SOg and SOj. But it
is advisable to estimate the fixed impurities as well, since otherwise
their weight, multiplied by 4i'A4A!, is erroneously deducted from
the free SOg.
Setlik (Chemiker-Zeitung, 1889, p. 1670) proposes to substitute
the following method for the titration of Nordhausen acid by
caustic-soda solution : — 50 or 100 grms. are weighed out in a long-
necked flask, and water is dropped in very slowly from a burette,
divided into ^^ c.c, till the fuming has ceased. During the
operation the flask must be well cooled. In order to observe the
246 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
finishing pointy the flask must be agitated after adding each drop
of -water till the fumes have been entirely absorbed by the acid.
When no fumes whatever are formed at the surface and a drop^
falling into the middle of the acid, dissolves quietly^ the end is
reached. Acid of more than 35 per cent. SOg must be previously
diluted with monohydrated sulphuric acid. It is claimed that this
plan is much more expeditious and quite as exact as the alkali-
metrical way ; but the analytical proofs adduced by Setlik himself
do not bear this out (there are deviations up to 0*9 per cent. SO3) ,
and we can regard his method only as a test for individual use, but
not between buyer and seller.
Rabe (Chem.-Zeit. 1901, p. 345) estimates the strength of fuming
(or ordinary) jsulphuric acid by utilizing the fact that Nordhausen
acid loses its property of fuming in contact with air as soon as all
its SOj has been converted into SO4H8 by the water present in
ordinary sulphuric acid containing less than 100 per cent.
SO.Hj:
a Hj^SO^ + c H3O + b H2SO4 + c SO4
=a H2SO4 + AH2SO4 + CH8SO4.
We require, to begin with, to know the percentage of a certain
sample of strong chemically pure sulphuric acid, say 95 per cent.
H3SO4 + 5 per cent. H2O, which is ascertained in the usual way
by titration. This acid we run from a glass-tap burette into a
beaker containing 25 c.c. of Nordhausen acid, repeatedly agitating
and blowing upon the acid (cooling is generally unnecessary), until
the acid in the beaker does not form any more fumes on shaking.
Suppose we require for this 24*8 c c. of the acid A^ containing
95 per cent. HjO and 5 per cent. HoO. If we wish to ascertain
the strength of another sample (B) of concentrated ordinary acid,
we run this from a burette into 25 c.c. of the same Nordhausen
acid as before. Suppose we now require 30*5 c.c. of acid B ;
24*8 X 5
this proves that acid B contains =4'07 per cent. H2O.
If no impurities were present, this would mean a strength of
95-93 per cent. HjSO^.
On the other hand, the percentage of SO3 in Nordhausen acid
can be ascertained as follows : — Starting with pure (say, 30 per
cent.) Nordhausen acid (C), we run in concentrated ordinary acid Z>,
ANALYSIS OF FaMINO SULPHURIC ACID, 247
and find that 59*4 c.c. of acid D is required to make the fuming of
25 c.c. acid C disappear. We now try an unknown Nordhausen
acid E, and find that 25 c.c. of it require 49'8 c.c. acid D. This gives
us the proportion 30 : 594= a? : 49*8 for the percentage of free
SOjin acid E, ^= — ^^74- = 25-15 per cent. SO3. 1 c.c. of
30
acid D had indicated 59^=0*505 per cent, free SO3; hence we
need only multiply the c.c. of acid D required for suppressing
the fumes of 25 c.c. of any unknown Nordhausen acid by the same
coeflScient, viz. 0'505.
[Rabe's method yields very quick results, and is probably quite
suitable for rough tests in the ordinary routine of acid-making.
There is no weighing, only measuring, and any colour or opacity
of the acids does not interfere with the test. But as the differ,
ence of specific gravities is neglected, this forms an element of
uncertainty, and this is greatly increased by the evident difficulty
of keeping a stock of exactly analyzed and chemically pure con-
centrated and Nordhausen acid without any change. As com-
mercial acids always contain certain impurities, these influence
the results as well. The only real advantages of this method over
that of Set.lik are that no cooling is required during the operation,
and that the large amount of concentrated acid is more accurately
read off than the small amount of water, but otherwise Setlik's
method is preferable.]
Nordhausen acid is always sold by the percentage of uncom-
bined sulphuric anhydride it contains (not taking any account of
the pyrosulphuric acid, which is considered = SO3 4- H2SO4) . Thus
'^ 30 per cent. Nordhausen acid '^ means a mixture of 30 parts by
weight of SOs with 70 parts of H2SO4. The price of SO3 is rela-
tively higher in weak than in strong acids, as in manufacturing it
the 5 or more per cent, of water contained in ordinary ^' rectified
oil of vitriol '' must be saturated with SO3, each part of water
requiring 4'444 parts of SO3 to form H8SO4. It is therefore
decidedly more advantageous to dilute strong Nordhausen acid
with the strongest obtainable rectified O.V., or still better with
monohydrated sulphuric acid.
The estimation of sulphuric acid in sulphates has been described,
pp. 64 et $eq.
248 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
Estimation of the Impurities of Sulphuric Acid.
The impurities of sulphuric acid are recognizable qualitatively in
the following manner : — A residue on evaporating sulphuric acid in
a platinum crucible may contain sulphates of sodium (more rarely
of potassium), of calcium, aluminium, iron, lead; copper, zinc, or
other metals occur rarely in sensible quantity. Ammonium sul-
phate is sometimes present in somewhat large quantities (Gintl,
Chem.-Zeit. 1879, p. 653). The individual substances are sought
for by the ordinary analytical methods. Iron is already betrayed
by the colour of the residue after ignition, and can also be detected
in the acid itself, without evaporating it, by the ordinary reagents,
such as potassium ferrocyanide, potassium sulphocyanide, &x;.
Lead is often shown as a white precipitate of sulphate on diluting
concentrated vitriol with water — ^further, by adding one or two
drops of hydrochloric acid, by which white clouds are formed,
which vanish on addition of more hydrochloric acid or on heating ;
with more certainty it is shown by diluting (even weaker acid)
with three or four times its volume of strong alcohol. The pre-
cipitate must, of course, be examined further — for instance, with
the blowpipe, by reduction on charcoal to metallic lead, by
moistening with ammonium sulphide (which blackens it), &c.
■ Arsenic is recognized by sulphuretted hydrogen in a dilute
solution, more safely by Reinsch's test — diluting with equal
volumes of water and pure hydrochloric acid, and immersing a
bright copper foil, which, after gentle heating, is covered with a
fast-adhering slate-grey precipitate, which, according to Lippert,
is a compound of copper and arsenic, Cu6As2 (if the arsenic is
present as arsenic acid, the reaction only sets in after longer
heating) ; further, by Marsh's apparatus, in which, by means of pure
zinc and water, the arsenic is given off as arseniuretted hydrogen,
and is proved by reduction in a red-hot tube (Berzelius), or by
lighting the gas and holding a piece of porcelain in the flame, on
which any arsenic appears as spots. Since it is difficult to procure
zinc absolutely free from arsenic, it is well to substitute aluminium
foil for it. There may be arsenic acid as well as arsenious acid
present ; this can be proved by neutralizing with ammonia and
adding magnesia mixture : any precipitate must contain the
arsenic acid, the filtrate the arsenious acid.
Selmi (Gazz. Chimica, x. p. 40) asserts that arsenic can be
detected in acid which gives no reaction by Marsh's test, by
DETECTION OP IMPURITIES OF SULPHURIC ACID. 24:9
adding to 1000 grams of the acid 300 grams water and some lead
chloride^ distilling and testing the first portions of the distillate
with sulphuretted hydrogen.
Seybel and Wikander (Chem.-Zeit. 1902, p. 50) prove the
presence of arsenic in sulphuric or hydrochloric acid by the
yellow precipitate of AsJg, produced by the addition of a solution
of potassium iodide. Sulphuric acid should be diluted to 45' B.,
hydrochloric acid should be employed in the concentrated state.
The reaction is interfered with by free chlorine, ferric salts, nitrous
acid (which equally cause a yellow coloration by the formation of
free iodine) , and by lead, which forms yellow Pb J2 . (Unfortunately
commercial acids mostly contain one or the other of these
impurities.)
The Committee of the Society of Chemical Industry, and of the
Society of Public Analysts, in 1901, recommend the Marsh-
Berzelius test as the most reliable and sensitive^ being capable of
recognizing 1 part As in 7 millions. Their detailed prescriptions
refer more particularly to the detection and appropriate valu-
ation of arsenic in beer and food- materials (Journ. Soc. Chem.
Ind. 1902, pp. 94 et seq,).
Of volatile substances sulphuric acid may contain : — hydrochloric
acid (from the common salt present in the nitrate of soda), to
be proved by nitrate of silver, after having diluted the acid, silver
sulphate being also very soluble ; hydrofluoric acid, to be proved
by heating in a platinum dish covered by a glass plate coated
with wax and containing scratched-in figures; sulphurous acid,
to be proved by the decolorization of a weakly blue solution of
iodized starch, or very accurately by reduction with zinc or
aluminium to sulphuretted hydrogen, which is recognized by its
turning lead-paper brown or by colouring purple an alkaline solu-
tion of sodium nitroprusside (comp. pp. 159 et seq.). The oxygen
compounds of nitrogen are nearly always present in the sulphuric
acid of trade. They are recognized in the simplest manner, and
with nearly as much precision as by any other test, either by the
decolorization of a drop of dilute solution of indigo on heating,
or by carefully pouring a solution of ferrous sulphate on the acid
contained in a test-tube, so that the liquids do not get mixed.
In the presence of traces of nitrous acids or of higher nitrogen
oxides a brown ring will be formed at the point of contact; if
more be present, the iron solution is coloured brown or black ;
250 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
but after some time it loses colour again^ especially if it has
become warm by the reaction. Selenium also gives a red ring
similar to that given by traces of nitrogen oxides ; but the colour,
instead of gradually vanishing, after standing for a time turns
into a red precipitate at the bottom of the test-tube. Nitrous
and hyponitric acids are also recognized by turning blue a solution
of starch containing potassium iodide.
The most sensitive reagent for nitrogen acids is diphenylamine,
which is most conveniently employed in a solution of 0*5 grm. in
100 c.c. concentrated sulphuric acid, diluted with about 20 c.c. of
water. A few c.c. of this solution is poured into a test-tube or
conical glass, and the solution to be tested is carefully poured
on the top, so that the liquids mix only gradually. If traces
of nitrogen acids are present, a fine blue colour is produced at
the point of contact. But as all other oxidizing substances, also
selenious acid, produce the same blue colour, errors may occur
through the (very frequent) presence of selenium, which gives
the blue reaction with diphenylamine even in the absence of any
trace of nitrogen acids. It is therefore necessary to test first with
ferrous sulphate, as described above.
As I have shown in Zsch. f. angew. Chem. 1894, p. 345, the
diphenylamine reaction is best employed in the following manner :
0'5 gram white diphenylamine is dissolved in 100 c.c. pure strong
sulphuric acid, adding 20 c.c. water ; the heat assists in dissolving
the substance, and the reagent keeps in well-stoppered bottles a
long time without turning brown. When testing for nitrogen
acids, pour a few c.c. of the specifically heavier liquid into a test-
tube and carefully pour the specifically lighter liquid on the top,
so that the layers only gradually mix. The presence of as little
as ^\) milligram nitrogen in the shape of nitrogen acids per litre
is indicated by a blue ring forming at the surface of contact of
both liquids, most easily perceived by holding the glass sideways
against a white background. Both nitric and nitrous acid are
indicated in this way.
Brucine indicates only nitric acid if there is a great excess of
strong sulphuric acid present ; neither selenium nor nitrous acid
interfere with this test, but nitrous acid equally reacts with
brucine if there is but little sulphuric acid and much water
present, say 1:2. In order to detect nitric acid by itself, an
aqueous solution to be tested should contain at least § of its
DETECTION OF NITROGEN ACIDS IN SULPHURIC ACID. 251
volume of strong sulphuric acid. The brucine can be added either
as powder or dissolved in pure strong sulphuric acid^ say 1 c.c. of
a solution of 0*2 gram brucine in 100 c.c. strong acid^ for 50
c.c. of the solution to be tested, of which | must consist of strong
sulphuric acid. If as little as y^ milligram nitric nitrogen be
present^ a pink colour is produced which gradually^ on heating
very quickly, passes through orange into yellow. In Zschr. f.
angew. Chem. 1894, p. 347, I have shown how this test can be
utilized for a quantitative colorimetric estimation of small
quantities of nitric acid. Gomp. also ibidem, 1902, pp. 1, 170,
and 241.
Most reagents, like diphenylamine, ferrous sulphate, and indigo,
indicate both nitric and nitrous acid. There are far more reagents
which prove the presence of nitrous acid (or nitrites) alone, not
that of nitric acid : for instance, a mixture of starch solution with
a solution of iodide of zinc (a blue colour being produced), and of
various organic amines, which with nitrous acid form corresponding
azo-colours (Griess, Berl. Ber. xi. p. 624) . Of these the most
frequently used are : metaphenylcne diamine, which produces a
yellow colour with 0*1 milligram nitrous acid in a litre, or else a
combination of sulphanilic acid and a-naphthylamine (reagent of
Griess). I have shown (Zeitschr. f. angew. Chem. 1889, p. 666)
that it is best to mix both substances, dissolved in dilute acetic
acid, at once, and to keep this solution ready for use ; any nitrous
acid getting in from the laboratory air is thus betrayed from the
outset by the reagent turning pink. This colour can be removed
by shaking up with zinc dust and filtering. For actual use, the
solution to be tested for nitrous acid is heated up to about 80^ C,
and a few c.c. of the mixed reagent added to it, when a rose-colour
will be developed with less than f f)\7o mgr. N2O3 in one or two
minutes. Solutions containing too much nitrous acid give only
a yellow colour. In order to obtain a reagent which is not dis-
coloured on keeping, a little of the a-naphthyl amine is boiled with a
few c.c. of water, the hot solution is poured oflF, and only this is
used, mixing it with dilute acetic acid and a dilute solution of
sulphanilic acid.
If any nitrous acid present is carefully destroyed by treatment
with urea, the ordinary reagents, like diphenylamine, ferrous
sulphate, and indigo, will indicate any nitric acid present, this not
being acted upon, by urea.
252 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
Selenium can be detected by the red colour imparted to a solution
of ferrous sulphate, which after some little time turns into a red
precipitate (not vanishing on heating like the brown colour produced
by nitric oxide), or by means of sulphur dioxide. According to
Jouve (Chem. Centralbl. 1901, i. p. 1889) codein or morphine
prove the presence of selenium, but only when 0*5 per cent, is
present, SO2 when merely O'Ol per cent. ; all of these act only
on selenious, not upon selenic acid. Both acids, however, are
proved by the red colour produced by the action of acetylene,
if O'OOl per cent. Se is present. A little HCl hastens the sepa-
ration of Se, which dissolves in the hot sulphuric acid with green
colour.
Orlow (Chem. Centralbl. 1901, i. p. 480) also rejects codein
and prefers SO2, especially on heating. 5 parts H2SO44-IO parts
water + 10 parts SO2 solution give a red precipitate at once
with 0*3 per cent. H2Se203, but also with 0*03 per cent, after
standing a few days or heating a few hours. Even 0*003 per
cent, gives a rose-colour. Rosenheim (ibid. 1901, ii. p. 234) dis-
cusses at length the influence of selenium on the ordinary tests
for arsenic.
The quantitative estimation of the impurities contained in sul-
phuric acid is best carried out with various portions of the sample.
Usually only the following are looked for. Lead is estimated by
diluting the acids, if concentrated, with its own volume of water
and twice the volume of absolute alcohol, when all of it is
precipitated by PbSO^. Iron is estimated by reducing with pure
zinc and titrating with potassium permanganate ; not leaving out
of sight its action upon SO2, N2O3, &c. A very convenient
colorimetric method for estimating traces of iron has been de-
scribed by me in Zsch. f. angew. Ch. 1896, p. 3 ; comp, also my
^Chera. techn. Untersuchungsmethoden,' i. p. 325. Arsenide
estimated by reducing any arsenic acid to arsenious acid by a
stream of SO2, expelling this by CO2, and precipitating by H2S.
The presence of lead, antimony, copper, platinum, &c. makes this
process very complicated (comp. thereon McCay, Amer. Chem.
Journ. vii, no. 6). If the quantity of As is somewhat considerable,
it can be reduced to AS2O3 by SO2, followed by CO2 ; the liquid
is then neutralized by soda, and the AS2O3 titrated by iodine
solution (Kisling, Chem. Ind. 1886, p. 137). Further particulars
are given in my * Untersuchuugsmethoden,^ i. p. 327.
ESTIMATION OF NITROGEN ACIDS IN SULPHURIC ACID. 253
The volatile impurities of sulphuric acid are estimated as
follows : —
Sulphurous acid, if at all present ia weighable quantities^ can be
estimated by a solution of iodine according to Bunsen's method.
The acids of nitrogen (nitrous, hyponitric, and nitric) cannot easily
be present along with sulphurous acid in sensible quantity ; their
quantity is very considerable, however, in certain intermediate
manufacturing products ('^nitrous vitriol''); and the methods for
estimating it are therefore of great importance. Also in chamber-
acid and in more concentrated products there is much oftener
nitrous or even nitric acid present than sulphurous acid ; and in
this case the estimation of even minute quantities is sometimes of
importance, because they exert a very injurious action during the
concentration of the acid in platinum.
Nitric oxide, as shown on p. 212, is soluble in sulphuric acid
only in extremely slight quantities, inappreciable in any ordinary
mode of testing. In practice accordingly no account need be
taken of nitric oxide, especially in the case of the stronger acids,
since in any case it cannot be present in sufficient quantity for
estimation ; and the latter need only refer to the proper acids of
nitrogen. Of these, again, only nitric and nitrous acid need be
taken into account. Nitrogen peroxide, NjO^ (formerly called
hyponitric acid), when dissolved in sulphuric acid behaves exactly
like a mixture of equal molecules of nitric and nitrous acid (p. 223).
Nitrous acid itself does not exist in any but rather dilute sulphuric
acid ; in somewhat concentrated acid it exists as nitroso-sulphonic
acid, S05j(OH)(ONO) (comp. pp. 215 et seg.). The solution of
this compound in sulphuric acid behaves, however, towards oxi-
dizing agents and in most other respects exactly like a solution of
nitrous acid, which, in fact, is formed from it by dilution with
water. Ordinarily in doing this, part of the NOgH is decomposed
into NO and NOjH (p. 219), but this decomposition, which would
interfere with the analysis, can be prevented by proper precautions,
as we shall see later on.
First of all we must describe the methods for estimatins: the
total nitrogen acid, that is nitrous and nitric acids together, in
which case the result can be calculated as NzO^ N2O5, NO3H, &c.
Frequently, for technical purposes, the N is calculated as NOaNa.
Of the many methods proposed for this end I only mention those
which are employed for technical purposes.
254 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
The method of Pelouze, raodiiied by Fresenius and others, is
only adapted for the estimation of nitric acid ; it is, however, some-
times used for estimating a mixture of this and of nitrous acid,
after the latter has been converted into nitric acid, for instance, by
chlorine, potassium bichromate, permanganate, &c. It is founded
upon the fact that free nitric acid oxidizes ferrous chloride or
sulphate, according to the equation
6 FeClg + 2 NO3H + 6 HCl = 6 FeCU + 2 NO + 4 H^O.
By means of potassium permanganate the ferrous salt, not oxidized
by nitric acid, is estimated, and the quantity of the latter is calcu-
lated from that of the ferrous salt consumed.
This method is described in great detail in our first edition,
vol. i. pp. 54 to 58, and second edition, vol. i. pp. 173 to 176; it
is not repeated here, as the much handier nitrometer method has
made it obsolete.
Another class of methods based upon the action of ferrous salts
on the nitrogen acids is that first proposed by Schloesing, and
subsequently modified by many others. In this class of methods,
the process is carried on in such manner that all nitrous and
nitric acid present is converted into nitric oxide, NO, which is
then estimated in various ways — mostly by measuring its volume
as a gas. This method is very much used by agricultural chemists,
especially in the modification introduced by Grandeau. A table
for reducing the volumes of NO to weights of N2O5 for various
temperatures and pressures has been calculated by Baumann
(Zeitschr. f. angew. Chem. 1888, p. 662 ; reprinted in Journ. Soc.
Chem. Ind. 1889, p. 135). For nitrous vitriol this method is
seldom used, because it is far more troublesome than the
nitrometric methods to be subsequently described.
Many methods are based on the reduction of the nitrogen acids
to ammonia by means of zinc, iron, or a combination of both.
All the older forms of these methods have been superseded by
the modification introduced by Ulsch (Zsch. angew. Ch. 1891,
p. 241 ; Lungers ' Untersuchungsmethoden,' i. p. 275), which is
extensively used for the estimation of nitrate of soda, but seldom
for the nitrogen acids present in sulphuric acid.
The process which is now mostly used for the estimation of the
total nitrogen adds in sulphuric acids (as well as for that of nitrate
of soda, comp. p. 96, of nitro-glycerin, and for many analogous
purposes) is the nitroineter method, founded upon a reaction
NITROMETER METHOD. 255
discovered by Walter Crum (Phil. Mag. 1840, xxx. p. 426). It
consists in agitating the substance in question with mercury in
presence of a large quantity of sulphuric acid, by which means all
the nitrogen acids are converted into nitric oxide, NO, whose
volume is ascertained by gasometric methods. Crumbs process had
been occasionally employed for the estimation of nitrates, e.g., by
Frankland and Armstrong; and had been specially recommended
for nitrous vitriol by G. E. Davis (Chem. News, xxxvii. p. 45).
But it attracted very little attention, least of all with technical
chemists, because it was cumbersome and expensive (requiring a
mercury trough) and withal gave no very trustworthy results^
owing to the diflSculty of manipulation. I drew fresh attention
to this process (Berl. Ber. xi. p. 436), and made it generally
accessible, both by proving its accuracy under the circumstances
here mentioned, and by devising for it a special instrument, which,
made its manipulation extremely easy and simple. This instru-
ment, which has since found a great variety of applications in
gasometric and gas- volumetric analyses, some of which are men-
tioned in other parts of this work, has been called the ^'Nitrometer',''
It is made in various shaped for various purposes; the shape which
is used in testing nitrous vitriol is shown in fig. 44.*
Its principal portion is a glass tube, a, of a little over 50 cub.
centims. capacity, divided into tenths of a cub. centim. At the
bottom it tapers to fit an elastic joint ; at the top it ends in a
funnel c communicating with the inner part of the tube by a three-
way tap. Its plug has one bore, through which the measuring-
tube communicates with the funnel, and another bore through
which the contents of the funnel can be run off. The division of
the measuring-tube, a, begins from the tap itself, and goes from
the top downwards. The tube a hangs in a clamp e, which can be
instantaneously opened by a spring, so that the tube can be taken
out. Another clamp/, sliding on the same stand, carries a plain
cylindrical glass tube, b, tapering below, of the same contents and
about the same diameter as the measuring-tube. The lower ends
of the two tubes are connected by a thick elastic tube, b slides
up and down in its clamp with friction. In order to use the
* Since I first published my above-quoted paper, Campbell, Davis, Dupont,
and others have made known apparatus very similar to mine, for which they
have adopted my name, ** Nitrometer." Not one of these, however, combines
all the advantages found in the instrument constructed by me.
XOb PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
apparatus, b is placed so that its lower eiid is rather higher than
the tap rf, and, the latter heing opened, mercury is ])oured in
through b till it just comes up to the funuel c. Aa'it flows
iuto a from helow, it will not allow any air-bubhles to remain in
the tabe. The tap d is now closed ; i is lowered ; and the acid
NITROMETER. 257
to be tested is run into the fannel c by means of a fine pipette. * Of
course it is necessary to have an idea of the maximum quantity of
NO which may be given off without expelling the mercury from the
tube altogether^ and the quantity of sulphuric acid must be taken
accordingly. By carefully opening the tap d, the acid is run into
a without any air being allowed to enter; in a similar way the
funnel c is washed out twice by means of pure concentrated sul-
phuric acid. It is not advisable to put more than 8 to 10 cub.
centim. of acid into the apparatus ; much better only 4 to 5 cub.
centim. altogether are used; but in any case there must be an
excess of strong sulphuric acid present. Now the tube a is taken
out of the spring clamp and well shaken up. The evolution of gas
in the case of nitrous acid commences at once — ^the acid taking a
purple colour^ in the case of nitric acid^ after a minute or so.
The reaction is ended by violent shaking for one or two minutes.
Sometimes it takes a long while before the acid clears and the
froth subsides, but generally this is effected in a very short time ;
anyhow it is necessary to wait a little, so that the apparatus may
assume the temperature of the air. Now by sliding b up or
down^ the level of the mercury in this tube is so placed that it is
as much higher than that of a as corresponds to the vitriol ; say,
for each 7 millim. of acid 1 millim. of mercury; or else the level
of the mercury is made the same in both tubes^ and the height of
mercury corresponding to the layer of vitriol in the tube is
deducted from the barometrical pressure. In the former case, it
is easy to ascertain after reading off whether the proper compensa-
tion for the height of the acid column has been made or not. It
is only necessary to cautiously open the tap d^ over which a drop
of acid has been left standing. If this is sucked in, and the level
of the acid falls, there has been too little pressure, and vice versd.
The volume of the nitric oxide can be read off to .^^ cub. centim. ;
it is reduced to 0° and 760 millim. mercurial pressure, and the
percentage of the acid calculated from it. Each cub. centim. of
NO^ measured at QP and 760 millim., corresponds to 1*343
milligr. NO, or 1701 railligr. N^Os, or 2-417 milligr. N^Og, or
4*521 milligr. NO3K, or 3*805 milligr. NOjNa. By this process, of
coui*se, nitric and nitrous acids cannot be distinguished, but are
always estimated together.
After reading off, b is again placed higher, the tap d is opened
so that tube a communicates with the small outlet tube, and thus
VOL. I. 8
258 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
first the nitric oxide and then the sulphuric acid^ muddy irith
mercuric sulphate, is driven out. When the mercury begins to
run out as well, the tap is closed^ and everything is again ready
for a new testing. If any sensible quantities of sulphurous acid be
present (as proved by the smell), it is best to add a very little
powdered potassium permanganate to the sulphuric acid, avoiding
any considerable excess.
The nitrometer was first provided with a three-way tap on
CI. Winkler's principle, with one transverse and one longitudinal
bore ; but now another arrangement, known as the ^^ Greiner-
Friedrichs '' or the " Geissler-Miescher '' tap, and shown in the
diagram, is preferred, which admits of much easier manipulation
and is far less liable to leakage.
It has been stated by T. Bayley that it is necessary to dilute
the acid contained in the nitrometer at the close of the experi*
ment, in order to expel the nitric oxide dissolved by the sul-
phuric acid, otherwise an error of about 0*5 c.c. is committed.
I have proved this to be wrong, no measurable quantity of
NO being dissolved by the acid contained in the nitrometer
(Joum. Soc. Chem. Ind, 1885, p. 447, and 1886, p. 82). This
could not be contradicted by Mr. Bayley, who, however, con-
tended that the iron contained in the acid as ferrous sulphate
acted as solvent for NO. I replied to this (Chem. News, 1886,
liii. p. 289) that the quantity of iron found in any commercial
acid would never lead to any appreciable error of this kind,
more particularly as it would be present as ferric sulphate.
From the volume of NO read o£E, and reduced to 0° C. and 760
mm., the nitrogen compounds present are calculated by the
foUowing table (p. 259), in which column a gives milligrams, and b
per cent, by weight, when employing in the nitrometer 1 c.c. acid
of 140°Tw.
Ill spite of the very great convenience, speed, and accuracy
of the nitrometric estimation of the nitrogen acids, many chemists
might have abstained from using it because the unavoidable
reduction of the volume of NO to 0° C. and 760 millim. pressure
appeared too tedious to them. In order to overcome this ob-
jection, I have calculated tables which admit of reducing any
volume of gas from 1 to 100 from any given temperature to 0° C,
and from any given pressure to 760 millim., by simple reading ofi*.
These tables were given in the Appendix to my first edition ; they
GAS-VOLUMETER.
259
N.
NO.
N,03.
NO3H.
cc.
NO. !
a.
1.
0-627
2.
1-254
3.
1-881
4.
2-508
5.
3135
6.
3-762
7.
4-389
8.
5016
9.
5-643
A.
a.
b.
a.
0-0366
i a0732
01098
01464
01830
0-2196
02662
0-2928
0-3294
1-343
2-686
4-029
5372;
6-715;
8-058'
9-401,
10-744
12087
0-0785
01570
0-2355
0-3140
0-3925
0-4710
0-5495
0-6280
0-7065
1-701
3-402
5103
6-804
8-5061
10-206
11-907
13-606:
15-3091
00995
01990
0-2985
0-3980
0-4975
0-5970
0-6965
07960
0-8955
2-8201
6-6401
8-460
11-280
14-100
16-920
19-740
22-560i
25-380
0-1648
0-3296
0-4944
0-6592
0-8240,
0-9888
11536
1-3184
1-4832,
3-805
7-610
11-415
15-220
19025
22-830
26-635
30-4401
34-2451
0-2225
0-4450
0-6675
0-8900
11125
1 -3360
1-5575
l-7800;
2-0025!
are also contained in Lunge and Hurter's ^ Alkali-Maker's Pocket-
book/ and are also separately published for use as wall-tables
by F. Vieweg & Sohn, Brunswick. Other tables, requiring very
little more time for use^ are found in Winkler-Lunge's ^ Hand-
book of Gas- Analysis/ 2nd edition^ pp. 177 et seq.
I abstain here from giving these or any other tables^ as an
instrument invented by myself, and called the *^ gas-volumeter '^
(Zeitschr. f. angew. Chem. 1890, p. 139; Berliner Berichte, 1890,
p. 440), has made unnecessary all calculations and tables in
connection with the reduction of volumes of gases to QP and 760
millim. This instrument, as shown in fig. 45 (p. 260), consists of
three glass tubes, all joined by very strong elastic tubes to a three-
way pipe D, and sliding upwards or downwards in strong clips.
Tube A is the measuring-tube, B the reduction-tube, C the level-
tube. A is divided into tenths of cub. centim. and generally holds 50
c.c. ; where larger volumes of gases are to be measured it is shaped
like B, and holds 60 or 100 c.c. in the upper, wider portion, and
another 40 c.c. (divided into ^ c.c.) in the lower, narrower portion.
The ^' reduction-tube " B holds 100 c.c. in the upper part, and
another 30 cc. (divided into ^-^ c.c.) in the lower part. This tube is
set once for all in the following way : — After putting the apparatus
together and partly filling it with mercury, the temperature close
to B and the barometric pressure are taken, and it is calculated,
by the well-known formula
(273 4-0760
273x/5>
(where / denotes the temperature in 0® C, b the height of the baro-
meter in millimetres), what would be the volume of 100 c.c. dry
air under the then atmospheric conditions. (This calculation can
s2
260
PROPERTIES OF OXIDES AND ACIDS OF SULPHUR,
be abridged by using any of the above-mentioned tables^ if they are-
at hand.) Suppose ^=20^^ 6=^750 millim. In this ease 100 c.c.
of dry air would occupy the volume 108*8 c.c. We now move B and
C so that, tap /being open, the level of the mercury in B is at 108*8,^
Fig. 45.
whilst the mercury in C is, of course, at the same level. Previously
to this we have introduced a drop of strong sulphuric acid*into B,.
but not sufficient to reach over the meniscus of the quicksilver^
which would be an impediment to taking the readings ; this is done,
because gases have afterwards to be measured in the dry state.
GA8-VOLUMETEE. 261
(In the more frequent ease in which tliis instrument is employed
for measuring moist gases, in lieu of sulphuric acid, a drop oE water
is introduced into B, and the calculation is made by deducting from
the barometric pressure the tension of aqueous vapour correspond-
ing to the existing temperature.) Now tap / is closed, and is
secured so that no air can enter or escape through it. In lieu of
this tap a capillary tube may be provided which is sealed by a
small flame, after having put a perforated piece of asbestos card-
board over the top of tube B, to prevent its temperature rising
during the sealing-operation. The best way of closing tube B
is by means of a mercury-scaled tap as described by me in Berl.
Bar. 1892, p. 8158.
It is quite evident that every time the level-tube C is raised so
that the mercury in B rises to the point 100, the air within B is
compressed to the volume it would occupy at 0^ and 760 millim.
independent of the temperature and barometric pressure actually
existing. Now suppose we have evolved or carried over into tube
A a certain volume of gas, and we adjust the position of the three
tubes so that the mercury in B stands at lOO*^, and that in A
exactly at the same level, it is evident that the gas in A is under
the same pressure as in B ; and, supposing its temperature to be
the same (which will be the case if the two tubes are close
together), the gas in A will be equally compressed as that in B to
the volume it would occupy at 0^ and 760 millim. barometric
pressure. The reading taken in A thus yields at once the corrected
volume without having to look at a thermometer or barometer, or
use any calculations or tables.
Tube A might be an ordinary nitrometer ; but it is far prefer-
able to use it only as a measuring-tube, and thus to keep it always
clean and dry, whilst the nitrometric operation proper is carried
out in the auxiliary '^ agitating-vessel " E. This is a non-
graduated vessel, holding 100 to 150 c.c, and connected by a
strong elastic tube with the level-tube F. The vessel E bears at the
top the usual three-way tap and cup c. The side-tube a can be
closed by a small ground-on cap b, or else by an india-rubber cap.
Before commencing the analytical operation, the tube F is raised
so that the mercury just issues out of a ; cap b is now put on and
tap c is closed. Now the nitrous vitriol (or solution of nitrate of
soda, comp. p. 96) is introduced through c, by carefully lowering
F, so that only the liquid, but no air, enters into E; strong
262 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
sulphuric acid follows^ to rinse out cup c ; the tap is now entirely
closed^ and E is violently shaken till the decomposition is complete
and no more NO is given off. The cap b prevents the mercury in
tube a from being thrown out in the shaking. The instrument is
allowed to cool down^ and is then put in the position shown in the
diagram^ so that the small tubes a and d are on the same level.
Previously a short piece of india-rubber tube has been slipped over
d, and by raising C the mercury has been forced right to the end
of d. Now cap b is taken off, and a is introduced into the short
elastic tube, till the glass tubes a and d touch. Now tube C is
lowered and F raised (as shown in the diagram) , and tap o is
cautiously opened {e having been left open before). The gas
will thus be transferred from E into A ; at the moment when the
sulphuric acid has entered into the bore of e^.but before it has got
inside of A, tap e is closed. Now the reading is taken as described
above ; the apparatus EF may be detached at any time and
cleaned as occasion requires.
The readings of the volume of NO taken in tube A may be
converted into grams of N2O3 or NaNO^, &c., by means of the
table given on p. 259. If nitrate of soda has to be analyzed^ each
c.c. will indicate 3*805 milligr. NaNOs ; hence, if 0*8805 gram of
nitrate were employed for the test, the numbei* of c.c. of NO
would at once indicate the percentage of NaN03. In the case of
nitrous vitriol the quantity will usually not be weighed, but
measured by means of a pipette, and the results obtained must
then be divided by the specific gravity of the acid to reduce them
to weight percentage. If the acid is near 140^ Tw., this is
unnecessary; for in this case a 1 c.c. pipette will deliver 1*70
gram acid, and as each c.c. of NO indicates 0*0017 NjO^, this
means that the number of c.c. read off is exactly = tenths of a per
cent, of N2O3 by weight of the nitrous vitriol.
We now proceed to the estimation of nitrous acid, or, more
properly speaking, of nitrososulphuric acid, present in sulphuric
acid, which is mostly sufiScient for testing the " nitrous vitriol "
from the Guy-Lussac tower.
Among all the analytical methods founded upon the oxidation
of nitrous acid, both the most convenient and the most accurate
is that Yiith potassium permanganate, first proposed by Feldhaus.
Even for scientific purposes there is not a more accurate method
for estimating nitrous acid in an acid solution than this, if other
oxidizable bodies be absent.
PERMANOANATE METHOD FOR NITROUS ACID. 263
Even nitric oxide is oxidized by this reagent^ according to this
equation :
10NO+6MnO4K + 9SO4H2=10NO3H + 3SO4K2 + 6SO4Mn
+4H2O.
Accordingly the seminormal solution of permanganate, each cub.
centim. of which corresponds to 0*004 O, will show 0*005 NO for
each cub. centim. Thus^ on the one hand^ nitric oxide can be
estimated quantitatively by this reagent : on the other hand, the
nitric oxide would make the estimation of nitrous acid inaccurate
if it were present at the same time, which, fortunately, is not the
case in sulphuric acid to an appreciable extent.
Nitrous acid itself is oxidized by permanganate, according to
the equation :
5 N2O3 + 4 Mn04K + 6SO4H, = 10 NO3H + 2 SO4K2 + 4 SOiMn
+ H2O.
Here every cub. centim. of seminormal permangauate solution
corresponds to 0*0095 gram N^Os.
The process formerly in use, where the permanganate solution
was run into the nitrous vitriol, has been shown to be quite
inaccurate by my investigations (owing to the formation of NO and
HNO3) ^^d has been replaced by the plan proposed by myself,
namely, manipulating in the following way : — ^The permanganate
is not run into the acid, but, on the contrary, a certain volume
of permanganate solution is taken, and the nitrous vitriol is run in
from a burette slowly, and with constant shaking, till the liquid is
just decolorized. In the cold there is some loss of time, since the
very dilute solution of permanganate is no longer acted upon instan-
taneously. This loss of time can be avoided by working at 30^
to 40° C, but no higher. When working with concentrated suU
phnric acid, this temperature is attained without any special means ;
otherwise the permanganate solution is heated up beforehand. If
seminormal solution is employed, it is diluted with about 100 c.c.
of tepid water. Sometimes a brown precipitate (of hydrated man-
ganese peroxide) is formed in the operation ; but this dissolves
later on, and the final result is quite as correct in these as in any
other cases.
In testing chamber acid, at most 5 c.c. of seminormal per-
manganate should be employed ; otherwise the quantity of sul-
phuric acid required for decolorizing it will be inconveniently
large. For proper " nitrous vitriol '' from the Gay-Lussac tower
264
PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
up to 50 C.C. permanganate may be taken« If the number of c.c.
of permanganate is called Xy and that of the acid required for
decolorizing it y, the quantity of N2O3 present, in grams per
9*5 X
litre of acid is — , calculated as
y
NO.H =
or as
NaNO«=
8
15-75^
y '
21-25 a?
y
The following table saves the calculation in all cases in which
a?=50. The column y gives the number of cub. centim. of nitrous
vitriol used^ a the percentage in grams per litre^ and b the per-
centage by weight for acid of 140° Tw. (For other strengths the
percentage by weight is calculated by dividing the figures of column
a by 10 times the specific gravity.) The figures in column a also
indicate 0*01 lb. avoirdupois per gallon^ or^ as nearly as possible^
ounces per cubic foot.
Table for testing Nitrous VitrioU
Acid
N,0,.
NO.H.
NOjNa.
consumed.
1
y-
a.
b.
a. h.
a.
/>.
c.c.
griu8. per
per cent, j
grins, per per cent.
grins, per per cent. |
litre.
1
litre.
litre.
10
47-5
2-80
78-8 4-62
106-2
6-22
11
432
2-54
71-6 4-20
96-5 ' 5-65 1
12
396
2-33
65-7 3-85
88-0
518
13
! 36-5
216
60-6 3-55
81-7
4-78 !
14
340
200 1
56-2 3i^
75-9 4-44 1
15
i 31-7
1-86 '
52-5 307
70-8 414
16
29-5
1-74
49-3 2-89
66-4 391
17
27-9
1-64 ,
46-3 2-71
62-5 ; 3-6.')
18
26-4
1-55
43-7 2-56
590 346
19
25-0
1-47
41-5 2*43
559 327
20
23-7
1-38
39-3 1 230
.^•1 311
21
22-6
1-33
37-5 ' 219
50-6 2-96
22
21-6
1-27
85-7 2-09
48-3 2-82
23
20-6
1-21
34-2 200
46-3 2-71
24
19-8
117
32-8 1-92
44-4 2-60
25
190
112
31-5 1-84
42-5 2-49
26
18-3
108
30-3 1-77
40-8 2-39
27
17-6
103 '
291 1-71
39-4 230
28
170
100
281 1-64
380 222
29
1(V4
096
27-1 1-58
36-7 2 15
30
15-8
0-93 '
26-3 1-54
a5-5 208
*
ANALYSIS OP NITROUS VITRIOL.
265
Table' for testing Xitroue Vitriol (continaed).
Acid
yfir
yo,n.
NO,Na.
consumed.
1
y.
a.
b.
a.
b.
a.
b.
cc.
flrms. per
litrer ,
percent
grms. per
litre.
per cent.
gnof. per
litre.
. per cent.
31
15-3
0-90
25-5
1-49
343
2-01
32
14-8
0-87
24-6
1-44
33-3
la'i
33
14-4
0-85
2:^9
1-40
323
1-89
34
13-9
0-82
232
1-36
313
1-84
;i5
13-6
0-80
22-5
1-32
304
1-78
m
132
0-78
21-9
1-28
29-5
173
tyi
128
«J-75
21-3
1-25
28-7
1-68
38
12-5
0-73
207
1-21
280
1-64
39
12-2
0-72
20^
118
27-3
160
40
11-9
0^70
19-7
115
266
1-56
41
11-6
0-68
19^2
112
25-9
1-52
42
11-3
i>66
18-8
110
25-3
1-4S
43
11-0
0-65
18-3
1-07
247
14.5
44
108
0-63
17-9
1-(X5
24-2
1-42
45
10-6
0-62
175
1-02
236
1-38
46
10-4
061
171
11)0
231
ia'»
47
10-1
059
16-8
0-98
226
1-32
48
9-9
0.")8
16-4
iym
222
V3{)
49
9-7
0-57
161
0-97
21-7
1-27
50
9-5
0-56
15-8
0-925
213
1-25
.w
86
050
14-4
O&i")
19-3
113
60
7-9
0-46
131
0-7a3
17-7
CK^i
65
7-3
0-43
121
o-7a>
16-4
(K«
70
68
044)
11-2
0-65.)
15-2
0-89
75
63
0-37
10-5
0615
1415
OS27
80
5 9 1
0-35
9-85
0-575
133
0-778
So
r}-(i
0-33
9-2
0 538
12 5
0-730
90
rrS
031
8-7
0-510
11-8
0 692
95
5-0
0-29
8-3
0-4rt5
11-2
0-(>.V)
100
■^■^ !
0-28
79
1
0462
10-6
0-020
In the presence ot other oxidizable substances, such as sulphurous
acid^ ferrous salts^ organic substances, &c., all oxidation methods
•are of course inexact — whether the bleaching-powder, or the
bichromate, or the permanganate process. Generally those im-
purities are too insignificant to do any harm; but, especially
where large quantities of nitrous acid are present, as in the
nitrous vitriol from the 6ay-Lussac towers, the permanganate
process is quite sufficient for the purpose of checking the course
of manufacture. Of the oxidizable substances only arsenious
acid sometimes occurs in sufficient quantities to affect the results
sensibly, but to a small extent only, in nitrous vitriol, where it is
mostly changed into arsenic acid.
266 PROPERTIES OF OXIDES AND ACIDS OF SULPHUR.
The estimation of nitrous acid by means of 'aniline^ which is
converted into a diazobenzol salt, the end of the reaction being
shown by potassium iodide and starch, has been practised for some
time at several colour- works, long before it was published by Green
and Ridea] (J. Soc. Chem. Ind. 1886, p. 633). According to
comparative tests made in my laboratory, it offers no advantage
whatever over the very much less troublesome permanganate
method, and may lead to serious errors (Zsch. angew. Ch. 1891,.
p. 629, 1902, p. 169).
Minute quantities of nitrogen acids cannot be quantitatively esti-
mated by the above methods, but the colorimetric estimation of slight
quantities of nitrous acid, as I have shown in Zsch. f. angew. Ch.
1894, p. 848, can be performed by Griess's reagent, modified as
follows : — 0*1 gram white a-naphthylamine is dissolved by boiling
in 100 c.c. water for a quarter of an hour; then 5 c.c. glacial
acetic acid (or its equivalent in ordinary acetic acid) and a solution
of 1 gram sulphanilic acid in 100 c.c. water are added. The
solution is kept in a well-stoppered bottle ; if it turns pink it is
decolorized by shaking with zinc-dust and filtering. A very slight
colour does not interfere with its use, as only 1 c.c. is employed for
50 c.c. of the solution to be tested. 1 c.c. of the reagent indicates-
-J J^^ milligram nitrous nitrogen in 100 c.c. water by turning the
water pink in 10 minutes. Strong mineral acids retard or stop
the reaction, but this can be remedied by adding a large excess of
pure sodium acetate.
For quantitative use a standard solution is prepared as follows : —
0'0493 gram pure sodium nitrite, containing O'OIO gram nitrogen,,
is dissolved in 100 c.c. water, and 10 c.c. of this solution is drop
by drop added to 90 c.c. pure sulphuric acid; the resulting
mixture contains y^^ milligram of nitrous nitrogen in a perfectly
stable form. Two colorimeter cylinders are charged as follows : —
Each of them receives. 1 c.c. of the Griess-Lunge reagent, 40 c.c.
of water, and about 5 grams of solid sodium acetate. To one of
these is added 1 c.c. of the standard solution, to the other 1 c.c. of
the acid to be tested. The contents of each cylinder are at once
thoroughly mixed, and after 5 or 10 minutes the colours are
compared. If they do not correspond, the more strongly coloured
liquid is diluted up to the point where layei*s of equal thickness,
show the same depth of colour in both solutions, and the percentage
of nitrous nitrogen is calculated from the amount of dilution.
COLORIMETRIC ESTIMATION OF NITRIC ACID. 267
Very minute quantities of nitric acid are best estimated by the
colorimetric brucine process, described by me, Zsch. angew. Ch.
1894, p. 347, the principle of which consists of comparing the
yellow colour obtained by heating to about 70^ with brucine
with a standard solution of nitric acid. As a rule, nitric acid
is not estimated by itself in sulphuric acid, but indirectly, by
estimating the total nitrogen acids by means of the nitrometer
(p. 253) and deducting the nitrous acid found by the permanganate
method (p. 262).
268 PRODUCTION OF SULPHUR DIOXIDE.
CHAPTER IV.
THE PEODUCTION OF SULPHUR DIOXIDE.
A, Brimstone-burners.
Already in the historical part attention has been drawn to the
point that important progress was made in the manufacture of
sulphuric acid when the periodical combustion of sulphur within
the acid-chambers was replaced by continuous work in special
apparatus attached to the chambers. This led to making the
sulphur-burners altogether independent of the chambers^ and con-
veying the gas generated in the former by a flue into the latter.
Whilst in the old periodical style of working only the oxygen
actually present in the chamber could come into play, and there-
fore after every combustion the chamber had to be supplied with
fresh air by opening the door and a special valve, of course at the
expense of much inconvenience and loss of gas, in the continuous
method of work the necessary air constantly enters the burner by
suitable openings at the same rate as the products of combustion
are aspirated into the chamber by the draught prevailing through-
out the apparatus. The continuity of work must be further
aided by employing a combination of several burners^ so that there
shall always be burning sulphur present. There are also burners
to which the brimstone is continuously supplied, in order to avoid
the drawback of irregular supply of air and gas occurring even with
the combination of several ordinary burners.
The plainest sulphur-burners, such as were the most usual in
England, are represented in figs. 46 to 48. The burner consists
of a brick chamber covered by an arch, the bottom being formed
by a cast-iron plate, a, separately shown in fig. 48. This plate
at the two long sides and one of the ends has a somewhat slanting-
up flange of 3 inches height — in front, however, only 1 inch, so
as to get out the ashes more easily. The plate does not go right
iNE-BURNBKS.
through the burner, but leaves the last third oE it free ; in this
part of the chamber the sulphur vapour, which is always formed,
can mix with the excess of air and be burned. Rarely, how-
ever, is this completely performed, and there is generally some
Fig. 46.
sulphur carried away ucbumed. This not only causes a loss,,
but also easily leads to the chamber-acid getting muddy and being
covered with a thin film of sublimed sulphur, which prevents the
contact between the bottom acid and the gas, very necessary for
the chamber process. The burner is further provided with aa
270 " PRODUCTION OF ISULPHUR DIOXIDE.
iron door^ b^ sliding in a frsnne and adjustable by a chain and
balance-weight; also witb a pipe^ c, for taking away the gas.
An air-channel, d, below the plate is in connection with a small
chimney, or sometimes only with the open air, in order to
cool the metal plate to some extent and prevent the sublimation
of sulphur. There are always several furnaces of this kind
combined together ; each of them, with plates of 8 feet x 4 feet,
can bum 5 cwt. of brimstone in 24 hours, which is put in in
6 portions, one every 4 hours ; if four furnaces are combined, one
of them is charged every hour, , Sometimes, however, much larger
and more frequent charges are made (see below *) , In these burners,
usually at the same time, the nitric acid is liberated by placing cast-
iron pots, provided with three feet and containing a mixture of nitre
and vitriol, amidst the burning sulphur by means of large tongs.
For a start the plates are heated by a small fire of wood shavings,
the door being left open, not till the iron becomes red-hot, but
only till the first charge of sulphur ignites of its own accord or
can be easily lighted by a red-hot iron; the further charges always
find the burner sufficiently warm. A special fire-grate below the
plate, to be used only at the start, is sometimes provided, but is
quite unnecessary. The admission of air is regulated by opening
the door. A, more or less widely; and its position is fixed by
putting a wedge underneath it, or by hooking the balance-chain
to a nail driven into the brickwork outside. At the commence-
ment, when the chambers are filled with air^ the damper in the
draught-tube is only opened gradually, to drive away the air more
thoroughly.
The style of working is generally rather rough; it must be
mentioned that such burners are nowadays seldom found in large
works, where more supervision can be used. Before the attendant
pulls up the door, he convinces himself regarding the state of the
chambers so as to judge how much nitre he is to ^'pof along
with the brimstone. Often the nitre is merely measured by guess-
work. First the brimstone is thrown in ; the door is immediately
let down, then a sufficient quantity of chamber-acid is poured
into the nitre-pots, always by guesswork; the door is opened
again^ and the pots are placed in the brimstone, now already on
♦ Davis (Chemical Engineering, ii. p. 123) states that up to 21b. of brimstone
cBn be burned per square foot per hour. This is more than the maximum
I have ever found in practice. It is best not to reckon upon much more than
1 lb. per square foot per hour.
OLD BBIUSTONE-BUBNEBS. 271
fire, by meaoB of au iron fork made for the purpose. The heat
produced by the progress of the combustion drives off the nitric
scid, and this enters the chambers along with the sulphur dioxide.
When the time is up, the door is raised agaiu, and the ashes are
raked out; first, however, the uitre-pots are Lifted out and emptied
of their liquid contentH. Then a new charge is made as above,
and so forth. In all other sulphur-burners, excepting the con-
tinuous ones, the work is carried on in the same manner : only the
iatrodnction of the nitre sometimes takes place in a less rough
way, or nitric acid is used directly in the chambers.
It is a sign that the burner is working well if the brimstone
bums with a pure blue flame ; as soon as the flame takes a brown
tinge, it indicates that much sulphur is Bubliming, and care must
then be taken to cool the plate,by the flue underneath.
A somewhat more perfect apparatus is shown in fig. 49. A is
Fig. 40.
the foundation, B the chamber for burning the sulphur, C the gas-
pipe. The foundation carries a cast-iron plate which covers the
whole furnace-bottom and is inclined a little forward. The
combustion-chambers are at the sides bounded by brick walls,
but in front, at the back, and at the top by cast-iron plates ; in
front also are the charging-doors, b b, and small openings, a a,
provided with slides for regulating the access of air. The furnace-
bottom is divided by 3- or 4-inch iron bars into three compart-
ments, corresponding to the doors and draught-boles, which are
served by turns. Inside the pots are visible, into which the mixture
of nitre and sulphuric acid is charged. The details of construction
272 PRODCCTION OP SULPHUK DIOXIDE.
are often very different from those shown in the diagram ; there are
humers with more or fewer working eompartments, with different
regolatiou of air, with rails over the bottom plates for pushing in
a box instead of the nitre-pots, &c. The nitre-pots must receive
a very small chaise : otherwise the danger of boiling over is
considerable, and the sodium sulphate among the sulphur is very-
troublesome. When nitric acid is used in the chambers, the
nitre-pots are not required at all.
Sometimes the iron sides of the sulphur •burners are made double,
and an air-channel is left in the space between. Thns, of course, the
temperature of the burner can be regulated to a nicety by opening
up a draught through the double iron wall when the burner gets
too hot, and shutting up the draught when it gets too cold.
Fi^. "lO. Fig. 1)1.
The diagrams figs. 50 to 53 show a set of two burners free from
most of the drawbacks mentioned. Fig. 53 is a sectional plan
taken at two different levels ; tig, 52 a longitudinal section ; fig. 50
half front elevation, half cross section ; fig. 51 back elevation.
a is the cast-iron bottom plate for burning the sulphur; it is
carried hollow on pillars ; and the channels b b formed thereby
underneath the plate communicate with the outer air by the open-
ing c, so that the plate can be cooled from below. The channels,
d d, left in the foundation a little further below, communicate with
IMPBOVED BKIUSTONI-BUBNEKI.
this system, aod ultimately end outside </. Owing to the differ-
ence of level and temperature, the air most always enter at rf' and
get out at e; its quantity can be easily regulated by partially
closing c. The door e is hung in the usual way. The gas of the
burner does not go straight to the chamber, but first ascends
through the opening / iuto a space separated from the burner
proper by an arch. Just above the opening there is a grating, on
which the nitre-pots are put, being introduced by the door ff.
VOL. I. T
274 PRODUCTION OF SULPHUR DIOXIDE.
There is here a small hole h, lined with an iron tube^ for admit-
ting a little more air to the upper compartment and burning any
sublimed sulphur. The gas first returns to the front, then back
again through the hole t and the second half of the upper com.
partment, and at last escapes through the cast-iron pipe k^ common
to two burners, whose upper stories are accordingly not built
alike, but are symmetrical.
This burner (known to inyself from actual use) admits of
very good regulation; the sublimed sulphur on its long course
through the upper story is either deposited as such or burnt, and
cannot get into the chambers. The boiling-over of the nitre-pots
can here be rendered harmless by simple contrivances. This
burner is in some points analogous to that of Harrison Blair (see
p. 277)^ but it is much simpler and adapted for a smaller scale of
work. As a rule the working-doors are closed within a very small
fraction, and the admission of air to the upper story is regulated by
more or less closing the hole b. Four such furnaces work together ;
every half-hour one of them is charged with ^ cwt. of brimstone.
Fish (B. C. p. 7757, 1891) makes the burner-bed incline to one
side, so that the sulphur can be gradually moved towards that side,
where the ashes are raked out.
In order to escape the drawback common to all sulphur-burners,
viz. the high temperature causing a sublimation of sulphur (which
some have tried to avoid by wetting the brimstone with water),
and even to turn it to some use, the cover of the burner is
occasionally employed for drying wet materials ; it has sometimes
been made in the shape of a pan for heating water or for con-
centrating acid, which is the most rational plan.
A large brimstone-burner, covered in with evaporating-pans, is
shown in figs. 54 to 56. Fig. 54 is a sectional elevation on the
lines E F G H of fig. 55 ; fig. 55 is a plan on lines A B C D of
fig. 54 ; fig. 56 a front view. These diagrams, representing a
furnace at work in America, have been kindly supplied by Dr.
Karl F. Stahl, of Johnstown (Pa.). Its bottom, roof, and sides
are formed of cast-iron plates, 1 in. thick, with 6-in. flanges bolted
together and caulked with rust-cement. A brick wall, 9 in. thick,
is carried all round, leaving a hollow space of 2 in. from the plates ;
a few of the bricks near the top and bottom are put in loosely, which
permits of air-cooling in very hot weather. The lead pans placed
on the iron roof are 3' 6" x 10' \0f' x 7", the weight of the lead
STAHL S BKIMSTONE-BUHNEK.
276 PBUDtlCTION OF SUI.PRVR DIOXIDE.
beiug 8 to 12 lb. per super, foot. The furnace bottom consists of
five plates, as seen in fig. 55 ; on each of the three front plates
(3' 6" X 12') from 1000 to 1300 lb. of brimBtoue can be burnt in
24 hour?.
An arrangement, made by Kuhlmann, for combining a steam-
boiler with a sulphur-burner (1st edition of this work, pp. 139,
140), did not answer at alt, and has been discontinued.
All the Bulphur-burnere hitherto described are built on the
intermittent plan ; and unless a number of them were working
together, they would yield a very unequal current of gas. As the
sulphur must, of course, be allowed to bum off as completely
Fig. r,fi.
as possible, the furnace in the final stage, and especially just
before being recharged, yields very little sulphur dioxide, whilst it
is not possible to regulate the draught so that exactly so much less
air is introduced as less sulphur is burnt. When at last the door
is opened for a new chai^, a very large quantity of air rushes
into the burner and further on to the chambers, without any
sulphur dioxide whatsoever. This irregularity, very prejudicial to
the chamber process, is certainly to a great extent neutralized
by the fact that always sevend furnaces (three, four, five, or
more) work together in such a way that they are charged in
turns ; for instance, with a four hours' shift and four (iirnaces
CONTINDOUS BRIM3ION£-BURK£a3. 277
one furnace is charged each hour, and thua gives out least gas
when its ueighboun are fully burning. It has, however, been
several times attempted to construct really coTitinuous burners,
which would save much labour, and, moreover, give a much
better-regulated current of gas than can be obtained with single
burners.
Two such continuous burners have been constructed by Petrie ;
we refer for diagrams and descriptions to our 1st edition, pp. 141
and 142.
Fig. 57.
The object pursued by Petrie is attained in a more perfect way
by the furnace of Harrison Blair, in which the volatilization of
the sulphur, which otherwise is a source o£ inconvenience, is
utilized to make the burning coutinuoua. The apparatus consists
of three parts, one of which serves for partly burning the sulphur
and entirely volatilizing the unburnt portion ; the second serves
278 PRODUCTION OF SULPHUR DIOXIDE.
for completely burning the latter portion, the third for decom-
posing the nitre. Although the two former compartments are
at a full red-heat during the process, sublimation of sulphur is as
good as impossible ; and the process is as nearly continuous as
possible, since the residue need only be withdrawn once in 24
hours. Fig. 57 shows a plan, fig. 58 a sectional elevation, of this
burner. A is the space corresponding to an ordinary burner-
plate, which has rather high sides and a descent towards the door ;
but 2 feet from the door it rises again a little, so that the residue
raked to that part can burn out completely before it is removed
by the door B, which takes place once in 24 hours. When this
has been done, the residue raked together from the other parts of
the burner is brought to the same place and allowed to burn for
24 hours again. The bottom of the burner is not made of iron,
but of closely-set bricks with well-grouted joints. This space A is
9 feet long, 6 feet wide, and 1 foot high. The door B is an iron
plate, loosely sliding in a frame, but a little slanting so that it
closes almost air-tight, and is easily removed. It is perforated by
a number of holes, which can be either partly or entirely closed
by a slide. The brimstone is either put in once every 24 hours
through the working-door, or gradually through a funnel C. C is
continued by a 7-inch cast-iron pipe to within 6 inches from the
bottom of the chamber; it is surrounded by a wider pipe to
protect it against being burnt too quickly. The funnel and its
continuation are always filled with brimstone ; and this is con-
tinually replaced as it melts off at the bottom. The simpler method
of charging once every 24 hours through the door seems after all
to have succeeded best. The admission of air through B is regu-
lated so that only sufficient sulphur is burnt for keeping up the
heat of the furnace ; the largest part is simply evaporated. At
the same time the regulation of the access of air allows of spread-
ing the process evenly over the whole day. The walls of the
furnace are made 1 ^ brick thick, in order to retain the heat.
The mixed gas and vapours enters through an opening of 9 X 9
inches (which can be closed by a fire-clay damper D) into
the combustion-space proper, E £, 8 x 6 feet, divided by three
partitions into four compartments, communicating alternately in
front and back by openings 9 inches square. Here at the same
time fresh air enters by the opening F, which is provided with a
damper of 3 x 8 inches. Now sufficient air is admitted for burning
CONTINUOUS BRIMSTONE-BURNERS. 279
all the sulphur^ which can be recognized with certainty by the
fact that on opening the plug G the entering air does not produce
a new flame. The roof of the combustion-space^ E^ is formed of
fire-tiles^ above which a second story, the nitre-oven, is situated.
There are three rows of nitre-pots, N, separated by reticulated
brickwork^ which also serves to support another roof of fire-tiles
for covering the nitre-oven^ altogether 18 inches high. The
diagram shows how the hot gas circulates round the nitre-
pots. The pots are renewed every six hours, so that every two
hours another row of pots has its turn. The hot gas, mixed with
the nitre-gas, first passes underneath the cast-iron dome, H^ for a
partial cooling, then through an iron pipe^ 24 feet high, into
a small cooling-chamber 18 feet long, 5 feet wide, and 1^ foot
high (whose bottom and top are covered with water) ^ and then
into the lead-chambers. Sometimes steam is admitted into the
combustion-furnace, which is said to hasten the formation of sul-
phuric acid. With a furnace of the dimensions stated, 26 tons
of brimstone per week are said to have been burnt in a per-
fectly satisfactory way, corresponding to the work of 15 ordinary
burners ; by cutting off part of the air it was possible to reduce
the sulphur burnt to 5 or 6 tons per week. For an equal chamber-
space much more sulphur can be burnt than with ordinary burners
without any damage to the process, owing to the even work and
the avoiding of any excess of air. Indeed Blair's burner is much
commended, and probably would have been more extensively em-
ployed, but that soon after its invention most large works (and only
such can do with it) have passed over from brimstone to pyrites.
At the present time, of course, nobody would think of such a
way of cooling the gas as is shown here in the cast-iron dome H.
We would employ its heat in a Glover tower, or previously for
concentrating acid. We would also replace the potting arrange-
ment shown in the diagram by the more perfect arrangements to
be described later on in connection with pyrites-kilns ; or we
would leave it out altogether, and supply the chambers witli
liquid nitric acid through the Glover tower.
A modification of the principle of burning the subliming sul-
phur by introducing air behind the burner was patented by
H. Glover (No. 3774, 1879). He arranges behind the burner a
chamber, loosely packed with bricks, in which the vaporized
sulphur deposits before it can get into the lead-chambers. This
280 PRODUCTION OF SULPHUR DIOXIDE.
brick chamber, when it is partially filled with sublimed sulphur,
is burned out by admitting air into it. The heat is utilized for
concentrating acid, and the gases are eventually passed into a
Glover tower, where they do all the necessary denitrating work.
This system is at work at a Philadelphia factory, and gives entire
satisfaction, as observed by myself, no repairs having been required
after the lapse of five years.
This arrangement is shown in fig. 59. A is the usual burner-
plate, B the feeding-apparatus (on the same principle as used in
Blair's burner) ; the burner-gases, with the subliming sulphur,
pass into the chamber C, where they meet air entering through the
pipes F, either cold or previously heated by waste heat. The
mixture further passes through chamber D, containing a network
of fire-bricks like that used in a Siemens' recuperator ; the mix-
ture and combustion here become perfect, and the gases, now
entirely deprived of free sulphur, pass away through H and the
flue IK. On their way a platinum dish, E, for concentrating sul-
phuric acid, is placed on the top of chamber D, and further (leaden)
pans, J and G, are employed for a first heating of the acid. From
here the acid gas passes into a Glover tower, where it is still hot
enough to concentrate all the Gay-Lussac acid (equal to 1^ times
the daily make of the chambers) up to 150° Tw., and impart to it
a temperature of 127° to 132° C. The lead pans G J and platinum
dish E produce daily 9000 lb. of acid of 91 or 92 per cent. HjSO^
from chamber-acid of about 123° Tw. ; that is, two-thirds of the
acid made from the 4000 lb. of brimstone burnt on plate A.
Since this acid is taken directly from the chambers (the Glover-
tower acid being used exclusively in the Gay-Lussac tower), and
since the concentration is not driven to a higher point than 92
per cent., the platinum dish never requires any cleaning out of
iron salts &c.
Another sulphur-burner, on the principle of continuously sup-
plying liquid sulphur, is that employed at the works of M. de
Hemptinne, at Brussels, and shown in figs. 60 and 61 (taken
from the ' Bulletin du Musee de ^Industrie de Belgique,' January
1882, sent to me by the Author). It consists of a cast-iron arch
A, strengthened by bracings, and resting on a large flanged
wrought-iron plate with flat rivets which can be heated or cooled
by flues underneath. On this plate there are placed, side by side
and as level as possible, four cast-iron plates with 3-inch upright
H. olotbb's beihstons>burner.
PftODttCTION OF SULPaVK DIOXIDE.
BRIM8TONE-BUANERS FOR WOOD-PULP WORKS. 283
flanges^ intended for burning the sulphur, which is supplied by
four spouts from a cast-iron box C, divided into four compartments
and built into the front wall^ as shown in the diagram. Perpen-
dicular partitions D serve as lutes for preventing the burner-gas
from blowing out in front ; if the combustion should spread to
the fronts a cover (not shown here) would at once put out the
flame. Thus the supply of sulphur takes place regularly ; the four
hinged doors a a in front serve merely for the entrance of the air
and for clearing out the cinders. The arch A consists of ten
pieces bolted together ; it is covered by light sheet-iron shutters,
E E, bent to the same shape and covered with a mixture of loam
and straw, which can bs raised or lowered by a chain, pulleys,
and counterpoises. This admits of regulating the heat of the
chamber ; if it rises too much, one or more of the shutters E E
are raised. An alarum thermometer, I, in a copper tube, indicates
the temperature. The gases escape through the metal pipe F,
resting on a thick-walled cast-iron box G, from which the deposit
formed can be withdrawn through J. Throujgh H an extra supply
of air can be let into the tube F. [This arrangement for supple-
mentary combustion is decidedly imperfect.]
For producing cold and dry sulphur dioxide, free from sulphuric
acid, such as is specially useful for preparing liquors for manu-
facturing wood-pulp^ N^methy (G. P. No. 48285) recommends the
combination of a sulphur-burner, cooled from the outside by
water running down the sides, with a chamber, placed underneath,
filled with iron borings, in which the sulphuric acid is retained.
From here the gas passes through a number of flat, perpendicular,
iron boxes, cooled by water running down their sides, and then
into the apparatus, where it is to be absorbed by milk of lime, &c.
W. Maynard (patented as a communication to A. M. Clark,
No. 6982, 1884) draws the gas out of the chamber where it is
generated (by burning sulphur in cups) by means of a goose-neck
pipe leading from the top of the chamber to a closed box provided
with a funnel delivering into another box below. Water is
discharged by a pipe into this funnel, which has grooved sides,
so that the liquor running round as well as downwards forms a
vortex, and draws away the vapours generated in the burning-
chamber. This arrangement is evidently not intended for sul-
phuric-acid making, but for preparing a solution of sulphurous
acid.
PRODUCTION OF BOLTBDB PIOXIDK.
Fig. 6-2.
BRIMSTONE-BURNERS FOR WOOD-PULP WORKS^ ETC. 285
The following arrangement^ by Korting Brothers^ serves for
preparing comparatively small quantities of sulphur dioxide from
sulphur, for bleaching, in the manufacture of glue, for saturating
the liquors in sugar-works, and the like. It is also used in wood*
pulp works. A (figs. 62 and 63) is a cast-iron retort provided with
a perforated dish a, in which the sulphur is placed- B is a
Korting^s injector, made of regulus metal (5 lead, 1 antimony),
which, by means of a steam-jet, aspirates air through the holes £6
into A and causes the sulphur to burn. The vapours are forced
downwards in the inner tube of the cast-iron cooler C, whilst cold
water flows in the annular space between the two tubes, entering
at the bottom and running out at the top. The box D, on which
the cooler is mounted, serves for retaining any sublimed sulphur
and other impurities. From here a tube leads the purified SO2 to
the place where it is to be utilized.
The N^methy burners, Korting burners, and others, are
especially used in Germany and Austria for the manufacture of
bisulphite of lime, to be employed in the manufacture of paper-
pulp (called ^' cellulose ") from wood. A number of other burners
serving for this purpose are described in the ^ Papier-Zeitung '
for 1894, pp. 1478 & 1830.
In Germany, in 1900, 33 paper-works made sulphite cellulose
by means of brimstone, of which they consumed 15,000 tons
against 33 works employing 70,000 tons pyrites. Both together
made by this process 350,000 tons wood-pulp.
The following analysis of the residue from the sulphurs-burners
has been made by Richardson (Richardson and Watts, ' Chemical
Technology,^ vol. i. pt. v. p. 198) : —
Sodium sulphate * 13*77
Calcium sulphate t 28*49
Calcium silicate t 15*91
Sodium silicate 1*10
Ferric oxide and alumina . . . 2*80
Water and sulphuric acid*... 13*05
Insoluble 24-29
99-41
* The fiodium sulphate and the free sulphuric acid (or rather the acid sul-
phate) evidently come from the nitre-pots boiling over,
t The lime no doubt partly comes from the brickwork of the furnace.
286 PRODUCTION OF SULPHUR DIOXIDE.
A special cooling of the gas from sulphur-burners for manu-
facturing sulphuric acid (as distinct from that of bisulphite of lime)
is^ as a rule; not only unnecessary^ but even injurious ; so that, for
instance^ in the furnace shown in figs. 50 to 53 the vertical metal
pipe conveying the gas to the chamber had to be protected against
cooling by a brick jacket. Even where no cooling takes place by
water-pans^ steam-boilers^ &c.^ the gas gets into the draught-pipe
sometimes at only about 100° or 120° G. temperature^ which is just
sufficient not to allow the nitric acid to condense before it gets
into the chambers^ a contingency decidedly to be avoided. Where
water-tanks, acid-pans^ &c. are used, the temperature of the gas
is said to come down as low as 40° C. ; in this case only liquid
nitric acid can be used for the chambers. In Blair^s or
Glover's continuous burner the temperature certainly rises much
higher; and in this case a cooling-arrangement^ such as that
described, was formerly thought indispensable, before means had
been found of utilizing the heat of the gases in a Glover tower or
otherwise.
B. The Production op Sulphur Dioxide from Pyrites.
1. Breaking the Pyrites.
The pyrites, as it comes into the market, is always sufficiently
pure to make a separation from gangue unnecessary, except in
the case of pyrites picked from coals (^^ coal-brasses,'' p. 41) ; but
this is only a locally-used by-product.
It is, however, always necessary to break up the larger lumps in
order to bum the pyrites completely ; and this is always done at
the works themselves — except in a few cases, where they buy smalls
direct from the mines. The majority of the factories break the
ore by hand; and it is found that various descriptions of ore
behave very differently in that respect. The Norwegian ore is the
hardest ; here the large lumps have to be broken with great labour
by means of 20-lb. fore-hammers. The Westphalian ore is much
more easily broken — still more easily the Spanish aftd Portuguese
and some of the French ores ; these, however, make a good deal
more smalls, 10 per cent, and more. The softest ore is that from
Chessy, consisting of loosely aggregated individual crystals, which
BREAKING THE PYRITES. 287
by a blow of the hammer fall to powder. Some of the Spanish
ores are equally roughly crystallized ; these ores are very trouble-
some for use as lumps.
In England the ore is generally broken so that all the pieces
pass through a sieve with 3-inch holes. On the other hand^ as
few smalls as possible are made. The broken ore must be sifted
again to separate the smalls^ for which purpose some works pass it
through a half-inch^ others through a quarter-inch riddle. What
remains on the riddle is lumps ; what passes through, smalls ^ fines,
or dust. Each of them has to be treated separately. It is very
important that the ore be used neither in too large nor in too small
pieces. In the former case it does not burn right through ;
there remain green cores in the interior of the cinders, which
can be seen on breaking them up. These large lumps also get
too hot on burning, and may cause the formation of slags (scars)
by production of FeS, as will be subsequently explained. If, on
the other hand^ the pieces are too small, they prevent too much
the access of air, and similar results follow in this as in the
former case.
It is quite obvious that the pyrites-burners can be worked to
full advantage only if the ore is in pieces of as uniform size as
possible, and it would hence be the best plan, although it is hardly
practicable in reality, to separate the broken ore into a number of
sizes, to be burnt in separate kilns. At Oker formerly the ore
was broken to walnut size for the grate-burners as shown later on.
For deep kilns, as now altogether used there, the ore is broken into
pieces of 2^ inches side. The fines, sifted through ^-inch sieves,
are burned in a Rhenania blende furnace (comp. later on).
Owing to the great manual labour required for the breaking of
pyrites, the same mechanical stone-breakers have been introduced
for this purpose which originally were made for road-metal. One
of these machines is that of Blake, built by Messrs. Marsden of
Leeds, which is shown in figs. 64 and 65, This machine is made
of various sizes, and accordingly varies in the amount of work
turned out and in the size of stones it can attack. A and B are the
two active parts, the ^^jaws.^' A is fast and perpendicular, B
movable, and makes with A an angle of 72^, by oscillating a little
round the fixed shaft D. This movement is communicated to the
jaw B from the main shaft, H, by means of the angle-lever, EE',
and the crank motion, 6 H, so that the angle-lever presses the jaw
FRODtlCTION OF SULPHUR BIOXIDE.
MKCHANICAL STONE-DREAKERS.
B against the atones Rharged^ the return motion of B being caused
by a. spring, F, cased in india-rubber. The angle-lever is adjust-
able by the wedge, N, lying behind its arm E'. The rollerj C,
causes the broken stones to be regularly ejected ; it receives its
motion by a belt from the main shaft, by means of the pulleys,
Kj L, and the expandiug roller, M. The crank-shaft, H, is also
driven by a belt from the fast and loose pulleys, 1 1'. The machine
is mounted on a four-wheeled bogie. It makes a great deal of
noise and needs frequent repairs ; but the jaws, which principally
suffer, are so arranged as to be easily replaced.
Blake's engine has been improved by Broadbent & Son, of
rig. (HI.
Staleybridge, who have replaced the spring bedded in india-rubber
by a simple, easily-adjustable lever arrangement, which saves
labour as compared with the original contrivance. Output, accord-
ing to size, from 40 to 130 tons in teu hours ; price £140 to £375.
At Oker, a steam-engine of 12 horse-power drives two stone-
breakers, mounted one above the other. The higher one breaks
the large stones roughly, the lower one down to the proper size.
They supply, in the case of very hard ore, of 2J inches, 60 tons,
with milder ore, 72 tons per shift of ten hours.
A crushing-mill was invented by Jfotte, at Dampreny, near
290 PRODUCTION OP BULPBUB DIOXIDE.
Charleroi, which has been improved by the Mftrkische Maaehinen
Pabrik (G. P., October 16th, 1877; 'Dingler's Journal/ ccxxvii.
p. 58). The principle is that of a peculiar kind of mortar, vith
hollow bottom, in which the crushing is done by a pestle, as seen
in fig. 66. Whether this mill is really preferable to the older
stone-breaking machines experience will show.
Durand and Chaptal's etone-brcaker consists of a number of
hammers attached to a horizontal revolving shaft. It is said to
make less dust than other atone-breakers. The smallest apparatus
breaks from 8 to 25 tons o£ stone in 10 hours, with an expenditure
of 2 or 3 horse-power, the larger size from 80 to 130 tons, with
6 horse -power.
Vapart's breaking-mill (address, "Chinee, Vieille Montagne")
works centrifugally,
The Humboldt Engineering Company at Cologne (Germ. Pat.
1906, Jan. 12, 1878) manufacture stone-breakers which do twice
PYRITES-BURXEKS FOR LUMP9. 291
«
the work of those formerly in use, with the expenditure of the
same force.
Other improvements in stone-breakers have been invented by
Brown (' Scientific Americau/ 1879, p. 194) and Welter (Qerman
Patent, No. 7494, March 5, 1879).
A machine very much recommended is Breuer's '* Sectorator ''
(G. P. 30477), supplied by Ernst Maetz, Berlin, S.W. As shown
in fig. 67, it contains a straight breaking-jaw, a, firmly connected
with the solid frame, whilst the movable jaw, &, is suspended in
two steel trunnions, and is partially revolved against the jaw a,
thus crushing the material. Tlie angle between the two jaws is
rather acute, so that large pieces are easily caught apd carried
forwards towards the bottom slit. If both jaws were arched, the
material, especially large pieces, would be able to escape upwards
for a long time, until gradually broken up. The width of the
bottom slit is adjustable by a wedge even during work, so that
any size can be obtained. The plate c behind the excentric sheave,
which is easily exchanged, is of cast-iron and of a smaller section
than any other part of the machine subjected to a breaking-strain,
so that in case of excessive strain by the passage of a foreign body
(iron, &c.), that plate must give way before any other part of the
machine.
Even at some large works they prefer dispensing with mechani-
cal stone-breakers, principally for two reasons. The first of these
is that they make more dust than breaking by hand ; but since
dust is now even more profitably burnt than pieces, this reason is
no longer valid. The other reason is, that at large works there
is always a certain number of men who are incapacitated for other
work, or who are temporarily unoccupied, and these are best set
to stone-breaking.
2. Pyrites-burners for Lumps.
Among the apparatus for burning pyrites in the manufacture of
sulphuric acid, a distinction has to be made between those intended
for lumps and those intended for smalls. It is indispensable to
keep both kinds apart, and to employ dificrent apparatus, or at least
processes, for them ; for if the broken ore is put into the burner
without separating the smalls, the air-channels, which ought to
remain between the pieces, are soon partly stopped up with powder.
292 PRODUCTION OF SULPHUR DIOXIDE.
and the access of air becomes irregular; thus scars are formed
and proper iirork is then impossible. Apart from the coarser and
finer powder obtained on breaking, a great deal of smalls comes
into the trade direct from the mines^ obtained there by the use of
water for separating the ore from the gangue.
Where cupreous pyrites is roasted without any regard to the
utilization of the sulphur, the only object being the extraction of
the copper, usually no regular kilns are employed at all, but the
ore is burnt in '^ heaps/^ This is done on an enormous scale in
the south of Spain ; but the damage to health and vegetation has
been so great that a law has been passed compelling manufac^
turers to abate this nuisance. In order to avoid the necessity
of building the very large number of closed kilns which would
be required for that purpose, various proposals have been made.
We quote that of Fleming (E. P. 10153, 1887). Above the
roasting-heap, and extending downward over the whole portion
(about one-third) which emits the fumes, is suspended an iron
hood, lined with tar and painted outside with a non-conducting
material. The hood is supported by chains from two pairs ot
shear-legs, and the whole is strengthened by iron stays. At one
end of the hood is a pipe, by which the roaster gases are led to
condensing-flues, to separate arsenious acid, and thence into vitriol-
chambers. If the gases have excess of air, they are led through
calcining-furnaces ; if insufficient, more can be supplied by regu-
lating-dampers in the flues. [Apart from all other objections to
this process, the " iron hood lined with tar " is sure to be not
very durable.]
The burning of pyrites in lumps (pieces) for the manufacture of
sulphuric acid is always done in such a way that the combustion
heat of the pyrites is utilized for maintaining the process without
employing any extraneous fuel. The apparatus used for this purpose
are called "kilns" or "burners.'^ In the first edition of this
work (pp. 151-154) there will be found a description of the old
and now abandoned kinds of kilns, with many diagrams — as the
burner originally employed by Farmer, the first shape of tall kilns
without grates, the Oker kilns for roasting the Bammelsberg ores.
These kilns (except Farmer's) are constructed without grates ; they
are still used for roasting poor ores, lead-matte, &c. In the
same edition there is a description of the Belgian hearth-furnaces
MODERN FREIBERG KILNS.
and of the Marseilles furnaces, both of them very faulty and quite
antiquated (pp. 157 & 158). This also holds good of the yarious
descriptions and illustrations of " Freiberg kilns," still given in our
second edition (pp. 214^318). Instead of these, I am enabled,
through the kindness of the respective authorities, to show the kilns
used at the present day (1902) at Freiberg and at Oker, at the
Government vrorks carried on at those localities.
PRODUCTION OF SULPHUR DIOXIDE.
Tig.oa.
MODERN VREIBERO KILNS. 295
For poor ores and intermediate products which must be treated
at metallurgical works (pp. 81 et seq.) furnaces are required of a
different kind from the grate-burners now uniTersally employed
for good pyrites in lumps. The style of kilns now used at the
Muldenhiitten near Freiberg is shown in figs. -68, 69^ 70. They
serve for poor iron-pyrites containing blende and arsenical pyrites^
as well as for lead- and copper-matte.
The grate formerly employed at Freiberg has been replaced
by slanting cast-iron plates^ ff. The air do^ liot now enter
through special channels^ but through the discharging and working
holes, a shows the charging hopper^ b the channel through which
the charge gets into the kiln^ c the erit channel for the roasting
gases^ d openings for spreading out the charge, e working-holes^
f discharging-holes. Each kiln roasts about 25 cwt. of pyrites
per 24 hours ; 5 kilns are combined in a set. The sulphur is
roasted off to 4 or 5 per cent, in the cinders.
For richer and purer pyrites, at Freiberg grate-burners are
employed consisting of three kilns, with 25 square feet grate sur-
face each, and a distance of 4 feet from th^ movable grates to the
crown of the arch. Each set roasts about 86 cwt. pyrites per
24 hours down to 2 or 3 per cent. S.
The kilns now used (1902) at Oker are exactly like the Freiberg
kilns just described. They are of two different sizes — deeper
kilns (with a layer of ore 9 ft. deep) serve for the poorer ores,
shallower kilns (the ore lying 6 feet deep) for the richer ores
(described p. 83). The English grate-burners, formerly employed
at Oker, have not been found suitable for this class of ores.
Kilns of the just-described kind have been found indispensable
for roasting poor ores, matte, &c., where the sulphuric acid is a
by-product and where the heat generated in the process is less
than when roasting ordinary pyrites, containing at least 40 per
cent, sulphur, usually a good deal more, such as is now universally
employed for the manufacture of sulphuric acid as a principal
product. For such richer ores the kilns or burners ought always
' to be constructed with grates and ash-pits. This causes a con-
siderable improvement in the working of the furnaces. Where
the air has merely to pass through a mass of burnt ore, its
quantity cannot possibly be regulated at the inlet, but only by
dampers at the other end of the furnace. It is even a more serious
disadvantage that in this case the subdivision of the air inside the
396 PRODUCTION OF SULPHUR DIOXIDE.
burner must be very irregular. According to the greater or smaller
■ I'esistauce offered fay the individual portions of the layer of pyrites,
the air vill pass through Tery unequally, and in less quantity
at the places where most pyrites is lying and where it is most
required. The addition of a grate and a closed ash-pit alters
the state of the case at once, in this way, that only a definite
quantity of air need be admitted into the ash-pit, and that, more-
over, this air must first spread equally uudenieath the grate and
rise all over the area of the burner. Thus the ore is much more
completely burnt, and at the same time richer gas is obtained,
which leads to a better chamber-process, higher yield of acid, and
smaller consumption of nitre ; the operation of drawing out the
burnt ore becomes much more regular and oders a greater guarantee
against raw ore getting into it ; lastly, it does not happen so
often that fused masses ("scars") are formed in the burner,
although also in the case of grates this easily happens if the
method of working ie faulty.
The introduction of grates led to further improvements — to
begin with, a diminutipn of the height of the burners, which made
them much handier for working, and wbicb acted especially well
with more easily fusible ores, although in some places the other
extreme of too low layers o£ pyrites has been resorted to.
Fig. n.
The different kinds of grate-burners which were introduced into
England about 1860, and have been employed up to this day both
there and in many factories abroad, are shown in figs. 71—75.
ENGLISH ORATB-BU&NEBS.
Figs. 71, 72, 73 show a somewhat simple eoastructioa, which
cao t>e made with open sand-castiogs ; figs. 74 and 75 a more
expensive kind of front plates^ requiring planing, turning, &c. : the
latter are much neater and cleaner, because no putty is required
for the doors. Sometimes these front plates undoubtedly become
8 little warped, and then the doors are not tight without putty.
Fig. 71 shows two burners in front elevation and one in section,
the first burner without doors. Fig. 73 is a cross section, showing
Fig. 73.
two rows back to back; fig. 73 a sectional plan, half taken just
over the grate, half through the middle of a door, a is the working-
opening, with the door b, which^slides in the grooved ledges, c c.
298 PRODUCTION OF STTLPnUR DIOXIDE.
cast to the front plate. The small door d, only to be used excep-
tionally, is arranged in precisely the same way. The openings of
the brickwork inside are protected by small metal plates ; e is the
movable cover of the ash-pit, provided with air-boles ; // are the
grate>bearers ; the front bearer /, at the same time carries the
bottom plate for the front wall, and is perforated with round holes ;
while //are cut out in semicircles. The arches are sprung parallel
with the working- doors, and, by the draught-holes g g, are in
connection with the gas-flues, A h. The latter, like the burners,
are cased in metal plates ; they are covered with fire-tiles.
Fig. 74.
A somewhat more costly but more perfect arrangement is shown
in figs. 74 and 75, in front elevation and two sectional elevations.
a is the working-door, with the small slide b for observing the in-
terior of the burner; it turns on hinges, and, as shown in fig. 75,
lies on a projection of the front plate, slanting forward towards the
bottom ; all the metal parts coming into contact are planed and
faced, so as to close air-tight. The doors c c for the grate aud d
for the asb-pit are constructed in the same way, whilst the rarely
ased doors e and/ (the latter for the gas-fiue) are made in the
simple manner shown in fig. 71. The burners are supposed to
be the last of the row ; so that the nitre-oven g, with the semi-
cylindrical trough h, the saucer t, and the hopper k are immediately
joined to them. The diagrams are all on a scale of 1 to 50.
ENGLISH FYKITES-BI'B.MERS, 299
Euglislt pjiites-bumers geaerallf have a moderate area of
gratca, about 4 or 5 feet wide, and 4^ to 6 Eeet from f roDt to back.
The inner walls sometimes rise quite perpeudicnlarly ; more
frequently the two sides and the back slant a little outwards, up
to the level of the working-door, to the extent of about 9 inches,
sometimes only 6 inches, in width, and half as inacb in the back ;
from that level the walls rise again perpendicularly up to the roof.
The front wall, which is only 9 inches thick, and mostly protected
by a 1-incb or l^-inch metal plate, rises perpendicularly, and is per-
forated with several working-holes. The ash-pit has either vertical
sides or, more rarely, sides converging towards the bottom, in
order to facilitate the removal of the cinders. Its depth varies
from 16 to 24 inches. The level of the working-doors, which
determines the depth of the layer of pyrites, varies from 1 foot
8 inches to 2 feet 6 inches ; but the former depth is considered
by most English acid-makers too little, at any rate for average
ores, and they prefer a depth of between 2 feet and 2 feet
4 inches, but nearer the higher than the lower limit. In
(iermany, the depth of ore is only 1 foot 6 inches, even
down to 1 foot 4 inches (comp. p. 305). The reason for this is
the fear of scarring, which English experience with the same ores
proves to be unfounded. The height from the upper level of the
ore to the abutment of the arch is usually about equal to that
of the working-door, say 9^ to 12 inches, and from there up to
the crown of the arch another 8 or 9 inches. The arch itself is
either sprung from side to side, as is the custom on the Tyne
300 PRODUCTION OF SULPHUR DIOXIDE.
(whereby the walls are made to bear the weight more equally and
the working through the doors is facilitated)^ or, as is usual in
Lancashire, from front to back (which is more advisable in the
case of two rows of burners being built back to back, in which case
the arch is sprung over both burners together, with a supporting
wall in the centre). In any case it is advisable to build the
burners back to back, even with arches sprung from side to side,
wherever it is locally possible ; thus one back wall is saved, the
heat is kept up better^ and a common gas-flue can be employed.
The gaS'fiue of the English burners is always at the top, each
burner-arch having a hole of- 4 to 5 inches square leading into it.
These holes are not always provided with dampers; but by gradually
increasing the size of the whole as the distance from the main
shaft becomes greater, evenness of draught is produced. The flue
itself can be made of bricks set in tar and sand, and covered with
fire-tiles. Most modern works prefer forming it by a second arch,
about 6 or 12 inches above the burner-roof, reaching right across
the whole burner, and supported by the front plate being made high
enough.
The principal feature of the English pyrites-burners, which is
used in all continental works as well, except in some burners for
metallurgical purposes (Mansfield or Freiberg kilns), is the employ-
ment oi grate-bars of square or oblong section, movable in bearings,
and leaving larger or smaller spaces between them, according to
Fig. 76.
Fig. 77.
^ -i
their position. (According to Hasenclever, in Hofmann's 'Bericht,'
1875, i. p. 158, movable grate-bars have been used in Prance ever
since 184*.) Fig. 76 represents such a grate-bar, showing ako
the parts which are forged or cast round, so that they can easily
turn in the respective hollows of the bearers. Bars 2 inches
square are usually made of wrought iron ; the oblong bars, 2 by
GRATES FOR PYRITES-BURNERS.
301
3 inches^ which^ being turned on edge^ leave a larger space^ and
therefore only suit larger pieces, are mostly of cast iron. The
grate-bars rest on cast-iron bearers, as shown in fig. 17 ; in the
shallower kilns (4^ to 5 feet from front to back) there are two
such, in the deeper kilns (5 feet 3 inches to 6 feet from front
to back inside) three. According to this, of course, two or three
rounded places must be made on the bars themselves. Lest these
should be weakened too much, the diameter of the round parts in
the square bars is equal to the side of the square, in the oblong
ones equal to the smaller side. In any case the front piece of each
bar, where it projects beyond the bearing-bar, remains square
or oblong, so that it can be turned round its axis by means oE a
suitable key (fig. 78) . The intervals between the grate-bars are
mostly mafiaged so that with 2-inch bars they are about 2 inches
when the bars are in the situation shown in fig. 79 ; but if they are
turned 90 degrees, as in fig. 80, the intervals will only amount to
1^ inch. In another actual instance the diameter of the bars
was ]| inch, the clear distance in the straight position 1^ inch,
in the diagonal position 1 inch. If, lastly, the situation is as
in fig. 81, where half of the bars are turned, the intervals will be
Fig. 78.
between the two above limits; and as each bar can be turned
separately, many combinations can be produced. Usually the bars
stand as in fig. 80 — that is, all with their diagonals in a horizontal
plane, or with the smallest possible intervals, so that the pieces of
ore cannot fall through. As soon as a portion of the ore has to
be removed, the attendant takes hold of the front end of the bar
with his key, and moves it a few times from side to side. Thus a
kind of crushing action will be exercised on the cinders getting
between the two bars, the intervals arc momentarily enlarged, and
304
PRODUCTION OF SULPHUR DIOXIDE.
the cinders jammed between the bars are forced downwards. Of
coarse great strength is required for this work. At the same
time, by the action of the key, the pyrites is loosened up to
a certain height. The workman now goes from one bar to
another, generally missing one, and shakes them, according to
his perception, so far that an equal quantity of burnt ore is
^.y. «
Fig. 79.
Fig. 80.
1
\:
Fig. 81.
drawn out all over the area of the grate. That which has fallen
through is allowed to lie in the ash-pit till the time comes,
once every 24 hours, for opening the bottom door and taking
away the cinders. A new shape of bars, which was said to
possess great advantages over the ordinary angular ones, was
patented by W. Helbig (Dingl. Journ. ccxxvii. p. 67) and is shown
on p. 222 of our second edition, but omitted here, as it seems to
have found no practical application. It was a cast-iron bar with
a worm-thread all round it.
A burner patented by Harlan & Grenshaw (G. P. 100,243)
contains hollow grate-bars, with tapering, narrow chambers, rising
vertically nearly to the top of the layer of pyrites, so that the air
enters not merely at the bottom above the grate, but also higher
up, nearly to the top of the pyrites.
Dr. Burgemeister (private communication) employs bars made
of a cruciform section. When turning these round, the smaller
pieces easily fall through ; the larger lumps get between the bars
and either pass through on turning back or are crushed. The
following diagram shows the different positions of the bars :
+ + + +
X X X X
It is a great improvement (but rarely met with, because it
• DISCHARGING THE BURNT ORE. 303
necessitates a somewhat complicated plant) if the ashpit is deep
enough for introducing an iron bogie below the grate whose top
equals the whole surface of the grate in size ; the ash-pit door^ of
course^ must be correspondingly large. The cinders in that case fall
direct into the bogie, and can be wheeled out in a few moments ;
usually they have to be raked out by hand, during which time
the door must stand open, and much false air gets into the
burner. Where there are not two rows of burners built back to
back, it is possible to charge on one side and discharge on the
other; but there is not much advantage in this arrangement, which
takes a great deal of space.
The discharging of the burnt ore {cinders) is sometimes expedited
by iron bogies running on tramways, which are introduced into the
ash-pits, and into which the cinders fall as the grates are shaken.
This very suitable plan necessitates a system of tramw^ays and
turn-tables, as well as a lowering of the whole floor. The follow-
ing simple and equally efficient plan seems therefore worthy of
recommendation. It is a tilting-box (figs. 82 & 83) . There are
two independent parts : — first, an iron box, of suitable dimensions,
with two outside pivots near the upper edge ; secondly, a light
but strong wheeled frame, which ends in forks fitting under the
pivots. As shown in the diagram, the whole is used like an
ordinary iron wheelbarrow on any hard ground. But by lifting
up the handle the box is first lowered to the groimd, then the
forks slip out and the frame can be run back. Similarly the box
is taken up again by running the frame in, and depressing the
handles till the forks take hold of the pivots. The boxes can
be made to fit into the ash-pits, and the cinders discharged into
them directly from the grates. Of course other applications of
this barrow will present themselves in chemical works.
In order not to be obliged to open the whole ash-pit when
shaking the bars, all the best furnaces are provided with a slit
in the front plate, through which the ends of the bars are
accessible. Except whilst the bars are being shaken, the slit is
covered by a door, which is best made in two halves.
Norrington (£. P. No. 4131, 1878) makes the ash-pit doors to
slide in horizontal frames, and connects all the doors of a set of
burners by jointed rods, so that they can all be moved together in
a horizontal plane by means of an endless screw and gearing at
one end. All the ash-pit doors are thus opened and shut at the
304 PKOOUCTION OF 8ULPHUH DIOXIDE,
same time. This is always done whenever any one of the working-
doors is opened bo that no gas can blow out, as the air cannot
enter in any otlier way. According to information from Messrs.
Charles Norrington & Co., Plymouth, this arrangement effects a
llg. 8-2.
considerable saving of nitrate of soda and of sulphuric acid ; owing
to the regularity of draught, all kilns burn equally well. This
statement is confirTned by Dr. Ballard (Report to the Local
Government Board, 1879, p. 180) .
According to a communication from Mr, K. Walter, a simple
means of preventing the blowing out of gas at the working-door
during charging is this : to arrange a flue underneath the burners,
in connection with the chimney, which is opened during the
charging just suflicient to prevent any blowing out at tl\e
PREVENTING NUISANCE DURING THE DISCHARGING. 305
working-door. Less gas is lost and less nuisance is produced in
this way than is otherwise the case from the working-doors. With
this arrangement the ash-pits require only loosely put-on wrought-
iron doors.
In England it would probably not be allowed to discharge the
gas into the chimney, even for a short time : this objection is
overcome by the following plan : —
Hasenclever (Chem. Ind. 1895, p. 494) describes the following
arrangement for preventing any escape of gas from the burner-
doors into the working-shed during the time the doors are open
for charging. In some cases the chambers are placed high enough
over the burners (say, at least, 18 feet above the floor) to secure
sufficient draught at all times; but even then the false air
entering during the time of charging may be troublesome in the
lead chanjbers. The device proposed by Jurisch (Ueber die
Gefahren f iir die Arbeiter in chemischeu Fabriken, p. 37) that a
by-pass should be made from the burners directly into the
chambers, which should be opened by means of a damper every
time a burner is charged, is altogether impracticable, and is evi-
dently not practiced anywhere. The difficulty, however, is solved
by connecting, during the time the kilns are being charged, the
ash-pit (by means of a special flue) with a chimney producing but
little draught, so that no burner-gas is drawn downwards into the
flue and the chimney, but the burning of pyrites in the kiln is
interrupted.
This is done at the Rhenania Chemical Works by means of the
arrangements shown in flgs. 84-87, which, at the same time,
illustrate the form of burners (p. 299) frequently employed on
the Continent, where the burners have a common gas-space, the
pyrites lies on the grate in rather shallow layers (19^ inches),
and the cinders are removed by means of bogies run under-
ground into the ash-pit. The burner-gas in the ordinary way
passes from the dust-chamber through pipe h into the Glover
tower. Each time the cinders are let down into the ash-pit
by shaking the grate-bars, damper b c d e (fig. 86) is shut down,
so as to close flue / against the outer air, but not hermetically,
and damper S is opened, which leads through g into a chimney,
which is not connected with any furnace, and is only 25 or 30
feet higli, so that it just projects over the roof of the shed.
The small quantity of air which enters round damper b c d e
VOL. I. X
DOORS AND BKICKWOKK OP PYRrrBS-BUKXERii. 307
18 partly drawn into this chimney, and the pyrites consequently
bums so slowly that no gas issues from the doors a, ai, a^, a^.
The men now charge the burner with fresh pyrites through these
doors, shut the doors, let down damper S, and raise damper b c de,
whereupon the evolution of sulphur dioxide recommences, and
goes on all the more regularly the more burners are united in the
same set. Experience has shown that no sulphur dioxide is drawn
down into fff during the operation, so that none can escape
through the special chimney.
The Stassfurter Chemische Fabrik, vormals Vorster & Griineberg
(G. P. 100,708), arranges a flue connecting two sets of burners,
or the single burners of a set, below the level of the grates. In
this case, when the charging-doors of one burner are open and
those of the other burner are shut, the air-openings below the
grate being shut in both, the second burner will draw its supply
of air from the first, through its open doors, so that no gas will
blow out of these.
In properly constructed pyrites-kilns, all doors for charging,
working, shaking of the bars, and getting out the cinders either
run horizontally in grooves, or, still better, they are hung on
hinges ; and the door-frame, cast upon the front plate, is made to
slant forwards below^ sometimes also sideways, so that the door
lies fast upon it by its own weight. As both the door-frame and
the edges of the door touching it are planed, the doors close tight
without any luting, whilst those running in grooves must be made
tight with lime-putty.
All brickwork^ so far as it is affected by the heat (that is, the
walls above the grates, the arch, and the gas-flue)^ is lined with
fire-bricks; the total thickness in front is one brick, behind (or as
the partition between two rows of burners) two bricks. The side
walls dividing each two burners of a row are 1^ or 2 bricks thick,
but they diminish upwards to one brick. The roof need only be
4i inches thick. The mortar is fire-clay, as usual; in the colder
parts, such as vertical gas-shafts, flues, &c., this docs not stand so
well as boiled-down tar and sand.
F. J. Falding (Min. Ind. vii. p. 666) constructs the first layer
behind the cast-iron front plates of hollow bricks, so that air-
channels are formed from the grates upwards to the top of the
burner. This keeps the burner-room cool and easier to work in,
at the same time retaining heat for concentration in acid pans on
x2
308 PRODUCTION OF SULPHUR DIOXIDE.
the top of the burners, or for increasing the efficiency of the
Glover tower, or for supplying the burners themselves with hot
air, which is an advantage in some cases.
Of course the burners are well bound together, either by special
uprights and tension-bars, or by flanges cast to the front plates,
provided with holes for the cross-bars (fig. 73) .
Opinions as to what size the burners are to be made vary a
good deal. Mostly smaller burners are met with, about 4 feet
6 inches to 5 feet from the outside to the inside of the back wall.
The reason given for this is that longer grates cannot be served so
well, and that in a larger burner the newly-charged ore forms too
shallow a layer (the depth of the whole layer of pyrites, including
the partially burnt ore, is not in question here). I have, how-
ever, worked for a good many years with larger burners, nearly
6 feet from the front to the back end of the grate, and have burnt
my ore better than the majority of other works using the smaller
burners. Certainly, the usual 7-cwt. charge had to be all put in
at once, whilst in the smaller burners it is introduced in two halves
every 12 hours; and many practical men assert that a 12-hours'
is preferable to a 24-hours' charging ; but this is not borne out
by experience. It is, however, a decided mistake to try burning
a much larger charge on the larger grate, say 8 or 9 cwt. This
can only be done with poor ores, such as are not in use at the present
day, except locally ; richer ores, especially those containing copper,
are sure to be fluxed by the heat getting too high, and cause the
greatest trouble. As a result of long experience, I am inclined to
consider a grate-surface of 4 feet 6 inches by 5 feet 8 inches, and
a depth of pyrites of 2 feet 3 inches, very favourable for burning
7 cwt. of 48-per-cent. Spanish ore, changing once every 24 hours.
The rate of burning just mentioned equals 30 lbs. of 48-per-cent.
pyrites per superficial foot of grate in 24 hours. With poorer ore
(40 to 42 per cent.) I have certainly burnt in the same grate
8 cwt. (=35 lbs. per square foot), and with 38- or 40-per-cent. ore
even 9 cwt. (= nearly 40 lbs. per square foot). In England the
maximum quantity of pyrites burnt per square foot of grate will
very rarely exceed 40 lbs. of 48-per.ccnt. pyrites ; reliable figures
from one of the largest works arc 35^ and 39 lbs. But in Grerman
works, according to Hascnclevcr, using Westphalian pyrites, the
proportions are ll'G, 445, 57'8, CO'3, and G5'0 lbs. (Wagner's
Jahrcsb. 1871, p. 212). Bode (ib. 1874, p. 245) quotes for
\
SIZE OF FYaiTlSS-BURNEHS. 309
Westphalian pyrites of 41 or 42 per cent., 50*7 fo 640 lbs.; for
Norwegian ore of the same strength, 38*3 lbs. ; for Valais ore,
with 35 per cent, sulphur, up to 92 lbs. per square foot in 24
hours. Favre (Monit. Scient. 1876, p. 271) states as the most
favourable ratio in his experience 55*3 lbs. of 40-per.cent. pyrites
per square foot in 24 hours.
At the Oker works, in 1901, the roasting area of the three deep
kilns (p. 295). serving for a chamber set of 5289 cub. metres
amounted to 33*55 superficial metres, that is 000634 sup. met.
per cub. met. of chambers. The quantity of ore roasted per
square metre in 24 hours was 0*45 ton copper-ore, 0*50 ton pyritic
lead -ore, 1*40 ton copper-matte, 1'60 ton lead-matte (comp.
p. 83 on the composition of these ores).
In 1902 the Rhenania works at Stolberg considered 200 kils.
Westphalian pyrites per superf. metre grate-surface as the normal
quantity ; with 230 kils. the degree of desulphurization was
somewhat less. The former is = 40 lbs., the latter = 46 lbs. per
superf. foot.
In England some thirty or forty years ago many pyrites-burners
were made about 33 inches wide and 26 inches from front to back
at the fire-bars, but 42 inches square at the level of the bottom of
the charging-door, giving a slope of 6 inches back and 9 inches
sideways. Later on burners were made larger, say 39 x 42 inches
at the bars and 48 X 51 inches at the charging-doors, many going
up to 54 inches wide and 60 or 66 inches from front to back at
the fire-bars. The smallest sizes burn about 4 cwt. of Spanish
ore = 44 lbs. per square foot in 24 hours ; the largest sizes 7 cwt.
= 31 lbs. and upwards per 24 hours. The most usual size is
about 16 to 18 square feet at the fire-bars and 22 to 24 square
feet at the base of charging-doors, burning 7 cwt. daily =37 to 41
lbs. of ore per square foot. This is about the maximum for rich
ores consistent with proper freedom from fluxing, but poor ore
may be burned in larger quantity (Thorpe's Diet, of Applied
Chemistry, iii. p. 719) . With the very well-burning Agnas Tenidas
ore Davis (Chem. Engineering, ii. p. 124) states the proper charge
= 2*3 lbs. per square foot per hour = 55 lbs. per day, in which
case the cinders tested 1 per cent. S.
It is hardly necessary to say that the pyrites-burners are always
built in sets. Usually 12 to 24 burners are served by the same
set of men ; and they must be worked so that every one gets its
810 PRODUCTION OF SULPHUR DIOXIDE.
regular turn, as is evident from the necessity of a regular evolution
of gas. Frequently the burners are built underneath the acid-
chambers. Not only must they in any case be protected against
rain (if not underneath the chambers), but they must not stand
in a space open at the sides^ since strong winds would put their
draught wrong, and cause them either to go too fast or to blow out
at the doors. It is best to protect them by light walls or by a brat-
lire with shutters adjustable according to the direction of the wind.
As the drawings of the English burners (pp. 298 et seq.) show, each
burner is independent of the other, and they do not communicate
one with another, but only with the common gas-flue. Each
burner, then, ought to have its own damper, which is not always
the case. On the Continent, frequently the single burners are
separated merely by low walls ; the ore in this case lies only about
18 or 20 inches deep on the grates, and the whole set is like one
large burner with a divided grate (p. 306). It is, of course, not
possible, as on the English system, to treat each burner individu-
^^J> to give it more or less draught, to isolate it for repairs, &c.
Nevertheless this system is in favour with some of the more
experienced Continental manufacturers, who say that 18 inches
depth is quite sufficient for the rich ores, now universally employed,
and that the connection of the gas-space of all burners into one
whole is preferable to the English system, because one burner can
aid another and the whole is visible at a glance. Evidently this
system, of which good illustrations are given in figs. 84 to 87, p. 306,
answers its purpose as well as the English ; and in a special case,
in which I saw a set of English burners working alongside a set
of burners of the kind just described, the manager informed mc
that he preferred the latter, because it was easier to regulate the
draught than with the English burners.
In Continental works possessing no Glover tower it is usual
to concentrate the chamber-acid up to 144° Tw. in lead pans^ which
are mounted on the top of the burners, and are heated by their
waste heat. Of all plans for concentrating sulphuric acid this is,
as we shall see later on, the cheapest, only excepting the Glover
tower. There is no reason why such pans should not be placed
on the English burners as well ; but even before the Glover tower
did away with most lead pans, the above arrangement does not
seem to have been practiced in England, where, however, the space
on the fnrnace-ai'ch is otherwise usefully employed for drying
FOTTING SPECIAL PYRITES-BURNERS. 311
*^ balls '^ from pyrites-dust, &c. There are also, as we shall see
m Chapter IX., sometimes reasons against placing the pans on the
top of the burners.
The ^'poiting " of the mixture of nitric and sulphuric acid (liquid
nitric acid is not used in England for this purpose) is now nearly
always done in such a way that the burners are not disturbed by it.
Twenty years ago the- pots were frequently put on pillars between
two burners, with a common gas-space ; these were provided with
special potting-doors in the burner-front, and cast-iron dishes as
saucei*s for receiving the stuff that boiled over; these saucers were
inclined towards the doors, so that the nitre-cake could not so
easily run into the burners ; but it got out of the doors, which
made them look very dirty ; and ultimately it also got into the
burners themselves. None of the better factories have this
arrangement now, but all pots belonging to a set of burners are
placed in a separate ^' nitre-oven/' which is nothing but an enlarge-
ment of the gas-flue, and either situated over the burners or on
pillars outside the same. The latter is preferable ; for also here
there is always a metal saucer provided for catching the boiling-
over nitre-cake; this may be cracked before it is noticed, and
much nitre-cake may get into the burners, doing great damage.
If the arrangement is similar to fig. 74, p. 298 (where, by the way,
the nit re- pots are replaced by a better contrivance to be sub-
sequently described), no risk of the above-mentioned kind is run.
Favre (' Moniteur Scientif.,^ 1876) reports that the works in the
south of Prance use pots of 2 ft. 7^ in. x 1 ft. 6 in. x 12 in.,
standing on a bridge between two burners ; and he also mentions
the drawback of boiling over into the burners. This would show
that those works, even in 1876, were in a backward state.
Special kinds of Pyrites^umers.
Addie (E. P. 180, of 1886) describes a peculiar pyrites-burner.
It consists of a cupola, brought to a white heat, in which the
pyrites is charged together with sandstone or other slag-producing
material, and is burnt by a hot blast, the cinders being reduced to
a molten slag which is run off from time to time. Unless this
apparatus was intended for some very special purpose, not
apparent at first sight, it must be pronounced as very impracticable
so far as the chemical manufacturer is concerned*
The methods described in the patent of Hargreaves, Robinson,
812 PRODUCTION OF SULPHUR DIOXIDE.
and Hargreaves (E. P. no. 5681, 1886) for treating pyrites are
evidently less intended for the manufacture of sulphuric acid than
for the recovery of arsenic, antimony, copper, silver, and gold.
A whole set of burners is combined in such a way that the
air or the gaseous products at first formed can be successively
passed through them in regular rotation. The air, previously
heated in recuperators, passes downwards, first through nearly
exhausted ore, afterwards successively through ore containing
more sulphur. The residual ore is treated with acid gases, in
order to bring the metals into a soluble state. For details we
must refer to the patent specification.
Bvrners for roasting copper-matte. — Haege (Berg- u. Hiittenm.
Zeit. 1893, p. 883) describes the process introduced by him at
Britonferry, near Swansea. The copper-matte produced there
could not be roasted in Mansfeld kilns, nor in ordinary pyrites-
burners. The desired result was obtained by increasing the heat,
ill the first instance by a suitable treatment of the matte, and in
addition to this by improving the construction of the burners.
The matte was rendered porous by tapping it on to a sand-bed
slightly moistened and dusted over with fine coal. It was then
crushed by a Blake's stone-breaker, in which one of the corrugated
faces had been substituted for a smooth one, so that flat, more
tightly lying pieces were obtained, which were separated from the
smalls by a riddle with meshes of | inch width. The burners are
of the ordinary shape of English pyrites-burners, described above,
but of slightly different dimensions : — Grate-surface 4 ft. 8 in. by
4 ft. 4 in. ; area at the level of the upper working-surface 5 ft.
by 4 ft. 9 in. ; height from grate-bars to the upper working-
surface 2 ft., to spring of the arch 8 ft. 4 in., to the crown of the
arch 8 ft. 8 in. ; smoke-flue at the lowest point 6 in., at the
highest 1 ft. 4 in. ; total outside height 7 ft. 10 in. The heating-
up takes place from the top, exactly as described in the text ; the
burners are ready for work in two or three days. Each burner
then receives a charge of from 6 to 7^ cwt. of crude matte every
12 hours. After three hours everything becomes red-hot, after
six hours a bright cherry-red heat is attained. Now the interior
of the burner is worked up through the middle door by means
of a steel poker, 2^ in. thick, pointed at one end ; any lumps
formed are broken up; and this working over is continued
through the upper door. After another two or four hours the
WORKING PYRITES LUMP-BURNERS. 318
heat is at its maximum ; the upper working-door is now mostly at
a dark-red heat. Then the heat decreases. 11^ hours after
charging the grate-bars are turned and shaken, in order to remove
the roasted matte, and after 12 hours a new charge is made.
There is no picking out and recharging of imperfectly roasted
matte, since everything is well finished. The draught must be
well regulated ; there should be a slight plus- pressure within the
burner. In this way mattes containing from 20 to 47 per cent,
copper are treated. The poorer matte yields rather hotter and
better gas and more sulphuric acid than the richer. With 20.per-
cent. matte the roasted product contains 9 per cent, sulphur, with
rich matte it contains 11 per cent, sulphur ; both are at once
ready for the concentrating work. From 40 per cent, matte about
47 or 48 per cent, of the weight of roasted matte is obtained in
the shape of chamber-acid of 110° Tw., with consumption of 0*8
to 10 nitre per cent, of chamber-acid. The gases are hot enough
to thoroughly decompose the mixture of nitre and sulphuric acid
in the nitre-oven and to denitrate the acid in the Glover tower ;
the acid flows from this with a temperature of 140° to 155° C.
Working of the Pyrites-burners for Lumps,
In order to start a burner it is first, if newly built^ dried by
a slow fire in the usual way, and then filled with burnt ore to
within 3 inches below the working-door. If no burnt ore can
be procured, ordinary road-metal, &c., may be takeu^ broken suffi-
ciently to pass between the grate-bars when they are turned. The
draught-hole of each burner is closed by a damper^ and the work-
ing-door is left open. Then ordinary fuel (wood or rough coals)
is heaped on the ore and lighted. After twelve or twenty-foui
hours the burner and the uppermost layer of the ore will have
reached a dull red heat ; the rougher parts of the fuel still present
are then drawn out and an ordinary charge of green pyrites is put
on. By the heat of the burner- walls, that of the ore below, and
the fuel still present, the fresh ore will soon be lighted ; when it is
fully burning, the working-door is closed, the damper closing the
access to the gas-flue is opened, and the gas allowed to go to the
acid-chambers. Care must be taken in lighting up not to go too
far, which would damage the burners.
Thus the process is started ; and it is now continued regularly
314 PRODUCTION OF SULPHUR DIOXIDE.
and uninterruptedly till it has to be stopped for external reasons.
Repairs are very rarely necessary in pyrites-kilns ; but those of
other parts of the acid-making apparatus may compel their
stoppage. At some English works the dampers are put in every
Saturday at midnight^ and are opened only on Sunday at midnight ;
in the meantime all other openings are well closed; and the
burner thus keeps its heat so well that the new charges at once
take fire when brought in. If any temporary interruption of work
does not last beyond four or six days, usually the burners can be
kept hot enough in this way to be started without any fresh
lighting-up by means of fuel.
The regular burning-process has a double object, from which
follow all the precautions to be observed. In the first place, the
sulphur contained in the ore is to be burnt as far as possible ; and,
secondly, a just sufficient quantity of air is to be employed, no
more and no less than is required for the chamber-process. This
means, besides the air necessary for burning the sulphur to
sulphur dioxide, as much more air as is required for oxidizing the
latter to sulphuric acid, and, moreover, a certain excess of air
found necessary in practical work. Anyhow, therefore, the air
will be more than just sufficient for burning all the sulphur con-
tained in the pyrites; and the second condition seems thus to
imply the first. But this can be said only for brimstone and for
pure pyrites not containing any zinc-blende or galena, &c. ; for
only the former can be desulphurized completely by their own heat
of combustion. The sulphates of iron, which are always partly
formed as intermediate products, arc decomposed again at a com-
paratively low temperature into Fe203, O, and SOj, or into Fe203
and SO3, for which the heat of the burners is quite sufficient. This
is a little more difficult with the sulphides of copper; but the
temperature of decomposition of CuSOj is also within a red heat.
Moreover the copper-extraction works do not want all the sulphur
to be burnt, but allow a residue of from four to at most six per
cent, sulphur in the cinders. If, however, the ores contain blende
or galena, which on burning are transformed into zinc and lead
sulphates, the burner cannot possibly effect a total desulphurization;
for these sulphates are only decomposed at a strong white heat,
which is not allowable in a pyrites-burner, and they must thus
remain as such in the cinders. Furthermore, if the pyrites contains
WORKING PYRITES LUMP-BURNER?. 315
calcium sulphate or carbonate, a corresponding quantity of CaS04
will remain in the residue. Any barium sulphate present would
not be taken notice of in the testing of the pyrites or the cinders,
being classed among the '^ insoluble."
In the case of the usual description of pyrites, not containing
any considerable quantity of zinc or lead, the burning of ore in
lumps will reduce the sulphur in the cinders with good work to
3 or 4 per cent. Less than 3 per cent, of sulphur rarely occurs
on an average of the whole year ; but with very good ores it may
go down to 2^ per cent. At the Rhenania works, at Stolberg,
even pyrites with 6 or 7 per cent. Zn is burnt down to 2 or 3 per
cent. S, exceptionally 4 per cent, (information received in 1902).
Most frequently the limit stated above for good work is exceeded ;
some works leave 6 or 8 per cent, of sulphur, and even more, in
their cinders, whilst their neighbours only leave 4 or 5 per cent,
in the same ore. The fault of this may be due either to the
description of burner employed or to tlic style of work. If, by
the construction of the burner, the pyrites lie in too shallow a
layer, and this is let down too soon on shaking the bars, it will
easily come out badly burnt, liut even if the burner is correctly
built, much still depends upon the skill and care of the burner-
men.
Excellent results are obtained with some of the very rich de-
scriptions of Spanish non-cupreous pyrites recently come into the
market (p. 56). These pyrites can he burnt down to 0*5 percent,
of sulphur in the cinders, so that the latter can be sent straight to
the blast-furnaces. But if this degree of purity is to be obtained,
the place in front of the kilns^ where the burnt ore is drawn
out, must be kept perfectly clean, so that no green ore can get
mixed among the cinders; and any portions of these which
have been spilt from the discharging bogies on to the ground
must not be shovelled back into them, as they will have some
admixture of dust from green ore, but they must be put back into
the kilns.
How much depends upon employing the ore in neilher too large
nor too small pieces, and upon keeping the pieces of as uniform a
size as possible, has been mentioned on p. 287. Only then will it
be possible to regulate both the depth of layer and the draught in
a satisfactory way.
316 PRODUCTION OF SULPHUR DIOXIDE.
Whether pyrites is properly burnt or not can be recognized to a
great extent by the eye. By the burning-process the pieces swell
out and burst in some place ; they become light and porous, and
assume the red colour of ferric oxide, in the case of cupreous pyrites
a more blackish-red colour. The burnt ore ought therefore to
consist of light porous pieces of the proper colour, apart from the
powder always present in large quantity, which is generally suffi-
ciently burnt off. Already, on taking up the larger pieces, their
weight will allow a rough judgment of the state of the burning ;
and this can l)e more distinctly recognized by breaking the pieces
and observing whether they contain a raw core in the centre. Also
the presence of many slags (scars) on the cinder-heap is a proof of
bad burning.
Important as these empirical signs are, no well-managed
factory will be satisfied with them, but will from time to time,
daily or at least twice a week, have the cinders tested, after having
drawn a large sample and reduced it properly. At all events
the above-mentioned empirical signs have hardly any value for
small ore.
The chemical testing of pyrites cinders (burnt are) can take place
by exactly the same methods as described in the second Chapter
for the analysis of pyrites itself. It is there shown that in the
case of burnt ore more expeditious methods may be used, and that
among these that of Watson-Lunge (igniting with sodium bicar-
bonate and titrating the undestroyed sodium carbonate) seems to
be the most accurate (p. 76).
The sulphur contained in the burnt ore is no longer in the
shape of FeS2, apart from any quite raw cores in large pieces.
But even fine or quite porous cinders, burnt as well as possible,
also those from pure pyrites free from lead, zinc, and lime, always
contain sulphur ; and as this cannot be in the shape of FeS^, the
question can only be whether they contain FeS or sulphates of
iron (most probably basic ferric sulphates), or both. According to
Scheurer-Kestner and Rosenstiehl (Bull. Soc. Chim. 1868, ix. p. 43),
the cinders contain essentially FeS ; they give two analyses —
(i.) of properly burnt ore^ (ii.) of an operation carried on too hot,
so that the ore bad fluxed. Both are from Sain-Bel pyrites,
containing 46*1 per cent, sulphur in the pieces and 49*28 in the
smalls.
PYRITES CINDERS. 317
I. 11.
Moisture 1*0
FeS ',. 8-5 27-2
Fe 5-4'> 17-31
S 3-lJ 9-9J
Oxide of iron 720 624*
Fe 50-41 4101
O . 21-6J 21-4/
tluarlz 18-5 10*4
1000 1000
From these analyses^ it would appear that there would be no
ferric sulphate whatever in the residues^ which is very improbable^
as such can be proved by washing with water (comp. Bode^
Dingler's Journal^ ccxviii. p. 327, and further analyses by Phillips^
Gibb^ Brauniug, Wedding, and Ulrich in the Chapter treating of
the recovery of copper from the cinders).
According to Troost, the first reaction is 8FeS2 = 2S + Fe3S4.
Uegnault holds that the sulphide formed has the formula Fe^S^.
Lemoine (Fischer's Jahresb. 1899, p. 355), from observations made
in a Maletra shelf-burner (see infra), believes that in the upper
layers there is always a distillation of sulphur, which afterwaixls
bums with a blue flame. Lower down this is no longer the case,
either because here most of the pyrites is already reduced to FeS,
or because the temperature is too low and. the supply of oxygen
too great for the formation of free S. Ferric oxide appears only
after roasting for 2^ hours. The action on pyrites seems to
proceed from the outside, where there is combustion into FcsOj^
and the heat acting on the inner, part produces a decomposition
into FeS and S. The S distils and burns outside, the FeS is
oxidized by the FejOg, which is again reformed by the outer air.
Richters (Dingl. Journ. cxcix. p. 292) gives the following
analysis of burnt ore from the Silesia works at Saarau : —
Water 4*35
Iron 43-36
Manganese 0*16
Silica 13-92
Carried forward ,... 61*79
* The calculation does not agree liere ; 62*4 Fe.p, would contain 43-68 Fe.
318 PRODUCTION OK SULPHUR DIOXIDE.
Brought forward 61 '79
Alumina 4*84?
Lime 0-02
Zinc oxide 8-83
Sulphur trioxide 4*35
Sulphur 1-53
Oxygen and loss 18*64
Nickel and arsenic traces
10000
Phipson has published the following analysis of residue from
Irish pyrites ('Chemical News/ vol, xviii. p. 29) : —
Zinc oxide 5'50
Cupric oxide 2'86
Manganese oxide 1*60
Nickel and cobalt oxide 0*12
Cadmium oxide 0*01
Lead oxide 1-67
Antimony oxide 0*04
Ferrous oxide 1'17
Alumina 3*25
Sulphur 2-60
Thallium traces
Indium traces
Gaugue 15-00
Ferric oxide 65*99
Lime ! 0*11
Magnesia 0'08
100 00
All that has been said {cf. p. 315) respecting the maximum of
sulphur in the cinders to be aimed at, only refers to the burning of
pyrites proper — that is to say, of ores containing essentially FeSj,
and got exclusively as a raw material for vitriol-making, in which
case the cinders are us good as worthless. Just in the same line are
those cupreous pyrites (with less than 4 per cent, of copper) whose
copper can only be extracted by the wet process ; for these the
above-mentioned rules for the sulphur in the cinders are equally
valid. But the case is quite different with a number of ores where
the residue from the burning is regarded as by far the most important
SUPPLY OP AlU TO PYRITEV-BUKNEKS. 319
product, and where the gas is only a by-product, often only con-
verted into sulphuric acid in order to get rid of it. To this
category belong blende, copper-pyrites, coarse metal, &c. Here
the burning-down to the above-mentioned minimum of sulphur
is partly not possible, partly not even desirable (as for copper-
pyrites) ; and there exist for each case definite rules, which,
however, do not belong to the domain of acid-making, but to
that of metallurgy. Even where a larger percentage of sulphur
is required for further metallurgical operations, it is more rational,
so far as concerns the acid-maker, in order to save labour,
burner-space, &c., to bum the material as well as possible, and to
supply the necessary sulphur afterwards by adding a little green
ore ; thus, for instance, the copper-extraction works proceed when
they receive the cinders too far desulphurized. The case of zinc-
blende is a special one and will be dealt with later on.
We now pass on to the second fundamental condition of proper
work in the pyrites-burners, viz., that neither too little nor too
much air be employed. At this stage we leave out of consideration
the absolute quantity of air required, and only treat of the prac-
tical rules and of the appearances observed in the burners them-
selves. If too little air is admitted, whether from too few holes
in the bottom door being opened, or from the damper in the
draught-hole not being enough drawn, or because the pipes arc
stopped up with dust, or the draught in the whole chamber system
is insufficient from one cause or another, the same thing will
happen as with sulphur-burners when they get too hot : sulphur
will sublime as such, and will be deposited in the flues, the dust-
chambers, the Glover tower, or the chambers themselves. It is,
however, a more frequent and serious consequence that, in the
case of insufficient draught, the often-mentioned slags or scars are
formed. These, as Scheurer-Kestner and Rosenstiehl have shown
(/. c), consist mostly of iron monosulphide, PeS, which is necessarily
formed when FeSg is heated, with exclusion or insufficient supply
of air, along with free sulphur. It is easily fusible, and fluxes in
the burners to more or less large cakes or scars, upon which the
air has practically no action. The FeS fluxes all the more easily,
as in the case of insufficient supply ^f air, where no cooling by the
excess of air takes place, and more heat is developed in certain places
than when the supply of air is abundant. The scars mostly enclose
some green pyrites, and in this way cause a further loss of sulphur.
A much greater loss is occasioned by their stopping the passage of
820 PRODUCTION OF SULPHUR DIOXIDE.
air^ SO that the ore above and below a scar is very incompletely burnt.
The heat is locally increased and driven further down than it ought
to be ; the zone of combustion is removed further downwards ;
and on letting down the ore the pyrites partly comes out incom-
pletely burnt. If scars have formed in the burner^ they naturally
descend as the cinders are let down^ and they would ultimately lie
immediately on the grates and entirely stop them up. This^ how-
ever^ must be prevented by all means. A careful workman always
breaks up the surface of the old ore before putting in a fresh
charge; and thus he finds out whether any scars have formed,
which mostly takes place near the surface : they can then be easily
brought to the surface by means of hooks and pulled out at the
door. But if they have been overlooked at first and have got lower
down, in doing which they constantly increase in size^ their removal
is more difficult. Then a very large and heavy poker of the best
tough iron (these are made up to 12 feet long and 2 inches
thick), bent in the way shown in fig. 88, is introduced into the
Fig. 88.
burner through the charging-hole, and the men work it till they
have got the point a underneath the scar. Several men, working
at the end i, then try to lift up the scar, in spite of the resistance
of the superjacent mass of pyrites. This labour is very disagreeable,
exhausting, and difficult. The middle doors, between the charging-
door and the grate, found in all pyrites-burners, are only used in
extreme cases.
With the low burners mentioned on pp. 305 & 310, where the ore
lies only fit a depth of 20 inches, scarring is next to unknown, at
least if the ore is very carefully sized, so that all passing through a
^-inch riddle is kept out. This agrees with the facts which
will now be explained.
Apart from other causes, the supply of air in a burner may be
insufficient because the ore lies too deep. As the depth of the ore
depends upon the vertical distance between the grate and the
working-door, it follows that for ores behaving very differently in
WORKIXO PYRITES-BURNERS FOR LUMPS. 321
this respect differently built burners must be used. Thus the deep
burners built for Irish pyrites had at once to be given up when
Spanish pyrites began to be used. With the same height of pyrites
which was just right for the poor ore, in order to keep the heat
better together, the rich cupreous ore, in itself more fusible, be-
came far too hot, and, moreover, the air could not pass through
quickly enough to make a complete burning of the ore possible at
every point ; from both causes combined followed this effect (easily
comprehensible after what has just been said), that the scarring
became excessive. It is always much more feasible to burn poor
ore in a shallow than rich ore in a deep burner.
Insufficiency of draught, if very considerable, will be easily
lecoguized by the gas blowing out of all the joints of the burners,
and especially coming out in force whenever the working-doors or
the bottom door are opened. On the other hand, the draught
should not be so strong that too much air will get into the chambers ;
the gas ought to be kept as rich as possible, as will be shown here-
after. It may be assumed that the draught is just right, if, on
opening the small slide in the working-door, neither gas nor flame
issues from it, nor, on the other hand, the flames inside the burner
perceptibly tend towards the draught-hole. They ought to rise up
perpendicularly and quite steadily ; and on opening the door they
may even tend slightly towards it. As, however, the exact regu-
lation of the draught can only be effected by regulating the holes
in the bottom door, and as on each opening of the doors above or
below the grate much more air must get in than is necessary, of
course the times during which the doors are opened are restricted
as much, and the charging, shaking of the grates, and discharging
are managed as quickly, as possible. It is very advisable to close
the holes in the bottom door completely while the top door is
open, or to proceed in the way described on p. 303 et seq. If the
draught is not very copious, whenever the door is opened, there
will be no room for so much air rushing in, in consequence of
which a portion of the gas will rush out and get into the burner-
house ; this is both a loss and a nuisance to the workmen, and,
in greater quantity, also to the neighbours.
For regulating the snpjdy of air several plans are possible. The
regulation takes place either before the grate, by the holes in the
bottom door, or behind the grate, by means of the damper in the
draught-hole or that in the large chimney behind the chambers,
VOL. I. Y
322 PRODUCTION OF SULPHUR DIOXIDE.
or else by fan-blasts (comp. Chap. V.). Regulation behind the
burners is only available where all the burners have a common
gas-space ; otherwise the draught through the chimney or fan must
be equal to the maximum amount required for all the burners^ and
must be changed according to the atmospheric conditions^ each
burner being regulated separately. This could be done best and
most safely by the dampers in the draught-holes connecting each
burner with the gas-flue j but these are rarely used for this pur-
pose ; they would have to be made very tight-fitting, and then
would easily be set fast by flue-dust. Therefore here also the
draught is made suflScient for all eventualities. The real regula-
tion of the air, at least generally in England, takes place by means
of the holes in the ash-pit door, of which a sufficient number are
closed by plugs or otherwise. Scheurer-Kestner went so far as to
pass all the air through a Combes^ anemometer ; but this can only
have been done for isolated experiments, since such a delicate
instrument can hardly have been kept fit for use for any length of
time in an atmosphere thus exposed to acid vapours and to dust.
It is therefere left to the burner-men to open or close the holes in
the door as required. At the Rhenania works (1902) they regu-
late the supply of air, apart from the exit damper, merely by a
slide damper, shutting off the flue below the grates in which the
cinder-bogies are placed.
The supply of air is usually regulated by the following practical
rules : — At the beginning (that is, immediately after making a
fresh charge) the burner does not require very much air, till the
ore has caught fire, which will take half an hour or an hour. Then
more air must be admitted, always with the above-mentioned
restriction — that the flames rise perpendicularly, and tend slightly
towards the slide when this is opened. When, however, the prin-
cipal portion of the sulphur is burnt and the flames become scarce,
the air is altogether shut ofiF, and further action is left to the heat
of the burner. About two hours before it is time for recharging,
the working-door is opened and the ore is well raked and turned
over by means of a hook to a depth of 3 or 4 inches, and any small
scars are removed. If herewith blue flames appear to any extent,
this proves the burning not to have been sufficient, and a little air
must be admitted. When the whole time is up, be it a twelve-
hours' or a twenty-four hours' turn, the air is entirely shut off^ at
the bottom, the small doors covering the grate-bars are opened, and
the latter are turned two or three times, leaving each alternate one
WORKING FYRITB9-BURNEES FOR LUMPS. 323
out. During this the workman must look through the working-
door, to see whether the layer of ore is let down evenly all over ;
he can easily manage, determined by the eye, not to let the ore
down too much or too little. Then, as quickly as posible, the
new charge of ore in pieces (usually with a little dust), which
must have been lying ready in front of the burner, is put in, and
the process begins again. It is evident that there must be a regular
rotation, so that a fresh burner comes in turn every hour or so ;
this is both indispensable for a regular evolution of gas, and con-
venient for distributing the labour over the day.
The burner-men ought to shake the grates quite equally for the
purpose of discharging, so that the ore does not come down more
quickly in one place than in another, and to take care that only
cold, thoroughly exhausted cinders, but no red-hot ore, comes
down. They ought then to work up the ore on the top through
the doorway with their pokers, and rake the surface so as to
make it even again. Then they must charge the new ore equally
all over, starting about two inches towards the door from the back
and the sides. Two men can attend to a set of 18 or 20 burners,
consuming from 6 to 8 cwt. of pyrites each every 24 hours,
including the wheeling away of the cinders and the potting.
The interior of a burner, after the throwing-in of a fresh charge,
is, of course, at first quite black. Gradually small blue flames
appear, which become larger and more lively and cover the whole
mass. After a few hours they become scarcer again ; but the mass
in the meantime has become red-hot. Later on it cools again ;
and towards the end of the period there is no glowing visible at
the surface ; but as soon as the mass is stirred up the glowing
appears again.
The men like to employ a practical test, to convince themselves
that the burner is not too hot for recharging, in the shape of
strokes made with brimstone on the burner-door : so long as
these take fire at once, the burner is still too hot ; only when they
remain is it cold enough for charging. Frequently it is necessary
to wait a short time, even for a few hours, after shaking the grate-
bars and letting down the burnt ore, in order that the burner may
cool a little before recharging it; this has the advantage that
the top layer, by turning it over, is caused to burn its sulphur
more thoroughly than it can be burnt after cold pyrites has been
thrown in.
y2
324 PRODUCTION OF SULPHUR DIOXIDE.
Generally it takes some time before the men get used to a new
kind of burner or of pyrites. If even skilled men are taken from
other places, they require special supervision^ and still more if a
new kind of pyrites has to be tried. If at all possible^ different
kinds ought not to be tried mixed up^ but one kind adhered to for
l^ome time, because only in this way do the men get used to a
thoroughly proper treatment of the burner. Each kind of pyrites
requires a little different treatment as to supply of air, breaking
up, &c.
An extremely great help in regulating the burning-process is
the analysis of the gas, which, however, is nearly always made for
a whole set of burners together in their common flue. We shall
enter into the details of this later on.
If a pyrites-burner is working properly, it will, if touched out-
side, be so hot in its upper part (say 6 inches below the working-
door) that the hand cannot be borne upon it ; further down it
must be cooler; and immediately above the grates it ought to be
cold, or at most hand-warm. This is one n>f the most important
practical signs of the proper working of the bulkier. If a burner is
too hot below, this may be due to insufficient draught, or (which
in the end comes to the same thing) there has either been too
much pyrites charged or there is too much dust in the burner,
which has stopped up the interstices. Too much dust may come
from bad riddling, from too much having been added on purpose,
from the falling of the " balls '' inside the burners, or from the
decrepitation of " explosive'' ores (comp. pp. 55 & 77),
In any case, the first thing to be done, apart from removing
the cause of the evil, is to agaiu cool the excessively hot burner.
Above all, more air must be admitted; and, in order to drive
up the heat more certainly, the new charge must be kept back
a little, and no fresh ore put in the middle, but only along
the sides and the back of the burner. It always takes one
or two days, sometimes longer, before a burner has recovered
its normal temperature. In specially obstinate cases there is
nothing for it but making very small charges for a day or two,
till matters have come right again. Some manage by taking out
the ignited top layer, allowing it to cool a little, and putting it
back into the burner, which in the meantime has received more
draught, owing to the lower depth of ore, and thus has become
cooled.
WORKING TH£ PYRITES-BURNERS. 325
If a kiln is allowed to go too hot for any length of time,
whatever may be the cause (want of air, too large charges,
stopping up by dust, bad breaking up), the consequence will
always be the same, viz., increased scarring, with all its un-
pleasant accompaniments. I have had to deal with cases where
the scars became so abundant that the burner had to be put
out, the grates had to be drawn, the whole of the stone taken out,
and the burner freshly filled up.
Of course it also sometimes happens that a burner goes too cold
and the fresh charges take fire too slowly. This may be caused
equally by a want of draught or by too small charges, and can be
easily remedied in either case. If it has, however, got so far that
the new pyrites will not take fire at all^ nothing remains but to
put in very hot ore from some of the other working-burners ; in
this way the matter may always be put right with some patience,
unless large scars are lying on the grates, or there are other
serious faults, which must be removed before the burner can be
expected to work properly. A frequently used but objectionable
remedy against cold burners is to put live coals on the pyrites.
Carbonic acid is a great enemy to the chamber process, probably
not so much by diluting the gas (for its injurious action is far too
great to be attributed to that alone), as by lying in the lower part
of the chambers and preventing contact between the chamber-gas
and the bottom acid, till it is removed by diffusion. This point,
however, is not yet cleared up.
A pyrites-burner may also go too cold if there is too much
draught — if, therefore, besides the air required for its intense work-
ing there is an excess, which only acts as inert cooling gas. This
is a very great fault ; for in this way the consumption of nitre is
increased and the yield of sulphuric acid very much diminished.
Long before the burners cool from this cause, an excess of air may
become injurious in this way; and by observing the flames in the
burners (much better, however, by the analysis of the gas), it must
be ascertained whether the proper proportion of air is present
or not.
Balar^l reports (^ Rapports du Jury International,' 1867, vol. vii.
p. 29) that in the first trials of Perret and Olivier for employing
pyrites in manufacturing sulphuric acid, thtyat last succeeded in
properly conducting the combustion, but obtained a very small
yield of acid. They ascribed this to an insufficient draught, and
326 PRODUCTION OF SULPHUR DIOXIDE.
applied a fan-blast ; but the yield instantly became minimaL
Now the other extreme was tried — the air-channels were quickly
stopped up with boards covered with sheep-skins and fastened by
stays. At once the chamber process became regular, and the key
to employing pyrites in the manufacture of sulphuric acid was
found. Probably the previous endeavours of Cl^ment-Desormes
in this respect were fnistrated by his allowing too much air to enter.
Some have objected to the employment of damp pyrites (Kerl-
Stohmann's ' Chemie/ 3rd ed. vi. p. 197), because in this case,
on burning, more sulphates are formed, which give oflf sulphur
trioxide ; this takes up moisture, and condenses as sulphuric
acid before getting into the chambers, destroying the flues and so
forth. Even with dry ore in damp weather similar phenomena are
said to be observed, and a smaller yield is alleged to be the con-
sequence of the moisture in the air. On my own part, I have never
noticed such an effect of damp weather, nor have I been able to
learn anything about it elsewhere, in spite of numerous inquiries.
It is very desirable that this point should be. specially examined.
The formation of sulphur trioxide, as well as the occurrence of
liquid sulphuric acid, in the connecting-tubes has certainly been
established ; but it has not yet been proved that the moisture of
the air acts in a way to increase that formation, and the contrary
is more than likely for the formation of SO3.
He who has no practical acquaintance with the matter, looking
at these numerous sources of mishaps, may be inclined to think
that the working of a set of pyrites-burners is a most difficult
task. But it is far from that. If once the burners are in order,
they remain very long so if the burner-men know and perform
their duty to any considerable extent, and if proper supervision
is exercised over them ; the pyrites-buraers then give even less
trouble than sulphur-burners. Certainly, when they do get
wrong, it takes energetic and experienced management to put them
right again.
It will now hardly be necessary to explain in detail why there
are only narrow limits for each given burner and style of charging,
within which the quantity of pyrites charged may vary (comp.
pp. 308 & 309). If too much is charged, the burner scars; if
too little is taken, it gets cold. When, therefore, for any reason
the daily quantity of pyrites has to be cut down, it is necessary to
put out a corresponding number of kilns and to fully work the
BVRNINO PYRITES-SMALLS. 327
remainder. Only in the case of brief temporary interruptions
is it possible to charge rather less than usual for a few days;
but I would recommend even in this case rather to allow the bulk
of the burners to go on as usual, and to keep the necessary number
hot without fresh charges by closing all openings. Then these
kilns will be much more easily put in order by the assistance of
the other burners in full work than if they had all cooled down.
3. Burning Pyrites-smalls.
We have seen above (p. 287) that the fine powder which passes,
say, through a ^-inch or at most a ^-inch riddle must be kept out
of the ore going into the ordinary pyrites-kilns. The '^ smalls,"
'^ fines,'' or " dust '' produced in this way, whether it be on
breaking the ore at the works or already at the mine, must be
dealt with separately.
This can be done in very different ways, according to circum-
stances. Where pyrites-smalls are not bought as such, the
question is only about the dust arriving along with the bulk of
the ore and also that made in breaking. Much more dust is
made when breaking by machine than by hand — in the former
case up to 20 per cent, with middling hard ores, and even more
with soft ores. Formerly, before rational and really satisfactory
contrivances for the burning of smalls were known, some large
works, which had already mounted stone-breaking machines, went
back to hand-breaking, in spite of its costing three to six times
as much, merely in order to avoid the excess of dust. This was
especially the case with works using soft ores, such as the Tharsis
ore ; with Norwegian ores the advantage was always on the side
of the mechanical breaking, because these are much harder and
make less dust. If the quantity of dust going through the smaller
riddle does not exceed 1^ cwt. to the ton, it can be got rid of,
according to my own experience, without any special contrivance,
in the following way : — ^The dust is sifted off as usual, and a
certain quantity of it is laid down for each burner alongside the
pieces. K, for instance, the whole charge is 7 cwt., 6^ cwt. of
pieces are used and ^ cwt. of dust ; if more than this is used^ the
burner easily gets out of order. First the coarse ore is charged as
usual ; and then the man throws the dust with his spade along the
sides and the back of the burner, leaving the whole central part
328 PRODUCTION OF SULPHUR DIOXIDE.
free. Anyhow, the ore ought to be levelled with a hook, after
throwing in the charge, in such a way as to make it higher along
the sides and back than in the centre of the burner. The reason is
this : the air entering from below meets with much less resistance
at the comparatively smooth walls than in the centre of the layer
of ore, and it will preferably rise along the former ; the centre
thus will get less air than the parts next to the walls. If, bow-
ever, the latter lie at a higher level, and especially if the passage
of air is obstructed by the dust lying at those places, the draught
will be more nearly equalized, and the burning will take place
evenly all over the area of the burner. Of course it is not well
to proceed too far in this way ; nor can it be expected that the
result is as good as when lumps and smalls are each treated in
the best way suited for them.
The arrangement just described does not answer if more than
l^ cwt. o£ smalls to the ton of pyrites has to be dealt with ;
and special arrangements must then be resorted to. Probably
the oldest method, now almost obsolete, is the following : — The
small ore is, without further grinding, mixed with sufficient clay
to make it plastic, made into a puddle with water, formed into
ballSy and dried on a steam-boiler or pyrites-burner. Rarely less
than 10 per cent, of clay will be required for this, often more, up
to 25 per cent. The balls are then charged along with lumps into
the ordinary burners, but never too many at a time (at most one-
sixth part of the whole charge), because they fall to powder in the
burner after a time, and if used in a greater proportion would stop
the draught. Only locally is such rich clay found that the balls
stand pretty well in the burners and can be well burnt off. The
workmen dislike them very much, because they disturb the working
of the burners, even when the above-mentioned restriction of
their quantity is observed ; if a burner is not quite warm, they
must at once be left off. Some, in order to get rid of them with-
out disturbing the burners, burn them by themselves, mixed with
"coal brasses," — that is, the pyrites picked out of coals, which
always retains some of the latter, and therefore burns more vividly
and gives out more heat than pure pyrites ; but then it sends the
injurious CO3 to the chambers. Usually not much is gained by
making the balls with clay, since they so quickly fall to pieces in
the burner ; and nearly as much can be done by throwing the
dust at once into the burner and saving the cost of making the
BURNING PYRITES-SMALLS IN COAL-FIRED FURNACES. 329
balls. Only by a very strong admixture of clay can the disinte-
gration of the balls be prevented ; but then the loss of sulphur
and the contamination of the burnt ore is all the greater. In
both cases the sulphur left in the burnt ore rises very much^
from 6 to 8 per cent, and more. Where the cinders go to copper-
extraction works^ the use of clay for balls is quite inadmissible.
These clay balls are connected with so many drawbacks^ that
something else was soon looked for. This was indispensable
where nothing but smalls could be obtained^ or where they could
be procured so cheaply that acid-makers wished to dispense
entirely or partially with using lump ore. At the pyrites-mines
there were formerly enormous heaps of smalls^ which were not
saleable at all and would sometimes have been given away for
nothing, just to make room. In other places pyrites only occurs
in a loose, roughly crystalline shape ; and, again, in others it is
obtained by wet preparation altogether in the state of smalls.
Thus there was great encouragement for constructing apparatus
for burning small pyrites in large quantities.
The contrivances for burning pyrites-smalls may be divided
into three classes, namely, those working by external heat, those
utilizing the heat of ordinary burners for pyrites in lumps, and
those arranged for burning the smalls by themselves without any
extraneous apparatus.
(a) Burning Pyrites-smalls in Coal-fired Furnaces.
Apart from the use of " balls," the oldest plan of dealing with
pyrites-smalls is that of spreading them on the bed of a furnace,
heated by flues underneath, the fireplace being arranged at one
end and the pyrites-dust being introduced at the other, and
being gradually moved forward towards the fire end, as room is
made for it by drawing out the burnt ore. In thus being turned
over many times on its way from one end of the furnace to the
other, the sulphur was supposed to be thoroughly burnt.
This is, however, but imperfectly the case, even if the furnaces
are made 100 feet long. Moreover, the cost of fuel in the best
case is very heavy (at least 10 cwt. of coal is consumed for a ton
of pyrites, usually much more), so is the cost of labour ; the con-
tinuous opening of the working-doors causes very much false air
to get into the chambers, even fire-gases sometimes leak through
330 PRODUCTION OF SULPHUR DIOXIDE.
the furnace-bottom^ and therefore the consumption of nitre and
the yield of acid are very bad. We shall^ consequently, not
go into any details respecting these " muflfle-furnaces," but refer
to the first edition of this work, where, on pp. 186 to 190, the
Belgian furnaces^ and those of Spence, of Godin, of Imeary and
Richardson, are described and partly illustrated by diagrams.
Since it has been recognized that no extraneous heat is necessary
for burning pyrites-smalls^ such furnaces must be looked upon
as altogether irrational, and they are practically obsolete now.
This, of course, has nothing to do with the fact that similar fur-
naces are in use for roasting galena and other ores which do require
external heat for the purpose.
(b) Burning Pyrites-dust by the heat of Burners for Lumps.
This was considered a great improvement upon the older
methods^ but it must be* equally pronounced obsolete now, at all
events in the case of ordinary pyrites. We shall therefore treat
these processes very briefly, referring for details to the first
edition of this work.
The first furnace for burning pyrites-smalls by means of the
heat from lump-burners seems to have been that patented in
France by Usiglio and Dony, Jan. 24, 1852, which, however, did
very imperfect work. Much more important is the furnace con-
structed by Olivier and Perret, which was introduced into the
majority of French works and was in use there for many years,
until replaced by the Maletra burner [vide p. 334). Olivier and
Ferret placed above an ordinary him p. burner a number of shelves
made of fire-clay, and charged with a thin layer (not above
f inch) of pyrites-dust (for exact description and diagrams, vide
first edition of this work, pp. 193 to 196). In this way it is
possible to bum about 1 cwt. of dust to each 2 cwt. of lumps,
the sulphur in the cinders being reduced down to 4 or 5 per
cent. The whole furnace was originally about 20 feet high, which
necessitated a second working-stage above the ground. There is,
of course, a good deal of labour connected with this system.
Later on it was made lower, and so arranged that all the doors
were on one side, so that a number of furnaces could be grouped
into a set.
In a simpler form, namely, that of a single cast-iron plate above
BURNINO PYRITES WITHOUT EXTERNAL HEAT. 331
ordinary lump-burners, this system was introduced into some
Tyneside works, first by MacCullocb, but was soon abandoned
again (comp. 1st edition, pp. 191 to 193).
Another way of carrying out the same principle was the funiace
of Hasenclever and Helbig (Ist edition^ pp. 196 to 201). Here,
at the end of a set of lump-burners, a tower-like appiaratus was
arranged with eight inclined shelves of fire-clay, over which the
dust was gradually to slide down and to be burnt on its way.
Thus from 10 to 16 cwt. of smalls were to be burnt for each
48 cwt. of lumps ; but the principle of automatic sliding down did
not answer; the motion of the dust had to be aided by hand-
work^ with much introduction of false air ; and although a large
number of these furnaces were erected, principally in Germany,
they have been almost or entirely abandoned for some time past,
and we abstain from describing them in this edition.
(c) Burning Pyrites-smalls without external heat.
We must, in the first place, mention a plan which does with-
out any special dust-burners, and only represents an improvement
in making " balls/' It is based upon the fact that pyrites, if it
is in the shape of very fine powder mixed with water, coheres
to a solid mass without the aid of any plastic substance. The
fine pyrites-dust, in the presence of water and air, begins to
oxidize very soon, even at the ordinary temperature ; thus basic
ferric sulphate is formed, which firmly cements together the
separate grains of dust. This only takes place to a sufficient
extent if the grains of dust are very fine and the mixture with
M ater very perfect ; and this can never be attained by merely
sifting and moistening the fine ore. The ore must therefore be
ground finely with water in a mill, for which purpose usually the
so-called pug-mills are used, a kind of vertical mortar-mill, some-
times with revolving bottom dish, or, if the dish is stationary,
with a mechanical arrangement for throwing out the mixture as
soon as it has reached the proper consistency. The pyrites-smalls
are thrown into the mill, water is run on, and the mill is run till
a homogeneous mixture similar to fine mortar has been formed,
Avhich by itself has somewhat plastic properties. This mass is
dried in layers of ^ inch thickness on the top of the pyrites-
burners, often in cakes about 1 8 inches square ; and after twenty-
four or thirty-six hours it has hardened sufficiently for use. It is
332 PRODUCTION OF SULPHUR DIOXIDE.
broken up into pieces of the same size as the lump ore, and charged
along with this into ordinary pyrites-burners. In this it is not
necessary to observe a certain proportion ; for the balls made in
this way are so hard that they can be thrown to the ground with-
out being broken, they do not fall to powder in the burners, and
burn out as well as lumps ; their cinders are, of course, of the
same value as those from lump ore, whilst those mixed with clay
make the utilization of the ferric oxide, at last obtained at the
copper-extraction works, very difficult.
The principal drawback of the process is this, that the mills
suffer very much wear and tear from the hard pyrites. In spite
of this, it was formerly the most usual in the large English works.
The labour of grinding, carrying to the top of the burners for
drying, taking down, breaking up, and laying down in front of
the burners amounts to I*. 4rf. per ton. To this must be added
6d, for coals for working the mill, and wear and tear of the same,
altogether about 2*., apart from the wages for the burning itself
(another 2*. per ton).
The same result as is attained by the ^^ pugging ^^ process is
obtained in perhaps a more complete, but decidedly far more costly^
way by the process of H. Wurtz (U.S. Pat. 252,287). He mixes
the fine ore with powdered metallic iron, moistens this with a
solution of sulphate of iron, and allows the whole to agglomerate
by rusting.
The process just described is not applicable in cases where the
great bulk or the whole of the pyrites employed is in the shape of
dust. In such cases formerly the only available contrivance was
the above-mentioned muffle-furnace, with all its great drawbacks.
The first who proved that the heat generated by the combustion
of ferrous bisulphide is sufficient for keeping the process going
without any external aid^ quite as well in the case of pyrites-
smalls as in that of lumps, was Moritz Gerstenhofer, whose furnace
is described at length and shown in several diagrams in the first
edition of this work, pp. 205 to 215. We here give only one
diagram, fig. 89, and a short description. It consists oE a shaft,
17 feet high, 2 feet 3 inches long, and 2 feet 7^ inches \iide inside,
provided with a large number of prismatic fire-clay bars, so dis-
posed that the intervals of each upper tier are covered by the bars
of the next lower tier. The pyrites-dust is fed in by means of
fluted rollers, and drops from tier to tier, forcing down the particles
OBRSTENHOFFH t
Fig. m.
t
334 PRODUCTION OF SULPHUR DIOXIDE.
previously lying on the bars according to the natural slope of the
ore. Before starting the feed of the ore, the furnace is brought
to a bright-red heat by means of a coal-fire. Afterwards the
combustion of the pyrites by means of the air entering from
below is quite sufficient for keeping up the heat.
The two great drawbacks of the Gerstenhofer burner are : the
very large amount of flue-dust produced in it, and the very in-
complete desulphurization of the ore (8 or 10 per cent. S in the
cinders). Principally for these reasons this ingenious furnace has
been abandoned again nearly everywhere, and is now only used for
roasting '' coarse metal '^ in a few copper-works. At the Freiberg
works, where it was used for a variety of mixed ores (25 to 36
per cent. S), it has also been replaced by the Rhenania furnace,
to be described below.
According to Scheurer-Kestner (Bull. Soc. Chem. xlv. p. 228),
Perret later on constructed a furnace, resembling Gerstenhofer's,
but free from the defects of the latter. The pyrites was stated
to be thoroughly utilized and the cost of labour reduced to one-
half as against Maletra's " shelf -burners " (see below) . At the time
of Scheurer-Kestner's report the furnace in question was evidently
still in the experimental stage; and as nothing more has been
heard of it, its success cannot have been so great as anticipated.
The object but imperfectly attained by Gerstenhofer's invention
has been realized by a very simple plan — so simple, indeed, that it
was not thought worth patenting at the time, although it has sub-
ficquently proved to be of immense importance. Mons. Maletra,
owner of the works of Petit Quevilly, near Rouen, after having
for some time burnt his smalls by means of an Olivier-Perret fur-
nace, conceived the idea of separating the upper part of this furnace
from the lower, and working the dust by its own heat of combustion
without any aid from a lump-burner. This idea, which was worked
out about 1867 with the aid of M. Tinel, proved entirely success-
ful; but in spite of this, and also of the *^ shelf -burner*' being the
jsimplest and cheapest of all dust-burners, it became comparatively
slowly known ; but since 1873, when it became better known
through the Vienna Exhibition, it has spread on the Continent with
4Jxtraordinary rapidity, whilst for a long time it attracted little
Mtention in England. The first burner out of France seems to
have been erected at the works of Schnorf Brothers, at Uetikon,
near Ziirich, in 1870; in Germany the first was erected at Kun-
\
halktra's shelf-burner. 335
helm's vorks in Berlin. Even if, as it would seem, some form of
these simple shelf-burners had been previously in use here and
there, their successful application for burning pyritea-smalls scema
first to have been effected by MaMtra's works.
Fig. 90 gives a longitudinal, iig. 91 a cross section, the latter
through two furnaces. Usually a whole set is built in a row. In
order to start it, a coal-grate, a, and fire-door, b, are provided, which
are walled up when the burner has got up to a white heat. During
Fig. 90.
this time the top working-door remains opeu, Tlieu the five
plates, c,d,e,f, g are charged with small ore through the doors A, i,k,
whereupon the pyrites takes fire at once. The air enters through /,
and is regulated at will. The gas travels over all the plates
in a serpentine mauner, indicated by the arrows, escapes through
m into the dust-chamber, », and through o into the acid-cham)>cr
or into another dust-chamber. The chamber n is covered by u
metal plate, p, upon which lead pans, r, r, are placed, iu which ail
the chamber-acid can be concentrated from 113° to 14i°Tw. The
336 FRODUCTION OF SULPHUR DIOXIDE.
acid of one pan commuaicates with that of another (as usual) by
siphons or by simple run-overs. Each of the shelves, which are
8 feet long and 5 feet wide, consists of eight plates in two rows of
four each ; they rest, at the sides, in the walls of the burner, in the
middle on fire-clay bearers, *, g, whose shape is better shown in
fig. 92. They are not equidistant, as can be seen in the drawiug;
Fig. i).l.
the upper shelves, where more gas is evolved, are wider apart
than the lower ones, where the radiant heat of the shelves is all
the more useful. The best distance for the upper shelves is
4.^ inches. In order to burn a larger quantity of pyrites, it is not
possible to leave the ore lying quietly, as in
Olivicr-Perret's burner : since here the ex- ^'?' "-■
tenia! heating hy the lump ore is missing,
the combustion would be too incomplete,
and tiie heat would soon get so low that the
burning would cease. The mass must there-
fore be moved, which is done in the follow-
ing way : — Every four hours the contents of the lowest plate, g, are
MALKTRA S SHELF-BURNER. 337
drawn through the door k on to the arch t (which is level at the
top, but slopes behind), after the burnt ore lying on the arch has
first been pushed through the door k to the opening into the
pit u. Then through the door i the contents of / are pushed
down to the plate g, and there levelled again. Thus the higher
plates are successively treated^ till the highest plate, c, is emptied
and can be charged with fresh ore. If four furnaces go together,
one of them is on turn every hour. The contents of the pit u are
removed once a day by the door v. The movement of the ore by
removal from one shelf to another causes its thorough combustion,
and thus also raises the heat. Four furnaces of the above dimen-
sions burn daily 3 tons of pyrites. From 6^ to 7 lbs. of ore are
calculated for each superficial foot of shelving.
Sometimes the shelves are made in the shape of a very flat arch,
for the sake of greater stability ; or at least the bottom is arched,
especially in the case of wide shelves. Some prefer building the
furnaces in such manner that the fire-clay slabs forming the shelves
are nowhere enclosed within the brickwork of the walls; they are
then more easily replaced when broken.
The Maletra burners have recently been improved by making
the fire-clay shelves strouger and doing away with the middle
bearers, s, s (fig. 92), which give much trouble in working the
burner. These single shelves are made from 3 feet 3 inches to
3 feet 7 inches inside.
Maletra^s burner, which has obtained general acceptance in
Germany, has been improved there by Schaffner, P. W. Hofmann,
Bode, and others.
Through the kindness of Dr. Max Schaffner, of Aussig, I was
enabled to obtain detailed drawings of the shelf-burners as modified
by him, and these are reproduced and described in our second
edition, pp. 255 & 256. This furnace has seven plates, each
served through its own door — three on one side, four on the other.
On the first side there is also the ash-pit door, 18 inches square,
for drawing out the cinders, which is thus done in the usual way,
not by the rather inaccessible pit of Maletra. The doors all slide
with their planed margins on equally planed ledges cast on the
front plates, so tliat luting or plastering is not necessary. A
certain number of angle-pieces are bolted to the front plates;
these, between their outer bend and the planed ledges, leave
sufficient room for the doors t to slide each way on the inclined
VOL. I. z
338
PRODUCTION OF SULPHUR DIOXIDE.
Fig. 93.
face a b ; and there is a sufficient number of such pieces present
for each door to be always held by three of them (fig. 93). This
style of work is evidently much cheaper than
casting everything in one piece, because the
planing is much easier ; it is also cheaper than
the English style, shown on pp. 298 and 299, and
quite as substantial as the latter. There are
no special openings for the air, as, in spite of
the planed surfaces, sufficient air enters to sup-
port the combustion. The regulation of the
draught is here effected entirely by the chimney-
damper.
In this burner diLst and peas are burnt to- * ^
get her, and the sulphur is burnt down to 1 per
cent. ; thus the grinding of the smalls, which
is still practiced at some works, is done away
with.
The Aussig or Schaffner dust-burner has
been erected in many works from the plans given in the previous
edition of this book, and that with entire success. Of course at
some places minor modifications have been introduced, but the
principle is always the same, also in the plans given by Falding
in 'Min. Ind.^ vii. p. 668.
Most manufacturers now consider that burners worked from
both sides allow too much false air to enter, and therefore prefer
arranging two rows back to back. I am enabled to give full
drawings of the most modern dust-burners from the designs of
Mr. H. H. Niedenfiibr, as shown in figs. 94 to 96. They are
clear enough to require no further description. '
According to communications from Mr. Benker, he still (1902)
builds his Maletra furnaces on the old system (p. 335), all the
compartments in one line. On the top he places a collecting-flue
and dust-chamber, 5 feet high ; at the end of the set a large dust-
chamber, of the same height and width as the furnace and 20 to
30 feet long, according to the description of ore. One man serves
two compartments. Each of these burns 20 to 24 cwt. of
50 per cent, pyrites per 24 hours, but up to 32 cwt. of poor ore,
such as he had to work in Italy, containing 26 per cent, sulphur
and 3 per cent, copper, of which 90 per cent, was soluble in water
and 95 per cent, soluble in dilute sulphuric acid after roasting.
.4
{
WORKING SHELF-BURNERS. 339
The ore is spread on the plates by means of a tooth-rake^ pro-
ducing an undulated surface and not leaving any bare places.
From such poor ores Benker obtained gases with 7 '7 per cent. SO2
on the average^ and produced 6*2 kg. acid of 106° Tw. per cubic
metre (=0*39 lb. per cubic foot) in 24 hours^ with a consumption
of 0-7 parts nitric acid 66° Tw. to 100 sulphuric acid 106° Tw.
One of the principal advantages of the shelf-burners is that
the ore is burnt out to a much larger extent not merely than with
any of the older forms of dust-burners^ but even with the best
lump-burners. Even without grinding the smalls, it is quite easy
to keep the sulphur in the cinders down to 1*5 per cent. At many
works, e.g. at Uetikon, the average amount of sulphur in the
cinders never exceeds 1 per cent., and frequently it is below. At
Mal^tra^s own works they get down to 0*6 or 0*8 per cent., but
this can be done only by crushing the smalls down to an almost
uniform fine powder. The amount passed through the burner also
influences this. Sorel states that a set of burners, charged with
34 or 35 kils. of ore per square metre every 24 hours, was regularly
burnt down to 0*75 per cent. \ with 36 kils. the sulphur in the
cinders rose to 1 per cent., with 32 kils. it fell as low as 0*42 per
cent. Jurisch, in his ' Schwefelsaurefabrikation ' (p. 80), quotes
30 kils. pyrites per square metre of plates, with variations from
25-6 to 35-8 kils. Stolzenwald (Chem. Zeit. 1901, p. 22), when
burning Hungarian pyrites (47 per cent. S), was not able to burn
more than 24 kils. per square metre of Maletra plates, in order to
get down to 1*7 per cent. S in the cinders.
I have seen in Germany shelf-burners working up " peas '' of
Spanish ore down to 28 per cent. S, and even real '* lumps''
of Westphalian ore down to 35 per cent. S.
Such results can, of course, only be obtained with pure ores,
free from zinc, lead, &c. Hence the cinders from shelf-burners
are readily bought by iron-works, both for blast-furnaces and
other purposes.
In a six-shelf burner there ought to be scarcely any purple
flame visible when pulling the charge down from the top shelf to
^the second shelf. The second shelf is at a bright red heat, the
bird one less so, and so forth ; the back part of the fifth ought to
oe visible only at night by the light radiated downwards from the
fourth, and the sixth ought to be perfectly black. Krutwig and
Dumoncourt (Chem. Zeit. Rep. 1898, p. 242) found the temperature
z2
340 PRODUCTION OF SULPHUR DIOXIDE.
on the top shelf =680% on the second 750°, on the third 720% on
the fourth 650° C.
Sorel found the following percentages of sulphur on the diflFerent
shelves : —
Sulphur in green ore 50 per cent.
Firstshelf 32 „
Second „ 17 „
Third „ 7 „
Fourth „ 5 „
Fifth „ 2 „
Sixth „ 0-75,,
He regularly detected half of the sulphur in the cinders in the
shape of PeS, the other half in that of sulphate.
Crowder (J. Soc. Chem. Ind. 1891, p. 298), in working with
shelf-burners containing seven beds, charged once every eight
hours (so that the charge takes 7x8^8 56 hours to complete the
course), found the following percentages of sulphur on the
different shelves (nearly agreeing with SorePs results mentioned
in the text) : —
Average of 23 trials. Ditto of 26 trials.
Ore charged 50 50 per cent. S.
No. 1 shelf 32-27 32*81
2 „ 21-41 17-55
3 „ 12-77 1109
4 „ 6-39 505 „
5 „ 4-08 3-42 „
6 „ 2-35 2-56
7 „ 2-27 1-96
If there is too much draught, the lower shelves cool down and
the upper ones get hotter. This may cause the process to appear
as going on very well ; but it soon turns out bad. If, on the con-
trary, there is too little air, the bottom shelf becomes luminous
and the sulphur in the cinders rises rapidly. In both cases there
is incipient fusion on the second shelf, which prevents the roasting
from being carried through. This can be remedied by admitting
a little air at the door of the second shelf, or even mixing a little
WORKING SHELF-BURN BBS. 341
dead ore with the charge. The admission of air to the inter-
mediate shelves serves also for bringing forward any burners
wliieh have got behind, and to burn any sulphur subliming from
the first shelf, in case the burners are going too hot^ or from damp
pyrites giving off sulphuretted hydrogen ; but this expedient^
useful as it is when properly handled, must be employed with
caution lest the bottom shelves get too cold from want of air.
In the normal style of working all the air required for convert-
ing the sulphur into sulphuric acid enters at the bottom shelf, and
this large quantity of cold air may lower the temperature of the
nearly burnt-out mass to such an extent that no more ferric
sulphate is decomposed. It was at first sought at Maletra's works
to avoid this by leading the burner- gas downwards underneath the
bottom shelf, thus heating tke latter and employing the ground-
space as a dust-chamber ; but this plan did not answer and was soon
given up again. It has even been tried to utilize the heat of the
burner-gases for a previous heating of the air serving for the
burning-process. But evidently this must most seriously interfere
with the draught, and will hardly answer in the long run. The same
advantage could be secured more easily by admitting at the bottom
only the quantity of air absolutely necessary for completing the
roasting of the air, and allowing the remaining air to enter by a
regulating-slide in the top working-door. In this case the bottom
shelf will be visibly red-hot in the dark. This plan can be carried
out only where the draught is very good, for instance by making
the gas to rise to a considerable height before entering the chambers,
and never leading it downwards in any part of its course. The
burner walls should in this case be made thick or hollow to
prevent loss of heat in the lower part ; on the contrary, an over-
heating of the top shelf should be avoided by making the gas-flue
rather high and causing the heat to be dissipated there, most
rationally by means of evaporating-pans for sulphuric acid.
Another plan is, introducing the requisite excess of air into the
first chambers by means of an injector.
The shelf -burner answers best for rich ores. With 50 per
cent, ore good results are obtained when burning from 32 to 37
kils. of pyrites per superficial metre (= 6*5 to 7*5 lbs. per square
foot) in 24 hours. One may go down to 28 kils. (=5*7 lbs.),
but only exceptionally, because the burners cool down too
much.
342 PRODUCTION OF SULPHUR DIOXIDE.
For poor ores^ say below 38 per cent, of sulphur^ the shelf-
burner does not answer.
The management of shelf-burners is really easier than that of
lump-burners^ but it involves a little more labour. It is generally
assumed that one man can charge^ burn^ and withdraw a ton of
pyrites every day ; but it is possible to get up to 25 cwt. It
seems best to give five burners to each man^ so that each burner
is charged every five hours. The phenomenon of scarring is
hardly ever noticed here. The working-doors must never be left
open any longer than is absolutely necessary for the work ; in this
case both the yield of acid and the consumption of nitre are just
as favourable when working dust on a shelf-burner as with the
best lump-burners. This is the uniform testimony of the many
works I have visited.
In order to start a new burner (which^ of course^ must have
been thoroughly dried first in the ordinary manner) , the commu-
nication with the chambers is stopped and a fire is lighted on the
shelves^ beginning with the bottom^ sometimes by means of a
temporary chimney. After four or five days^ when the burner is
moderately red-hot (it is unnecessary and even injurious to get it
up to a bright red heat), the remainder of the fuel is cleared away^
pyrites is charged on the three top shelves^ and communication so
made with the chambers^ whereupon the regular service is started
as previously described.
During the last twenty years the shelf-burners have also beeti
introduced into a number of English works, and everywhere with
great success. This has been done on the lai*gest scale at the
Newcastle Chemical Works (AUhusen's), where 129 shelf-burners
on Schafiner^s plan, described in this book, have been erected,
which consume from 600 to 650 tons of pyrites-smalls (Ma«on and
Barry's) per week. Each burner is charged, once in 8 hours, with
from 4^ to 4| cwt. of smalls.
The objections made to the shelf-burner in its employment
for sulphite paper-works by Harpf have been refuted by the
author in Zsch. angew. Ch. 1896, pp. 65 & 157.
Combination of lump-burners and dust-burners for the same set
of chambers, — Such a combination is generally avoided, as the
conditions of draught are very different in each case. I have,
however, seen a combination of the above-mentioned kind in
excellent operation at the Stassfurt potash- works managed by
MECHANICAL DUST-BURNERS. 343
Dr. Bemhardi^ the good result being brought about by placing
the shelf-burners so low that the top shelf is on a level with the
charging-door of the lump-burners. This causes an upward
draught in the shelf -burners, and prevents their blowing out, even
when the doors are opened.
Other Descriptions of Shelf -^burners.
The furnace of Finch and W., J., & S. Willoughby (E. P.
2913, of 1883) diflfers from a Maletra burner only in that the
shelves, instead of being placed horizontally^ are inclined alter-
nately in opposite directions.
A modification of the Maletra burner has been patented by
Mactear (no. 3701, 1878).
A furnace, combining some of the features of the Gerstenhofer
and the Maletra burners, has been patented by Hasenclever and
Helbig (description and diagrams, 1st edition of this work, pp. 220
to 222). It has never been carried out in practice, and is not likely
to be so now.
E. Bramwell (G. P. no. 22,758) has constructed a somewhat
complicated pyrites-burner, in which, contrary to the ordinary
dust-burners, the fresh air meets with the green ore, the products
of combustion being gradually led over partially roasted ore,
and at last over the almost spent cinders. This is effected by
placing five calcining-beds in a row, each of them provided with
an outlet for the gas connected with a gas-main, to be connected
or disconnected by means of a throttle-valve, so that the current
of gases can be directed at will. The last burner of the series
communicates with the first by means of a flue underneath, so
that a regular rotation can be kept up, as is done in lixiviating
vats for black-ash.
Mechanical Dust-burners.
The necessity of frequently opening the doors in Mal^tra's and
all similar furnaces is certainly a drawback ; it necessitates much
labour and cannot but introduce some false air. These drawbacks
have been overcome in a most ingenious way in the mechanical
pyrites-dust burner constructed by MacDougall Brothers, of
Liverpool ; but unfortunately fresh troubles have arisen there
344 PRODUCTION OF SULPHUR DIOXIDE.
which have caused these humers to be abandoned again. Still, as
in theory they are the most perfect of all dust-burners, we will
describe them here, especially since the drawbacks connected with
them have been overcome by later inventions. The MacDougall
burner is shown in figs. 97 to 99.
The burner consists of a metal cylinder, 6 feet in diameter and
12 feet high, formed of seven rings (ad) bolted together, and
provided with a solid bottom, but open at the top. The rings are
cast in such a way that the lower and inner edge of each can serve
as an abutment for one of the flat arches 6^ to b^, which divide the
inner space of the cylinder into seven chambers, the uppermost of
which is open at the top. The arches, as well as the cast-metal
bottom of the cylinder, are pierced in the centre, and allow the
passage of a cast-iron shaft, c, 6 inches thick, which is turned by
means of the toothed wheel rf, the pulley e, and the steam-engine
/. The shaft carries at top and bottom the lutes f/ and ffi, into
which the cups h and Aj, fixed to the top arch and the cylinder-
bottom, enter ; the latter are fast, whilst the lutes g and gi turn
round with the shaft, and a hydraulic joint prevents the escape of
gas at the places where the shaft enters and leaves the cylinder.
To the shaft are fixed the cast-iron arms, t'l, i^ . . . «7, provided
with teeth along their lower margin. The teeth are placed
alternately in opposite directions ; so that the arm ij moves the
ore-dust from the centre to the periphery, (2 the same from the
periphery to the centre; h acts like ii; i^ like i^, and so forth.
Corresponding to this, the arches are perforated alternately — 61,
is, and b^ near the margin, b^, ^4, and b^ in the centre. The latter
have a large central opening, 1 foot 3 inches wide, lined with a
metal pipe, which gives free play round the shaft to the gas and
the ore-dust ; whilst in the other arches the shaft is so tightly
surrounded by a metal pipe that scarcely any dust^ and still less
gas, can get through. The small ore (which need only be passed
through a 1-inch riddle, and therefore contains pieces up to the
size of a walnut) is lifted by the elevator k (also moved by the
engine/), and is emptied on to the top flat, Ai, where the arm i^
takes it round and gradually moves it towards the periphery.
During this time the ore is completely dried by the heat of the gas
below. The ore dropping down the edge at / from the open top
chamber is continually pushed into the first closed chamber by a ram
at A, The ram A can be moved reciprocally either by the rod B
[To iaU t- 344,
MACDOUOALL^S DUST-BURNER. 345
or C, and can be moved more or less quickly ; so that the feed of
ore can be regulated to a nicety. The arm 1*2 moves the ore towards
the centre of Jg, where it drops down ; tg moves it towards the
periphery of b^, where it drops down again, and thus quite gradu-
ally and constantly^ being directed by the teeth of the arms^ arrives
at the bottom^ and is emptied out through the pipe m. The two
slides^ n and 0, allow the contents of m to be got out without any
loss of gas or any air entering the other way. As the furnace
during the operation is in full heat, most of all near the top, the
ore ignites as soon as it arrives on the bottom of the first closed
chamber^ b^ ; Rnd in its gradual zigzag way towards the bottom
the sulphur is completely burnt off. The air is continually supplied
by the air-pump p in exactly the necessary quantity ; and the
gas escapes through the pipe r to the acid-chambers.
An apparatus such as is here figured is sufiicient for burning
3^ tons of ore in 24 hours ; with eight closed chambers instead of
six^ it can bum 5 tons. It is also very well adapted for burning the
spent oxides of gas-works; but then it must have only four cham.
bers. In a factory on the Tyne, where this apparatus was at work
for a while^ the consumption of coals for driving the engine
amounted to 4 tons per week. *A two-horse-power engine and a
1^-inch steam-pipe are said to suffice for the largest burner. The
wages amounted to £4 os, per week ; but this rather high amount
was explained by the fact that two other furnaces were building,
which were expected to be served by the same men who attended
the first. Of course this apparatus is quite independent of the
skill of the burner-men, which is mostly acquired only after some
years^ practice.
For heating up, the engine is started and the cold furnace
is gradually filled, care being' taken to regulate the thickness
of the layers of ore on the different floors. When the ore has
arrived at the bottom, the engine is stopped, and the flame of a
temporary fireplace, built against the cylinder, is allowed to enter
it, until the ore lying on the bottom and the floor &2 has taken
fire. Then the engine is started, the temporary fireplace is taken
away, the man-hole is closed, and nothing remains but to see that
the ore arrives at the bottom properly burnt. If this should
not be the case, the speed of the feeding-ram A, that of the
air-pump, or that of the agitating-shaft is altered till everything
is in order. It is easy to get the sulphur in the burnt ore
346 PRODUCTION OF SULPHUR DIOXIDE.
down to 1 per cent. ; in forced work only 3 to 4 per cent, can be
attained.
One objection will at once be made to MacDougall's burner, viz.
that the machinery in its interior must wear out very quickly. In
order to obviate this, all parts of the machinery are made of thick
cast-iron ; and when one of the. arms is worn out it can be renewed
through the man-holes, s, 3, without allowing the apparatus to cool
down. That otherwise this burner has many very great advantages
over all others is evident. The turning of the small ore is per-
fect without any opening of the doors and working by hand. Not
even during charging and discharging does false air enter the
burner; and by means of the air-pump exactly the necessary
quantity of air can be admitted (this, however, in practice was
found to be very difficult). This work, indeed, is done under such
favourable conditions as are realized by no other burner, whether
for pieces or for smalls ; and it might be assumed that the con-
sumption of nitre would thus be reduced to a minimum, and the
yield of acid raised to a maximum. Nevertheless MacDougall's
burner had to be given up again in the above-mentioned factory,
because the quantity of flue-dust was so great that it could not be
managed in any way, and the chamber-process was seriously inter-
fered with. Employment of the Glover tower was not to be
thought of. It does not appear that really efficient dust-chambers
were employed. The air-pump acted so violently that the dust
was carried away a great distance *. Probably this drawback might
have been counteracted by some alteration in the construction ;
but, altogether, the machinery caused endless trouble, continually
requiring repairs, and there is no doubt that it would have to be
altered a good deal before it could become a real success.
The drawbacks ascribed to MacDougalFs furnace were sought to
be obviated by a new patent of the same inventor, E. P. no. 3985,
of 188a. Dust-chambers are provided with perforated baffle-plates
for the interception of dust carried over by the draught, arrange-
ments being made for drawing out the settled dust without allowing
gas to escape or air to enter. The shaft and rake-arms are con-
structed of cast-iron, having a central wrought-iron tube so as to
obviate warping or bending from the eflFect of heat ; and in order
that the shaft may be readily withdrawn, the arms or rakes are
* Davis (* Chemical Engineering/ ii. p. 120) mentions that with mechanical
draught out of 25 tons of dust burned per week 4 tons were carried away as
flue-dust !
MACDOVOALL^S DUST-BURNER. 347
fixed thereto by a fork-shaped end and cotter. A second modifi-
cation is described^ which is to avoid the dust occasioned by the
vertical passage from floor to floor. The furnace is constructed of
an oblongs horizontal* floor or chamber, provided with a series of
vertical shafts, having rakes similar to those above described and
revolving in opposite directions. The teeth are so placed as to
draw the material towards each shaft, and thus pass it from one to
the other and from end to end of the furnace. Where prolonged
roasting or burning is requisite, a similar furnace, or the first
modification, may be superposed above the last mentioned, the
material being first passed through the upper furnace.
A further improvement was patented by the same inventor as
E. P. no. 22,504, 1891. Here the central shaft is made in several
lengths, coupled together by widening one end of the shaft to form
a square socket, and fitting into this the square end of the other
shaft, the two being secured by a square key wedged between
socket and square end ; a tight-fitting spring-clip protects the
coupling from the action of the burner-gas. The furnace-rings
are joined together by half check-joints secured by set pans and
rust jointing ; the ends of the pins are not exposed to the corrosive
fumes.
The subject of mechanical furnaces generally has been treated
in detail by Bode, Dingler's Journal, ccxix. p. 55, and Wagner's
Jahresb. 1876, p. 296.
A furnace very much like MacDougalFs was patented by
Mr. Perret in France, on June 23rd, 1875.
A mechanical pyrites-kiln, greatly resembling MacDougalPs in
principle, has been patented by R. Mackenzie (E. P. no. 4418,
1881). It is provided with a water-cistern at the bottom, in the
hope of promoting the process by the presence of aqueous vapour.
Also a similar furnace, differing only in details of construction,
has been patented by Black & Larkiu (E. P. no. 4456, of 1881),
and another by Johnson in America.
A much more important modification is the burner constructed
by Herman Prasch, which avoids most of the difficulties of the
above system by the introduction of water-cooling. The Frasch
burner has been described by me in Zsch. f. angew. Ch. 1894,
p. 15, from which parts of the drawings are reproduced here as
figs. 100, 101, 102. We notice the hollow shaft C, 8 inches outside
diameter, 5 in. bore. Above the cylindrical burner it is connected
with the fixed water-pipe D by means of a stuffing-box. As we
PRODUCTION OF SULPHUR DIOXIDE.
Pig. ]00.
see from fig. 102, pipe A' takes
the water from the cistern E to
the top of D, and pipe k' , starting
from the bottom of E, reaches
nearly to the bottom of shaft C.
This eauses a continuous circula-
tion of water from E thi-ough k
dowQwardSj then upwards in the
annular space between k' and C,
and back through k to E. Braucli
water-pipes reaching from C into
the hollow stirrers H protect these
also from fusing or deformation.
Shaft C is, moreover, protected
on the outside by wire gauze,
covered with a paste of fireclay.
The remainder of the figure
can be easily understood by
reference to the MacDougall
HERKESHOFF DUST-BURNER. 349
furnace, but attention must be drawn to the very efficient dust-
chamber shown in sectional elevation fig. 100, and sectional plan
fig. 101.
I have convinced myself by personal observation of the excellent
function of this apparatus, especially also of the fact that the
protective-water circulation between the hollow shaft C and
the hollow side-arms H is quite sufficiently effected by the steam
formed in the latter.
According to a report received from Mr. Frasch in 1902,
his furnaces, which are now regularly made 16 feet wide, bave
done perfect work ever since ; thirty of them are in operation at
eight different works of the Standard Oil Company for the purpose
of roasting metallic sulphides. The heat produced in the interior
of the shaft and arms is utilized by attaching a steam-drum to
the highest portion of the water circulation, and the steam is used
under two atmospheres' pressure for distilling benzine out of light
petroleum oils.
Another modification of the MacDougall burner is the Herres-
hoff burner, where the cooling is performed by air (figs. 103
& 104). This principle had been carried out exactly in the same
manner already before 1883 by the blende-furnace of the Societe
Vieille Montague {vide infra, p. 364). This burner and its working
are described in ' Mineral Industry/ vi. p. 236, and by Gilchrist
( Journ. Soc. Chem. Ind. 1899, p. 460) . The central shaft a, fig. 103,
is hollow, 14 inches outside diameter and about | inch thickness of
metal. Above every shelf there is a cross channel or socket passing
directly through the shaft, about 4 inches wide and 5 inches high.
Into this socket the side-arms are inserted. In the top ot the
channel, at the centre of the vertical shaft, is a pocket running
across the channel ; into this a rib at the inner and top edge of the
ai*m locks, when the arm is forced into its proper position, and the
weight of the arm keeps it locked in the channel. By raising the
outer end of the arm about 3 inches, the top edge of the rib is
brought below the pocket and the arm can be pulled out. Each
arm, weighing about 100 lbs., can be unlocked and removed and a
new one inserted in about one minute. The greatest strain is at
the point of contact with the central shaft ; but the cooling effect of
the air in the shaft prevents any damage, as the temperature never
exceeds a dark red heat. About six of these arms, at a cost of less
than £1 each, cover the renewals per annum. The arms are hollow.
PltOIIl'CTlOK OF 8DLFUUK DIOXIDE.
MECHANICAL DUST-BURNERS. 351
of a rectaDgular section^ and are made of cast-iron capable of
resisting the heat.
The cross channels do not take up the whole area of the shafts
but allow the air entering at the bottom to pass up the shaft; a
light steel stacks about 10 feet high, provides the draught.
The five shelves are fire-brick arches, 4^ inches deep^ levelled
off with ashes or similar materials. The ore is fed automatically
from the hopper c by means of the plunger d moving backwards
and forwards. The passage down the shelves takes place exactly
as in the MacDougall furnace. The central shaft revolves once
in two minutes, the plunger makes one stroke per minute. There
are two outlets for the burnt ore at the bottom^ 5x3 inches.
The furnace is about 10 feet wide and 10 feet high. The outer
casing is of ^-inch steel, riveted together, with a red brick lining,
8 inches thick. The total weight is about 7 tons; the heaviest
piece weighs a ton. The motive power is practically ^ H.P. per fur-
nace. Each furnace roasts 6000 lbs, of 44 per cent. Virginia ore,
or 49 per cent. Tharsis ore, in 24 hours. The amount of sulphur
left in the cinders depends upon the quality of the ore, may be
2 or 3 per cent., or down to 1 or 1^ per cent. The furnace takes
also low-grade ore, with 30 or even 25^ per cent, of sulphur.
With magnetic pyrites (pyrrhotite) the grain should not be larger
than wheat, with proper pyrites it may run from pea-size down to
dust. When some sulphur is to be left in for the wet copper
extraction process, up to 8700 lbs. per 24 hours can be treated.
Spent oxide of gas-works can be also worked, but it is best mixed
with pyrites fines.
At one American works where 70 of these furnaces are in operation
only two men attend to them. A prospectus issued in November
1901 enumerates ten American and fourteen German, French,
Austrian, and Italian factories where these furnaces were intro-
duced ; four furnaces had been ordered by British manufacturers.
During 1902 I heard great complaints about the unmanageable
quantity of dust produced by these burners (comp. p. 346)
A. P. O'Brien has further improved the HerreshofE furnace
(comp. Falding, Min. lud. ix. p. 623), but no details or definite
results can as yet be stated. Utley Wedge (U.S. P. 649,183 and
654,335) describes a similar furnace.
Other mechanical dust-burners. — Farmer and Hardwick's
mechanical pyrites-burner (1878) resembles to some extent Jones's
352 PRODUCTION OF SULPHUR DIOXID£.
and Walshes mechauical salt-cake furnace. It is automatically
charged and emptied; 5^ tons of pyrites are said to be burnt off
in from 7 to 9 hours. None of these furnaces seem to be in
practical operation.
The mechanical pyrites-burner of P. Spence (patented in Eng-
land Dec. 24, 1878 ; in Germany No. 9267 ; in America
No. 248,521) is a shelf-burner provided with mechanical stirring
arrangement. As this furnace seems to be constructed on rational
principles, and as it has met with practical success in America,
we give here a complete description of it.
Fig. 105 shows an exterior side view ; fig. 106 a plan ; fig. 107 a
cross section of one half of a double furnace^ the other half being
shown in outside view ; fig. 108 a longitudinal section of the furnace-
beds; and figs. 109, 110, and 111 views of the stirring and raking
instruments. The construction of the furnace-beds is best seen
from fig. 108. In this there are at 1 the walls of the furnace, in
which are fixed projecting fire-clay slabs, 2. Upon these are placed
tiles, 3, reaching from one side of the furnace to the other, a num-
ber of these composing the length of each furnace- bed. In figs.
107 and 108 the several beds are shown at 3, 3a, 3b, 3c; alternate
openings in these beds being shown at 4, 5, 6, 7. The pulverulent
material is thrown at H on to the floor 3 ; advancing rakes or
ploughs stir it and carry forward a portion of it through the open-
ing 4 on to the second bed 3a, The teeth of the rakes are formed
of a triangular section, as shown in fig. Ill, the apex of the triangle
being in the direction of the motion of each rake longitudinally
from end to end of the furnace, the flat sides of the teeth of the
rake being in the direction in which it is desired to traverse the
ore along the bed of the furnace. When the rake is advancing in
the direction of the pointed part of the teeth of the rake, the ore
will be only turned over; but when the rake is moving in an
opposite direction, a certain quantity of the ore will be carried by
the flat side of the teeth along the floor of the furnace. Thus the
ground material, delivered to the floor 3 at the point H, is stirred
and subsequently partially carried forward till it is delivered
through opening 4 on to the second bed, 3a, where the same opera-
tions take place, the material now passing down the opening 5
on to the bed 3b, and so through all the beds, until it is at last
discharged through the opening 7 into the receptacle 8. Since the
openings in the successive beds are on alternate ends of the furnace.
SPISNCB S DITST-ltl/HNEK.
PKOOnCTlON OF SULFHUK DIOXIDE.
Fig. 107.
*rnTTT
Fi^. 111.
[A ^ ^
tlie stirring and couvcyirig instrumeuts must be reversed as regards
their faces in succeeding beds. The teeth of the rakes are mounted
in angle-iron bars, 11, provided with rollers, 11*, wliich run upon
rails, 12, carried by the projecting supports 2, To eacli of these
aiigle-biu's are connected rods, 13, attached at their other ends to
a frame or carriage, 14, provided with wheels, 15, which run upon
rails, 16, on the floor, the said rods being supported and guided by
grooved pulleys, 17, On the carriage, 14, are fixed toothed racks,
18, situated outside the furnace, and supported at their outwai-d
ends by rollers, 19, and iu gear with these rods are piuions, 20, on
a shaft, 21, driven by motive power. Motion being communicated
spence's dust-burxer. 355
to the shafts 21> the pinions^ 20^ cause the racks^ 18^ to traverse
the frame^ 14^ which, as stated, travels on the rails, 16, and thus
the rods, 13, are caused to traverse the rakes or conveyers along
the several beds of the furnace. According to the positions shown
in the drawings, the carriage 14 is in its outward, or nearly out-
ward, position, and the flat ends of the instruments will have
delivered a certain amount of material through the opening 4 on
to the bed 3a, the same operation having taken place with regard
to the opening 6 and bed 3c. The carriage now running inward,
the sharp points of the ploughs will simply stir the material on the
beds 3, 36, while the blunt ends on the floor 3a will deliver a
certain quantity of material through the opening 5 on to the bed
36, and at the same time the instruments on the floor 3c will pass
an amount of completely calcined material into the receptacle 8,
to be removed at pleasure. The feeding of the furnace takes place
in the following manner : — At P is a channel leading to the top
floor 3, and above this channel is a hopper, 20, into which the
ground material is from time to time fed. The bottom of this
hopper is provided with a sliding plate, 26, having a ledge at
its inner end, as seen in fig. 107. This plate is connected with
rods, 22, swung upon arms, 23, and each having two stops, 24, 25.
According to the position shown, the material rests upon the ledge
of the plate, 26, which, when the carriage runs in, is pushed for-
ward by its arrival in contact with the stops 24, and this action
delivers a certain amount of material through the channel F. On
the return motion of the carriage the stops 25 shift the plate 26
back, and so on. Instead of plate 26, there may be a winged
bottom to the hopper. The furnace may be single or double, the
latter (which is shown in the drawings) being preferred. The
shaft 21 is connected by means of suitable gearing to any source
of motive power, so that it may be rotated first in one direction
and then in another, and thereby traverse the rakes alternately
from one end of the furnace to the other. The rakes may move
continuously ; but it is preferable that they should remain
stationary periodically in the position shown in the drawings, as
they are then clear of the ore and out of the direct action of the
heat, thereby sufiering less injury from corrosion.
A number of these furnaces have been put up by the Sulphur
Klines Company of Virginia, at Baltimore. According to a com-
munication kindly made to me in May 1888 by the President,
2a2
356 PRODUCTION OP SULPHUR DIOXIDE.
Mr. Crenshaw, three double furnaces were connected with a set of
chambers of 180,000 cubic feet capacity, with Glover and Gay-
Lussac towers, and two other furnaces with a set of 126,000 cubic
feet. The 5 furnaces were to burn 14 tons of 47/48 per cent,
pyrites each 24 hours, down to less than 2 per cent, of sulphur in
the cinders.
It is mentioned as a drawback of Spence's furnaces that at the
back end a collection of dust takes place by which the rakes are
prevented from doing their work properly. Bartsch, of Bridgeport
(Pischer^s Jahresb. 1886, p. 256), consequently applies to the
furnace ends a separate set of broad plates, moved by a second
moving-frame in sucli a w^ay that the dust is cleared out automatic-
ally every time the principal frame is made to work.
As stated in Zsch. f. angew. Ch. 1894, p. 134, I have met with
several Spence furnaces at work in America, but they did not give
general satisfaction. A variety of this furnace has been patented
in America by A. C. Johnson, of Baltimore, where the moving
parts are protected from heat (U.S. P. 642,334).
The ordinary shelf-burner has been combined with a mechanical
arrangement by Hering (G. P. 9634), who feeds the top shelf
continuously by means of a screen, the burnt ore being removed
from the bottom shelf by another screw.
T. Mason (E. P. 3196, 1880) employs a furnace with a bed
slanting slightly downwards, across which a number of fluted
rollers (say 20) are lying, made of cast-iron or stoneware, moved by
means of gearing outside the furnace. The pyrites-dust is fed
mechanically into a hopper, situated at the upper end, and is
gradually moved down the inclined hearth by means of the fluted
rollers, the cinders being discharged at the lower end. A furnace
placed below gives additional heat in the case of poor ores. (It is
not likely that such an arrangement would stand the wear and
tear unavoidable in this case.) Modifications of this furnace are
contained in the patent No. 1788, of 1881, and No. 2831, 1882.
Walker and Carter (E. P. 4000, 1883) employ for roasting
pyrites a set of eight horizontal cylindrical retorts (four tiers of
two each), heated outside by an ordinary coal fire, and communis
eating by openings at alternate ends. Hollow shafts, provided with
stirrers, pass through each retort ; cooling-water runs through all
these shafts from one to another and prevents their warping.
Special contrivances prevent the stirrers from touching the sides of
MECHANICAL DUST-BURNERS. 357
the retorts in spite of unequal heating. The broken ore is fed into
the top retorts and gradually finds its way into the three following
tiers ; a current of air traverses the retorts in the opposite direc-
tion. This apparatus is stated to have worked well for roasting
pyrites containing 42*2 per cent, sulphur (Engineering and Mining
Journal^ xxxvii. p. 29i), but it does not convey the impression
that it would go on for a long time without very heavy wear and
tear ; and the necessity of an outside fire also militates against it.
The report made on this burner by W. Martyn (J. Soc. Chem.
Ind. 1885^ p. 26) is not encouraging.
W. Briiokner (Engineering and Mining Journal^ xxxvii. p. 425;
Fischer's Jahresb. 1884, p. 221) employs for roasting pyrites-smalls
the principal of a horizontal cylindrical revolving furnace. As
this furnace is provided with internal firing, and the SO^ gets
mixed with all the smoke-gases, it is hardly intended, and certainly
not adapted, for serving in the manufacture of sulphuric acid.
This is, on the contrary, the aim of an apparatus patented by
R. & C. Oxland (E. P. No. 7285, 1885), who roast the pyrites in
a revolving cylinder of 30 feet length heated from the outside, so
that the SO2 is kept apart from the smoke-gases. At the lower
end of the rotating tube is a cast-iron prolongation, heated exter-
nally by a fire-grate and flues. The amount of air admitted to
the calciner is regulated by a contrivance in the end plate of the
prolongation, which is also fitted with a door for the removal of
the cinders. (This apparatus seems to be more intended for
roasting arsenical ores, and recovering the white arsenic in the
process.)
K, Walter has constructed a special burner for '^peas/' that is,
small pieces of ore between the size of a hazel-nut and such as
pas:} through a sieve with eight holes to the lineal inch. This
apparatus is described and figured in the 1st edition, pp. 225 to
229. It has grate-bars laid sideways across the burner, cast with
projections beneath, through which a bar passes. This makes it
possible to turn all the bars at the same time and exactly in the
same way. The spaces between the bars can therefore be made
very small, so that the " peas '^ do not fall through without the
bars being moved ; and as they lie only 6 inches thick, the air can
pass through with the ordinary draught. Each compartment,
with a grate surface of 32 square feet, burns from 14 to 20 cwt. in
24 hours.
358 FBODUCTION OP SULPHUR DIOXIDE.
Walter's burners require a strong draughty and must be worked
for a separate set of chambers ; with insufficient draught very large
scars are at once formed. As they are only adapted for a special
size of ore, they have not met with any extended application.
4, Furnaces for Koasting Zinc-blende.
Zinc-blende is now the most important of all zinc-ores^ and all
of it must be converted into zinc oxide by a thorough roasting.
This is nothing like so easy as with pyrites, as, on the one hand^
blende contains at best only about 33 per cent., sometimes down
to 18 per cent., of sulphur^ and, on the other hand, it is much
more difficult to burn out than iron or copper pyrites, zinc sulphate
being formed, which is very difficult to decompose. As the
manufacture of zinc requires nearly all the sulphur to be driven
off, and as the burners employed for pyrites are useless in the case
of blende, the latter was formerly roasted in reverberatory furnaces,
all the sulphur dioxide passing away with the furnace-gases. This
would probably be done even now, if the damage produced by the
acid smoke had not caused the sanitary authorities, both in Eng-
land and on the Continent, to impose upon the manufacturers the
duty of doing their best to condense the acids contained in the
smoke, the damage caused by which has been described svpra,
pp. 154 et seq.
The development of the processes for abolishing the acid-smoke
nuisance and utilizing the sulphur in connection with the roasting
of blende, at least in one of the most industrial parts of Germany,
has been described at length by Robert Hasenclever (Zeitschr. des
Vereins deutsch. Ingenieure, 1886, p. 83; Fischer's Jahresb. 1886,
p. 257), of whose paper we give a short abstract. Before 1855
all the blende consumed at the large Stolberg zinc-works was
roasted in ordinary open furnaces by a direct coal-fire, the gases
from which contained only about 0*75 per cent. SOj by volume
and escaped into the atmosphere. In 1855 the Rhenania chemical
works at Stolberg erected a furnace, intended to utilize the larger
portion of the SO2. This furnace, the invention of Friedrich
Hasenclever, was the first by which the gases from roasting blende
could be utilized for the manufacture of sulphuric acid. It con-
sisted of a long muffle-furnace in which the ore was shifted along
the hearth from back to front by manual labour (evidently identical
in principle with Spence's first furnace, supra, p. 330) . Thus half
MIEMDE-HTRNACES,
tlic sulphur could he uHlizcd for the manufacture of sulphuric
aeiti, but with great waste of nitre. Godiii improved this furnace
by placing a series of muflBea on the top, which the ore had to pass
gmdatim : hut this apparatus waa given up again for the Gersteu^
hiifcr furnace (siiprii, p. 3^2). By this, comparatively rich gases
were obtained, but the immense quantity of fine-dust was very
trouhleaome, and it proved too dilficiiit to combine this furnace
ivitli another for finishing the roasting. After this several orritnge-
mciits were tried by Hasenelever and Helbig, but these sho.ied
various drawbacks; and this led in J874 tn the construction of a
IV. 112.
new furnace, which has been fully described and illustrated in the
first edition of this work, pp. 201 to 204 ; we describe it only
briefly here {fig. 1 12) . It consisted of a reverberatory furnace, y,
on the top of which was placed a muffie, c, reaching all over its
length. The bottom of this muffle, forming the top of the rever-
360 PRODUCTION OF SULPHUR DIOXIDE.
beratory furnace, was heated directly ; the flame of the furnace,
passing over the roof of the mufl9e, heated this as well^ and then
passed under an inclined plane at an angle of 43°^ at the top of
which, at a, the fresh (powdered) blende was fed in. The ore, on
sliding down this plane (partitions, d, d, preventing its transit
from becoming too rapid), underwent a preliminary heating by
the hot gases passing underneath ; this heating was continued in
the muffle c c, where air was admitted and as much of the sulphur
was burned away as possible. At last the semi-burnt ore was
pushed down into the furnace ff, where it was burned completely
by the fire from the gas-producer k. The sulphur dioxide formed
here is certainly lost, and it constituted no inconsiderable portion
(generally more than one third) of the whole ; but that formed in
the muffle and on the inclined plane is strong enough to be carried
into lead chambers and converted into sulphuric acid, certainly at
no great profit, if any.
A large number of Hasenclever furnaces were erected on the
Rhine and in Silesia, but they were found to utilize on an average
only 60 per cent, of the sulphur as sulphuric acid. The remaining
40 per cent. SO2, being mixed with the fire-gases, still escaped
into the atmosphere and were blamed for even more damage than
they actually caused, the part played by the numerous zinc- works,
glass-works, &c. being generally overlooked by the public (comp.
pp. 82 & 83). We shall return to this subject in the 7th eectiou
of this chapter.
All these furnaces have become obsolete by the construction of
blende-roasting furnaces, which permit of utilizitig the whole of the
sulphur of the ore in vitriol-chambers , and thus do away with the
extremely troublesome processes for absorbing SO2 from fire-gases.
All these modem furnaces have one principle in common : they
combine the heat produced by the burning of the blende sulphur
with heat applied externally, but in such manner that the fire-
gases are kept entirely separate from the roasting-gases. It has
been found that this indirect heat suffices for completely roasting
the blende.
The furnaces which first realized the principle of utilizing all
the sulphur of the blende in vitriol-making, by completing the
roasting by means of indirect fire, were invented by M. Liebig, who
had the idea of constructing a sort of shelf-burner with shelves
partially hollow and heated by indirect fire. His invention (G. P.
ore is spread in layers twice as thick as on the bottc
small gaps between the charges, to prevent their
it takes three or four days before a hatch chargf
discharged at the bottom. The raking over anc
)f the blende in vitriol-making, by completing the
eaus of indirect fire, were invented by M. Liebig, who
of constructiDg a sort of shelf-burner with ehelvea
iw and heated by indirect fire. His invention (G. P.
RHENANIA BLENDE-FURNACB. 361
21032 of Eichhorn and Liebig) is described and illustrated (from
the patent specification) in our second edition^ pp. 273-275
Comp. also Eichhom's report in Fischer's Jahresb. 1889, p. 332.
We omit the description of the original style of these furnaces
in favourof a modification known as '' Rhenania furnaces/' because
they were constructed at the Rhenania Chemical Works at StoU
berg (managed by R. Hasenclever). These are now widely used,
not merely for blende^ but also for other poor sulphur-ores in the
state of smalls or even of *' peas '' (pieces up to ^-inch diameter),
and for metallurgical products (matte) of the same kind (p. 363).
The Rhenania furnace is shown in figs. 113 to 116, in a shape in
which it has been used for several years, and which embodies the
same principle as the furnaces actually at work at present^ which
merely differ from the above in details, to be mentioned below.
These furnaces require two men per shift, and roast about 4 tons
blende per 24 hours, with a consumption of 16 cwt. of coal. The
temperature of the top muf9e is 580° to 690^", that of the following
muffles 750° to 900°. The progress of roasting is shown by the
following percentages of sulphur: —
Raw ore 19-2 268 26 5
End of first muffle 17-6 191 to 21'9 15-9 to 214
„ second „ 120 112 „ 14*3 99 „ 12*4
„ third „ 3-4 1-02 „ 1*48 0 75 ,, 106
99
roasting 0*6 0-35 „ 1*02
In the case of richer ores the second muffle does not require
heating; the fire -gases may be conducted below the third muffle
without a partition, and then at once over the second roasting-
bed.
Jahne (Zsch. f. angew. Ch. 1894, p. 305) describes some ex-
periments made with these furnaces. They are greatly improved
by arranging a partition cutting vertically through all three
muffles and thus dividing the furnace into two parts^ without any
change in the firing. This admits of roasting 30 per cent, more
ore down to the same percentage of sulphur and saves labour. It
is best to build two furnaces back to back. On the top beds the
ore is spread in layers twice as thick as on the bottom beds, leaving
small gaps between the charges^ to prevent their getting mixed.
It takes three or four days before a batch charged at the top is
discharged at the bottom. The raking over and moving on of
362 PRODUCTION OF SULPHUR DIOXIDE.
the batches^ following upon the diBcharging of a finished batch,
requires 4 hours. The most careful work must be bestowed upon
the finishing of the batches in the hottest place ; this lasts at
best 1^ hours, usually 3 to 4 hour?, sometimes 7 to 8 hours.
Tests are made by adding hydrochloric acid to a sample^ and
looking for any evolution of HgS. The sulphur can be brought
down to 0*1 or 0*2 per cent. ; it ought never to exceed 1 per cent.,
except in the case of blendes containing limestone, which may
retain 2 or 3 per cent. S in the shape of calcium sulphate. In
regular work a charge of 9 to 12 cwt. roasted ore is drawn every
6 hours. 100 parts of raw blende average 85 of roasted ore, and
require 25 or 26 parts of good coal. The furnaces are provided
with large dust-chambers, which must be. cleaned every few weeks.
The dust consists mainly of zinc sulphate and calcium sulphate.
Lead oxide is carried farthest, even through the ten cylinders of a
Hargreaves plant connected with the furnace at Stolberg, right
into the earthenware receivers of the acid-condensing plant. If
the blende contains mercury, that metal is found in the mud of
the vitriol-chambers connected with blende-furnaces, and can be
recovered from it by distillation.
In Chem. Ind. 1899, p. 25, Hasenclever gives some more details
respecting the Rhenania blende-furnaces, which now roast 4700 kils.
blende in 24 hours with two men per shift. Only for poor ores
the fire plays round all the muffles ; generally the ore is roasted in
three superposed muffles and the fire plays only round the top and
bottom muffle. Two furnaces are built back to back. In order
to avoid annoyance to the men, there is a space of 25 feet between
two opposite tuniaces. Six furnaces are connected with one set
of chambers, their flues being built closely together, to keep the
heat, until they unite just in front of the Glover tower.
Pierron (Mon. Scient. 1900, p. 562) states that at the Vieille
Montague the consumption of fuel is only 18 kils. per 100
blende of 25-28 per cent. S (with hand-furnaces or mechanical
furnaces?).
Jensch (Zsch. f. angew. Chem. 189 A, p. 50) shows by analyses
that the sulphur in roasted blende is mostly contained therein in
the shape of ferrous sulphide ; when roasting down to 2 per
cent. S, no ZnS is present, and it is therefore quite unnecessary
to drive the roasting down to 0*5 per cent. S, as is sometimes
demanded.
RHEXANIA BLENDE-FURNACE. 363
At Oker (1902) all " fines'' of the ores used there (p. 85),
passed through a 7 millim. sieve^ are roasted in Rhenania blende-
furnaces. Each furnace roasts from 3*55 up to 5 tons ordinary
ores or matte^ according to whether the roasting is driven to 3 or
to 7 per cent. S in the residue; and from 3 to 4*5 tons regulus,
according to whether 2 or 6 per cent. S is left in the residue.
At Freiberg the Rhenania furnace is used (1902) for pyrites
smalls and " peas '' down to 5 millim. size (^ inch), for blende,
recently also for matte and speiss^ containing about 20 per cent. S,
25 per cent. Cu, and 20 per cent. lead. For ores containing rather
more sulphur the furnaces are provided with three mufSes directly
superposed. The fire-flue passes underneath the hearth of the
bottom muffle, rises up at the end of this and along the ends of
the second and the top muffle, and returns along the roofing-arch
of the latter, so that the fire-gases heat only the bottom of tlie
lowest and the top of the highest muffle. For poorer ores the
furnaces have only two muffles, the fire-gases passing both under
their top and over their bottom and between both muffles. With
such furnaces, whether containing two or three muffles, the sulphur
in the cinders from pyrites ^' fines '' is brought down to 2 per
cent., from pyrites *' peas '' to 3 per cent., from blende to 1 per
cent., from matte and speiss to 5 per cent. Each furnace takes 3^
to 4 tons ore per 24 hours, with a consumption of about 12 cwt.
lignite and 6 cwt. coal ; the work is done by two men per shift of
12 hours.
At Stolberg, in 1902, each furnace turned out 8 tons of roasted
blende per day, with four men per shift, two on each side, and from
14 to 18 per cent, coal of the weight of roasted ore. The blende must
be crushed to pass through a sieve of 2 millim. meshes. The product
in the absence of lime shows 0*5 to 1 per cent. S, in the presence
of lime correspondingly more. The roasting is stopped when
a sample, treated with hydrochloric acid, does not stain lead
paper. Sulphates (apart from calcium sulphate) are most easily
formed when the furnaces do not go hot enough, e. ff, after
stopping for Sundays ; once formed they are never completely
decomposed.
Bemelmans (Germ. pat. 7677o) describes a furnace for roasting
both pyrites and blende in separate compartments ; the sulphur
vapour obtained from pyrites, sometimes by addition of coal, is
utilized for removing from the blende (which must not be mixed
364 PRODUCTION OF SULPHUR DIOXIDE.
vfiih coal) arsenic^ antimony^ and phosphorus in the shape of
volatile sulphides.
Michel Perret (G. P. 37842) has modified his well-known
furnace for burning fuel in the shape of dust so as to roast blende^
without mixing the fire-gases with the gases evolved by the burn-
ing blende. The principle is very similar to that of Eichhorn and
Liebig.
Mechanical blende-burners are the following : — The burner
patented by J. Haas (G. P. 23080) is very similar to Mac-
Dougall's (comp. p. 343), but the single chambers^ in lieu of
having simple brick bottoms^ are separated by flues through which
pass the fire-gases from a coal-fire. Mechanical stirrers move the
ore from the top shelf over three others and ultimately into an
open hearth, where the last roasting takes place.
Hegeler's burner is a combination of an Eichhorn and Liebig's
burner with a stirring arrangement somewhat similar to Spence's
(p. 352), but difi^ering from it in some important practical details.
This furnace works most successfully at Mathiesen and Hegeler*s
zinc-works, La Salle, 111. It treats 35 or 40 tons blende with 28
per cent. S in 24 hours.
A mechanical blende-roasting furnace patented by the Chemische
Fabrik Rhenania (G. P. 61,043) has not been actually erected
(Chem. Ind. 1899, p. 26). Hitherto in Europe no other but the
Yieille Montague mechanical blende furnaces seem to be at work.
The burners patented by the Societe Vieille Montague (G. P.
24,155 and 36,609) are mechanical burners in which the flame of
the coal-fire is not separated from the roasting-gases. These
burners have been continuously at work at Oberhausen since 1883.
Their construction is shown in fig. 117. There are several super-
posed circular calcining-hearths, A, A, to which is attached a square
calciner B. The ground ore is charged through hopper a by means
of feeding-rollers and flues on to the top chamber and gradually
finds its way downwards and into B. The fire of the fuel burning
on grate T first passes over B, then over the circular hearths A, A,
into the dust-chamber C and into the flue S. The agitation is
procured by the vertical shaft b and arms e, e, the stufBng-boxes
being packed with asbestos. Shaft b is contained in an outer pipe ff^
and the air rising between them acts as a cooling medium. The
arms carry tooth-rakes m in a radial position for the purpose of
stirring, and slanting solid rakes / which move the ore from the
MECHANICAL BLENDE-FURNACES. 365
circumference to the centre, or the other way, as is required for the
purpose of gradually traosportiog the ore downwards and ultimately
on to hearth B. [These furnaces, which had to be frequently
repaired and mode much flue-dust, are being gradually replaced by
hand-worked mufBc- furnaces.]
The furnaccB or kilns for roasting ordinary copper-ores, lead-
ores, and ao forth, aa far as they are not mentioned pp. 311,312, and
357, cannot be described in this book, as they belong to tlie domain
of metallurgy proper, and in these cases the roas ting-gases are, if at
allj sent into vitriol-chambers merely to get rid of them, but with-
out the expectation of profitable work.
We merely mention a few modern attempts to obtain workable
roast iug-gases from such ores by new methods.
Fig. 117.
Huntington and Heberlein (E. P. 8064, 1896; 3795, 1897)
obtain SO^ by a new treatment of lead-ores, but in such a way
that their conversion into sulphuric acid seems to be almost im-
possible.
Sebillot (13. P. 21,6lt!, 1898) charges cuprous or other sulphur-
ores, mixed with fuel, into a furnace provided with air-blast. The
gasea are taken into a chamber containing coke, pumice, or a
suitable metallic oxide, where they are treated with air and steam,
and where sulphuric acid is formed (comp. Chap. XI.}.
Treatment of complex ores containing blende. — Hart (Journ. Soc.
366 PBpDUCTION OP SVLFHUR DIOXIDE.
Chem. lud. 1895, p. 544) proposes treating Buch o>es with sul-
phuric acid in a saltcake pot, and when tlie mass has become pasty
transferring it to a blind roaster and finishing it there, all the
gases going into vitriol-chambers. The zinc remains behind as
sulphate, which can be obtained by lixiviation aad crystallization;
or else it is mixed with poor zinc-ore and roasted, in which case
the oxygen of the sulphate combiaes with the sulphur o£ the
blende. The reaction seems to be :
ZnS + 4SOa=ZnS04-|-4SOj.
5. Burners por the Spent Oxide or Gas-works.
I'he spent oxide is now generally washed, so as to obtain
ammonia salts therefrom, and is also frequently treated for the
ferrocyanide or sulpliocyanide. At all events the sulphur, which
it contains in the free state, sometimes up to 50 per cent., is
ultimately burnt for the manufacture of sulphuric acid. This is
sometimes done in ordinary brimstone-burners, as shown supru,
p. 268 et aeq. ; but in this case it is difficult to bum it out com-
pletely, and there is loss of sulphur in the cinders. Ordinarily it
is burnt in apparatus very similar to "shelf-burners," as shown
in f)g. 118 (Hill's burner). Each chamber in this case is about
riy. 118.
10 feet long, 20 inches wide, and 9 inches high. MacDougall's
and Herreshoff's mechanical burners (pp. 343 & 349) have also
been employed for this purpose.
Cowen's burner, figs, 119 and 120, consists of a row of fire-clay
gas-retorts,* and requires no further explauation. Other works
BURNEHti FUR SPENT UXIOB AND SULFDl
ar3 said to burn that material in burners with very narrowly-
placed grate-bars.
Soiuttimcs tlie oxide is moulded into bricks and put into liimp-
buiners ; it burns off very well and the cinders fall tliroujjli the
grate-bars by themselves : in fact, the bars must be touched as little
as possible. This process does not answer so well as shelf-burners.
The asfh Alkali Report, p. 97, recommends not to discharge
the hot residue from oxide burners through the front working-
doors, as this causes a nuisance, but to push it into chambers
placed at the back, where they can cool oflf. The f^ases should be
led through loug, heated flues, in order to burn the tarry substances
and ammonia whieh destroy nitre.
6. Buhners for Svlphvhetted Hydrouex,
These are usually of a very simple description. As shown in
tig. 121, they consist of a brick chamber provided with sonic
battling-walls a, a. The sulphuretted hydrogen gas (which is
nearly always mixed with a large quantity of inert gas, chiefly
nitrogen) is introduced by the cast-iron pipe b, the supply being
regulated by an inlet-valve c. Air is admitted partly round the
pi|>e A, partly by n special opening d, which ought to be provided
with a slide or otiier means of regulating the amount of air. The
licat produced by the combustion of the sulphuretted hydi-ogcn
368 PRODUCTION Of SULPHUR D10XID£.
h quite sufficient for keepiug the temperature of the chamber
at a red heat, so that the ga%is always lighted again if by chance
the flame has been extinguished. This is aided by the baffliug-
walls a, a, which serve both the purpose of supplying a reservoir
of heat for the just-mentioned purpose, and of mixing the gases so
as to insure perfect combustion. The grate e is required only
for irregularly composed gases, like those formed in the saturation
of the gas from ammonia-stills by sulphuric acid ; especially for
re-lightiug the gas after stoppages over Sundays, and so forth.
With gases of regular composition and comparatively rich in
sulphurettcd hydrogen, like those given off in Chance's sulphur-
recovering process, the grate e is quite unnecessary, as these gases
are as easily lighted and kept burning as coal-gas. The doors //
serve for " potting" the nitre, where it is not preferred to employ
more rationally constructed apparatus for this purpose (comp.
Chapter V.). The size of the whole chamber may be about 10 to
12 feet long, 4 or 5 feet wide, and 3 feet high. Pans for con-
centrating sulphuric acid may be placed upon it, and even in this
case the gases will issue hot enough to do full work in a Glover
tower. Sublimation of sulphur is never observed with ordinary
BURNING SULPHUBETTED HYDROGEN. 369
care in admittiug the air. One very great advantage lu burning
sulphuretted hydrogen is this : that, contrary to the variations in
the amount of SOg in burning brimstone or pyrites, even when
keeping up a regular rotation of the burners, there is in this case
a perfectly continuous process, as the supply of HgS from the
gas-holder is continuous ; the amount of air need never be varied
when once regulated; the percentage of SOg in the burner-gas
is altogether uniform ; the chamber process is consequently much
more regular than with brimstone or pyrites, and the consumption
of nitre is correspondingly smaller. All this, however, holds good
only if the percentage of HgS in the gas is practically constant,
whilst with gases of very varying composition, such as those evolved
in ammonia-works, the very contrary is the case.
It must be remarked that at some works, in burning the sul-
phuretted hydrogen from the Chance process, an increased con-
sumption of nitre has been noticed, whilst at others a saving in
nitre in comparison with the burning of pyrites has been effected.
Evidently in the former case the quality of the sulphuretted
hydrogen has not been as it ought to be ; it has no doubt varied
in percentage, and may even have contained a notable quantity of
carbon dioxide, so that the chamber-process would not be as regular
as desirable. Sometimes it has been noticed that the combustion
has not been quite perfect, so that sublimed sulphur has been found
in the Glover tower or even in the chambers ; but this is evidently
owing to mistakes and careless work, and should not occur with
ordinary care.
In 1886 E. Lombard (Monit. Scient. 1889, p. 1281) described
a shelf-bumei* for sulphuretted hydrogen, consisting of two separate
compartments, 7 ft. 6 in. deep inside. There are four shelves,
6 ft. 6 in. long and 1 ft. 4 in. wide, formed of four fire-clay slabs
each. The top shelf is perforated with many holes, and occupies
the whole length and width of the furnace ; the other shelves are
not perforated, and leave at alternate ends a passage of 12 X 16 in.
for the gases. Each furnace is provided at the base with four
burners for HgS and two air-tubes, disposed in two tiers of three
each, the air-tubes occupying the central places. The burners
consist of fire-clay tubes, 6 ft. 6 in. long, 1^ in. wide inside, and
I in. thick, projecting two-thirds of their length into the furnace,
and provided on the top with slits or holes for dividing the gas.
They are connected in front by a cast-iron tube with a stop-cock for
VOL. I. 2 B
370 PRODUCTION OF SULPHUR DIOXIDE.
regulating the flow. The air-tube is 2^ in. wide, and is provided
with an iron thimble for regulating the quantity entering. The
gaseous products of combustion pass into a flue, 1 ft. 6 in. x 2 ft.,
on the top of the furnace, then into a small dust-chamber, and then
into the Glover tower. Total height 6 feet. The pressure of the
gas is =1^ inches of .water; it is said to work very well.
Simpson and Parnell (E. P. 14,711, 1886) regulate the supply
of air and gas in any desired proportion, so as to obtain either free
S or SO2, by employing two vessels, each of which is provided with
an inlet and an outlet valve. Both vessels are filled and emptied
simultaneously by a mechanical arrangement, the two vessels acting
in conjunction, so that the gas entering and leaving one of the
vessels beai*s a constant proportion to the quantity of air entering
and leaving the other vessel. [Such an arrangement, very useful
as it undoubtedly is for the production of free sulphur from HjS,
seems unnecessarily complicated when the object is that of burning
the H2S with an excess of air for the purpose of sulphuric-acid
manufacture.]
7. Processes for Absorbing Sulphur Dioxide contained in
Acid-smoke, Fire-gases, and the like.
The abatement of the nuisance caused by the acid-smoke given
ofl^in metallurgical and other operations presents special difficulties
where the percentage of acids is so slight that their utilization by
condensation or by conversion into sulphuric acid is out of the
question ; that is, if less than 4 per cent. SO2 by volume is present.
The damage done by such acid gases has been . described in
Chap. III. p. 154 et seq.
A survey of the processes tried a number of years ago at
Stolberg for dealing with acid-smoke has been given by Hasen-
clever (Fischer's Jahresb. 1881, p. 173). All of these will be
mentioned below ; they all have one common feature : they are
very expensive, and at the same time they hardly ever attain their
purpose completely.
The problem of dealing with the enormous quantity of sulphur
dioxide contained in ordinary coal-smoke has been hardlv ever
attacked in a serious manner, as the expense and inconvenience
of any imaginable measures for this purpose have hitherto appeared
to be quite unbearable ; and it does not seem as if this would be
diflferent in the near future. The only practicable remedy in this
case, as well as in some cases of metallurgical smoke, is to dilute
DEALING WITH SMOKE-GASES, 371
the gases with a large quantity of air, by erecting very tall
chimneys for carrying them up to a considerable height above the
surface of the earth. Such chimneys have been made up to 450
feet in height. In the case of hydrochloric acid they have entirely
failed in their object (comp. Vol. II.), but in that of sulphur dioxide
the dilution of air is more efficient. Freytag certainly estimates
(somewhat arbitrarily) that smoke is harmless only when it does
not contain beyond 0003 percent. SOj by volume; but as, for
instance, in lead-works the percentage of SO2 in the main flues,
where all the smoke and fire-gasses are mixed, rarely exceeds O'l
per cent., it is very likely that, if these gases are allowed to escape
only 200 feet or more above any vegetation, they get sufficiently
diluted with air in their descent to become harmless. This is
owing to the fact that sulphur dioxide diffuses pretty equally in
the air, whilst hydrochloric acid, sulphuric anhydride, acid salts,
&c., which form visible fumes^ generally reach the bottom in a
very little divided stream, and cannot therefore be made innocuous
by very tall chimney-shafts. In fact, this is the only explanation
why the scores of tons of sulphur dioxide daily belched forth in
certain localities by lead-works, zinc-works, glass-works, &c., have
not ere now destroyed all vegetable life around the works, which
is notoriously the case only in a few isolated instances. But as
such instances do occur, and as altogether the requirements of
sanitary authorities are constantly becoming stricter, the removal
of the acid-smoke to a high level by means of chimneys cannot be
pronounced a final solution of the difficulty, even where only SO2
is the acid concerned, all the more as in moist weather the acids
escaping from the very tallest chimneys are brought down to the
ground in a somewhat concentrated state.
A proposal has been made by Wislicenus and Isaachsen (G. P.
124,900) to dilute smoke-gases to such an extent that the amount
of acid contained in them becomes innocuous. This is to be
attained by building within the chimney-stack a second lower
stack, provided with pipes carried up and downwards. Air is
introduced into the inner stack and escapes partly at its top,
partly through the up and down pipes, thus intimately mixing it
^vith the chimney gases. [It is not likely that this process can
be carried out in many cases as prescribed above, but it is stated
in the 38th Alkali Report, p. 76^ that bringing down the per-
centage of acids in the exit gases from chemical works by means of
diluting them in the chimney with air is not contrary to the law.]
2b2
372 PRODUCTION OF SULPHUR DIOXIDE.
O. Schott (Dingl. Journ. ccxxi. p. 142) has proposed to utilize
the sulphur dioxide given off in making glass from sodium sulphate
for the manufacture of sulphuric acid. The gas is to be made
richer in sulphur by employing for the glass-mixture gypsum in
lieu of limestone. Sulphate of soda, gypsum, and coal are to be
mixed in proper proportions, and brought to a bright red heat in
muffle-furnaces or in elliptical glass pots, until the SO2 is driven
off. The fritted residue of sodium and calcium silicate is to be
powdered and used by glass-works; the gas is to be conducted
into lead chambers and worked for sulphuric acid. This process
seems entirely impracticable ; especially since such diluted gas
(mixed with a great deal of carbon dioxide) has not yet been
utilized.
Thirion (Fr. P. Feb. 28, 1874; Wagner's Jahresb. 1875. p. 391)
makes a similar proposal for heating sodium sulphate with coal
and silica, whereby a mixture of sulphur vapour, sulphur dioxide,
and carbon monoxide is cooled. The sodium silicate is to be
decomposed by CO3 or to be used as such. [As a proposal for
manufacturing sulphur or sulphuric acid this process is evidently
hopeless.]
We will now give a synopsis of the various methods for treating
ordinary acid-smoke, with special reference to the removal of SO2
and S.O3. For details we must refer to the sources quoted in this
work, to a special treatise by C. A. Hering, ' Die Verdichtung
des Hiittenrauches ^ (Stuttgart, 1888), and to one by Schnabel,
* Metallhiittenkunde,' ii. pp. 58 et seq.
Condensing by water seems to be the simplest and most obvious
process, looking at the great solubility of sulphurous and sulphuric
acid in water. But this process is in reality only practicable where
the percentage of acids is not too slight; dilute acid-smoke is not
sufficiently washed without employing a comparatively enormous
quantity of water ; and, surprising as it is, SO3 is even more diffi-
cult to condense in this way than SO2. It is quite certain that
condensation by water can be made to pay only where the gases
are sufficiently concentrated to convert them into sulphuric acid
in lead chambers ; it is therefore the interest of smelting-works,
&c., to conduct their processes in such a manner that the acids are
diluted with as little inert gases as possible. If the percentage of
SO2 reaches 4 per cent, by volume, they may be submitted to the
Scbroeder and Haenisch process {vide infra), or they may even be
converted into sulphuric acid, although this will hardly leave any
ABSORBING SO] BY WATER, SULPHURIC ACID^ OR CAUSTIC LIME. 373
profit at that percentage ; but it is enough to have removed the
nuisance. Where, however, there is less than 4 per cent, of SOg
in the gases, any utilization is out of the question; the thin acid
liquids obtained by washing the smoke must be run to waste (which
in most cases means a fresh nuisance, and is not permitted by the
authorities) ; nor is the SO2 and SO3 anything like completely
taken out of the gases j and the nuisance is at best only diminished,
but not remedied.
At all events the contact of the absorbing-water with the acid
gases must be made as intimate as possible. The condensiug-
apparatus used for hydrochloric acid, and described in Vol. II.,
act only for somewhat strong gases ; the weak gases which we are
here treating of require special means, such as paddle-wheels or
similar spray-producing apparatus, costly to work and to keep
in repair, and generally imperfect in their action. Haworth
(U.S. P. 268,793, of 1882) proposes to condense the SOj given off
in lead-smelting by water, boiling it out of the solution and taking
it into a lead chamber — an economically hopeless process.
In lieu of water, Freytag (G. P. 9969, 14,928, 15,546) and
Hasenclever (ti. P. 17,371) employ somewhat concentrated .ml-
phuric acidy in an ordinary coke-tower of large size. This agent
retains the SO3 much better than water, so that in some cases the
expense of working the process is paid by the sulphuric acid
gained. SO3 is also retained to some extent, but only if the gases
have been well cooled. The necessity of doing this and of
previously removing the flue-dust^ which is sometimes very
difficult to perform, is a great drawback to this, as well as to
all corresponding processes. In fact Freytag^s process has been
abandoned again ; at the best it could remove only a small portion
of the injurious constituents from the acid-smoke. (Schroeder
and Haeuisch, Chem. Ind. 1881^ p. 118.)
Absorbing the acids by caustic limey generally in the shape of a
cream of lime, is one of the oldest and, if properly carried out, still
one of the most efficient ways of removing the acid-smoke nuisance.
Where the quantity of acids is but slight, and the manufacture in
question is sufficiently profitable otherwise, this process is even
now applicable, and if properly applied it does remove practically
all the acids. The cream of lime should meet the gases in a finely
divided state, either by flowing down properly-constructed towers,
or, still better, by being converted into a spray by means of paddle-
wheels or the like (comp. Rayner and Crookes' patent, E. P. 2678,
874 FRODUCTION OF SULPHUR DIOXIDE.
]875). That this leads to the desired effect, even with the large
quantity of SOo emitted in roasting blende, has been proved by
working on the large scale in Upper Silesia (comp. Bernoulli,
Fischer^s Jahresb. 1880, p. 184). But unfortunately the expense
of this process, where large volumes of acid gases are concerned,
is very serious, more especially as nothing like the whole of the
lime can be utilized for absorption, and the attempt to utilize the
product as bisulphite of lime (Hasenclever, G. P. 10,710) has
failed (comp. Schroeder and Haenisch, Chem. Ind. 1884, p. 118).
According to Jensch (Fischer's Jahresb. 1889, p. 321) the
deposit forming in the milk-of-lime towers contains so much lime
that it can be used over again, and at least a product is obtained
containing 37'7 per cent, lime, 38*4 SOj, 2*8 SO3, 4*1 CO2, &c.,
which is very useful as an addition to animal manure for the
purpose of fixing the ammonia, in which respect it is equal to
gypsum.
It has been found by CI. Winkler and other observers that the
SO2 in smoke is much more injurious to vegetation if accompanied
by much steam, e. g, in smoke-gases from brick-kilns. Spitta
(G. P. 110,388) proposes to absorb SO2 and steam at the same
time by passing the gases up and down several flues, into which
slaked lime in the shape of dust is injected from the top. The
bisulphite of lime formed is removed from the bottom of the flues
by means of special doors.
Egestorffs Salzwerke (G. P. 70,396) describe a very efficient
apparatus, consisting of a series of chambers with inclined bottoms,
connected with collecting-tanks. The alkaline absorbing liquid
is pumped up over and over again and brought into contact with
the gases by means of a spray-producer.
Limestone is very much cheaper than caustic lime, and is almost
equally efficient if employed in the proper way; that is, if a very
large surface of limestone is exposed to the acid gases, and if this
surface is kept from being covered with a crust of sulphite by
being constantly washed with a stream of water. CI. Winkler has
constructed a special arrangement for this purpose (G. P. 7174),
which completely fulfilled its object at the Schneeberg ultramarine
works. It consists of three brick chambers filled with large pieces
of limestone, the roof being formed by plank covers perforated
with many holes, through which water is kept running on to the
limestone. The gases pass through these chambers successively
ABSOUBIXG SO3 BY LIMESTONE^ MAGNESIA^ ETC. 375
and in regular rotation. The absorption of SO2 is excellent, but as
each cwt. of sulphur requires 3 cwt. of limestone, it is still too
dear for most metallurgical purposes, especially as any utilization
of the sulphur is out of the question.
The limestone treatment is frequently employed at sulphate-of-
ammonia works for getting rid of the SO2 formed by combustion
of the HjS escaping from the saturators. This treatment is
frequently mentioned in the Alkali Inspectors' Reports, and in
the 36th Report (for 1899), pp. 25 & 26, Mr. R. Forbes Carpenter,
the Chief Inspector, states the following conditions as being
absolutely necessary for success : — Ist. There must be sufficient
draught at the furnace, and the suction at the condenser outlet
must be adequate to supply this at all times, to avoid sublimation
of sulphur. Such draught might be supplied by injecting air
under slight pressure in the furnaces. 2nd. The gases must be
completely cooled before and behind the furnace. If they enter the
limestone-tower above 38° C, much calcium sulphate is produced,
which forms a protective crust on the limestone. The hot gases
are to be cooled first by cast-iron pipes until some condensation
takes place^ when leaden pipes must be substituted for them.
3rd. The limestone-tower should be made of wood planks, tongued
and grooved, or of brick and cement, not of cast-iron. [The
latter material would surely not be dreamt of by a chemical
engineer in such a case, but possibly it has been used in gas-works !]
4th. The supply of water is best made in two forms, one constant^
the other intermittent (by flushes), especially in the case of lime-
stone, but with hard chalk the intermittent flush only may be used,
at not too long intervals. [Comp. the much better feeding methods
for Gay-Lussac and Glover towers described in Chap. VI.]
Precht (E. P. 3443, 1881) employs for absorbing SO2 from
gaseous mixtures either magnesium hydrate or aluminium hydrate,
especially the former. It is either spread upon trays moistened
with water, or is brought into contact with the gases (previously
cooled to 100°) in the state of a cream, in an apparatus provided
with a mechanical agitator, or in columns like those employed
for treating sulphuric acid by sulphuretted hydrogen (Chap. X.).
This produces a crystalline precipitate of magnesium sulphite,
besides a solution of magnesium sulphate. On heating the mag-
nesium sulphate to upwards of 200° the SO3 is split ofi^, and can
be condensed as such or converted into sulphuric acid, whilst
376 PRODUCTION OF SULPHUR DIOXIDE.
magnesia remains behind^ together with about 3 per cent,
magnesium sulphate. The latter is heated with coals^ and thereby
converted into MgO^ remaining behind^ and a mixture of SO2 and
CO2, which is utilized in vitriol-chambers [?] . M. Lyte (J. Soc.
Chem. Ind. 1882^ p. 165) gives a detailed description of this
process with diagrams. It has been tried at several places^ but has
evidently been found too little advantageous for most purposes.
Alumina is included in Precht's patent^ but is less efficient than
magnesia. Sometimes acid gases have been passed through layers
of clay-slate (schist)^ whereby sulphate of alumina has been
formed^ but this process is evidently only practicable under special
local circumstances.
Zinc carbonate or omde was proposed by Schnabel (Fischer's
Jahresb. 1882, p. 266), who had previously made manifold
attempts at the Lautenthal smelting- works for treating the acid-
smoke, all without any sufficient success. Ultimately a process
was adopted (G. P. 16,860), consisting in passing the gases over
basic zinc carbonate moistened with water. Zinc sulphate is
formed, which, on heating (preferably mixed with coal), yields
sulphur dioxide, to be converted into sulphuric acid in lead
chambers, and a porous residue, consisting of a mixture of zinc
oxide with basic zinc sulphate. SchnahePs apparatus is rather
complicated, and the result not very satisfactory ; the process is
very troublesome to carry on, and costs much more than the value
of the sulphuric acid obtained (a provisional protection, No. 5416,
1881, for this process was taken out in England by M. Lyte).
Fleitmann (G. P. 17,397) passes the sulphurous gases, together
with some air, through a kiln containing a mixture of ferric oxide
and coal. The latter, in burning, yields the necessary beat, and
at the same time reduces the Fe203 and SO2, so that FeS collects
at the bottom (the success of this process is more than doubtful).
Metallic iron, moistened by water, was employed by Winkler
(G. P. 14,425), but was not found practicable for dilute acid gases.
Thorp (E. P. 8862, 1889) again recommends towers filled with
scrap-iron, and kept moist with water or a solution of feiTOUS
sulphate, the temperature being maintained at from 49° to 71° C.
Metallic copper or zinc, in very finely divided form, was tried at
the Frankfort gold-parting works, but without sufficient success ;
but at the same works the following interesting process was
worked out.
TREATING 8O2 WITH COPPER OR WITH H28. 877
Rossler (Dingler's Journal^ ccxlii. p. 278; Fischer's Jahresb.
188]^ p. 184) showed that gaseous mixtures^ containing^ besides
a large excess of air, far too little SOg and SOg for being treated
in vitriol-chambei s and otherwise not treatable in any efficient
manner, can be completely deprived of both the above acids i)y
forciDg the gases, by means of a Korting's injector and a perforated
coil of pipes, underneath a column of water, holding some copper
in suspension or some cupric salt in solution. The cupric sulphate
acts as a carrier of the oxygen of the air upon the SO3, and large
quantities of sulphuric acid are formed in this way, so that this
process might even be employed for manufacturing sulphuric acid.
At Frankfort, however, it is carried out in this way, that the tank
into which the gases are passed is always supplied with precipitated
copper, from which by this process cupric sulphate is obtained
without any expense. Rossler has also applied this principle to
the treatment of ordinary acid-smoke (G. P. 22,850), by combining
a whole set of apparatus. This process is very adversely criticized
by Friese (Chem. Ind. 1895, p. 137), who has made a long series
of experiments with it, with totally negative result. No oxidation
of SO2 with air to SO3 by the catalytic action of CaS04 could be
proved. SO2 reduces a hot solution of CuSO^, with intermediary
formation of cupric sulphite, to metallic copper. A smooth and
easy oxidation of the cupric sulphite to sulphate does not take
place. Hence this process would be useless for the production of
cupric sulphate, and still more so for that of sulphuric acid.
A special class of processes utilizes the reaction between sulphur
dioxide and hydrogen sulphide, either both being in the state of
gases, or the latter being in the nascent state as evolved frt m
sulphides. The reaction in its simplest form is :
2HjS-l-SO2=2H30 + 3S;
but, apart from the fact that polythionic acids are formed by
secondary reactions, the above reaction is anything but complete
with very dilute gases. Details about it will be given in Vol. II f.
in the chapter treating of the recovery of sulphur from soda waste ;
in this place it may suffice to mention that a patent founded upon
the above reaction was taken out by Landsberg (G. P. 6364-) in
connection with the roasting of blende.
When sulphides are employed, the reactions are even more
complicated, but the absorption of SOj can be made more complete.
378 PRODUCTION OF SULPHUR DIOXIDE.
CI. Winkler (Fischer's Jahresb. 1880, p. 245 ; more details in
Chem. Ind. 1880, p. 126) describes a very interesting process
for dealing with the gases from an ultramarine-works containing
much SO^. They were brought into contact with a solution of
sodium sulphide, obtained from the sulphate going to waste in that
manufacture, by reducing it with coal. The SOj is completely
absorbed, with formation of sodium tbiosulphate, or, in another
modification, with formation of free sulphur ; but on the large scale
sodium tetrathionate was formed, which had to be decomposed by
heating into sodium sulphate, SOo* and free sulphur. Theoretically
nothing was consumed but coal, but evidently a very large amount
of fuel must have been used in the various evaporations and
furnace operations, with an amount of skilled labour out of propor-
tion to the value of the products obtained. After having been used
from 1868 to 1877, the process just described was abandoned for a
simple absorption by limestone moistened with water (p. 374).
Even before Winkler, in 1864, Jacob had carried out for some
years a similar process to that just described, employing either
sodium or calcium sulphide, at Miinsterbusch (Fischer's Jahresb.
1881, p. 181).
Calcium sulphide^ proposed many years ago by Dumas, forms
also the absorbing substance in Kosmann's process (6. P. 13,123).
By reducing calcium sulphate with coal and lixiviating a solution
of calcium sulphydrate is obtained [?], which in very finely
divided state is brought into contact with the gases containing
SOj. The result is the formation of sulphur and gypsum :
5S02 + 2CaH2S2 + 2H20=7S + 2CaSO^, 2HoO.
The sulphur is extracted from the mixture by superheated steam-
and the gypsum returns into the cycle of operations [it is very
doubtful whether this would succeed !]. From further com-
munications by Kosmann (Fischer's Jahresb. 1882, p. 270), it
appears that the absorbing medium was afterwards prepared
by boiling sulphur with milk of lime, that is, as the ordinary
'' liver of sulphur,'^ and that the whole process was entirely in
the experimental stage (from which it does not seem to have
emerged) .
Barium sulphide, which was experimentally tried at Freiberg,
proved much too costly.
Vegetable charcoal is proposed by A. H. Allen (B. P. 189, of
SULPHUR DIOXIDE IN THE PURE STATE. 379
1879), who passes the gases, freed from dust, through drying-
towers fed with sulphuric acid and then through columns filled
with charcoal, previously ignited in a stream of nitrogen, where
the SOg is retained, whilst the nitrogen passes on. By a vacuum
or by heating to 300°-400°, or by a combination of both, the SO2
is to be driven out and utilized. (This process, apart from the
prohibitory expense, is hardly practicable, because the gases in
question contain nearly always a large quantity of oxygen which
will convert the SO2 to a great extent into sulphuric acid within
the pores of the charcoal.)
A totally different way of employing coal is used in one of the
oldest processes for dealing with acid-smoke, namely, passing
the gases through red-hot coals, in order to reduce the SO^ to
sulphur. This has already been mentioned in Vivian's pamphlet,
'' Proceedings of the Subscribers to the fund for obviating the
inconvenience arising from the smoke produced by smelting copper
ores '' (London, 1833) , and in a pamphlet of Reich's, describhig the
experiments made at Frankfort in 1858, and it has been proposed
over and over again, with the same negative results. A new
apparatus, by Schroeder and Haenisch (G. P. 33,100), is said to
give good results, nearly the whole of the SO2 being reduced to S ;
but their process evidently works only with rich gases, and does
not deal with those poor gases which concern us here.
Bemelmans (Germ. pat. 77,335) converts the SO2 by reduction
with carbon and hydrogen into HgS, dries this by the process to
be described below, mixes it with dry SOj, and converts them into
H2O and SO2.
8. Preparation of Sulphur Dioxide in the Pure State ♦.
Formerly pure SO2, free from nitrogen and excess of oxygen,
was required only in very few cases for industrial purposes. The
methods employed for preparing that gas were various, one of the
commonest being the action of concentrated sulphuric acid upon
copper at a higher temperature. This is, of course, only appli-
cable where there is a sale for the cupric sulphate formed, and is,
moreover, hardly workable on a large scale. Cheaper and easier
is the process of heating strong sulphuric acid with charcoal, when
* A special treatise on the preparation, properties, and application of pure
sulphur dioxide is * Tliisaiges Schwefeldioxyd/ by A. Harpf (Stuttgart, 1900).
380 PRODUCTION OF SULPHUR DIOXIDE.
a mixture of SOj with CO2 (and CO) is obtained : —
2 SO4H, + C = 2 H2O + 2 SO2 + CO2.
The CO and CO2 are harmless in many applications of SO^. Sul-
phur dioxide, quite free from other gases^ is made by heating
concentrated sulphuric acid with sulphur: —
2S04H2 + S = 2H20 + 3S02;
this can be done by running a slow stream of sulphuric acid on
sulphur, heated to about 400° in an iron retort. It should, how-
ever, be noticed that melted sulphur acts very strongly upon cast-
iron ; hence another process, privately communicated to me frona
a trustworthy source, would seem preferable. The operation is
performed in a cast-iron pan, widening out at the top so that a
lining of acid-resisting bricks can be put in it. Concentrated
sulphuric acid is boiled with sulphur, which floats on the top and
is kept by the brick lining from coming into immediate contact
with the iron, whilst the lower part of the pan is fully exposed ta
the heating action of the fire.
Sulphur dioxide was made by the Compagnie industrielle des
precedes Raoul Pictet (G. P. No. 22,365), and was purified in a
special apparatus, utilizing the fact that the hydrates of SOj all
crystallize at —10^, and that gaseous S02atthis temperature loses
all its aqueous vapour. We refrain from describing this (some-
what complicated) apparatus, which is also described in the
Journal of the Society of Chemical Industry, 1883, p. 413, as
the condensation of liquid SO2 is performed in a much simpler
way by the Schroeder and Haenisch process, which in fact has
caused the above-mentioned process to be abandoned.
P. Hart (E. P. 13,950, 1885) prepares pure sulphur dioxide by
acting with strong sulphuric acid, of spec. grav. r750, on finely
ground iron sulphide, both being mixed in a cast-iron retort and
heated to over 200° C, when a steady stream of nearly pure SO,
is evolved.
An old and well-known process for obtaining pure SOgis: heat-
ing ferrous sulphate with sulphur, with a little air, the reaction
being :
2FeS04 + 2S+3 0 = Fe203 + 4S02.
This process was made the subject of a new patent by Terrell
(B. P. 5930, of 1884), who evidently lays the greatest stress
SULPHUR DIOXIDE IN THE PURE STATE. 381
on the ferric oxide remaining behind^ which famishes a good
paint.
Ford's process (Am. Tat. 363,457; Chem. Zeit. 1887, p. 721)
consists mainly in burning sulphur by means of air previously
dried with sulphuric acid^ and passing the gases through a worm,
where, by cooling and pressure, liquid SO2 is condensed. It is
difficult to see any novelty whatever in this process.
The only process for preparing pure liquid sulphur dioxide on
the large scale is at present that of Schroeder and Haenisch,
which allows of preparing that substance in a cheap way from
gases containing down to 4 per cent. SO3. It is unnecessary to
say that richer gases are better for this purpose. This process
has made liquid sulphur dioxide a cheap article, manufactured on
a large scale, and has rendered it possible to employ that substance
for many purposes for which formerly only the ordinary impure
gaseous SO3 was available.
The process of Schroeder and Haenisch, embodied in B.P. 2621,
1883; 6404 & 6405, 1885, was first taken up by the firm Wilhelm
Grillo (afterwards converted into the Aktiengesellschaft fiir Zink-
Industrie, vormals Wilhelm Grillo) at their zinc- works at Hamborn,
Rhenish Prussia, where, in 1885, an experimental factory was erected,
turning out about 12 cwt. liquid sulphur-dioxide per diem. The
gases, testing 6 per cent. SO,, were taken from a novel kind of
blende-roasting furnace, similar to that described on p. 361 (system
Julius Grillo, G. P. 28,458). In 1886 four such furnaces were
combined with a larger plant for 8 tons SO3 per diem. Very soon
after similar factories were erected at Lipine and at Chropaczow,
in Upper Silesia, and about 1899 another at Bound Brook, N.J.,
U.S.A. (The factory at Chropaczow has since been converted into
an ordinary sulphuric-acid works.)
The process consists in absorbing water in an ordinary coke-
tower, and expelling it again from the resulting weak solution by
the action of heat, in such manner that the latent heat of the steaui
carried along is fully utilized, and ultimately a very small amount
of coal is required. It is described by the inventors, apart from
the patent specifications, in ' Chemische Industrie,' 1881?, p. 120;
but this description is now antiquated, and we shall here go by later
descriptions (Paper Trade Journal, 1888 ; Zsch. f. angew. Ch.
1888, p. 488), by personal observations of the work as practically
carried out, and by notes received from the inventors in 1902.
382 PRODUCTION OF SULPHUR DIOXIDE.
The burner-gases arrive in the flue a a (fig. 122), and after having
lost part or most of their heat by passing underneath the lead
pans e e, they pass into the coke-tower b, where they are treated
with such a quantity of cold water that all the SO2 is condensed
(the exit gases contain only 0*05 per cent. SO, per volume), and
only O and N pass out at c. The solution of sulphurous acid,
containing about 10 kils. SOg per 1000 kils. of liquid, is run
out by pipe d, and first passes through an apparatus, shown sepa-
rately in fig. 123, where it receives a previous heating, and then
successively through the closed lead pans e e, where the heating is
continued by the action of the hot burner-gases acting in the
flue a a. The apparatus, fig. 123, serves for heating up the cold
acid solution by the heat of the spent liquor, resulting at a later
stage of the process. It is built up by superposing a number of
sheets of lead, 7 lbs. per superficial foot, of considerable surface,
corresponding to the quantity of acid liquor to be treated. These
sheets are combined in such a way as to form a corresponding
number of shallow lead chambers, about 1^ inches deep, superposed
Fig. 123.
■! J I I I I III IIIIUJIIIIII X^ILg«X 1 I I.I T T-r-m d'
[Li n JE i ill n i M I i ii i ii i i i bCc? * i rr
<:> «» 4» «» «i « « * H» «» «ii
j"{rTT IirrtlTIIIlItliliMi'B fWi j t t t j
d^., TTTt I m 1 1 f 1 1 1 1 I M n'f T ' '«'> I I i t T ¥ T
one over another and connected with each other in the following
way : — The acid liquor flows through d into the bottom chamber
from left to right; through a side connection rf', occupying the
whole length of the lateral edge, it is conveyed into the third
chamber, from here through Sf' into chamber 5, and thus further
cools the 7th chamber, 9th, and so forth, issuing at 8'. The
chambers Nos. 2, 4, 6, 8, and so forth, serve in the same way for
running down the hot spent liquor obtained at a later stage of the
process. This liquor, which enters at q, always flows in a direction
at right angles to that of the acid liquor rising up in the alternate
chambers, so that the connections for the spent liquor flowing down
are situate in the front and back part of apparatus fig. 123. In
SCHROEDER AND HAENISCH^S PROCESS. 38S
order to prevent a sagging of the platen, strips of lead are arranged
in each chamber as stays, running in each chamber in the direction
of the current of liquor. The thin sheets of lead being good con-
ductors of heat, the cold acid liquor, on rising through chambers 1,
3, 5, 7, &c., is gradually heated up, whilst the hot spent liquor,
descending through the intermediate chambers, gives off its heat.
Of course there must always be a certain difference and loss of
heat, depending upon the duration of contact, the depth of the
liquid, and the speed of the current. With chambers of 1^ or
2 inches depth, and counter currents lasting 10 or 12 minutes,
and sufficiently large surfaces, the difference of temperature will
be about 10° ; that is, the cold acid liquor will be heated up from
15^ to 85° C, whilst the hot spent liquor goes down from 95° to
25° C. In practice the heating up rises to 95°, and the cooling
liquid descends to 50°. The cold spent liquor is run to waste
through X,
Quite recently (1902) the flat lead chambers, where leaks are
not easily discovered and repaired^ are replaced by a set of small
lead cylinders working on the system of gradual heat-exchange
(counter- current) .
The heated-up acid liquor now travels successively through the
covered lead pans e e, where the heating is continued as mentioned
before, so that the boiling-point is attained. The gases and vapours
here evolved are conducted through pipe / into the water-cooled
worm g, and from here through the pipe h into the tower i, where
the last remaining admixture of moisture is taken out by dry
calcium chloride or (preferably) by coke moistened with strong
sulphuric acid. From here the dry sulphur dioxide passes through
pipe k into the pump /. The liquor heated to boiling in the pans
e e, which still contains some SO^, passes through pipe m into the
column n, where the steam is to a great extent condensed by
injection of cold water, whilst nearly dry SOg passes up in p, and
thus equally gets into the worm ff and further on into the pump /.
The column n is shown in some detail, as described in a further
patent by Schroeder and Haenisch (G.P. 36,721), which refers to
the separation of steam from its mixture with SO^, and as this is
a matter of general importance, we shall give their statements at
some length. It is not easy to separate large quantities of aqueous
vapour from a mixture with gaseous SO^. Indirect cooling by
outward application of cold water requires a very large leaden
384 PRODUCTION OF SULPHUR DIOXIDE.
apparatus^ and the effect is but partial^ as the vapours pass without
hindrance through the central parts of the worms or other kind of
apparatus. Moreover the condensed water^ unless the temperature
of the cooling-apparatus is kept nearly at a boiling heat^ carried
down very much SO^,. The new process effects the removal of
the steam from the aforesaid gaseous mixture by direct injection
of water, which certainly at first condenses a good deal of SO^.
But if the acid solution thus formed is made in a systematic
manner to meet the hot mixture of aqueous vapour and SO^^
its temperature will be gradually raised and will ultimately
attain boiling heat, and pari passu its percentage of SO^ will
decrease, so that at 100° it is nearly at zero. The following
table shows the diminution of the percentage of SO9 with the
rise of temperature : —
Percentage of a saturated solution of SOj : —
At 20° C. = 8-6 per cent. SO^.
30° C. = 7 4 „ „ „
40° C. = 6-1 „ „ ,.
50° C. = 4-9 „ „ „
60° C. = 3-7 ,, „ „
70° C. = 2-6
if » >}
, 80° C. = \-7
, 90° C. = 0-9
, 100° C. = 0-1
>^ ii y*
yy yy
yi y* jy
If the injection of water is so adjusted that the liquid running
off is at a temperature of 95° or 100° C, the latter cannot,
as shown by the preceding table, carry away any considerable
quantity of SOo. On the other hand, if the course traversed is
long enough, the steam must be completely condensed by the
eold water injected.
This process is carried out in the apparatus shown in fig. 124,
viz., a leaden column, filled in the lower part with stoneware
diaphragms, in the upper part with coke. The mixture of steam
and SOg enters through pipe a and rises in the tower. Cold
water is injected by the rose h, condensing both water and SO^,
and flowing down as an aqueous solution of sulphurous acid. On
reaching the lower parts, it meets continually fresh quantities
of hot gases and vapoui-s, and arrives at the bottom 100° C. hot.
It there yields up again the SOg absorbed higher up, and at the
bottom pipe c carries off botl
the injected water and thai
coudeiised from the steam
After some time of working,
and with proper regulation of
the feed, the temperature ol
the water from the top down-
wards rises regularly to a
boiling heat, and its per-
centage of SO3 diminishes at
the same ratio; but the quan-
tity of SOj in the upper region
is so considerable that the
injected water cannot retain
it all, and pipe d carries away
a continuous stream of gaseous
SO, deprived of steam. The
dish shape of the stDueware
parts iu the lower half of the
tower has the advantage of
retaining the deacendiug liquor
for some time, and exposing
it to the heat of the rising
steam; but instead of this,
coke may be used all over,
if the tower is made high
enough.
Returning to fig. 122, we see
that the water condensed in
the worm g finds its way
equally into column n, and
is deprived of its SO^ there.
The hot spent liquor runs off
at the bottom by pipe q, and
is utilizedj as explained before,
for heating up the cold acid
liquor in the apparatus, fig. 123,
where the entrance to the pipe
g is visible. In order to regu-
late the compression of the
SCHHOEDEK AND HAEXISCH's FROCESB.
386 FRODUCTIOH OF SCLPHDK DIOXIDE.
gaseous sulphur dioxide to a liquid, a taffeta bag (r) is interposed in
pipe k, and the motion of pump I is r^ulated according to the siie
of this bag. The compressed gas enters throagh » into the worm t,
and ia liquefied there by cooling and pressure, the latter depending
upon the temperature of the cooling-water, e, g., 1-26 atm. at 10°,
2-24 atm. at 20°, 3*51 atm. at 30°, &c. From t the liquid runs
into the wrought-iron reservoir u, Jrom which it is drawn off into
the iron bottles v or into tank-waggons. In order to get rid of
the carried-along portions of oxygen and nitrogen, the boiler « is
provided with an outlet-pipe w, connected with u by a valve ; the
gases from here are conducted back into the absorbing-tower b.
Haenisch has also described an improved column for boiling
the solution of SO, (O. P. 52,025). In 1899 a foctory in Silesia
produced 1266 tons liquid SO, by the above process from zinc-
blende gases; the production of the other works is not known.
The cost price is supposed to be £2 per ton (on somewhat arbitrary
data) ; the selling price ia £5 to £5 10«. at the works.
Fig. 125.
The liquid sulphur dioxide ia sent out in iron cylinders (bottles)
holding 1 or 2 cwt. each, or in tank-waggons of 10 tons capacity.
The former ave shown in figs. 125 aud 126, When sent out, the
outlet valve is protected by a cap a. Before use this ia removed,
as well as the small cap fixed on the neck b. If now the plug of
the screw-valve c is. turned by means of a key, the sulphur dioxide
SENDING OUT LIQUID SULPHUR DIOXIDE. 387
escapes in gaseous form through the opening in b. The stuffing-
box or the whole valve must not be removed. After a time the
evaporation of SO^ lowers the temperature to — lO*' C, so that no
more gas is given off till the apparatus has taken up heat from
without.
If the sulphur dioxide is to be got out in the liquid form^ the
vessel is put on its side (fig. 126) in such a position that neck b is
at the top side. The pressure of its vapour now forces the SO^
out of b. The bent tube within the vessel admits of emptying it
entirely of liquid SO2. This substance can be conveyed away by
a lead pipe, screwed on to b, or even by an india-rubber pipe. The
bottles are tested to 50 atmospheres pressure, so that there is no
danger whatever in their transit, as the vapour-tension of SO,
amounts to : —
0 atmosphere overpressure at — 10° C,
0-53 „ „ „ 0°
1-26 „ „ „ +10°
2*24 atmospheres „ „ 20°
3-51 ,, „ „ 30°
515 ,, „ „ 40°
Still, it is advisable not to keep the liquid in a place whose tem-
perature may rise upwards of 40" C.
To prevent over-filling, in view of the expansion by heat, the
bottles should not be filled more than nine-tenths, which is ascer-
tained both by weighing them and by prolonging the outlet-tube
inside to one-tenth of the depth of the vessel ; when opening the
valve only gas, no liquid, should escape. These bottles are usually
made to hold 100 kils., exceptionally 500 kils. Besides, tank-
waggons are used for 10 tons liquid sulphur dioxide, contained in
three wrought-iron welded cylinders, about 23 feet long and 2 feet
3 inches diameter, tested for 30 atmospheres.
Boake and Roberts (Engl. pat. 19,789, 1892) find that liquid SO^
does not act on tin or soft solder, and that therefore these can be
employed in the manufacture of carrying-vessels.
Soy (Engl. pat. 12,276, 1893) patents the process of sending out
liquid SO3 in glass vessels sealed by the blowpipe, to be broken by
a hammer in the rooms to be disinfected.
The principal uses for liquid sulphur dioxide are for refrigerating-
machines (Pictet's and others), for wood-pulp manufacture (to
2c2
388 PAODUCTION OF SULPHUR DIOXIDB.
bring the calcium bisulphite liquors up to strength), for the
purification of beet-root juice, for disinfecting^ for bleaching, for
the manufacture of glue and gelatine.
Recently liquid sulphur dioxide has. been applied by Behrend
and Zimmermann as a means for increasing the efficiency of steam*
engines by utilizing the heat of the exhaust steam for evaporating
SOj. The high-pressure vapours thus produced are utilized in an
auxiliary cylinder for generating motive power, and are afterwards
again condensed to liquid SO^. Hitherto this system does not
seem to have fulfilled its expectations.
The formerly rather extensive use for bringing calcium bisulphite
liquor (for the manufacture of wood-pulp) up to strength has very
much decreased, since the factories have improved their plant for
the direct preparation of strong sulphite liquor.
The employment of the strong SO3 obtained by the Schroeder
and Haenisch process for manufacturing sulphuric anhydride
(fuming O.V.) will be mentioned in the 11th Chapter.
The analysis of liquid sulphur dioxide is described in my
' Chemisch-technische Untersuchungsmethoden/ vol. i. p. 269.
9. Draught-pipes and Flubs.
The fines leading the burner- gases from the kilns into the
chambers or into the Glover towers may be constructed of brick-
work only so far as the gases keep hot enough not to allow any
moisture whatever to condense, that is especially in upright flues
and flue-dust chambers. From this point they must be made of
cast-iron, and further on, when they have got cooler, of lead.
The gas, which goes away red-hot from the burners, must
necessarily be cooled down to the temperature of the chambers^
say 60° or 80° C. ; otherwise the first chamber would be very
quickly destroyed. This cooling was formerly effected by con-
veying the burner-gas in very long fiues of cast-iron, or partly of
cast-iron, and, when partially cooled, of lead. Such cooling-flues
were made up to 300 feet long.
The cast-iron pipes are suitably shaped, as shown in fig. 127,
in order that the upper half may be replaced independently of the
lower, or taken away for cleaning; the latter can also be done by
means of man-holes. For a set of from 12 to 18 burners a pipe
of 2 feet diameter is sufficient ; but they are now and then made
DRADOHT-PIPES AXD FLUES.
Upwards ot 3 feet in diameter. Sometimes they are lined with
fire-bricks, as shown in fig. 128; the cooling in this case is very
imperfect and the cost higher. Occasionally, in very large works.
square or oblong fluca of wrought- or cast-iron are found. Brick
flues (for perpendicular shafts or for ffne-dust chambers) are made
of bricks boiled iu tar, and set with tar and snnd. Earthenware
pipes mostly crack too quickly.
Perpendicular stacks and pipes act as a sort of chimney, and
they are therefore carried up nearly to the top of the chambers
where there is no Glover tower. These pipes were sometimes
outwardly cooled by water, and even very complicated contrivances
were met with for this purpose. It has long been recognized that
the only rational way of cooling the burner-gas is this: taking
away its heat for some useful purpose, and this is almost everywhere
done in the Glover tower (Chap, VI.). Apart from this, the heat
of the bumer-gas is occasionally utilized for concentrating acid
in lead pans (Chaps. IV. & IX.^ Sometimes other ways of dis-
posing of this heat are employed. Thus at the manure-works of
Messrs. H. & E. Albeit, at Biebrich-on-tbe-Rhine (see fig. 129),
each pyrites-kiln A is surmounted by a gas-chamber B ; the burner-
gases enter by the holes a, provided with shut-off valves, and
through similar openings {6) get into the pipe d, leading to the
Glover tower C ; a third opening (c) permits sending the gas
direct into d and C. In each of the chambers B there are hori-
zontal cast-iron pipes [e), branching off from a main pipe/, into
which air is forced by a fan-blast. This air, being exposed to the
hot burner-gases, becomes hot itself and leaves the chamhers in
order to be carried away by the main pipe h. This heated air is
then conveyed to drying-stoves, where the superphosphate is dried
390 PRODUCTION OF SULPHUR DIOXIDE.
by its action. At Albert's works the temperature of the air is
brought to 100* C, but it may be kept cooler or hotter (up to 135°)
by regulating the speed of the fan-blast. The burner-gas, entering
through d into the Glover tower C, is still hot enough to do all the
concentrating and denitrating work required ; the Glover towers
Vig. 120.
are not so overheated and require less repairs than formerly; the
acid flows from them only at 115°-120°, against 140° as before
the new arrangement, and requires leas cooling for the Gay-Lussac
towers. The saving in coals for the drying-stoves is 5 tons ppr
diem (comp. the description, with further diagrams, in Zeitschr. f.
angew. Chemie, 1889, p. 287; also E. P. of Albert, FeUner and
Ziegler, No. 15,980, of 18H8}.
Flue-dusi.
In the gas-flues and draught-pipes ^ue-tfusf is always deposited,
much more when smalls are burnt than with lump ore, especially
in furnaces where the small ore is moved about. In such oases
special dust -chambers are indispensable, as haa been remarked iu
the description of those furnaces. Even with large lumps the
flues and pipes must be cleaned out from time to time, as they
would otherwise be stopped op entirely. At some factories this is
done monthly, at others more rarely. If the deposit is allowed to
remain too long, it hardens into a stone-like mass, which cannot be
got out without stopping tbe process.
The composition of this deposit varies, of course, very much;
and even its external aspect varies from that of a dry, light dust,
to that of a thick, strongly acid mud. Clapham analysed such a
7LUE-DUST. 391
deposit^ from a source not mentioned (Richardson and Watts^
Chem. Technol. i. 8, p. 70), and found : —
Sand, &c 2-833
Lead oxide 1*683
Ferric oxide 3-700
Cupric oxide trace
Zinc oxide trace
Arsenious acid 58'777
Sulphuric acid 25-266
Nitric acid trace.
Water 8000
99-759
D. Playfair (Chem. News, xxxix. p. 245) has examined flue-dust
from pyrites-kilns, in which he found chiefly arsenic, antimony,
lead^ copper, and iron ; of thallium 0*002 to 0*05, of tellurium and
selenium 0*001 per cent, was present. He describes in detail the
analytical methods employed.
Reich (^Erdmann's Journal/ xc. p. 176) found in the Mulden
Works a crystallized deposit consisting of equal molecules of
arsenious and sulphuric anhydride. Similar deposits have been
frequently observed since.
In other cases the deposit is dry dust, mostly consisting of
mechanically conveyed pyrites-dust, better burnt than that within
the burner itself (Bode, 'Beitrage,' p. 41), and nearly always con-
taining so much arsenic than its crystals can be seen with the
naked eye.
H. A. Smith (' Chemistry of Sulphuric- Acid-making^) found in
the dust 46*36 per cent, of As^Og, along with a large quantity of
sulphur in the pasty condition — the latter, of course, formed by
sublimation from pyrites.
The flue-dust is also a principal source of thallium, as we shall
see; and when selenium occurs in the pyrites it is found in the
flue-dust.
The flue-dust from the roasting of blende is, of course, quite
differently composed from that formed in burning pyrites. Such
flue-dust contains (Fischer's Jahresb. 1882, p. 273) : —
392 PRODUCTION. OF flULPHUR DIOXIDE,
}^«-2^ 12'?S}'^''«
I. IL
Zinc oxide insoluble 8-401 o/j.on 8*^
Ditto in soluble combination 17*80
Ferrous oxide, soluble 2' 16 . 2-52
Ferric oxide, ditto 2-40 4-20
Lead oxide 3-38 4-26
Sulphuric acid, insoluble 6*46 1 24'89 ^'^"^"l 26*88
Ditto soluble 20-43 J 18-84 J
Water 6-59 9-00
Residue (chiefly ferric oxide) 31-80 32-42
99-42 99-48
Flue-dust from the St. Mardy Tinto Santarossa pyrites (com p.
Lunge and Banziger, supra p. 55) contained sulphur: sulphur
free 0-13 per cent. ; ditto as sulphide 1-48; SO3 as free sulphuric
acid and sulphates 16-31; AsA 6907; SbgOa 1*68; CuO O'U;
FcaOs 2-03; sand 265 ; water, traces of other substances and loss
6-51.
Bellingrodt (Chem. Zeit. 1886, p. 1039) has found in the flue-
dust from roasting blende at Oberhausen (Rhenish Prussia) a
sufficient quantity of mercury to make its recovery profitable.
Krause (G. P. 55,676) washes the flue-dust from blende-furnaces
and tries to recover the zinc from the washings by precipitation
with alkaline carbonates.
Where the quantity of flue-dust is very large, as is generally
the case with arsenical ores, and with some of the burners for
pyrites-smalls, the ordinary dust-chambers, which form simply
enlargements of the gas-flue, are not sufficient, and special contri-
vances must be adopted here. This matter has been thoroughly
worked out in the lead-smelting works and other metallurgical
establishments, and a large number of apparatus has been con-
structed for this purpose. A very complete synopsis of this is given
in the pamphlet by C. A. Hering : ' Die Verdichtung des Hiitten-
rauchs ' (Stuttgart, 1888) , pp. 8 to 36. Many of the contrivances
employed at lead- works, &c., are unsuitable for pyrites on account
of being made of iron. But the general principles remain the
same: the flue-dust must be made to deposit by cooling, by
retarding the speed of the gaseous current and by offering to it
large surfaces to which it can attach itself. All these conditions
are more or less fulfilled by making the gas-flues adequately long
FLUE-DUST CHAMBERS. OUA
and wide, but this ib not aiifficient for " bad cases," especially
for arsenical ores. The case is here complicated by the fact that
the cooling of the gas may be injurious to the chamber process,
and that the long flues, especially those carried in ii zigzag way
or provided wit!i "baffle-wails," interfere very seriously with the
draught. The latter disadvantage has been greatly lessened since
it has been recognized that it is unnecessary to carry the gases in
Vig. l:».
flues like those skeV^hed in figs. 130 and 131 (in the former the
diagram may be taken either as plan or elevation), where the
current of gas is constantly checked by meeting solid surfaces,
but that the snrfaces may be disposed in the direction of the
current itself, where they cause the dust to be deposited on
tfaem without interfering with the draught. Fig. 132 shows how
this can be done in such a way that the flue-dust can be removed
without interfering with the process. The gases arriving through
a are, by means of dampers, sent either through chamber A or B.
In the present case, the dampers i b being closed, the gases travel
through A. Each chamber is divided into several longitudinal
394 PRODVCTION OF SULPHTTB DIOXIDE.
channels by tbin partitions d d, made of masony, fireclay slabs,
lead, or other suitable material. The gas thus travels id parallel
streams, without any check than t)iat of the indispensable friction,
and the streams collect again into one, issuing at e. Wheo
chamber A is too much choked up by dust, the dampers b b are
opened, the dampers c c are shut, and the ga^es now travel through
B, giving an opportunity to clean out chamber A by means of
suitable man-Iiulcs.
rig. 132.
In very bad cases, as, for instance, with mechanical dust-bumers,
longitudinal partitions are not sufficieHt, and real baffle-platea
must be employed, as shown in figs. 130 and 131, and very much
multiplied in figs. 100 and 101 (p. 347). In such cases the loss of
draught must be sometimes compensated by mechanical means.
In many cases, nhcre very large quantities of flue-dust have to
be dealt with, the gases must be cooled artificially. This was
done at the Freiberg works by a special kind of lead flue, cooled
by water, as sketched in figs. 133 and 134, where the first represents
a longitudinal section on the line A B, the second a sectional
plan uu the line C D. The sides of the flue are formed by a
number of oblong pipes, a a, joined together at their narrow
ends. On the top there is a shallow trough, h, supplied with a
constimt stream of water, which trickles down through holes, shown
in the diagram, into the space a a, and from these through other
holes into the common channel c, running lengthways. The
bottom of the flue is not water-cooled, but as it rests on the small
pillars d d '\i \& exposed to the action of the air. These flues are
very expensive to build, hut they have been found to do their
work lery well indeed, and they need next to no repairs (detailed
in the 'Freiberger Jahrbuch/ 1879, p. 151, table xii.). Of the
FLUE-DUST CHAUBEH8. 305
2 or 2| per cent, of arsenic contained in the Freiberg ores, by far
the greatest portion (97 per cent.) is condensed in these long flues,
where the gases are nltbnately cooled down to the temperature of
the outer air.
Kg. 133.
Bauer {Jalirb. f. Berg- u. Huttenw. 189i, p. 39) states that
the nine sets of chambers connected with the Freiberg smelting-
works (containing <{0 lead chambers) possess 8037 metres of flues,
of 3'8 square metres section. The flue-dust in 1893 contained
1137 kils. silver, 1656 tons lead, and 917 tons arsenic, valued at
j613,600. Eighty per cent, of the dust was recovered, 20 per cent,
has hitherto been lost. The damages to be paid had diminished
from £3050 to £180. In that year new flues on the Monier-
Freudenberg system were built for recovering the last 20 per cent,
of (lust. The rate of cooling of the gases was 1° C. per 8'3 metres
length In closed-in Monier flues, per 4*5 metres in freely exposed
Monier flues, per 3 metres in leaden flues, and per 6 metres in
brick flues.
In 1902 I am informed that the water-covered flues have been
abandoned at Freiberg as being too costly, and have l>een replaced
by simple lead tunnels. Where the beat is too great for the stability
of the latter, brick tines are employed.
396 PRODUCTION OF SULPHUR DIOXIDE.
Falding (Min. Ind. ix. p.623) describes adust-catcher, constructed
by A. P. O'Brien, of Richmond, Va., on the well-known centri-
fugal principle. It works in connection witli the cast-iron fan of
the same inventor, described in Cliap.Vl ., and receives the gas from
five-Herreshoff fines furnaces, retaining 75 per cent, of the dust.
At the same time it does very efficient service as a metre-oven. As
shown in figM. 1 35 & 136, it consists of a tapering, hopper-shaped
Fig. 135.
iron shell, 8 feet wide in the cylindrical part and 12 feet high, with
a 6-in. opening in the bottom for the discharge of flue-dust. It is
lined with 4 inches fire-brick. The gas enters through the top pipe
at a high rate of speed from the fan and strikes the cylinder tan*
gentially ; it leaves the apparatus through a central pipe. The gas
takes a rotary motion and deposits all the heavy dust, which is
automatically discharged through the 6-in. opening. Six tubular
metre-pots are arranged iu the manner shown, so that they can be
charged from the top and discharged sideways.
Sulphur Dioxide for the Manufacture of Wood-pulp ought to be
as free as possible not merely from flue-dust but also from sulphuric
acid (or sulphur trioxide, comp. next Chapter). For this purpose
BURNER-GAS. 397
very efficient diist-chambers must be provided ; the gas must also
be specially cooled, e, g. by perpendicular cast-iron pipes, 2 feet
6 inches diameter, running up for a height of 50 or 60 feet and
down again ; they end in a lead-lined pipe provided with a man-
hole and filled with iron borings. Here the sulphuric acid
which is condensed by the cooling, collects, and is taken up by
the iron borings. Provision must be made for removing the
solution of ferrous sulphate formed, and for renewing the iron
borings as they waste away. (A description of such an apparatus
is found in ' Papier-Zeitung/ 1894, pp. 2099 & 2130.) A more
efficient removal of the SOg and H2SO4 is effected by the apparatus
of Nemethy, G. P. 48285, of 1889, in which the gases from the
burner pass through a large bos. containing iron borings &c.
before entering the cooling-apparatus (comp. p. 283).
10. The Burner-gas.
(1) Composition of the Burner-gas from burning Brimstone.
Atmospheric air contains, roughly speaking, 28 per cent, by
volume of oxygen, and 79 per cent, nitrogen *; If it were possible
to convert all the oxygen into sulphur dioxide, the volume would
not be changed, since 1 mol. O2 furnishes 1 mol. SOg.
In the case of making sulphur dioxide for the manufacture of
wood-pulp, we want to render it as free as possible from a surplus
of oxygen. But for the manufacture of sulphuric acid we must
introduce into the burner sufficient oxygen for later on forming
SOs, and a further excess practically necessary in the process.
For the former object we must at once increase the oxygen by
50 per cent., as 2SO2 require O2 for the formation of SOg, and
the theoretical maximum of SO2 in the burner-gas would therefore
be 14 per cent, together with 7 per cent, oxygen and 79 per cent,
nitrogen. Practically we must have an excess of oxygen equal
to 5 per cent, of the exit-gases from the system, together with
95 per cent. N.
If we call the unknown volume in the-burner-gas of this excess
79
oxygen = a?, it must carry along — a? vols. N. To this are added
* Of course this item "nitrogen" comprises argon, helium, and ^11 other
indifferent gases recently discovered in atmospheric air. To simplify matters,
we shall throughout this book omit the special mention of these gases, which in
sulphuric-acid making play exactly the same inert part as elementary atmospheric
nitrogen.
398 FttODUCTION OF SULPHUR DIOXIDE.
79 vols. Ny entering along with the 21 vols. O required for forming
14 vols, of SO2 and converting them into SO3. The volume of
the total N and of the excess oxygen required in practice for each
14 vols, of SO2 introduced into the chambers thus amounts to
^n .79 ^^ . 100
X was stated to form 5 per cent. = ^^ of this volume.
We have thus the equation :
^_ 1/79 + 100 \^79J^
From this follows :
5 16 _n
*~2l'"°'^2l'^~20'
79x21 .-„ ,
•"=2()^l(i='^'^^^'°^^-'
that isj besides the theoretical quantities of gas mentioned above^
another 5*18 vols, of oxygen^ along with the corresponding
79
5*18 XoT= 19*50 vols, of nitrogen^ are necessary. The gaseous
mixture formed in the sulphur-burner accordingly ought to con-
tain upon each
14 vols, of SO2,
7-h 5-18=1218 „ O, and
79+19-50=98-50 „ N.
121-68 vols.
From this the following composition for 1 litre of this gaseous
mixture is computed : —
01 123 litre SO3
00977 „ O
0-7900 „ N
1-0000
ii
That is to say : The normal quantily of SO2 in burner^gas from
irimstone-burners is 11*23 per cent, by volume.
This normal quantity can be attained by proper care, but is
frequently not reached.
BUKNER*OAS. 399
(2) Composition of the Burner-gas in burning Pyrites,
The proportion of air required in this case differs very much
from the preceding. We shall calculate this for pure iron
disulphide. This body consists of 46*66 per cent. Fe and 53'33
per cent. S.
Although on burning dense pyrites sometimes the iron is not all
oxidized up to Fe^Os^ and a little magnetic oxide^ Fe304, is formed^
we must suppose the complete conversion of iron into Fe203 as the
normal state to be aimed at for complete utilization of the sulphur.
Consequently 2 mols. or 240 pts. of FeS^ require 3 atoms =48 pts.
O for oxidizing the iron^ and another 8 atoms=128 pts. O for
burning the S into SOj. Altogether 11 atoms=176 pt<. oxygen
arc necessary for burning, and another 4 atoms of oxygen = 64 pts.
for changing the formed 4 mols. = 256 pts. of SO2 into SO3. From
this we calculate that for each thousand parts of FeS2,
200 pts. oxygen are required for oxidizing the iron,
533J „ „ „ forming SO2,
^^^1 V 7> t* oxidizing this to SO3.
1000 .. in all.
}i
Now here also an excess of oxygen must be used^ even larger
than in the case of brimstone, which we will assume to amount
to 6*4 per cent, by volume in the gas leaving the chambers. If
we call the unknown volume of oxygen in excess to be introduced
for each kilogram of S employed as FeS,, x litre, the volume of
79
nitrogen accompanying it is ^ ^ litre. Both together and the
4933*3 liti*e N introduced along with the O requisite for combustion
and formation of SO3 form the gaseous mixture escaping at the
end, the volume of which is therefore
79 100
4933-3 + ^ -h^ a: = 4933-3 +;^.r.
As A* is TTTTL of this volume, we have
a? =454-1,
Accordingly for each kilogram of S burnt as FeSo, apart from
400 PRODUCTION OP SULPHUR DIOXIDE.
the theoretical 6244*7 lit. air, another 454*1 lit. oxygen along with
4D4ax79_jyQg.^ nitrogen— that is, 2162-5 lit. air— altogether
8407*2 lit. air at 0^ and 760 millims. pressure, have to be introduced.
Now each kilogram of free sulphur (brimstone) requires 6199
litres air at 0° and 760 millims. ; consequently a certain quantity
of sulphur, burnt as FeSg, requires
8407-2 , „^« ^.
-^^ = 1*356 times
as much air as if burnt in the free state«
This is not quite the proportion of the gas as it enters the
chambers. For on burning FeS2 a portion of the oxygen remains
behind with the iron, whilst on burning brimstone the whole
quantity of air gets into the chambers, and at equal temperature
and pressure retains its volume, since oxygen on combining with
S to SO2 does not change its volume.
The 8407*2 litres air entering the burner for each kilogram of
sulphur burnt as FeS^ furnished the following quantities of gas,
calculated for 0° and 760 millims. : —
699*4 lit. SO2 generated from the same volume of O,
349*7 „ O required for transforming SO2 into SO3,
454*1 „ O as excess,
4933*3 „ N accompanying the theoretically necessary oxygen,
1708*4 „ N „ ,, excess of oxygen.
81 44*9 ht. containing 699*4 lit. SO2,
803*9 „ O,
6641*7 „ N.
8144*9
For a certain quantitv of sulphur, burnt as FeS2, -^tt^ft times
6199
= 1*314 times as much gas must enter the chambers as if the
sulphur were burnt in the free state.
Consequently, in the case of burning pyrites, 100 volumes of the
normal gaseous mixture on entering the chamber ought to consist of
8*59 vol. SO2,
9*87 „ O,"
81-54 „ N.
In many factories the sulphur dioxide is much below 8*59 per
COMPOSITION OF GAS FROM BLENDE-FUBNACES. 401
cent., sometimes not above 6 per cent, of the volume of the
gas. In that, ease so much les^acid is made in the same chamber-
space^ unless the formation of sulphuric acid is increased by a
larger consumption of nitre. We shall return to this subject in
Chap. VII.
Sulphur Dioxide for manufacturing Calcium Bisulphite {in the
manufacture of wood-pulp, ^c). — In this case the conditions are
different from those just described. There is no question of having
to provide the oxygen for forming H2SO4 from SOj and the excess
of oxygen practically necessary in the lead chambers ; and the
formation of SO3 in the burners should be avoided as much as
possible. The operation should be conducted so as to exceed as
little as possible the amount of osfygen demanded by the equation :
2FeS2 + ll 0=Fe203 + 4S02, which corresponds to a theoretical
maximum of 16 per cent. SO2 by volume in the burner-gas.
Practically, however, 11 per cent, should not be exceeded, because
otherwise the burners get too hot, which causes the sublimation of
sulphur and the formation of scoria in the burners.
Harpf (Wochenbl. fiir Papierfabr., Biberach, 1901, nos. 23,
25, & 27) gives some calculatious referring to this special case,
containing nothing of importance.
(3) Composition of the gas from Blende- furnaces.
For burner-gas from zinc-blende the following calculation has
been made by Hasenclever (Chera. Ind. 1884, p. 79) : — Zinc-blende
(in the pure state), ZnS, consists of 63 parts Zn + 32 parts S.
For burning it into ZnO + S02, 3x16=48 parts O are required,
for converting the SO2 into SO3 another 16 O ; therefore for 95
ZdS, containing 32 S, altogether 64 O. This means that each
kilogram S in zinc-blende requires 2 kilograms O, or 1398*7 litres
at 0° and 760 millims., together with 5258*0 litres nitrogen =665(5
litres air. In order to make allowance for the 6*4 volume per cent.
of oxygen required to be in excess in the exit-gas, we find this by
the formula : —
6*4 /._^^ 100 \
'*=iooV^2^^+^i^>
n= 484-0 litres oxygen, corresponding to
1820*7 „ nitrogen
2304*7 „ air,
VOL. I. 2 1>
402 PRODUCTION OF SULPHUR DIOXIDE.
Consequently the normal gaseous mixture in roasting 1 kilo^.
blende consists of : — *•
699-4 litres SO,,
3497 „ 0 for forming SO3,
484'0 „ O in excess^
5288*0 . „ N entering with the theoretically necessary oxygen,
1820*7 „ N „ „ excess oxygen,
864r8 litres of gases.
This means that 100 volumes of the gas should contain :
8*12 vols. SO2
9*69 „ O
8219 „ N.
According to information received in 1902 the blende-gases
contain only exceptionally less than 6 per cent. SO2 ; ordinarily
6 to 7*5 per cent. SO2 apart from SO3 (see below).
So far, as we see from Hasenclever's calculation, theory would
show that the strongest obtainable burner-gas from blende is not
much inferior to that obtainable from pyrites (p. 400) . But apart
from the fact that here, as well as in the case of pyrites and to some
extent even of brimstone, the theoretical figures are undoubtedly
interfered with by the formation of sulphuric anhydride, there is,
at least with all the older blende-furnaces, a far more potent reason
why the practical percentage of SO2 in blende-gases should be far
below the theoretical one. Seeing that in those furnaces only ^ or
at most § of the sulphur was liberated as SO2 [and SO3] , that the
other ^ or J remained behind in the state of ZuSOi, and that the
nitrogen corresponding to the four atoms of oxygen contained in
ZnSO.1 dilutes the burner-gas, it is easily understood why formerly
it was considered good work if blende-gases contained 5 or at
most 6 per cent, of SO2. The modern furnaces (p. 360 et seq.)
undoubtedly yield better gases, not much inferior to the burner-gas
from pyrites.
(4) Sulphuric anhydride in Burner-gas.
In the pyrites-burner, besides sulphur dioxide, there is always
sulphuric anhydride formed during the burning. This fact has
SULPHURIC ANHYDRIDE IN BURNER-GAS. 408
long been known^ and was explained in 1852 by Woehler and
Mabla^ and again in 1856 by Plattner (^Die metaUurgisehen
Robtprocesse ') after many experiments^ in this way — that many
substances, one of which (ferric oxide) is present in large quantity
in the pyrites-burner, dispose sulphur dioxide to combine with the
oxygen of the air to form sulphuric anhydride. We have already
seen, and in Chapter XI. the subject is fully detailed, that this
reaction can be used for the production of sulphuric anhydride
itself. Another plausible explanation is, that in the cooler parts
of the pjnrites-burners sulphates of iron are formed, which in the
hotter parts again split up into FcaOa and SOs. This explanation,
however, is not sufficient for Fortmann's experiments (Dingl.
Journ. clxxxvii. p. 155), according to which the whole of the
fumes of anhydride appear the moment the pyrites take fire.
Scheurer-Kestner (Bull. Soc. Chim. 1875, xxiii. p. 437) explained
the matter from the well-known fact that ferric oxide can act as an
oxidizing agent by successively giving up and absorbing oxygen.
It is, however, established that even on burning pure sulphur
a little anhydride is formed, as we shall see.
In Fortmann's experiments, made on a small scale, on burning
pyrites far more SOg than SOo was formed, viz. in one experiment
4 times as much, in another as 5 : 3. His results were evidently
altogether wrong, in consequence of a faulty analytical method.
Scheurer-Kestner [loc, cit.) only found 2 or 3 per cent, of all the
SO2 converted into SO3, but a larger deficiency of oxygen in
the gas than corresponds to this amount ; and the later discussion
between Bode (Dingl. Journ. ccxviii. p. 325) and Scheurer-Kestner
(ib. ccxix. p. 512) did not clear up the matter.
In order to decide the question of the formation of SOy on
burning pyrites by more exact methods than those hitherto used,
especially by Fortmann, I made, together with Salathe, a series of
experiments (Deutsch. chem. Ges. Ber. x. p. 1824). It was found
that SO2 cannot, as Scheurer-Kestner had supposed, be absorbed
and estimated by barium chloride, because even chemically pure
SO2 with BaCl2 in the presence of O or atmospheric air at once
gives a precipitate of BaSO^. Check tests proved that exact results
were obtained by conducting the gas through an excess of standard
iodine solution, retitrating the latter by sodium arsenite, and esti-
mating the total sulphuric acid formed in another portion of the
liquid by precipitation with BaCL. By retitration the quantity
2 D 2
404 PRODUCTION OF SULPHUR DIOXIDE.
of SO2 absorbed was found, and by subtracting this from the total
sulphuric acid that of the SO3 was obtained. Two experiments
with burning Spanish cupreous pyrites, containing 4862 per cent,
of sulphur, in a glass tube in a current of air gave
I. II.
Sulphur obtained as SO,, 88'02 88-78 p. c.
SO3 5-80 605 „
;: ;::::::::::;::::;:; iT.} ->' ■■
in the residue 3'43
lost 2
Of the sulphur of the burner-gas itself there were present
I. II.
As SO. 93-83 93-63 p. c.
„ SO3 617 6-37 „
Two other experiments were made in this way : — In the glass
tube 50 grams of cinders from the same pyrites, in pieces about the
size of a pea, were completely freed from sulphur by ignition, and
fresh pyrites burnt as before, the gas passing through the cinders^
Pound : —
III. IV.
SurphurasSOa 79-25 7690 p. c.
SOa 16-02 16-84 „
Residue and loss 4-73 6-26 „
Of the sulphur of the burner-gas itself there werejpresent : —
in. IV.
AsSOg 83-18 82-00 p. c.
„ S0» 16-82 1800 „
On the large scale the formation of SO3 will hardly be as much as
in the last two experiments, because in the burners the gas passes
through much less ignited ferric oxide than in our experiments.
By later experiments in my laboratory (Chemiker-Zeitung,
1883, p. 29) it was found that in roasting pyrites by itself 5"05
per cent., when passing the gases through a layer of red-hot
pyrites- cinders 15-8 per cent, of the total sulphur reappeared as
SOa, which entirely confirms the above results. On burning
brimstone it was found that even here 2*48-2-80 per cent, of the
sulphur was converted irito SO3 ; and this quantity was increased
SULPHURIC ANHYDRIPE IN BUBNER-GAS.
405
to 9*5-13*1 per cent, if the gases were passed through red-hot
pyrites-cinders.
Hempel (Ber. d. deutsch. chem. Ges. 1890^ p. 1455) found that,
on burning brimstone in oxygen at the ordinary atmospheric
pressure, about 2 per cent, of it' was converted into SO3 (which
agrees with my results, as given above); but when effecting the
combustion under a pressure of 40 or 50 atmospheres, about half
of the sulphur was converted into SO3.
Further experiments were m^e by Scheurer-Kestner (Bull. Soc.
Chim. xliii. p. 9, xliv. p. 98) with the gases from pyrites-kilns as
given off in actual manufacturing. We quote here a series of
his results, obtained with samples of burner-gas taken at various
times — A, from a lump-burner; B, from a Maletra dust-burner.
Volume per
cent of bOj.
Sulphur converted
into SOg per cent.
of total S.
A. Lump-burner.
7-3
2-8"
7-5
5-8
6-5
G-6
8-3
1-2
10
00
Average
" 81
9-9
2-8
6-2
8-4
•J
B. Dust-burner,
8-2
30"
90
6-8
7-6
0-4
11-3
7-7
0-8
10
Average
" 8-5
8-7
2-5
8-7
9-3
7-6
41
The quantity of SOs formed is here found to be very irregular,
varying from 0 to 9*3 per cent, of the SO2 ; the average is decidedly
less than in our laboratory experiments with pyrites.
F. Fischer (Dingl. Journ. cclviii. p. 28) obtained the following
results, which at the same time give an idea of the difference in
the composition of the gases on the various shelves of a Maletra
dust-burner : —
1
406 PRODTTCTIOX OF SULPHUR DIOXIDE*
SOa SO, O
per rent. per cent. per cent,
A. First test (shelf-burner).
Second shelf from below 096 0*44 184
Fourth „ „ 1-52 068 16-6
Sixth „ „ 3-81 0-97 12-5
Main flue 8-26 1-34 59
„ „ 7-53 1-27 7-5
B. Second test (shelf-burner).
Sixth shelf from below 8-43 3-17 3-9
„ 4-92 068 10-7
Second shelf from below 2*48 1-42 14-8
Fourth „ „ 2-62 0*78 160
Main flue 580 0-65 10*6
C. Lujnp'buriier up to 9*3 2*1 50
These tests were made by an expeditious method which cannot
compete as to accuracy with that employed by me or by Scheurer-
Kestner. The much larger quantity of SOg in proportion to SOg
is perhaps explicable in this way *.
Blende-roasting gases, -when tested by my method at the Rhenania
works at Stolberg, yielded up to 25 per cent, of the total S as SO3.
If the burner-gases are not passed hot into a Glover tower, but
are cooled in the old way, most of the SO3 condenses in the shape
of sulphuric acid, more than enough water for this purpose being
contained in the air and the pyrites. Where the gases go into a
Glover tower, this, of course, retains all the SO3 previously formed,
also in the shape of SO2H4 (Scheurer-Kestner, loc. cit,). We
shall further on consider this fact in detail when speaking of the
Glover tower and the formation of sulphuric acid generally.
The constant presence of sulphuric anhydride in various pro-
portions in the burner- gas is, of course, a source of inaccuracy in
the testing process according to Reich (see below), which indicates
only the sulphur dioxide, as we shall see later on ; it causes, more-
over, a deficiency of oxygen and an excess of nitrogen in the
composition of the gases. Hitherto no satisfactory relation has
been found between the amount of SO2, SO3, O, and N in the
many analyses of burner-gases, as is apparent from the disputes
between Scheurer-Kestner and Bode (vide supra p. 403; comp.
* Harpf (Dingler's Journ. ccci. part 2) haa grossly misinterpreted Scheurer*
Kestner^s results, as shown by me, ibid, part 4.
t
PERCENTAGE OF SULPHUR DIOXIDE IN BURNER-GAS. 407
also Ber. d. deutsch. Chem. Ges. vii. p. 1665)^ as irell as from
Fischer's tests just quoted.
(5) Actual Percentage of Sulphur Dioxide in Burner^gas.
Another source of dilution of the burner-gas^ likewise not trace*
able quantitatively, is this — that the burnt ore does not contain
pure Fe203, but sulphates of iron, which must always retain more
oxygen than FcaOa, and the nitrogen corresponding to this excess
of oxygen must be found in the buroer-gas. On the other hand,
a little nitrogen will have to be deducted if in the burnt ore FeS
is present ; but this amounts to very little indeed.
Lastly, in the factories working with nitrate of soda decom-
posed immediately behind the burners, the dilution of gas caused
thereby must be accounted for. The calculated density of NO3H
is 2' 17823 ; we need only take this into account, as the NO3H
forms the largest portion of the gas given off by the nitre mixture.
It differs so little from that of SO2 (viz. 2'21126), that, looking
at the small quantities in question, we can take the two as equal
without any sensible error. Now in normal working order, and
using a Gay-Lussac tower, certainly not above 5 per cent, of nitre
on the burnt sulphur is consumed (corresponding to 3*7 per cent.
of NO3H), or 1*85 upon the SO2. Thus a gaseous mixture which,
without the nitric acid, contains 8*59 per cent, of SO2, contains
1*85 X 8*59
besides r - nitric acid vapour, which increases its volume
to 100' 1589: and diminishes the percentage of SO2 in the total
volume to 8*576 — a diminution too slio:ht to be traceable by
analysis. Even if the nitric acid is not calculated as such, but
as NO2 or N2O3, it has no sensible influence upon the analyses,
even if the sample of gas is taken in a place, where the nitrous
vapours coming from the Gay-Lussac tower have already entered
into the process.
Still all the above-mentioned causes unite in somewhat dimin-
ishing the percentage of SO2 in the burner-gas : so that the
percentages stated above : —
11*23 per cent, by volume in burning sulphur,
8-59 „ „ „ pyrites,
must be looked upon as the maximum, which in practice can only
be approached, but hardly ever reached, and which ought never
to be exceeded.
t
408 PRODUCTION OF SULPHUR DIOXIDE.
I£ the sulphur dioxide in the burner-gas be estimated, this will
sufficiently test the style of burning, since the oxygen of the gas
must necessarily be in inverse proportion to its sulphur dioxide —
although not exactly, as the sulphuric anhydride comes into play.
In practice, usually from 11 to 13 per cent, of oxygen is found ia
good burner-gas.
The innumerable observations made upon the percentage of
burner-gas in chemical works during recent years have proved
that with very good pyrites the above maximum figures can be
very nearly approached, whilst with other ores, badly burning or
containing unfavourable metallic sulphides, only 7 to 7^ per cent.
SO2 in the burner-gas is attained {e. g. Biichner, Dingl. Jourii.
ccxv. p. 557). Of course, looking at the difficulty of keeping the
evolution of gas exactly equal, the different observations made in
the coiirse of a day will frequently yield less than the above figures
{e,g, Scheurer-Kestner, in Dingl. Journ. ccxix. p. 117, in one day
found 6'5, 6*5, 6*0, 8*0, 9*0, 8*7 ; even greater differences occur in
his tests quoted supra^ p. 405} ; and they only signify the average
percentage of the burner-gas. As a minimum, below which the gas
of real pyrites ought never to fall, 6 — as ordinary average, 7 to 8 per
cent. SO2 by volume can be assumed. If less is found, the draught
should be cut off; if more, more air should be admitted.
Crowder (J. Soc. Chem. Ind. 1891, p. 298) quotes the following
observations on the volume-percentage of burner-gas (details in the
original; no account is taken of the SO3 present) : —
SO^. O.
Lump-kilns from 4*97 to 6-83 960 to 11*21
Old dust-kilns (shelf -burners) „ 603 „ 7'02 9*10 „ 10*00
Ditto, another ore „ 6-34 „ 7-43 7- 78 „ 8-82
New dust-kilns „ 4'86 „ 7*03 8*98 „ 1068
The temperature of the gas, where it enters the Glover tower,
in the case of lump-burners is between the melting-points of zinc
(412° C.) and antimony (432°). The gas from dust-burners, which
has first to traverse a series of dust-chambers, is generally hot
enough to melt lead ( = 326°), but it is sometimes rather less.
In the case of the Rhenania blende-furnaces (p. 361) the gases
arrive at the Glover tower above 300° C, sometimes up to 400° C.
All the above calculations refer only to pyrites proper — that
is, containing merely a few per cent, of other metallic sulphides.
C0MPABI80N OF BRIMSTONE AND PYRITES. 409
If the latter have to be roasted by themselves (for instance,
preparatory to their metallurgical utilization)^ only poor gas can
be obtained, partly because more sulphates remain in the residue,
for which the corresponding nitrogen is found in the gas, partly
because they must be roasted altogether with a larger excess of air.
Bode states (' Gloverthurm,^ p. 88) that at Oker poor ores with
27 per cent, sulphur, of which only 22 per cent, was combined
with iron,' the remainder being present as blende and barium sul-
phate, yielded gas with 5*5 per cent. S02* Lead-matte yields gas
with 5 to 5"5 per cent. ; coarse copper metal (with 34 per cent. Cu,
28 per cent. Fe, and 28 per cent. S), 5*5 per cent. SO2.
According to Wunderlich (Zeitschr. f. d. chem.Grossgew. i. p. 74),
the gas at Oker contains 5 to 7 per cent. SO2 ; its temperature in
the case of ores rich in sulphur reaches 360°, in the case of poorer
ores about 280°.
Attention must be drawn to a circumstance frequently over-
looked in analyses — that for technical purposes very rarely a reduc-
tion of the volumes of gases to QP and 760 mms. is effected. This
matters less in ordinary gas-analysis than in tests like that of Reich,
where the gases are compared with a fixed quantity of SO2 assumed
to be at QP and 760 mms. This causes most tests made by Reich's
method to indicate less than the real percentage of SO3 present.
Usually 4 per cent. SO2 in the gases entering the chambers is
considered the minimum at which it is possible to make sulphuric
acid without actually losing money by the process. Locally, of
course, this may be modified to some little extent. At Freiberg
4 to 3^ per cent. SO2 is stated as the minimum at which the
manufacture of sulphuric acid can be carried on without pecu-
niary loss. The average at those works, where a great variety
of poor ores, all arsenical, and ^' matte '^ is roasted, preparatory
to the smelting process, is from 5 to 7 per cent. SO2 in the
burner-gas.
(6) Comparison of Brimstone and Pyrites as Material
for Sulphuric-acid-makinff.
We have seen above that the burner-gas from brimstone is
richer than that from pyrites in the proportion of 1 to 1*314; that
is to say, under equal conditions, the gas generated in burning
pyrites occupies 1*314 times as much space as if the same quantity
of sulphur had been employed as brimstone. From this it directly
410 PRODUCTION OF SULPHUR DIOXIDE.
follows that the gas will also require much more chamber-
space; thus^ for an equal production of sulphuric acid^ the
chambers must be about one third larger if working with pyrites
than if working with brimstone. Usually it is assumed that the
consumption of nitre has to be increased in a similar ratio ; this^
however, is not the case, as a properly constructed Gay-Lussac
tower retains almost the whole of the nitre-gas, and the excess
volume of air is not of great importance. At the present day,
indeed, in well-managed works, less nitre is used with pyrites
than has ever been used with brimstone.
Leaving the nitre out of consideration, the advantages of using
brimstone are: — a somewhat higher yield of acid (see Chap. X.);
rather less cost of plant, and less trouble with the burners if any
thing goes wrong; and, above all, much greater purity of the
sulphuric acid, especially from iron and arsenic (this is important
only for sale acid, not for use in manure-works, alkali-works, &c.) .
If brimstone could be had at the same price as the sulphur in
pyrites, nobody would hesitate for a moment to employ the former;
and even a moderately higher price would not deter from this ;
but where the pyrites-sulphur, as is the case in most industrial
countries, only costs half the price of brimstone, or even less, the
latter cannot be employed, except for pure acid ; and even this,
when the difference in price is very large, can be more cheaply
made from pyrites than from brimstone.
Owing to this cause, the manufacture of brimstone-acid in
Europe is confined to small factories which make specially pure
acid for bleach-works, for manufacturing articles of food, &c.
A somewhat considerable number of such factories still exist in
England, whilst very few are found in other European countries.
But it should be noticed that sometimes acid is sold in England
as *' brimstone-acid'' which is in reality made from pyrites and
purified from arsenic, or else acid made from spent oxide of
gas-works, which is also practically free from arsenic.
In America formerly all sulphuric acid was made from Sicilian
brimstone ; but the notes given above (p. 58) as to the consumption
of pyrites in America show that this is rapidly gaining ground.
It is frequently asserted that sulphuric-acid chambers worked
with brimstone last very much longer (up to three times) than with
pyrites. It is not impossible that there is some difference in this
respect ; but even this is not certain, and at all events the difference
ESTIMATION OF SULPHUR DIOXIDE IN BURNER-GAS. 411
in the duration of the chambers is nothing like so great as was
formerly supposed^ and forms no item in a comparison of costs
(comp. Chap. V.).
(7) The Quantitative Estimation of Sulphur Dioxide in Burner-gas,
This is usually effected by Reich's process, which consists in
aspirating the gas through a measured quantity of a solution of
iodine, to which a little starch has been added. This is carried
on till the blue colour of the solution disappears ; the amount of
gas aspirated in proportion to the constant quantity of iodine
employed admits of calculating the percentage of SOo in the gas.
The reaction taking place is as follows :
2I + S02 + 2H20 = 2HI + H2S04.
The operation is carried on with the apparatus shown in fig. 137.
A is a wide-mouthed bottle of about 200 or 300 c.c. capacity,
provided with a three-times perforated india-rubber cork. Through
one perforation passes the glass tube a, which, by means of the
elastic tube b, serves for introducing the gas ; for this purpose a
hole is drilled in some convenient part of the burner- pipe, and
the india-rubber cork c exactly fitted into it. The second, some-
what wider perforation is closed by the small plug d; through
the third the elbow tube e passes, which is connected with the
corresponding tube / of the bottle B, holding 2 or 3 litres. The
latter serves as an aspirator — the glass tube g, reaching to its
bottom, being continued into an elastic tube h, closed by the
pinch-cock i, the whole when once filled serving as a siphon. The
graduated cylinder, C, holds 250 cub. centims.
When the sulphur dioxide in any gaseous mixture has to be
estimated, the cork c is inserted into a hole 6f the pipe conveying
the gas.
The vessel A is filled up to about half of its capacity, through rf,
with water; and B is filled almost entirely. A small quantity
(usually 10 c.c.) of standard iodine solution (12'65 grams I in 1 litre
of water) is added to the water in A, along with a little starch, by
which the water assumes an intensely blue colour. The pinch-cock
on the elastic tube b is shut so that no air can enter into A ; t is
opened, so that water runs out until the air in A and B is so far
expanded that the column of water in B is supported. The water
then ceases to run, provided that everything closes air-tight ; if
412 PRODUCTION OF SULPHUR DIOXIDE.
not, the water will continue to run. When the apparatus has
been thus tested^ i is shut and b opened ; then i is opened so that
the water runs out slowly^ the gas to be tested entering through a
in single bubbles and rising through the coloured water. As soon
as the SO2 contained in the gas gets into the water^ it converts the
free iodine into hydriodic acid ; and after a certain time the liquid
will be decolorized^ which at last happens very suddenly and can
be very accurately observed. Directly this happens^ the cock t
is closed. By this preliminary operation the whole of the inlet-
tube is filled with the gas to be tested.
Now d is opened, and a measured volume (say, n cub. centims.)
of standard iodine solution is put into the vessel A^ by which, of
course, a blue colour is again produced; d is closed again ; t is
cautiously opened, and water is run out till the liquid in a, which,
on opening d, had risen to the level of the outer liquid, has been
depressed to the point of the tube, in order to expand the gas in A
up to the degree of pressure at which the following observation
is made ; then i is quickly shut, all the water that runs out is
poured away, and the empty graduated vessel C is put back
into its place. Now i is opened, and, by the running-out of the
water, gas is slowly aspirated through A, till the liquid is decolorized
again, whereupon i is closed, and the volume of the water run
out into the graduated cylinder is measured. We will call it m cub.
centims.. In this process no sulphur dioxide escapes anabsorbed if
the bottle A is constantly shaken; it is best to do this with one hand^
holding the pinch-cock i open with the other hand, and letting this
go the moment the colour has vanished, or even when it is but
faint, as it generally disappears on shaking a little longer.
It is advisable to add to the iodine solution a little sodium bi-
carbonate, which will facilitate the absorption of SO2 (C. Winkler).
If a second testing is to be made, without any further alteration
a fresh quantity of iodine solution can be put in and the process
recommenced. When this has been repeated a few times, the
decolorized liquid in A, after a short time, again turns blue,
because then its percentage of HI has become so large that it
decomposes on standing and liberates iodine. This liquid must
then be poured away, and replaced by fresh water and a little
starch solution.
The calculation of the result is as follows : — ^The n cub. centims.
of iodine solution, provided it contains 12*65 grams per litre, by
ESTIMATION OF SVLPHDK DIOXIDE IN RURNER-OAS. 413
Fi^. 137.
its I decolorizatioQ bIiows 0-0032 gram SOj, whicli, at 0° C. and'
a barometrical pressure of 760 millims., oceupies a volume of
1*114> X R cub. centims. If the barometer shows b millims., and the
414 PRODUCTION OF SULPHUR DIOXIDE.
thermometer /° C, and the difference of water-level in the
aspirator is =A millims.^ equal to Yq:^; naillims. of mercury, the
exact volume of 0*0032 x n gram SOj is
1-114 xnx ^^^ X (1+0003665/) cub. centims.
*"I3^
As the water run out, and thus also the gas aspirated through A,
amounts to m cub. centims., the volume of the aspirated gaseous
mixture, before the absorption of the SO2 contained therein, must
have been
i«+l-114xnx T- X (1 + 0*003665 x^ c^b. centims.,
*-T36
and the percentage of SO3 in volumes of the gaseous mixture
100xlll4x»x — - X (1+0003665 x t)
^-Wa
»n+lll4xnx - -^V X (1+0003665 x 0-
*-13^
In many cases a correction for the barometrical and thermome-
trical changes will not be required ; and the formula is then simply
lll*4x« . , ^^
". 1 i 1 7~ per cent. SO9.
//i + l"114x /i^
If the percentage of SO2 in the gas is very small, and thus m
very large in proportion to n, the formula may be simplified into
111-4X
m
If 10 cub. centims. of a decinormal iodine solution ( = 12*65 grams
in 1000 cub. centims.) have been put into A, this quantity, accord-
ing to the above-given formula, will correspond to 0*032 gram, or
11*14 cub. centims. SOo, at 0°C. and 760 raillims. ; and this number
need only be multiplied by 100, and divided by the number of cub.
centims. of water collected in C, plus 11, in order to find the per-
<;entage of SOo in the gas. The barometrical and thermometrical
ESTIMATION OF SULPHUR DIOXIDE IN BURNER-GAS. 415
ijorrections are in this case, of course, neglected. The subjoined
Table will save this calculation. On employing 10 cub. centims.
of decinormal iodine solution, the following number of cub.
centims. collected in the graduated cylinder, C, show : —
Gub. cenlims. Volume percentage of SO.^.
82 120
86 11-5
90 11-0
95 10-5
100 10-0
106 9-5
113 9-0
120 8-5
128 80
138 7-5
148 7-0
160 6-5
175 G-0
192 5-5
212 50
Even if the gas to be examined is taken at a point where it is
already mixed with nitre gas, this will not exercise any practically
important influence upon the result. We have already seen that,
in ordinary work, for each 100 parts of SO2 only 1*85 NHOy, or its
equivalent as NjOg or NO2, exists in the gas. In such dilute
aqueous solutions as come into question here nitric acid hardly at
all oxidizes sulphurous acid ; this, however, is done by nitrous and
hyponitric acids. Even if we assume that only XO2 is formed
(which is going much too far), this could at most oxidize its
equivalent in SO2, according to the formula
N02+H20 + S02=S04H2 + NO.
46 NO3 thus oxidizes 64 SO3, or 1-35 NO2 (the equivalent of
1*85 NO3H) only 1*88 SO2 ; in other words, in the worst case, never
happening in practical work, of 100 parts,S02 188 part would be
oxidized by nitrogen acids instead of iodine. Even this maximum
error would, say, at 10 per cent., only amount to a deficiency of
0*188 per cent. ; but this is certainly reckoning it much too high.
416 PRODDCTION OF SULPHUR DIOXIDE.
(8) Lunge's Test for total Acids in the Burner-gas.
CoDsidering the inaccuracy peculiar to Reich's teat, owing to the
constant presence of sulphuric anhydride in the burner-gas (comp.
supra, p. 402 et seg.), the question arises whether it would not be
better to substitute for it a test showing the total acidity of the
burner-gases. There is no difficulty in doing this, either by the
method indicated for testing the exit-gases, or in a more expe-
ditious way by employing in the apparatus (fig. 138) a caustic-soda
solution tinged red by phenol ph thai eiu (litmus is not suitable ia
this case, nor is methyl-orange, which acta differently upon sul-
phurous and sulphuric acid, comp. p. 160). I have shown this
method to be quite practicable and accurate, and it is carried
out at many works for the regular control of the process. A
decinormal solution of caustic soda ia employed, of which 10 c.c.
are tinged red with phenolphtbalein and diluted to about 100
Fig. 138.
or 200 c.c. The gas is aspirated through it slowly, exactly as in
Keich's test, with continuous shaking. Especially towards the
end the shaking must be continued for a while {say ^ minute;
each time after aspirating a few c.c. of gas through the liquid,
till the colour has been completely discharged, which is best
ascertained by putting a white piece of paper or the like under-
ESTIMATION OF OXYOEN IN CHAMBER-OASES. 417
neath the bottle. The calculation is made exactly as \vith the
iodine test^ counting all the acids as SOj. A large number of
practical tests made in this manner have shown that the percentage
of total acids calculated as SO2 is always larger than the figures
found by the iodine test, owing to the presence of SOs, and that
the results of the former test agree with those of gravimetrical
estimations.
The absorption-bottle used by me differs from Reich's in
having an inlet-tube for the gas, closed at the bottom, and
perforated with many pin-holes, through which the gas rises in
many minute bubbles, instead of one large bubble, as shown in
fig. 138. Experience proves this to be greatly superior to fig. 137.
The otherwise excellent absorbing-bottle described in No. 34
of the Alkali Inspectors' Reports, p. 22, is not available in this
case, as it contains india-rubber rings, which act upon iodine.
(9) Estimation of Oxygen in Burner- and Chamber -gases.
Although burner-gas is not generally tested for oxygen, this
test being reserved for the exit-gases, we will here describe the
methods employed for estimating oxygen in any of the gases
occurring in the manufacture of sulphuric acid.
Oxygen is for technical purposes always estimated by means of
an absorbent, observing the contraction of volume produced.
Some of these absorbents are not now employed, as nitric oxide
(used by Priestley and recently by Scheurer-Kestner, Compt. Rend.
Ixviii. p. 608; also lately recommended by Wanklyn), ferrous
hydrate (Vogt, Dingl. Joum. ccx. p. 103), and others. Cuprous
chloride in ammoniacal solution may be employed for absorbing
oxygen, but it has no advantages over pyrogallol or phosphorus,
and several drawbacks, so that it cannot be recommended. The
choice really lies between the two agents just mentioned.
Of these pyrogallol must be used in an alkaline solution, and it
acts very promptly indeed*. Its use for this purpose was proposed
by Chevreul as early as 1820, but it became general only through
Liebig many years after. It is true that this reagent in the
presence of pure oxygen forms some carbon monoxide (Grace
Calvert, Proc. Manch. Lit. & Phil. Soc. 1863, p. 184) ; but this
never happens with gaseous mixtures containing no more oxygen
than atmospheric air (Poleck, Zeitschr. f. aualyt. Chemie, 1869,
VOL. L 2 E
418 PRODUCTION OF SULPHUR DIOXIDE.
p. 451). It is therefore altogether reliable in the present case^ and
is very much employed. Its drawback is that the solution (25
grm. potassium hydrate and 10 grm. pyrogallol to 400 c.c. water)
gradually thickens and becomes useless long before this somewhat
expensive reagent has been used up.
Phosphorus is one of the oldest absorbents for oxygen^ and it has
again come into use^ since the manner of employing it has been
properly studied^ and as it is now found in trade in very thin
sticks^ offering a great surface. Of course it must be kept under
water^ and must be exposed only to the gas to be analyzed. It
acts only at a temperature of at least \&^, better a little above;
this is the first condition to be observed in its use^ but it is
easy to attain if the apparatus is kept in a somewhat warm place
(near an acid-chamber) • The action of phosphorus on oxygen is
interfered with by traces of tarry matters and the like ; but such
do not occur in chamber-gas. As it is^ the gases, both for the
pyrogallol and the phosphorus test, are often previously washed
and freed from acids by passing them through a solution of caustic
potash j but there is generally no great error made by omitting
this treatment. The action of the oxygen on the phosphorus is at
once indicated bv the formation of a white cloud, and it is necessarv
to wait a few minutes till that cloud has completely disappeared,
when the absorption of oxygen will be complete. Once charged,
such an apparatus may serve for hundreds of tests, but it should
be kept from daylight.
The apparatus employed for estimating oxygen in gaseous mix-
tures may be of various descriptions. Those most used are Orsat^s
apparatus (comp. Lunge and Hurter's Alkali -Maker^s Pocket- Book,
p. 74, and Winkler- Lunge's Technical Gas- Analysis, 2nd edition,
p. 87), Lindemann's apparatus (ibid. p. 92), and M. Liebig's
apparatus (Dingl. Journ. ccvii. p. 37, and ccxxxiii. p. 396) ;
Younger^s apparatus (J. Soc. Chem. Ind. 1887, p. 348) is a slight
modification of the latter. The two former can be used for
phosphorus or pyrogallol, the latter for pyrogallol only.
Of course Hempel's, Bunte^s, or any other apparatus for general
technical gas-analysis may be employed as well.
Instead of taking only single samples of the escaping gas, it is
advisable, along with these, to collect an average sample (say, for
24 hours) by aspirating a certain quantity (say, about 50 litres) by
means of a large aspirator with the outlet»cock very slightly
ESTIMATION OF NITROGEN OXIDES IN OASES. 419
opened. Owing to the slowness of the aspiration^ the gas stand-
ing over the water in the aspirator will be thoroughly mixed up,
and by taking a sample from the aspirator the average percentage
of oxygen can be estimated with some degree of accuracy. We
shall treat this matter more in detail in the 7th Chapter, when
describing the testing of the exit-gases.
(10) Estimation of the Oxides and Acids of Nitrogen in Gaseous
Mixtures.
Burner-gas will contain the above acids if the ^^ potting '' has
been done either within or close to the burners, as is usual in
England. The acids and oxides of nitrogen are, however, but
exceptionally estimated in burner-gas, and not very often even in
chamber-gas ; the latter is not indispensable^ because the colour
of the chamber-gas on the one hand, and the testing of the
" drops '^ on the other, which we shall describe in the 5th Chapter,
are sufficient for guiding the manufacturer in his work.
In well-conducted works, however, the chamber-exits are tested
not merely for total acidity, but also for nitrogen acids separately.
For this purpose the prescriptions formulated in 1878 by the
British Alkali-Makers' Association mav be observed, which we
shall give in detail later on (Chap. VII.) , These do not extend to
nitric oxide, but we shall see that it is easy to estimate this at the
same time. In this place we shall give a short outline of the
methods employed by Lunge and Naef for their extended experi-
mental investigation of the vitriol-chamber process (^ Chemische
Industrie,' 1884, p. 5) for estimating nitrogen oxides and sulphur
dioxide at the same time. The pipe bringing the gas from the
chambers is continued into a Y-pipe, both branches of which are
connected with sets of absorbing- tubes. One branch is connected
with three U-tubes containing concentrated pure sulphuric acid
(for retaining H^^^ &^d N2O4) and a fourth tube containing an
acidulated solution of potassium permanganate (for retaining NO).
The other branch of the Y-pipe first leads into a long glass tube
filled with cotton-wool or glass-wool, where any drops of mechani-
cally carried-over sulphuric acid are retained, and then into two
U-tubes containing pure caustic-soda solution (for estimating
SO3). At the end of both sets of tubes there is an aspirator, con-
sisting of a large glass bottle, holding about 20 litres, divided into
2e2
420 PRODUCTION OP SULPHUR DIOXIDE.
single litres^ with a tap or tap^siphon leading into a similar bottle,
so that the quantity of water run out indicates the quantity
of gas passed through each set of tubes. The gas remaining in
the aspirators is tested for oxygen by absorption ; nitrogen by
difference. Sulphur dioxide is found by oxidizing the caustic-soda
solution with bromine- water and precipitating the sulphuric acid
formed by barium chloride. In the sulphuric acid nitrogen trioxide
and tetroxide are estimated in the way indicated in Chapter III.
(p. 253 et seq.), by testing both with permanganate and with the ni-
trometer. The tube filled with potassium permanganate must have
retained the nitric oxide, which is found by adding titrated ferrous
sulphate solution and re-titrating with permanganate. Where the
quantity of SO9 in the gas is considerable, this method cannot be
employed ; in this case it is not possible to separate the 'Nfiz and
Ns04, and the method otherwise used for the chamber-exits must
be employed (Chapter VII.).
Nitrous oxide has never yet been actually proved to exist in
chamber-gases ; but it probably does occur there in very slight
quantities. The former methods of its detection and estimation
were very inadequate and failed entirely in the presence of other
nitrogen oxides. It may be possible to apply the method proposed
by Enorre (Berl. Ber. xxxiii. p. 2186)| viz. burning with excess of
hydrogen in a Drehschmidt's platinum capillary (comp. Winkler-
Lunge's Technical Gas-Analysis^ 2nd edition, p. 162), or else
PoUak^s method (described in his inaugural dissertation^ Prag, 1902,
p. 52, and in Treadwell's ' Lehrbuch der analyt. Chemie,' 2nd ed.
ii. p. 538) of burning in a bright red-hot Drehschmidt^s capillary
with pure carbon monoxide, measuring the contraction and
estimating the CO2 formed. The contraction in burning for NO is
NO
; for N2O there is no contraction at all ; the volume of the
absorbed COj is equal to !N0 + N2O. Hence the volume of N2O is
obtained by deducting twice the volume of the first contraction
from the volume of COj found. This is made clear by the
following equations : —
2 vol. NO +2 vol. C0=1 vol. N2 + 2 vol. CO3.
1 vol. NjO + l vol. C0 = 1 vol. N2 + I vol. CO2.
NECESSITY OF LEAD CHAMBERS. 421
CHAFTER V.
CONSTRUCTION OF THE LEAD CHAMBERS.
We have already seen, p. 8 et seq., through how many stages the
construction of that apparatus has gone in which formerly all
sulphuric acid, other than fuming oil of vitriol, was made and
up to now is still made for the most part, viz. the lead chambei*.
The reader already knows that sulphuric* acid is fonned by the
oxygen of the air being transferred to sulphur dioxide through the
intervention of the acids of nitrogen and with the aid of a molecule
of water, thus :
SO2 + 0 + H20 = S04H2.
All the substances entering into the process are in the state of a
gas or vapour, except the water which is sometimes introduced as
a spray or mist. The reaction takes a certain time, as the
nitrogen compounds which serve as carriers of oxygen have to be
frequently reduced and reoxidized, and as the gases and liquids
are only gradually mixed so intimately that they can actually
enter into reaction. There must therefore be a space provided in
which large quantities of gas can remain for some time.
According to the calculations given on pp. 397 and 399, for each
kilogram of sulphur in the state of brimstone 6199, or in the
state of pyrites 8145 litres of gas, reduced to 0° and 760 millims.
pressure, must enter into reaction ; and these figures are a good
deal increased by the higher temperature, the steam, &c. In order
to harbour such vast quantities of gas, very large spaces must
be provided. Since the strongest acids have to be dealt with,
both in the liquid and the gaseous form, most materials other-
wise used in building are out of the question ; and since, of
those suitable, glass, earthenware, &c. are excluded by the large
size of the apparatus, practically only one material remains which
is sufficiently cheap and suitable for the purpose, viz. lead. The
disadvantages of this metal, such as its great weight, its softness
422 CONSTRUCTION OF THE LEAD CHAMBERS.
and lack of rigidity^ its . easy fusibility^ its comparatively high
price, cannot outweigh the advantages which none of the base
metals share with it for our purpose, viz. : — the great chemical
resistance to the acid gases and liquids ; its ductility, which per-
mits rolling it into large sheets ; its extraordinary pliability and
toughness, in consequence of which it can easily be shaped in
every possible way ; and, lastly, even its easy fusibility, which per-
mits the edges of two sheets to be so completely united by melting
together with a strip of lead, that they form a whole for all prac-
tical purposes, so that it is possible to make vessels of indefinitely
large size and any shape, provided care be taken to support the
walls of the vessel on the outside, lest they collapse by their own
weight.
A special advantage of lead is this, that even after a number of
years, when the chambers have become quite worn out, the greater
portion of its value can be recovered by remelting the material ;
even the mud containing lead can be utilized.
The attempts to make sulphuric-^cid chambers from other materiaU
than lead have completely failed. To this class belongs the proposal
of Leyland and Deacon (patents of September 10th and December
2nd, 1853) to make them of hard-burnt firebricks, slate, sandstone^
basalt, &c., set with a mixture of melted sulphur and sand. VuU
canized india-rubber or gutta percha are just as useless ; Krafft
(Wagner's ^ Jahresb/ 1859, p. 137) found that gutta percha in an
acid-chamber loses six times as much weight as lead, and half as
much again of its surface. It would be absolutely impossible
to use it, because it softens at the temperature of the chambers,
and in that state is even more easily acted upon by the gases.
Simon's zeiodelite (Dingler's ^ Journal,' civ. p. 100), a mixture of
1 9 sulphur with 42 pounded glass, to be employed in slabs of ^ inch
thickness, has, no doubt, never been so much as tried for this
purpose, no more than the sheets of glass proposed by Wilson and
others.
We will now describe the erection of lead chambers.
The chambers are always placed at some elevation above the
ground- level. At the present day chambers are probably nowhere
found placed on the ground itself, or on such low foundations that
one cannot at least walk underneath ; mostly their bottoms are
much higher than this. The first object of this is to give the
opportunity of ascertaining whether the chambers are tight. If
FOUNDATION ANB PILLARS. 423
their bottoms are not easily accessible, large quantities of sul-
phuric acid may get lost in the ground before any loss is detected.
And this means not merely a loss of the acid, but still more : the
foundations get corroded and undermined, and the whole structure
may collapse. The expense of building the chambers on pillars
&c. is not thrown away, as the whole space underneath can be used
as a warehouse which in winter time has always a moderately high
temperature ; or it may even, if high enough, be utilized for the
pyrites-kilns &c., although this course is not to be recommended.
In the latter case it should be from 17 to 20 feet high. At some
works, which are pressed for space, even the saltcake-furnaces,
ball-furnaces, &c. are built underneath the chambers; but the
space below them must then be at least 30 feet high.
In any case the soil must first be examined to ascertain whether
it affords a spfe foundation ; for if it settles more in one place than
in another, the chamber gets out of plumb and its bottom out of
level, which, owing to the acid lying on the latter and to the insta^
bility of the chamber-sides, causes great inconvenience. A rocky
or pebbly ground is best; next to this, sand or clay; marl or lime-
stone are bad, because sometimes acid will run over accidentally,
which acts upon it; and this may happen even with clayey soil.
In such cases the whole soil underneath the chambers must be
protected by a layer of asphalt.
The pillars upon which the chamber is erected must, of course,
go down to the " rock,'' as in auy ordinary higher building. If
the accumulation of made ground or loose earth is so deep that it
would be too costly to excavate and raise the pillars from below,
piles must be driven in, according to well-known building-rules,
and the pillars built upon these.
The pillars themselves can be made of brickwork, stone, cast-
iron, or wood. Sometimes, instead of single pillars, two longi-
tudinal walls are erected, connected by cross joists and interrupted
by doo«8, windows, &c., as shown in the sketch, fig. 139. Such
long walls take much material and make the room underneath the
chambers dark, in spite of the windows. They are only suitable
where the chambers are placed unusually high in ordei: to build
furnaces underneath. Up to a height of about 26 feet metal
pillars seem preferable.
The cheapest pillars are those made of wood or bricks ; very
rarely they are made of stone — much more frequently of the dearer
424 CONSTRUCTION OF THE LEAD CHAMHERS.
but much stronger and more durable material, cast-iron. If made
of wood, round or canted balks of at least 10 inches (better 12
inches) thickness must be employed. Mostly fir- or pine-wood is
Fig. 139.
used, especially Scotch fir; but the American pitoh-pine or yellow
pine, Bucli as is used for ship-building, is preferable (on account of
its much greater durability) in spite of its higher price. This
applies not merely to the foundation pillars, but even more to the
frame of the chamber itself. The pillars must vary in their thick-
ness, mutual distance, and manner of being stayed, according to
their height and the weight resting upon them (which may be taken
at 150 lbs. per superficial foot of the total chamber-area, for the
lead, timber, and acid, the latter alone in a full chamber amounting
to 120 lbs, per superficial foot}; but for an average height of 10 to
13 feet, which will not often be exceeded with wooden pillars, they
ought not to be further apart than 10 to at most 13 feet from
centre to centre. In any case they are put into a stone socket
projecting from the ground, lest the bottom of the pillar be
damaged by any moisture or acid ; the stone has at the top a
hollow of ^ to 1 inch depth, into which the foot of the pillar fits
exactly ; at first a little tar is poured into it. Wooden pillars do
not last for ercr ; they are not to he trusted very much, and are
rarely found now in larger works, at any rate as principal pillars,
except where wood is very cheap.
Frequently brick piUara are employed . These also are not often
made above 13 feet, at most 15 feet high; they ai« at least
18 inches (better 2 feet) square. They are made of common
bricks with a mortar very poor in lime.
The brick pillars io many works have
been replaced by cast-iron ones, because
they are not very durable, especially at
the top, where the beams rest. Even
the bricks themselves become rotten
by contact with the acid, and only
stand better if previously soaked in
hot tar ; but they take the mortar very
badly after that. They may also be
painted with hot tar afterwards.
On the Continent, where, in con-
sequence of the colder winters and
hotter summers, the chambers have to
be placed in a closed building, the
pillars may be built in a piece with the
main walls of this building; but it
is even then best to keep them seiia-
rate, as their settlement is different
from that of the main walls.
Chambers 20 feet and upwards in
width are sometimes built with mixed
pillars — viz. brick pillars for the two
long sides, and wooden pillars for the
centre row.
Stone pillars are not often used for
acid-chambera. Alade of rough stones,
they would be extremely clumsy ; and
hewn stone in most places is too dear.
On the other hand, of course, stone
pillars of the latter kind are very sub-
stantial, and last almost for ever,
unless the stone be very soft and
rotten.
In the larger works in England ca»l-
iri/n pillars are almost exclusively
employed, iu spite of their higher cost.
These can be made 30 or even 36
feet high; they take very little space,
and are almost imperishable if painted
from time to time. They can be
426 CONGTRDCTION OP THE LEAD CBAMBEBS.
■weighted a good deal more than any other pillars, unless these are
made very thick; and they can be used as supports for many other
purposes by means ot cast-on brackets or even of pieces bolted
on subsequently. A brick or stone foundation must be made for
them up to the level of the grouod or a little higher; the top
stone is made with a socket to receive the foot of the pillar, as in
the case of wooden ones ; or holes are drilled into the stone,
corresponding to other holes in the base of the column; and the
joint is made by iron cramps, fastened by pouring in melted lead,
or in some other way.
The cast metal columns are now frequently made of an H-
shiiped section and a little tapering upwards. Fig. 140 will
show this more distinctly tc^ether with a bracket on each side for
receiving a wooHen stay for the timber above. Another, cross-
shaped section is shown in fig. 141. These constructions are
better adapted for brackets, &c. than round columns. If higher
than shown in the figures (15 feet), they must be correspondingly
stronger — for instance, for 20 or 24 feet height, 12 inches
diameter at the base. Such columns can be placed at 20 feet
distance from centre to centre, if the beams resting upon them
are strong enough.
Sometimes the columns are made of wrought iron, of the section
shown in fig. 142. They are a little dearer than cast-iron
columns, but more durable and reliable.
The pillars are in most works placed so that F'g- 1-li
they stand directly under the side frame, which
has to carry the weight of the chamber sides, and
in the English systt^m also the whole weight of
the chamber top. This, however, in anycasesuf-
(ices only for very narrow chambers ; for chambers
ot ordinary width (from 20 feet upwards) a centre row of pillars
must be added to prevent sagging of the joints. But as the
weight of the acid in a full chamber may be up to four times as
much as that of the frame and lead combined, it seems more
rational to place the pillars more inside, in wjiich case two rows
suffice even for a chamber of ordinary width. -
Above the pillars there are generally, in England, longitudinal
sleepers. If there is a continuous wall in the place of pillars, to
cover this with a 2-inch plank will be sufficient ; but if there are
separate pillars, the sleepers must be strong enough to support the
SLEEPERS AND JOISTS. 427
whole structure of the chambers^ both wood and lead ; and their
strength will then depend on the distance between the pillars. With
chambers of 20 feet height^ and distances between the pillars of
20 feet from centre to centre, the longitudinal sleepers should not
be less than 12 to 14 inches high, and ought, besides^ to be sup-
ported by stays, as shown in fig. 143. With the pillars at shorter
distances (say 10 or 13 feet), timber of 9 by 12 inches, always on
edge, suffices for the longitudinal sleepers. The joints of the
beams of which they consist ought to be well connected, as shown
in fig. 148, and should be placed between the pillars, where they
Fig. 14:3.
JL
20
are supported from below by the stays. The upper face of the
sleepers must be levelled as well as possible from one end of the
chambers to the other. Above these the cross joists are placed,
running from side to side, and made long enough to carry the side
frames, and to leave, moreover, a passage round the chambers.
For the latter object only every third or fourth joist need project
about 5 feet on each side. The joists are mostly planks on edge.
If chambers are much less than 20 feet wide, which rarely
happens now, no centre longitudinal sleeper is needed, and the
cross joists should have 9 by 3 inches section and corresponding
length. Wider chambers require a centre row of pillars and
sleepers; and iu this case, as such long planks are not easy to get,
the joists can be made in two lengths, resting on a side and on the
centre sleeper. The horizontal distance of the floor-joists is
usually 12 inches from centre to centre. Some works have them
3 by 11 inches. The length of the joists is equal to the width of
the chambers j9/ti« the chamber- frame, plus the width of the passage.
The joists are covered with a 1 -inch ^oor, laid quite level in all
directions. As the flooring-boards might easily warp afterwards
from the heat of the chambers, this must be prevented by the
well-known methods of carpentry. The edges of the boards are
planed so as to form a perfectly smooth floor without any chinks.
428
CONSTRUCTIOX OF THE LEAD CHAMBERS.
Another system of building is more in favoar on the Continent.
Firsts from pillar to pillar strong sleepers are laid across the width
of the chamber; upon these a large number of longitudinal joists
are laid^ and the flooring-boards on the top of these, running from
one side of the chamber to the other.
Upon the whole the frame of the chamber is erected^ which serves
for supporting the lead. It consists, for each side of the chamber^
of a sole-tree (sill) and a crown-tree (capping), connected by
uprights or ^' standards/^ and further tied by cross rails or stays.
The sole- and crown-trees and uprights are either of square section
(say 6 inches square for a chamber up to 20 feet high) or oblong
(say 7 by 3 inches). The Sole- and crown-trees lie on the flat
side; and the uprights are mortised into them so that their
Fig. 144. Fig. 14.5.
,-5
longer side just covers the trees. In the corners the trees project
over and are rabbeted into each other. If no cross rails are
employed, the uprights are placed 3 feet 3 inches apart from each
other ; if they are connected by cross rails, they can be placed
4 feet apart. The cross rails are 3 inches by 2 inches ; they are
only partly let into the uprights, in order not to weaken these ;
and are placed at vertical distances of 4 to 5 feet from each other.
The chamber-lead is kept a little away from the woodwork in
order to expose the lead always to the cooling action of the air.
If this is not done, the lead is found to be quickly corroded in the
parts protected against radiation of heat by the wood ; it has
even been observed that insects from the wood have bored through
it"^. It is becoming more and more the fashion to shape the
woodwork so as to present the least possible contact with the lead,
as shown in figs. 144 and 145. Almost the same effect is obtained
by using round timber for uprights.
The best kind of timber for this purpose, as well as for all others
* 1 have described such a case, where the beetles in question belonged to the
gpecies Tetropium luridum and Hiflotrupes bq/olus (Zsch. ang. Chem. 1897^
p. 527). * ^ ■
TIMBER-FRAME. 42U
where acids are concerned, is American yellow-pine or pitch-pine ;
but as this is frequently too expensive, ordinary red-wood is also
very much in use. It is beneficial to protect it against the action
of the acids by a coating of whitewash, which is at the same time a
slight protection against the risk of fire. Another kind of pro-
tection from the former, although not from the latter, risk consists
in painting the woodwork with coal-tar, or preferably with a sort
of tar-vamish, made by dissolving coal-tar pitch in heavy tar-oils,
and known as "prepared" or "refined" tar (Lunge's 'Coal-Tar
and Ammonia,' 3rd edition, p. 374). The latter enters better into
all the pores of the wood and on drying does not leave so many
crevices; it is altogether preferable to raw coal-tar for painting
wood, iron, or brickwork, and is not much dearer.
The painting of the woodwork is best done twice, and before the
lead is put on, so that all parts can be reached by the brush.
Special care must be taken lest any acid gets into the mortise-
boles, where the uprights are joined to the sole-trees, 8sc. No
empty space should be left here where any acid could lodge, but
all interstices should be filled up with coal-tar pitch or the like.
It seems also another good plan to cut out the bottom of the
upright, and make it fit on to a corresponding saddle-shaped part
of the sole-tree, as shown in fig. 146. Two small lead pipes drain
Fig. 146.
away any liquid collecting in the low corners, so that no acid can
ever lodge there and cause the wood to decay.
In France sometimes the bottoms of the uprights are not at all
mortised into the sole-tree, but rest flatly upon them, being kept
in their places by pressure and friction only.
Whether cross rails are used or not, in any case there should be
diagonal stays, to give more stability to the frame. II. is not of
430 CONSTRUCTION OF THE LEAD CHAMBERS.
much consequence how the stays are put^ so long as this is done
according to the well-known rules of carpentry.
If, as is usual in England^ the chambers are in the open air^ one
side of the frame is made about a foot higher than the other, so
that the rain-water and melted snow can run off^ and on the lower
side a water-spout is arranged so that the rain-water cannot run
along the chamber-side down into the acid at the bottom.
At some places the chamber-frames are made of angle-iron.
This plan has the advantage of presenting an extremely durable
and clean erection and of avoiding overheating of the lead in any
part. Such frames may be constructed in the following manner.
The side-frames consist of thin angle-irons crossing each other at
right angles^ the uprights 9 ft. 6 in. and the horizontals 7 feet
distant from each other. No iron nails are employed at all; the
lead straps are simply bent round the angle-irons. The roof i&
suspended from angle-irons in exactly the same way. Of course
iron frames are more costly than wooden and must be kept in
order by painting from time to time, preferably with coal-tar
varnish (p. 429).
Now the chamber itself can be erected, and we shall therefore
now speak of the lead to be employed. For this, sheet-lead as wide
as the rolling-mills can supply it, and of convenient length, is used,
so as to have as few seams as possible. The usual thickness in
England is 6 lbs. to the superficial foot, sometimes 7 lbs., especially
for the ends and the top, or for the first chamber of a set.
This thickness is sufficient for a chamber to last upwards of
ten years ; the bottom lasts longest, because it does not get so
hot as the sides and the top, and it is also more protected by
the mud of lead sulphate which collects upon it ; only in cases of
gross neglect (for instance, if nitric acid gets to it) it is quickly
worn out, whether the lead be thick or thin. But where ctTic-
blende is used, the mercury contained in it may have a different
effect, especially since the blende-furnaces are driven at a higher
temperature, so that more mercury gets into the chambers.
According to information received from Dr. Hasenclever in 1902,
it has been noticed at Stolbergthat the mercury acts most strongly
at the lateral parts of the bottom w hich are less protected by the
sulphate-of-lead mud, and where the joint between the side sheet
and the bottom sheet causes the double layer of lead to collect acid
and mercury between the two sheets. Here sometimes mercurv
is visible in globules, and that part is worn out in less than three
QUALITY OF LEAD. 431
years. Hence they make the whole bottom of stronger lead,,
rolled extra strong at the part next to the sides, which is, moreover,
protected by a covering of acid-proof flags.
In America the usual thickness of lead is only 5 lbs. per super-
ficial foot, and even 4 lbs. lead is sometimes used (Journ. Soc.
Chem. Ind. 1885, p. 27); but this seems very bad economy indeed,,
and it can only be done when burning brimstone. In the best
American works I have found 6 lbs. lead.
On the other hand, at some of the best English works not
only is 7 lbs. lead used throughout for the chambers, but in the
most exposed places, such as the front and back ends of the
leading chamber and several feet of the sides adjoining these, 9 lbs.
is used. Sometimes the side sheets are rolled so that the upper
and lower two feet are made stronger than the remainder, because
these parts are more quickly worn out. Chambers built in this
careful way last upwards of 20 years.
The quality of the lead is certainly not indifferent. Opinions
were formerly not altogether agreed as to the point whether pure or
impure lead better resists the action of sulphuric acid ; and in the
present case it is no doubt not so much this acid as the nitrous
compounds to which the attack of the lead is due. Most manu-
facturers formerly inclined to the belief that " hard '' lead is better
adapted to vitriol-chambers than *' soft lead/' A test sometimes-
performed consists in trying which of several samples of lead in
contact with sulphuric acid gives off more hydrogen from a given
surface in a given time ; but this test is very apt to mislead, and
there is really no good test known as yet (comp. Journ, Soc. Chem.
Ind. 1884, p. 230).
This subject has been fully discussed before, p. 206 et seq., where
the conclusion was come to that for vitriol-chambers the purest and
softest lead is the most suitable material. In Zsch. ang. Ch. 1892,
p. 643, I have given the following analyses of lead specially suited
for vitriol-chambers: — 1. Soft lead of the Mulden lead-works:
0001 per cent. Cu, 0-044 Bi, 0-0004 S b, 00005 Fe, 00004 Sn, 0-0005
Ag, no As. 2. Soft lead from W. Leyendecker & Co., Cologne :
00034 Cu, 00019 Bi, 00029 Sb, trace of Fe, 00047 As, 0*00025
Cd, trace of Ni & Co, 0 0010 Ag, 00002 Zn, 00024 O.
G. E. Davis ('Chemical Engineering,' i. p. 142) quotes the
following analyses of " chemical lead '^ (from what source or by
whom made is not stated) ; the figures indicate milligrams per
kilogram (I omit the decimals) : —
432 CONSTRUCTION OF THE LEAD CHAMBERS.
A. B. C. D.
Antimony 17 19 64 32
Copper 21 13 14 19
Silver 22 14 39 16
Iron 3 2 12 2
Cadmium 1 1 1 2
Bismuth 4 2 22 4
Zinc 2 12 3
Sulphur 12 2 1
According to Eng. & Min. Journal, March 8, 1902 (Journ.
Soc. Chem. Ind. 1902, p. 510), a special brand of lead is designated
in Missouri '^chemical hard lead^' and is sold at 5 cents per
100 lbs. above the price of common Missouri lead. It is supposed
to contain a little copper and antimony, but no attempt is made to
keep the composition within precise limits. MUhlhauser, in Zscfa.
angew. Ch. 1902, p. 758, quotes an analyses of soft chamber-lead
from Chicago and St. Louis.
Leyendecker (B. P. 2756, 1901) prepares a special quality of
chemical lead by adding either 0' 1-0*5 per cent, copper or O'l-
0*5 per cent, copper and 0*l-0'3 per cent, antimony. (Such a
patent could hardly be maintained in the face of my researches,
p. 206 et aeq.y published a long time ago, for the purpose of
benefitting chemical manufacturers generally.)
The analysis of such lead, which 1 have obtained from the
Rhenania Chemical Company, as compared with ordinary soft lead,
gives the following results : —
Ordinary Leyendecker'a
soft lead. special quality.
Bi percent 0*00501 0-00605
Cu „ 001787 006683
Cd „ 000004 000003
As „ 0 0-00002
Sb „ 000039 005025
Fe „ 0-00089 0*00074
Zn „ 0-00082 0-00122
We notice, in the '' special quality,*' besides a certain amount
of copper and antimony (the latter in such small quantities a^
not to counterbalance the useful effect of the copper) also an
extraordinary small amount of bismuth, being only 1/100 of
that present in the '' ordinary soft lead.'* The firm using both
JOINING LEAD BY BURNING. 433
had not had a sufficiently long experience with them to judge of
any difference in behaviour.
Another ^^ special quality/^ particularly recommended for KrelPs
concentrating apparatus (comp.Chapt.IX.)^ is supplied by Graffweg
& Co.j Diisseldorf. It is probably something very similar to
Leyendecker's lead.
All sheet-lead before being used should be '^ mangled,'' in order
to beat out all inequalities and indentations casually produced in
transit &c. For this purpose it is tightly rolled round a wooden
roller, about 6 inches thick, and is beaten all the time with a
plumber's mallet.
The sheets of lead were, in the infancy of acid-making, joined
together by the ordinary soft solder , which is very convenient for
use, but is soon corroded by the acid. Places soldered thus are
also much more brittle than pure lead. So long as the chambers
had to be put together in this way, there was occasion for innumer-
able repairs.
Another plan (which is far better in this respect, but takes
much lead, and is only easily applicable for straight seams) is the
rabbet'^omi. The edges of two sheets of lead are turned over in
the way shown in fig. 147, placed one into the other, and beaten
Fig. U7,
down on a smooth surface. Such joints are gas-tight, and have
been used here and there in England till within the last few years.
The kind of joint now generally cm ployed is that made by burning,
with the lead itself — that is, by melting it with a hydrogen flame
fed by compressed air. In this way the two sheets are joined so
tightly, that with good work the joint, being thicker than the
sheets, is actually stronger than they are. If the work is rough
and uneven^ foreign substances will easily be deposited in the rough
parts, by which the lead may be damaged.
This mode of joining was invented by Debassayns de Richemond
in 1838. Two apparatus are required for this, whose construction is
seen in figs. 148 & 149. Fig. 148 shows the '' plumbers' machine"
— that is, the hydrogen apparatus — quite similar to an ordinary
laboratory gas-holder, but made of lead, often with a wood casing.
The lower vessel. A, contains a lead grating, KL, upon which
VOL. I. 2 F
^34 CONSTRUCTION OF THE LEAD CHAHBERS.
granulated or scrap zinc ia put. The upper veasel contains dilute
sulphuric acid. The counecting-tube with a cock /allows the gas
to pass out of the opening (', after it has first been washed in a
water-vessel. Often there is a plain outlet just above the cock /,
The outlet is connected with a long india-rubber tube, by means
of which the gas can be conducte<i to a distance. The tube G
serves for running sulphuric acid from B to A, It can only run
Kg. U-. Fig. 140.
in as gas is allowed to escape by opening the cock /; aiid thus
a continuous current of gas is obtained. The openings D, E,
and F serve for introducing acid and zinc, and for running off the
solution of zinc sulphate.
The second part of the apparatus, which is shown in fig. 149, is
siraply a portable smith's-bellows of cylindrical shape, the lever of
which, 0 a c, a buy works with his foot. The air is forced through
the valve D from C to the closed air-vessel B, and escapes through
the opening /, likewise connected with a long elastic tube. The
Fig. l-iO.
two tubes are united by a blow-pipe, tig. 150 ; and the mixture
is ignited. Each limb of the blow-pipe is provided with a stop-
cock, by turninj; which the plumber may admit moreair or hydrogen
BURNING THE LEAD TOGETHER. 435
at will, and thus can produce a flame of any size, which, however,
should never be an oxidizing one.
The mouth-piece of the blow-pipe itself is sometimes connected
with the fork-shaped piece by a short elastic tube, to make it more
mobile. Besides the ordinary mouth-piece, ending in an aperture
of about 2^5 of ail iiich diameter, the plumber also carries another,
provided with a small brass shield, to obtain a steady flame in windy
weather. The gases unite only immediately before escaping; and
thus the flame cannot strike back. By means of this machine a
pointed and very hot hydrogen flame is produced, which, at the
place where it touches, melts the lead immediately down to a
certain depth ; and the art of burning consists in touching and
melting parts of two sheets at the same time^ which, on cooling,
solidify to a whole.
It has happened that plumbers have been poisoned by arseni-
uretted hydrogen, produced by impurities in either the zinc or
the sulphuric acid. The hydrogen can be freed from this by
washing it with a solution of cupric sulphate, by which the arsenic
is precipitated. Hydrochloric acid should not be employed for
evolving the hydrogen, even when free from arsenic, as the
workmen are injured by gas thus made.
The burning itself is a kind of work requiring much practice,
because the plumber must not allow the flame to act a moment
too short or too long. If he does the former, the fusion is not
perfect and the seam is not tight ; if the latter, he burns a hole
in the lead. Wherever it is possible, one sheet is laid about
2 inches over the edge of the other, as shown in fig. 151. The
Fig. lol.
^^rMtM,a^*/MrMf,y '.
1(y/. '. MP".'
seam is made with the help of a strip of lead, about f by ^ inch
thick, which the plumber holds in one hand whilst he guides the
blow-pipe with the other. He works in this way : — He touches
with the flame the place a (fig. 151), where the edge of one sheet
lies upon the other, so that the surface of the lead (previously
scraped clean) just melts, but the back part of the lead does not
melt. At the same time he holds the above-mentioned strip in
the flame, so that drops fall from it on to the just-melted part
2f2
436 CONSTRUCTION OF THE LEAD CHAMBERS.
of the sheets, and the whole is united into a seam, b, all fusing
together into one mass. By a slight motion of the wrist the
plumber removes the flame for a moment, and the lead, which has
only just been melted, at once solidities ; in another second the
flame ia again directed upon the lead, and a new drop flows partly
over the firBt one ; so that at laat the whole seam takes the shape
shown in fig. 152.
Although all this is much more easily described than carried out
successfully, still the burning of horizontal seams is learned in a
comparatively short time, and can be done very quickly by a prac-
tised workman. In windy weather it is certainly much more
difficult, and in rainy weather it is not possible at all.
Fig. 162.
The burning of perpendicular (upright) joints is much more
difGcult, and, even in the hands of the most experienced workmen,
takes at least three times as long for the same length of seam
as horizontal buroiug, without ever being as strong as the latter.
This ia easily understood; for the melted lead, which quietly remains
lying on a horizontal sheet, iu upright burning at ouce runs down;
anil this can ouly be prevented in one way : the lead must be heated
exactly up to the melting-point, and the tlame instantly removed
till the seam has solidified ; and the burning must always be done
from the bottom upwards, so that to a certain extent the seam will
retain the drops of lead. In this case not much use can be made
of strips of lead for strengthening the seam.
A practised plumber can burn as much as 10 fee't upright or
25 feet horizontal joints in an hour; but such figures are only
reached in piece-work.
Recently the burning of lead has frequently been effected by a
pure oxykydrogetijlame, both oxygen and hydrogen being applied in
ERECTING A LEAD CHAMBER. 437
the compressed state^ contaiued in steel cylinders. The burning in
this way is done much more rapidly. Suitable burners are sold
by the Sauerstoff Fabrik, Berlin, O.
Water-gas can be employed for lead-solderiug without a blast
of air (or oxygen), but its poisonous properties must not be
forgotten.
7'he way of erecting a lead chamber in England is usually as
follows : — The commencement is made with the sides, for which the
sheets are made as wide as possible (most lead-rolling mills supply
them up to 7 feet 9 inches, some even wider), and so long that they
extend 4 inches beyond the height of the chamber, of course
taking into account that one side of the chamber is a foot higher
than the other. 6 inches are reckoned to turn over the crown-tree;
but 2 inches are saved at the bottom, because the lead afterwards
expands by the heat of the chamber.
Now, on the wooden floor before mentioned a wooden table (the
'' sheet-board '') is constructed, held together at the back by battens,
but quite smooth on the upper surface. It has the width of
two or three sheets of lead (that is, 15 feet 6 inches, or 23 feet
3 inches) and the height of the chamber — which, of course, can
only be done when (as is generally the case) the chamber is at least
as wide as it is high. On this table the sheets of lead are rolled
out flat, placed side by side, so that one overlaps the other 2 inches,
and burned together ; at the same time all the straps (of which we
shall speak directly) are burnt to the lead, which can be done
because the upper surface will afterwards be the outer one. The
upper edge is bent round the sheet-board, so as to hold it fast ;
and when everything is finished this end is wound up by a set of
pulleys, so that the sheet-board is raised along with the sheets of
lead, and lies flat against one side of the chamber-frame. Now the
upper edge of the lead is at once bent over the crown-tree and
nailed down, as well as all the straps. For this no cut or wire
nails are ever used, but wrought nails with broad heads (^' plate-
nails^^), about 1^ inch long, whose heads are all protected
against the acid by dipping a few at a time into melted lead.
When the lead has been completely fastened to the frame, the
sheet-board is lowered down, moved forward its own width, and
another piece of the chamber-side made upon it, till in this way
the chamber-sides and ends have been finished all round. Only
iu the corners it is preferable to use single sheets, which form
438 CONSTRUCTION OF THE LEAD CHAMBERS.
a rounded corner : this is much stronger than a sharp edge.
The object of the described process is this, to reduce the upright
burning to a minimum. It is much better than the former
plan of hoisting up each single sheet, turning its margin over the
crown-tree, and unrolling the sheet by its own weight. In this
case every single sheet had to be joined to its neighbour by upright
burning, and the straps had to be burnt on in an equally incon-
venient manner. If at all possible, the seams ought not to be
behind the uprights, so as to be better accessible for repairs ; and
for this reason also it is to be recommended to make the chamber-
frame as shown in figs. 144, 145, or 153, where the uprights do
not touch the lead at all.
The straps of the sides must be arranged according to the style
of the frame. If this only consists of uprights mortised into the
crown- and sole-trees, without any cross rails, the straps consist of
perpendicular pieces of lead nailed sideways to the uprights with
five leaded nails each. The strap ought to be long enough to turn
over the edge of the upright, so that two of the nails come to the
front (fig. 155, upper part). Such straps are placed alternately on
one and on the other side of the upright, one about every 4 feet.
These straps do not allow the chamber-lead to follow the changes
of temperature by extension or contraction. This easily leads to
deformation of the sides and tearing off of the straps; and it is
therefore better to avoid this, which can only be done by nailing
down the top strap in this way. For the lower straps longer
pieces of lead are burnt to each side of the upright, which meet on
its front, and are there joined by rabbeting (see fig. 155, lower part,
and fig. 156). There are no nails used here ; so that the lead walls
may move up and down the upright, whilst at the same time they
are all the more stiffened by being held fast in two places. This,
of course, takes more lead and labour than simple straps. In each
case the straps are about 8 inches in depth.
The object of keeping the lead clear of the wood, and of giving
it scope for expanding, is well attained in the form of strap shown
in figs. 153 and 154. The upright a is placed with one of its edges
pointing towards the chamber. The strap b turns round the edge
of a, and is fastened to it, not by ordinary nails, but by a broad-
headed pin c, which passes through a slit 2 inches in height. This
arrangement allows the strap to work up and down, as the chamber-
side expands and contracts.
440 CONSTRUCTIO.V OF THE LEAD CHAMBERS.
If the frame is provided with horizontal cross raib, odIt b
few upright straps are used — sometimes oooe, ooiy horizontal
ones, turned down over the rail, and nailed *
to it (fig. 157), two o£ 6 inches length for
each rail. This kind of straps protects the
chamber-sides much better against defor-
mation than the upright straps, and carries
the weight better upon the frame {this
is confirmed by information from Stolberg in 1902) ; it nlso
permits the lead to he kept further apart from the wood, since the
straps may leave about \ inch (not more) space between tlie lead
and the rails. The diagram shows this.
The chamher-sides can also {as at the Tbanu Works) be made of
horizontally disposed sheets of lead. The overlap in this case is
nailed to the horizontal cross rails in lieu of straps, as shown in
fig. 158 ; but first the whole height of the cbamber-sido is fiDished,
Kiji. 157.
the whole is rolled upr»n a ivooden roUei-, and allowed to unwind
itself by its own weight from the fop. In this way there is not so
much pull upon tiie seams as if the chamber were made of sheets
hanging down by their length, since each sheet is supported just
in the place where ihei-e would be a pull. This plan, indeed, seems
to be worthy of general recommenda-
tioQ ; for it saves the lead and lahour
of all the straps, and supports the
chamber very well.
At least as substantial is the plan used
at Aussig. There are no side-straps at
all ; but to each upright of the frame
corresponds a strip of lead burnt to the
chamber-side along its whole height,
probably the lap turned outside. This
is nailed sideways to the upright. Be-
tween this and the lead there is a wooden
lath, to increase the contact of air with
the chamber-lead as much as possible.
Fig. 159 shows this in horizontal sec-
tion.
FifT. 150.
Mr. Benker ascribes great advantages to his pefforated straps,
sliown in figs. 160 to 162. Fig. 160 is a plan, showing the
chamber-side a, the uprights b, the cross-bars c, the small wooden
bars d, and the straps e. The same parts are seen in vertical
section in fig. IGl. The chamber-lead is kept 2 or 2^ inches apart
from the cross-bars; the perforations of the straps (which may
extend the whole width of tl)c cross-bars, as in fig. 1G2) cause a
strong curreut of air to rise upwards and cool the lead, without
allowing any quantity of dust to accumulate on the straps. This
system is especially good for chambers which are driven hard for
" forced work."
lu the first-described case, now generally used in England, at
first only about a yard of the scams is burned, and that at the top,
so that the chamber can be covered in and the remainder can be
done at leisure in bad weather. The next thing, therefore, is the
chamher-top. For this we need a temporary scaftbldiug, movable
CnXSTRDCTlON OF THE LEAD CHAMBERS.
on wooden rollers, made of high trestles joined together at the top,
equal in height and width to the chamber, and in length to at
least two (or, better, three) sheets of lead. This scaffold is put
together within the chamber itself, its separate parts being got in
by bending back one of the side sheets. It is covered on the top
with a flooring of boards ; and upon this the sheets serving for the
chamber-top are flattened out. I'^ese are a little wider than the
chamber, so that they project 3 inches on each side. Thus they
do not project quite as far as the overlap of the side sheets
(6 inches), and there remains a joint Buitable for burning
(fig. 163, fl), which is burned very strongly. Now the sheets
themselves are joined by burning, and all the top straps are burnt on.
The latter, in England, serve for fixing the chamber-top from above
to the top joists carrying it. The latter, for a chamber 20 to 26 feet
wide, are 3 to 4^ inches thick and 10 to 12 inches high, and are
placed at distances of 14 to 18 inches from centre to centre. Their
length is at least suflBcient to reach to the outside of the crown-
trees; it is better if they even project a little beyond, to have a
good Bupport. The straps themselves are made ^ indies square,
and stand alternately on both sides of the top joists, about 3 feet
apart ou each side. At other works there are fewer but longer
straps. They are bent up and nailed to the top joists laid above
them on edge, with five leaded nails each. When all this has been
done, the top joists, by the help of the straps, carry the lead of
the chamber-top, and the joists themselves rest upon the side
frames, but separated from them by the overlap of the chamber-
sides. The joists should be well clear of the chamber-top (farther
than is shown in the figure), so that air can circulate between
lead and wood.
The top joists are protected from canting over by a few boards
nailed across them, which at the same time serve as a passage on
the chamber-top. Where the chambers are roofed in, longitudinal
sleepers may be laid on the top, joined to the top joists by iron
clamps, and the whole suspended from the timber of the roof
441 CONSTRUCTION' OF THE LEAD CHAMBERa.
v'hici) must be made strong enougli for this purpose ; but some
hold thnt even with roofed-in chambers it is sftfer to keep the
chamber-top independent of any movemeat of the roof-
Where the cliamljer is too wide for any single cross joists, two
lengths of these must be joined together and trussed, accordiDg to
the rules of carpentry ; in this case trussed girders may rua across
the width of the chamber, and the proper joists, to which the top
lead is fastened by straps, mn parallel with the long sides of the
chambers ; they are either mortised into the girders, or (which is
the stronger plan) they rest in cast-iron shoes bolted to the girders.
This, however, is only required for chambers standing in the open
air ; it is not very convenient, as the side frames have to be
weighted very much. Such wide chambers, as we shall see below,
have not altogether turned out well.
Fifi. 104.
Quite different from tlie just-described chamber-tops are ihose
found in many continental works. There are no wooden top joists,
but, in the place of these, thin iron rods about J in. thick, fastened
to the chamber-top by a lead covering burnt on each side to the
chamber-lead. These horizontal rods themselves are suspended from
the roofing by means of J-inch rods placed at short distances from
each other. This system cannot be employed for chambers standing
in the open air, as it makes the chamber-top dependent upon the
beams of the roof ; it is shown in fif;. IGJ-.
Another system, which may or may not be connected with the
roofing, is the following (fig. 165): — The chamber-side a is carried
CHAMBER-TOP AND -BOTTOM. 445
somewhat higher up, aud bends round an iron rod, b, | inch thickj
the part coming back over the iron being burnt to the other lead.
Here and there holes are left for the passage of the iron hooks, c,
which are bolted to BtroDg joists, d. The latter may form part of the
roof, or they may be supported quite independently on the crown-
tree e. The straps // hold up the cliamber-top (/, aud prevent it
from sagging ; the clear space between g and d is about 8 inches.
The object of this arrangement is to prevent all contact between
lead and wood even at the top edges of the chambers.
At the Grieeheim works, in Germany, the following very rational
plan of erecting lead chambers is followed : — On a staging of the
whole area of the chamber-bottom, but raised over its top, first the
chamber-ends are made; over these, without removing the ends,
the chamber-sides are made, firat one, then the other, and last of
all the sheets composing the top are laid down and burnt together.
Thus ultimately five layers of lead are lying on the staging one
above auotlier. Then the top straps are burnt on and are joined
to the top joists, which are put in their respective places. The
whole lead-work is now hung from sis differential pulleys, and the
staging is removed. As this is done, tlie ends and sides drop down
into their places, and need only be joined iu the corners, where
they are bent in at an obtuse angle. This mode of procedure
causes nearly all the burning to be horizontal, so that the work is
done more quickly, cheaply, and substantially.
The chamber-bottom is left to the last ; and it happens no doubt
very rarely (in England probably never) that, according to older
prescriptions, the bottom is laid down first aud protected by straw
and boards while the remainder of the chamber is being made.
416 coNsinircTioN op the lead ci
It is, on the coutrary, made last of all, but not always in the same
way. Id some works the side sheets are burned to it all round,
and openings are left iu a few places for drawing off the acid, for
taking sample.", &c. In the majority of works the bottom is
independent of the sides, and forms an enormous tank with turned-
up sides, into which the chamber-sides bang down, dipping into the
bottom-acid, and thus forming u hydraulic joint. This allows the
Fig. 166.
chamber-sides to expand and contract with the temperature, and
also makes the bottom-acid accessible from all sides, so that it is
generally preferred in spite of the larger expenditure of lead ; but
recently a good many works have adopted tlie first-mentioned plan
of making the chamber as a closed box, which saves both lead and
the trouble unavoidably connected with the second system. Often
the upstand, or "lag," which should not be less than 14 inches
CHAMBER-BOTTOM.
447
high, so as to afford a good deal of room for acid^ is made from a
narrow sheet of lead of douhle width, by bending up one half and
leaving the other half to form a portion of the chamber-bottom ;
the latter is then finished by burning it together with other sheets^
of lead. This is more convenient for the plumber than taking
sheets equal in length to the width of the chamber^ along with
the height of the upstand on each side. The latter must not be
left loose, because it is easily bagged out by the side pressure of
the acid ; but a 1-inch board is placed all round the chamber-
floor, over the edge of which the upstand is turned round and
nailed down. This is shown in fig. 166. Instead of a solid board,
it is preferable to employ merely a number of perpendicular or
horizontal rails, which admit the cooling-action of the air upon
the lead. At Stolberg the ^' lag ^' is made up to 2 ft. 6 in. high,
wdth a correspondingly strong plank to resist the side pressure of
the acid.
Fig. 167.
UUUUUUUUUUUUUULILJUUUUUUUUDnL?
In some works the bottom is divided into two, three, or four
parts by partitions the whole height of the upstand. The object
of this is, not to empty the whole chamber in case of repairs ;
but it is very rarely done, as this arrangement prevents a free
circulation of the acid, and as the bottom mostly suffers less than
any other part of the chamber — excepting through gross neglect,
by the formation of nitric acid, which ought not to happen at all.
448 CONSTRUCTION OV THE LEAD CHAMBERS.
Palding (Min. Ind. vii. p. 679 et seq.) gives details of chamber
construction which refer to the usual English plau^ mostly followed
in America as well. Fig. 167 shows part of the chamber-side
in elevation. The pillars (posts) a a are of 14 x 14 inches wood ;
corbels bb 14x14x5 in.; stringers c 14x14 in.; joists d
3x15 in., 16 in. centre to centre, 3 x 2 in. herringbone-strutting;
gangway-floor e with 2 x 12 in. joists. Of the chamber-Arame
itself, the sill /is 6 x 10 in., with a dowel-pin at each post and
toe-nail to each intermediate upright. The strong uprights (posts)
^y 6x6 in., 13 ft. 9 in. from centre to centre; the intermediates
hh 6x2 in., 33 in. centres; the bracing ii 6x2 in., with
lag-screw to each post and spike to each intermediate. Crown-
tree k 6 X 10 in., with lag-screw at each post and toe-nail to each
intermediate upright. Top joists II 3x15 in., 14 in. centres,
with three lines of solid 2x12 in. board- bridging.
Fig. 168 represents a portion of chamber-ceiling, seen from
above, and fig. 169 the same in sectional elevation on a larger
scale, which clearly shows how the chamber-lead m at the top ib
joined to the sides n and turned over the crown k, and how the
straps 0 0 (24 inches centre) suspend the top from joists / /.
Fig. 170 makes this clearer by a side elevation, and fig. 171 shows
the way the straps are cut from a strip of rolled lead.
Fig. 172 gives a plan-section of a chamber-corner, and fig. 173
a sectional elevation of the lower part of a chamber, showing how
the straps r r are fastened to the uprights g, hy and how the lead
side m and the saucer p stand off. Lastly, fig. 174 shows how the
upstand p of the saucer is turned over ledge * (2 X 1 in.), and held
at the bottom by another strip / (3x2 in.), spoked to the sill.
In England, where the winters are not severe, lead-chambers
are hardly ever roofed in, but are only built so that the rain-water
can run off as described above. But even then the space between
each two chambers must be covered by a light roof, and the whole
set must be surrounded by a wooden shed, because a gale might
tear the lead off the frames, or even throw down a chamber
altogether. These wooden houses have windows or Venetian
blinds, changed according to the wind. In windy places they are
always made first, as soon as the foundations and the frame are
finished, but before the lead has been fastened to the latter,
because during the building the incomplete chamber is even more
exposed to being thrown down by a gale than after completion.
EMGLISH CHAUBEB-PLANT.
rig. IGrt.
D
mr^v
[Ly^^J-
M
g Pig. i;o.
450 CONSTRUCTION OF THE LEAD CHAMBERS.
Thus the chamber-tops are exposed in England to the heat of
the sun in summer and to the snow in winter ; this is tolerable,
because neither of them occurs to an excessive d^ree. In the
less windy places even the chamber-sides are sometimes left without
protection against the weather, but never so in well-arranged works.
In the south of France, on the other hand, the chamber-tops are
always protected against the sun and the rain by a roof; but the
sides are generally exposed, which, on account of the heat of the
sun there, is certainly very wrong. In the north of France, in
Belgium, and in Germany the chambers are always completely
enclosed in buildings^ usually of a very light construction, and it
must be said that this would be decidedly preferable also in the
English climate.
Niedenfiihr recommends placing the chambers on brick pillars,
and filling up the spaces between these on the outside with a light
wall. The chamber-sides are surrounded by a wooden shed and a
light roof, employing roofing- felt as a cover for this. He reckons
a square foot of such a building, including the chamber-frame, to
cost from 3*. to 4*. 6d.
Renewal of the Chambers. — ^The greatest wear and tear is
experienced in the first chamber, more especially at the front end,
and, as some assert, even more so at the back end and the
immediately adjoining parts of the sides. Hence the first (leading)
chamber is often made of stronger lead than the others. Besides,
it must be noticed that any angular parts of a chamber wear out
more quickly than round or straight portions. The upright
corners are therefore always broken or rounded off ; but this is
not easily managed with the horizontal top corner. Hence, at
some w^orks they make the lead stronger in that place (p. 431).
The plan of making the roof partly slanting (p. 454) may do
some good in this respect as well, as this avoids a sharp corner.
The "curtain,^^ or part dipping in the acid, and alternately
subjected to this and to the action of the air, is also liable to
quicker wear. There is general agreement on the point that any
part of a chamber which gets hotter than the remainder will wear
out much more quickly ; and this should be guarded against in the
construction of the chamber-frame {supra, pp. 438, 441, & 453).
We have constantly laid stress on the fact that the lead should
be clear of the woodwork at all points, both because it is thus
longer preserved by the cooling-action of the air, and because it
RENEWAL OF CHAMBERS. 451
is accessible to the plumber. But this condition can^ of course^
be realized only^ for the sides and top^ not for the bottom.
Fortuimtely the latter suffers leasts being protected by the acid
itself and by a layer of sulphate of lead. If, however, a leak
occurs here after all, it is very awkward to repair. Sometimes
this can be done by measuring its distance from the sides, cutting
a hole in the chamber-top and dropping down a bucketful of
plaster of Paris or, preferably, of a mixture of fresh and burnt
pyrites-dust, which quickly hardens into a cake and may stop the
leak for years. But if this does not succeed, there is nothing
left but to stop and empty the chamber^ and to enter through the
manhole in order to get at the bottom.
A chamber will last very much longer if the frame be substan-
tially made, and the straps be well burnt on and nailed down and
numerous enough, so that they will not be readily torn off. Should
this happen, the mischief must be repaired at once : nowhere does
the saying come more true ^'that a stitch in time saves nine.''
If the repair is put off too long, the chamber-lead, pulled by its
own weight, wrinkles irregularly, and the chamber becomes unfit
for work much too soon. Especially those parts of the frame
most exposed to the action of the acid must be carefully looked
after, and, in case of need, at once repaired, before the parts of
the lead sides dependent upon them have lost their support and
have collapsed. This is most necessary at the junctions of con-
necting-pipes, at the places where the acid is siphoned off, &c.
The wind must also be kept off, and any loose pieces in the brat-
ticing round and between the chambers promptly put right; a
gale of wind may tear off the straps of a whole chamber-side at
once or force the frame to one side. The gangway round the
chambers ought to be wide enough (say 5 feet) to admit of easy
control and repair.
It used to be reckoned that with 6 lbs. lead in normal circum-
stances a chamber would generally last from eight to ten years, but
with many repairs during the latter years. But since the art of
building, and more particularly of managinff, vitriol-chambers has
become better understood, they have been made to continue much
longer in use. On the Continent, where they are not (or formerly
were not) so much strained as is frequently the case in England,
vitriol-chambers generally last much longer than the above term,
viz., 30 or even 30 years ; but even in England this is found to
2q2
452 CONSTRUCTION OF THE LEAD CHAMBERS.
be the case at some works where the chambers are built with more
regard to durability than to economy in first cost.
There is no doubt whatever that^ all other things being eqiial^
a chamber lasts longer in proportion as it is less heated ; it is
not so much the heat itself^ but the intensity of the chemical
reactions going on within the chambers which produces the heat,
and moreover the increase of the action of all chemicals by the
elevation of temperature brings about the same result. It is
only another way of stating this fact^ if we say that a chamber
lasts all the less time the more nitre is sent into it and the more
acid is made in it.
In the case of chambers without a roof the top generally wears
out firsts then the parts dipping into the bottom-acid and the
ends ; the bottom remains good up to the last^ unless nitric acid
gets to it, which most easily happens in the last chamber, if its
strength is allowed to run down too much.
When a chamber requires so much repairing aud patching that
it does not seem likely to pay, and when, after all, the escape of
the gas from the too numerous chinks and rents can no longer be
kept down, it is very bad economy not to pull it down at once ; for
the yield of acid must fall off very much. In this case a temporary
connexion is made between the two apparatus on either side of it,
the acid contained in the chamber is worked down as long as it
will run, a hole is cut into its side, and men provided with india-
rubber boots are sent in to shovel up the mud lying at the bottom
into a heap, from which a good deal of acid is still obtained by
draining. The mud must now be removed ; if the space under-
neath is free, a receptacle is formed by low banks of clay, a
hole is cut in the chamber-bottom, and the mud pushed down.
If this is not possible, it must be removed in a much more
troublesome manner, by thickening it with sawdust and washing
with water. In either case it is dried in a reverberatory furnace,
sometimes with the addition of a little lime in order to prevent
the escape of acid vapours. Notwithstanding this, the operation
usually causes a very disagreeable stench, probably owing to
arsenic, selenium, &c. The dried mud, principally consisting of
lead sulphate, is either smelted for lead in a small cupola heated
by coke, or simply sold to the lead-workers.
After taking out the lead-mud, the chamber-lead is detached
from the frame, and the good whole pieces rolled up for use as
sheet-lead ; the others are melted in an iron pan^ the dross is
RENEWAL OF CHAMBERS. 453
scummed off^ and the lead cast in the usual pig -moulds; at the
lead-rolling mills this lead is much liked for other chemical
purposes (see p. 431). Including the pig-lead, the dross, and the
lead sulphate, usually nine tenths or upwards of the original weight
of the chamber is recovered ; the remainder has disappeared in
one shape or another \rith the acid made.
If the frame has been substantially made, it stands a second,
sometimes a third lead chamber, with a few repairs, putting in odd
beams &c. Of course, in case of any doubt, it would be extremely
bad economy to run the risk of having to stop a chamber because
its frame would not hold out as long as the lead.
Mr. G. E. Davis has sent me the following observations: —
A set of three chambers (20 x 18 x 120 feet) had been at work at
high pressure for seven years, when the first two chambers were
pulled down. They were built of 7 lbs. lead ; at the end of the
time the top was still between 5 and 6 lbs., the bottom between
2 and 3 lbs., the sides were almost nil. In the first chamber
sulphate of lead equal to 19 tons metallic lead was found, in the
second 16 tons metallic lead. The chamber-top had not been
repaired all this time, the sides had had new lead all round, and
the bottom had been repaired in places.
Special observations on the wear and tear of lead chambers have
also been made by Burgemeister (Chem. Zeit. 1889, p. 1633) . A set
of two chambers was observed after 23J years, during which time
the larger chamber had been at work with brimstone for 32 months,
with pyrites (first Westphalian, then Rio Tinto) for 178 months,
idle for 75 months. The thickness of the lead was originally
2*57 mm., at the end of the period on an average only 1*88, that
is a loss of 0*69 mm. or 26*8 per cent. The part dipping into the
bottom-acid was most worn ; next the places where the lead was
double or where it was protected against cooling by the wooden
frame. For this reason it is best to burn the joints outside, because
the inner part of the lap-joint is then eaten away first without
injuring the joint ; if the joint is burned inside, the lap is loose
on the outside, and as soon as the inner part is eaten away the
chamber must leak. The bottom of the chamber, which is pro-
tected by the acid, suffers least. A small chamber which was
placed between the Glover tower and the large chamber, and which
was kept at a higher temperature (from 65° to 90P C), had lost in
120 working months as much as 17*65 per cent, of the thickness
of lead.
■W* CON8IEOCTIOX OP THE LEAD CHAMBERS.
Shape of Lead Chancers.
Tlie sAape of the chaiDbers is nearly always that of a long bos
of square or approximately square transverse section. At some
places, in order to save lead, the chambers have been made up to
60 feet wide; but this is not to be recommeoded on any account.
It causes difScuIties in coustmcting the wood frame, and, what is
more serious, the yield of acid in such large chambers is not so good
as in those of ordinary shape, say between 20 and 30 feet wide.
This is easily understood, as in such very large sections the gases
do not get properly mised, and there are too few surfaces offered
for contact and cooling (comp. Chap, VII,),
For the purpose of saving lead, the chambers belonging to the
different works of the Rhenania Chemical Company are constructed
in the way illustrated in fig. 176 — that is, with the top comers cut
Fig. 176.
off, to suit the slope of the roof. This admits of putting the
largest possible height of chamber into a roofed building; and
Mr. Hasenclever also contends that the " dead corners " of square-
sectioned chambers are thereby avoided. There is also less wear
SHAPE OF LEAD CHAMBERS. 455
and tear than in the sharp corners of chambers of the ordinary
square section (comp. p. 450),
The usual width of vitriol- chambers is rarely below 20 or above
30 feet ; their height varies from 16 to 25 feet, or exceptionally a
few feet more. Their length (always speaking of the principal
•chambers, not of the small chambers or '' tambours " arranged
before and behind these in the French system) is rarely below
100 feet, but may attain 200 or even 300 feet.
H. A Smith, in a pamphlet on the Chemistry of Sulphuric-acid
Manufacture (1873), endeavoured to prove that the upper space of
the vitriol-chambers did no work at all, and that chambers of 3, 6,
or at most 8 feet in height would be most suitable. His experi-
ments (described and refuted in detail in our first edition, p. 285
et seq,) were decidedly inconclusive, and a practical test of his
theory at the Oker works led to its entire rejection.
E. and T. Deplace (E. P. 5058, 1890) describe an annular
chamber, in which the gaseous current is continually changing
its direction, owing to that shape. Siphon-shaped tubes placed
on each side of the chamber produce a circulation and mixture of
the gases. These chambers occupy less space than the usual form,
and are stated to produce up to 6 kils. acid of 52°B. (=3*7 kils.
H2SO1) per cubic metre = 1 lb. sulphur to 13*2 cub, feet per
24 hours. A few sets of this kind have been erected in France
and England. According to the 28th Alkali Report, p. 55, the
shape of chamber actually built differs a good deal from that
described in the patent. According to information received from
manufacturers, the production from these chambers per cubic foot
does not exceed those of ordinary chambers.
Th. Meyer's ''tangential chambers'' (E. P. 18,376, 1898)
are also devised as a means for inducing a better mixture of
the chamber-gases. The chambers should have a circular or
polygonal section, and the gas inlet-pipes be placed tangentially
on the upper part of their sides, the outlet-pipes in the centre of
the bottom ^, This imparts a spiral motion to the gases, rapid at
the circumference, slower towards the centre, and thus causes
them to travel through a greater distance and to get much better
mixed than in ordinary chambers. The inventor gives more details
* The patent specification speaks of the centre of tlie ceiling j but in practice
ihe gas-exit is from the centre of the chamber-6o^^om.
456 CONSTRUCTION OF THE LEAD CHAMBERS.
concerning his system in Chem. Zeit. 1899, p. 296 ; Zsch. augew.
Ch. 1899, p. 656, and ibid. 1900, p. 739. His system has been
carried out at the Norddeutsche Chemische Fabrik, Harburg, and
at the Chemische Dungerfabrik, Rendsburg. The chambers are 10
metres in diameter and 8 metres high, in a set of three, with Glover
and Gay-Lussac towers. The draught is very good, the yield from
92-5 to 95-2 of the theory (291 H2SO4 per 100 S burned) ; per
cubic metre the production is from 366 to 3*87 kils. H2SO4,
with a consumption of 1*34 to 1-44 nitric acid 36° B. ( = 1*0 to 1*07
nitrate of soda) for 100 H2SO4, or, say 3 NaNO» to 100 S burned.
[This production is good, but still inferior, not merely to French
*' forced work '^ with ordinary chambers, but also to that of some
of the German works, comp. p. 468.]
Recently (Zsch. ang. Ch. 1900, p. 742) Meyer has improved his
chambers by arranging in the first (and hottest) chamber a cooling
system, consisting of 43 lead pipes, 2 to 2^ inches wide, suspended
in water-lutes from the chamber-top all round the circumference
and reaching 8 to 10 feet down into the chamber. They are
closed at both ends ; through their tops enter thin lead pipes,
reaching nearly to the bottom of the larger pipes, for introducing
the cooling- water, as shown in fig. 177. The whole offers a
cooling-surface of 23 square metres ( = 250 square feet), i. c. 7 per
cent, of the heat-radiating surface of the chamber sides and ceiliog.
The water, of which 8^ tons is used per day, issues at a temperature
of 67° C. The heat evolved by the process of converting SO*,
O, and H2O into H2SO4, as far as it goes on in that chamber, is
calculated =2^ millions metrical heat-units per 24 hours, of which
500,000 are removed by the cooling-water =20 per cent. This
is not shown by the chamber thermometers, as the loss of heat is
made up by that newly generated by the chemical process, and it
is manifested by the increased production of sulphuric acid
(according to the theories of Lunge and Sorel, comp. later on).
The pipes last a long time and can be immediately renewed by
taking them out of the hydraulic seals. The hot water is used for
feeding the steam boilers.
I have shown in Zsch. f. angew. Ch. 1902, pp. 151 et seq., that
the advantages to be realized by Meyer's proposals by his own
showing are not very considerable. It is very doubtful whether
for equal weights of lead his tangential chambers produce any
more than even moderately well-managed ordinary chambers.
MEVER a TAKGENTIAL CHAMBERS.
Ill another paper (Zsch. augew. Chem. 1901, p. 1245), Meyer
acknowledges the " Lunge towers " as the best solution of the
problem of bringing the misty particles floating in the chamber to
Fig. 177.
act upon each other, and he advises to combine such intermediate
towers with his " tangential " chambers.
I am informed that in 1903 seiious attempts had been-made to
introduce circular chambers into England, but after inquiry
and consideration this plan was given up.
458 CONSTRUCTION OF THE LEAD CHAMBERS.
Combination of chambers to sets. — Sometimes the whole working-
space is contained in one chamber. Scheurer-Kestner (Wurtz,
Diet. iii. p. 147) mentions a single chamber of 142,000 cubic feet
<;apacity, and quotes the experience of different works, according
to which it is quite unnecessary to divide the set into several
chambers. More frequently, however, several chambers are com-
bined to form a set, which, to begin with, affords this advantage
—that for repairs it is not necessary to stop the whole set.
A great diversity of opinion exists as to how the single chambers
are to be combined to form sets. Among the hundreds of
vitriol. works very few will be exactly alike in this respect ; and
frequently even in the same works different combinations are
found. We may, however, consider it as established that it is
almost indifferent in which way the chambers are combined, if they
are, in the first instance, properly built (that is, not too high or
wide), and if, secondly, they possess a certain cubical capacity for
the quantity of sulphur or pyrites to be consumed. Within these
limits those combinations are best which require least lead^ and
which are laid out so as to afford the greatest facility for super-
vision. Of course there is also an extreme limit to the capacity
of the whole set ; but opinions differ upon this point also. At some
works a set consists of nine or eleven chambers of 35,000 cubic
feet each; at others, equally large, it is limited to three chambers
of 42,500 cubic feet each, &c. Thus at Hebburn-on-Tyne several
sets of three chambers each are employed, each chamber 20 feet
wide, 125 feet long, 17 feet high on one side, 18 feet on the other;
each set serves for 18 burners, burning 7 cwt. daily. At Gates-
head there are several sets of three large chambers^ each so
arranged that two of them communicate separately with a set of
kilns (''working-chambers'^), and both of them are connected
with the third chamber ; the whole set has a capacity of about
200,000 cubic feet. More usually the gas passes through all three
chambers in succession. Very often four chambers are combined,
each of them about 20 x 20 x 130 feet, — or five chambers, two of
them working-chambers, thus, ^ 4/^9 — o' even six chambers.
In America (Zsch. f. angew. Cliem. 1894, p. 133) I found at
one place a set of twelve chambera of equal size, each 24 feet
SETS OF CHAMBERS. 459
long, which, it was asserted, combined very good yield with small
oonsuroption of nitre. At another place they had three chambers,
60, 50, and 40 feet in length, with plate-columns between.
The chamber-set described by Hasenclever (Chem. Ind. 1899,
p. 26) consists of two very large and two small back chambers
of a total capacity of 7250 cub. met. =267,C00 cub. feet. He
mentions the existence of sets comprising 12,000 cub. met.
=420,000 cub. feet, which form decidedly a very undesirable
exception.
At one of the most modern German works I found sets con-
sisting of three chambers, all of them 10 metres wide and 7 metres
high ; the first had a length of 41, the second of 31, the third of
10 metres. They produce 2*8 kils. H2SO4 per cub. metre (equal
to about 18 cub. feet chamber-space per lb. of sulphur burnt in
24 hours).
The Rhenania Chemical Co. prefers throughout sets of two
equally large chambers, followed by two small back chambers
(1902).
In France it is usual to combine three chambers in a set,
exceptionally four or five. The total capacity of a set hardly ever
exceeds 6000 cub. metres (=210,000 cub. feet).
According to Mon. Scient. 1900, p. 563, Benker has for a
number of years built chambers (in France) on the following
plan : — Besides the Glover tower he employs a dry filtering chamber
in order to retain flue-dust and arsenic. His chambers are three
in number, of a total capacity of only 2000 to 3000 cub. met.
(70,000 to 105,000 cub. feet). Into the last of these he introduces
SO2 (burner-gas) ; then comes a small intermediate chamber and
then two Gay-Lussac towers. The draught is produced by a fan-
blast. Benker claims to produce 6 to 7 kils. acid of 52° B.
( = 3'7 to 4*3 kils. H2SO4) per cubic metre ( = 1 lb. sulphur burnt
upon 13*2 to 11*3 cubic feet !), with a consumption of 0*8 to I'O
nitric acid 36° B. per 100 acid of 52° ( = 3 to 37 lbs. NaNO^ per
100 S burned). The chambers are fed with a water-spray in lieu
of steam (vide infra), and he always injects some SOj (burner-gas)
into the last chamber (G. P. 88,368 & 91,260 ; comp. later on).
Benker, according to direct communications received from him
in 1902, employs only narrow chambers, say 18 to 20 feet
wide, and from 25 to 33 feet high. Such chambers are, in the
first instance, better adapted for water-spraying, but they also
460 CONSTRUCTION OF THE LEAD CHAMBERS.
afford a better mixture of the gases, especiallj if the sides are
cooled, by employing an open, bratticing for the side passages
(2^ in. laths with 1^ in. clear spaces) and a roof -rider. The
cold gases descending along the sides must rise again iu the
centre ; but in the case of very wide chambers a dead space
remains where the velocity is very slight, and where the mist of
nitrous vitriol sinks down without acting on the gaseous con-
stituents. This cannot produce '^high-pressure work'' (comp.
later on). Benker objects to Meyei-^s tangential chambers that
there is no question of tangential action, that in the centre of
these circular chambers gases of very different concentrations get
mixed up, and that they would be too expensive if the only
proper way was followed, viz. building many small circular
chambers in a set. We shall later on, when describing the
system of water-spraying, give a complete diagram of Benker's
chambers.
The size of the chambers varies very much. Apart from the
'' tambours " of the French system^ the proper chambers are
made with as little as 10,000 and as much as 110,000 cubic feet
capacity. Such small chambers are no longer built as main
chambers ; the usual capacity of these may now be taken as
ranging from 25,000 to 70,000 cubic feet, more frequently nearer
the upper than the lower limit. Smaller chambers cost ranch
more, comparatively, than large ones, and it is doubtful whether
they afford any corresponding advantages.
The different chambers of a set are either placed on the same
level, or, more suitably, each following chamber is placed 1 or 2
or, better, 3 inches higher than the preceding one, so that the
acid of the back chambers can be run more easily into the working-
chamber. In the first chamber the acid is both strongest and
most free from nitre ; and it is therefore preferable to draw off any
acid from this, whether it be for sale, for use, or for concentration.
The acid drawn off is replaced partly by that newly formed in the
same chamber, partly by the weaker acid run over from the other
chamber. If there is only one long chamber, the acid is always
found strongest near the entrance ol* the gas.
In England, all the chambers of a set are generally of equal size^
apart from local circumstances ; and this plan is now more fre-
quently found on the Continent as well than formerly, when the
French system, even now the more usual, was the only one to
SIZE OF CHAMBERS.
461
be met with. In this system there is a "large chamber/' C
(fig. 178), placed at the lowest level, combined with a few small
chambers at a higher level, both before and behind the large one.
Thus the first small chamber or '' tambour" A, serves for deni-
trating the nitrous vitriol by hot water ; the second one, B, for
introducing fresh nitric acid ; the third and fourth tambours, E
and F, for finishing the reaction.
In the south of France (Favre, ' Monit. Scient.' 1876, p. 272)
there is mostly a large chamber of 135x26x20 feet, or of
100 X 16 X 22 feet, combined with two small chambers, together
¥\
i«?-l
78.
A
B
about 140,000 cubic feet. At Aussig each large chamber is
200 feet long and 24 feet wide, and is combined with a small
tambour for catching the fine dust, and two small end chambers,
not receiving any steam, but only serving for cooling the gas pre-
viously to its entering the Gay-Lussac tower.
Some manufacturers reject all preliminary chambers (tambours)^
because the chamber process is carried on best if a large space is
afibrded at once for the mutual reaction of the gases. Thus in a
462 CONSTRUCTION OF THE LEAD CHAMBERS.
large French works two thirds of the whole chamber-space are
occupied by the first chamber, two ninths by the second, and
one ninth by the third ; this system is also adopted at Uetikon.
At the Government works at Oker (official commanication,
1902) there are five sets of chambers, the best working of which
have the following dimensions : —
Chamber I. 35 met. long, 800 m. wide, 6*50 m. highs 2800 cub. met.
„ II. 30 „ 8-00 „ 6-50 „ =1560 „
„ Ur. 3015 „ 5-49 „ 6-25 „ = 869
5289
The chambers communicate by pipes 1 met. wide, entering
about f^ of the height and leaving 1*20 above the bottom. Small
front and back chambers have been designedly left out in this
system, but small back chambers exist by chance at the older sets.
Small chambers, of course, require more lead and space than
large ones of equal capacity. A preliminary chamber, however, is
serviceable, where no Glover tower is present, for catching the dust
and cooling the gas, so as to save the large chamber. For the
same reason the nitric acid was once usually introduced into
a special tambour ; but it is best, as we shall see, to run it down
the Glover tower. •
A small chamber at the exit end is certainlv serviceable for
drying the gases previously to their entering the Gay-Lussac
towers in cases where there is no long tube or tunnel for the
above-named purpose.
The waste of lead in small chambers is more easily understood by
a definite example : — A chamber of 100 X 20 X 20 feet has a cubical
capacity of 40,000 cubic feet and a surface of 8800 square feet. A
tambour of 16 X 10 x 10 feet has a capacity of 1600 cubic feet and
a surface of 840 square feet. Its contents are therefore ^, but
its surface almost ^^ of that of the large chamber; and conse-
quently its surface is nearly 2^ times as large, in comparison with
its capacity, as that of the large chamber.
Whilst, of course, there is no doubt that a given cubic space
of chamber-room is more cheaply obtained with a few large than
with a greater number of small chambers, it is, on the other band,
very easy to overstep the mark in this direction. We have seen
above that, in the case of chambers of an excessive section, the
gases do not get properly mixed ; but the same principle applies
SIZE OF CHAMBERS. 463-
evea to the division of the chamber-space in the direction of its
lengthy since every time the gas has to be compressed into a
comparatively narrow connection-tube in order to pass from one
chamber to another^ this must bring about a good mixture supe-
rior to that produced in the same length of undivided chamber-
space. For this reason, to begin with, it seems expedient to
subdivide the chamber-space by multiplying the number of cham-
bers ; and we shall further on meet with another strong reason
for the same purpose, namely, that the cooling down of the con-
tents of the chamber, essential for the reaction among them, is
promoted by their contact with the comparatively cool end-walla
of the chambers.
In England it is taken as a practical rule that for every cubic
foot of chamber-space there should be about 0*2 foot of total
surface (top, bottom, sides, and ends). A chamber 20 x 25 x 100
feet would contain 50,000 cubic feet and have a total surface of
10,000 square feet, which is exactly the above-stated proportion.
Sets of chambers in England are rarely made larger than
200,000 cubic feet ; if more is required, the whole is broken up
into two or more sets.
When speaking here, and elsewhere, of '^ chamber-gases," we
always comprise in them not merely the vapours of water, nitrous
anhydride^ &c., but also the misti/ particles of liquid sulphuric
acid, uitrosulphuric acid, &c.^ floating about in the atmosphere of
the chambers.
Schertel, in fact, starting from the principle adduced by my
experimental and theoretical basis, to which he agrees, proposes to
multiply the number of chambers, keeping them rather short
(Chem. Ind. 1889, p. 80). Bode (Zeitsch. f. angew. Chemie, 1890>
p. 11), on the same principle, proposes chambers of half the usual
length, but twice the ordinary width — say 40 feet. This would
involve some difficulties, although not insuperable, in constructing
the chamber-frames. Later on (Sachs. Jahrb. 1890, p. 148)
Schertel described practical experiments bearing out the theo-
retical considerations just mentioned ; and further experiments on
the manufacturing scale^ entirely confirming my own results and
conclusions, have been made by Retter (Zeitsch. f. angew. Chem.
1891, p. 4).
In the usual case, where a set of several chambers is combined
to form the acid-making apparatus, the question arises how the
464 CONSTRUCTION OF THB LEAD CHAMBERS.
single chambers of the set are to be connected. One thing about
this is certain : that the connecting-tubes must be placed at the
small ends^ so that the gas shall travel right through the length
of the chambers, and no dead corners are left. But the next question
is^ at what part of the section the connecting-tubes are to leave or
enter the chambers. There is general consent as to this^ that the
gas should enter the first chamber near its top. Some proceed in
this wa^ : they take the gas away at one end near the bottom
and introduce it into the next chamber near its top. Others main-
tain just as strongly that this is wrongs and that; on the contrary^
the gas-pipe ought to leave each chamber near its top and enter
the next chamber near its bottom. Others^ RgRin, contend that
it matters very little where the gas enters and leaves^ and that
it is therefore the simplest plan to make straight connecting-
tubes about midway in the height of the chamber. This last
view seems to be borne out by the practice of several practical
men of very large experience, and it agrees very well with the
investigations of Lunge and Naef {vide infra), who found that the
composition of the chamber-gases in any given cross-section of the
chambers does not diflFer very materially between top and bottom, so
that it must be indifferent where the connecting-tubes are placed.
This is confirmed by information from the Rhenania works in
1902.
The connecting-tubes may be round pipes or angular flues
(tunnels). The former are preferable, because they can be made
without a frame, and because they stand better. They must,
however, be made of strong lead, say 9 to 12 lbs. per square foot,
and bound here and there with iron hoops, between which and
the lead wooden staves are placed in order to keep the pipes in
shape ; but if the weight of the lead amounts to 15 lbs. per square
foot, no staves are needed. Figs. 179 & 180 will make this
clearer.
The iron hoops serve also for suspending the pipes from beams
&c. The width of the pipe introducing the gas into the first
chamber, whether it comes from the Glover tower or from the
burners, must be adapted to the quantity of gas conveyed. For a
combustion of 7 tons pyrites daily a pipe of 2 feet diameter, for
9 tons one of 2^ feet, upwards of that one of 3 feet diameter will
do ; more than 10 tons are rarely consumed for a single set in
twenty-four hours. Since the volume of the gas decreases in its
CONNECTINQ-PIPES. 465
onward journeyj the eounecting-pipes between the single ehatnbers
may be successively a little smaller; but it is not well to grudge
any thing here, since no harna is done if the pipes are too large,
but very much if they are too small.
Fiy. 179.
At Griesheim several coiiucctiug- tubes are introduced between
the cliambers instead of one. This seems very rational, and at
the same time serves for partially cooling the gases in their transit,
which we shall find further on to be an important feature. " Dead
corners " are most easily avoided by this plan.
The total cubical contents of a set of chambers must bear a certain
proportion to the quantity of acid to be produced, several special
circumstances modifying that proportion. Thus it is certain that
for pyrites more chamber-space is needed than for sulphur; wc
have seen above (p. 400) tliat the relative proportion may be stated
VOL. I. 2 b
466 CONSTRUCTION OF THE LEAD CHAMBERS.
as 1 : 1*314. But now the question is. What is the absohite amount
of space needed ? Properly speaking, the connecting-pipes, if they
are of great length, and the Glover and Gay-Lussac towers should
also be included in the calculation, and that to a larger extent
than coiTCsponds to their cubical contents.
The consumption of nitre also influences the chamber-space ;
within certain limits a larger consumption of nitre may compensate
for a smaller space.
Partly from this the widely divergent views on this point may be
explained, but not entirely ; for some manufacturers obtain about
the same yield as their neighbours possessing half as much more
chamber-space, although both the pyrites and the general con-
struction of the plant and their consumption of nitre are as nearly
as possible the same. In the following remarks we shall reduce all
measures to cubic feet of chamber-space required for burning 1 lb.
of sulphur daily, taking, in the case of pyrites, the sulphur bought,
not that actually burnt.
For some particulars concerning older methods, comp. 2ud edition
of this work, pp. 371 & 372; for recent ones, pp. 456 & 458.
From sundry English alkali-works I can state the following pro-
portions (1879): —
I. II. III. IV. V.
28 25 20 18 16 cubic feet.
I. and II. were considered too high by the chemists of the respective
works themselves ; but it should be stated that the same space was
employed in 1864, when 30-per-cent. Irish pyrites was used, for
which it was more suitable. III. (viz. 20 lb.) is a proportion era-
ployed at many large works ; but IV. and V. are found in works
having as good a yield of acid (270 to 288 o.v.) and no larger
consumption of nitre (3^- to 4 per cent.). In all cases rich Spanish
or Norwegian ore was burnt, and both Gay-Lussac and Glover
towers were used. From this it follows that under the same con-
ditions 20 cubic feet per lb. of sulphur charged is amply sufficient,
and 18, or even 16, will do; but the latter certainly is generally
assumed to be the lowest allowable limit. This agrees with a
statement of Wright's (Chem. News, xvi. p. 94), who demands 16
to 19*2 cubic feet.
From the Inspectors' Alkali Reports it will be seen that the
amount of chamber-space actually employed at English works
varies in a most extraordinary way, and not merely in consequence
TOTAL CHAMBER-SPACE.
467
of the fact that very small works generally employ an excessive
chamber-space. It is also seen from the same source that the
usual assumption that less chamber-space is used with brimstone
than with pyrites is altogether erroneous. We will here give
merely a few figures obtained by taking averages of the single
works enumerated, leaving out those burning both pyrites and
brimstone, or coal-brasses, or " oxide.^'
In the 20th Report, pp. 48 & 49, we find the average of 18
works burning pyrites to be 23"1 cubic feet of chamber-space
(minimum 15*5, maximum 38*4) ; the average of 10 works burn-
ing brimstone 29*7 cubic feet (min. 21*7, max. 44*8). In the 21st
Report, pp. 20 & 21, 21 works burning pyrites average 26*6 cubic
feet (16-40) ; 15 works burning brimstone average 26*2 cubic
feet (21-48). On pp. 64 & 65, 22 works burning pyrites average
29*2 cubic feet (17*3-43'2) ; 18 works burning brimstone average
31'4 cubic feet (19*3-46*2). But as the great majority of these
works are too small to afPord a real guidance in this respect, we
will quote in detail (from 21st Report, p. 81) the figures of fifteen
medium and large-sized works in the Widnes district, comprising
some other interesting information :—
1
1
Pyrites burnt
per week.
Tonp.
175
52
Oubio feet oham-
ber-spaoe per lb.
1
i Nitrate of soda
per cent, of
sulphur burnt.
1
Capacities of Gay-
Lussac towera per
TotAl acidity of
waste gases as
of S burnt in
24 hours.
ton pyrites used
per week. Cub. ft.
grains SO, per
cub. ft.
210
3-50
360
180
500
65-8 [ 0-87
i 210
17-8
4-70
18-7
2-10
i 125
28-0
4-00
361
0*65
98
17-8
4-20
32-4
2-88
240
210
426
24-0
1-71
250
28-3
3-75
15-7
2-34
150
21-0
37-8
0-79
250
19-3
6-00
20-5
1-90
60
22-3
27-5
389
260
22-0
3-30
33-7 1-60
117
210
400
53-5 i 1-30
183
200
21-5 2-94
70
17-5
79-6
0-70
Total 1
.
and 2590
21-0
417
44-4
1-82
averages
The usual proportions in the south of France were stated by
Favre (Monit. Scient. 1876, p. 271) as follows : — Each square
2h2
468 CONSTRUCTION OF THE LEAD CHAMBERS.
metre of grate-surface in the pyrites-burners daily receives 270
kilograms of 40-per-cent. pyrites^ and corresponds to 180 cubic
metres of chamber-space. This means 1*66 cubic metre for each
kilogram of sulphur charged^ or 26*5 cubic feet per lb.
In the north of France 1 found, in 1878, about 8 cubic feet
per lb. of pyrites, or about 17 cubic feet per lb. of sulphur chained,
with good yields and low consumption of nitre, but only for low
or medium temperatures; in summer ^ to ^ more chamber-space
is required.
Recently a new style of working has been introduced into several
French works, called '^production intense/* say " forced or high-
pressure work.'' It consists in supplying the chambers with a
greatly increased stock of nitre, without losing any of it, by means
of largely increased Gay-Lussac and Glover towers ; in this way
the production may be increased to almost twice the usual amouut^
so that, in winter at least, a maximum of yield and a minimum
consumption of nitre are attained with the extremely small chamber-
space of 0*7 cubic metre per kilogram, or 11*2 cubic feet per lb.
of sulphur burnt. We shall have frequent occasion in later parts
of this book to speak of this '' forced style,'' which at the time of
writing is nowhere practised in England or Germany.
In 1 900 Pierron (Monit. Scient. 1900, p. 367) stated as a minimum
production in 24hours,2"34 kils. H2SO4=0'78 kils. sulphur burned
per cub. met.=:l lb. S per 19 cub. feet, but at Euhlmann's works
the normal production is 2*9 kils. H2SO4, and Benker (p. 459)
claims obtaining 3*6 to 4"2 kils. H2SO4 with ordinary chambers.
The ordinary production can be increased by the use of artificial
draught (fan-blasts), by ^^ tangential chambers " (corap. p. 456), by
'^ plate columns," and by other means described in their places.
For Germany, Niedenfuhr (1902) states the usual cubic space
= 1*2 cub. met. per kil. of sulphur = 19 cubic feet per lb., which
is just the same as the maximum space allowed in France, but
decidedly more than the average employed in that country, even
where no *' intense production " is aimed at. But Niedenf iihr's
statement is decidedly not valid for the most carefully managed
German works, which, according to direct information, manufacture
3*5 H2SO4 per cubic metre of chamber-space, which is about 0*85
€ub. metre per kil. of sulphur burnt or about 14 cubic feet
per lb.
For the Bhenania works, producing their SO3 from zinc-blende,
J
TOTAL CHAMBER-SPACE. 469
Dr. Hasenclever states as the normal production 2*5 kils. of
60° B. per cubic metre in 24 hours, with a consumption of 0*5 to
1*0 per cent, nitre per 100 acid of 60° B.
The following (hitherto unpublished) data have been kiudly
supplied to me by Mr. 6. E. Davis respecting the results obtained
in a set of three chambers, each 120 x 20 x 18 feet, with a Glover
8^x22 feet and Gay-Lussac 12x12x60 feet. This set was
worked : — A. One month without the towers : pyrites burned
135 tons, potted 96 ewt., 25 cb. ft. chamber-space per lb. of sulphur
in 24 hours ; 79 lbs. nitrate of soda sent into the chambers per
ton of pyrites. B. One month with the towers in full operation :
pyrites burned 180 tons (19 cb. ft. chamber-space per lb. sul-
phur), 101 cwt. NaNOa introduced by Gay-Lussac acid, 56*2 cwt.
nitrate potted, 98 lbs. nitrate introduced into chambers per ton
of pyrites. C. One month with towers : 240 tons pyrites
( = 144 cubic feet per lb. sulphur), 205*6 cwt. NaNOs in Gay-
Lussac acid, 86 cwt. nitrate potted, 135 lbs. nitrate in gases per
ton of pyrites. D. One month with towers : 302 tons pyrites
( = 11*5 cub. ft. chamber-space per lb. S), 277*3 cwt. NaNOg in
Gay-Lussac acid, 135*1 cwt. potted, 152 lbs. total nitrate per ton
of pyrites. Lastly, E. One month with towers : 380 tons pyrites
(^8*1 cb. ft. chamber-space per lb. S), 394 cwt. NaNOs in Gay-
Lussac acid, 203 cwt. nitrate potted, total nitrate 176 lbs. per
ton of pyrites. Under the last conditions the chambers were
worked for nearly a year; oxygen at burners 10 per cent., at
dust-burner 12 per cent. Average acidity of gases going into
Gay-Lussac 5*5 grains H2SO4 per cubic foot, and gases leaving
the tower 1*4 grains H2SO4.
The yield varied very little in all these cases, viz. from 40*2 cwt.
acid of 123° T. per ton of pyrites in A to 39*8 cwt. in B. (The
above quoted consumption of nitre is very high, even for easy
work, let alone for high-pressure work.)
Mr. Davis states the general opinion of English acid-makers
as follows : — " If you go on in the old way, working with 25 cubic
feet of space per lb. of sulphur per 24 hours, the process goes on
absolutely by itself. Decrease your chamber-space to 15 cb. ft.,
and you want a chemist and clever foreman, while with 8 cb. ft.
neither foreman nor chemist knows what peace is either by night
or day.^*
The preceding statements refer to ordinary pyrites, but when
472 CONSTRDCTION OF THE LEAD CHAMBERS.
acid in a given chamber- space than usual is the employment of a very
large quantity of nitre {intense production^ or high-pressure work) .
This means providing the chambers with unusually large Gay-
Lussac and Glover towers, which absorb a considerable portion
of the saving in the size of the chambers themselves. Still, this
style of work has obtained much favour in France, where it has
led to a sensible reduction of the chamber-space at some works,
as already mentioned (p. 468). But it is evident that the limit of
reduction by that means is soon reached.
A third way of increasing the production of acid for a given
chamber-space is by contrivances for a better mixture of the gases
within the chambers. Some of the proposals in this direction are
combined with the cooling action demanded by the theories to be
explained in Chap. VII.
Most proposals for manufacturing sulphuric acid in a diminished
space start upon the assumption that in the ordinary vast chambers
the gases are not sufficiently well mixed ; some of them also on the
supposition that there is not enough '^ condensing " surface for the
sulphuric acid, and that this should be artificially increased. So
far as it was assumed that the sulphuric acid required to be con-
densed from a vapour into a liquid, similar to the condensation in
distilling alcohol, &c., this theory is, of course, wrong, inasmuch
as the sulphuric acid is liquid as soon as formed, and does not
exist at all in the chambers in the state of vapour. But we shall
see further on that for other reasons it is certainlv true that a
large amount of surface, for the chamber-gases to impinge on, is
indeed a most important factor in the chamber-process, and that,
moreover, a certain amount of cooling is also of great importance
in this respect. We shall see that this proceeds from the necessity
of bringing about the reaction between the nitrososulphuric acid
and the liquid water or dilute sulphuric acid floating about in the
chamber. Whilst, therefore, we must acknowledge that former
inventors were on the right track when increasing the surfaces of
contact, it is a fact that their efforts were unsuccessful; but this
was caused by the circumstance that they did not (and could not
in the then state of the subject) properly understand the essence
of the process, and that they consequently chose the wrong means
for their ends. Partitions within the chambers, if made of lead,
are most quickly corroded ; if made of glass, they soon collapse
{vide p. 474) .
DIMINISHING THE CHAMBER-SPACE. 473
An arrangemeDt of Ward^s (E. P. 1006, of 1861) consists in a
kind of mixing-chamber, for the combustion of 7 tons pyrites in
twenty-four hours, 64 feet long, 16 feet high, and 20 feet wide,
followed by a second lead chamber, or flue, 200 feet long by 3 feet
high and 3 feet wide, almost filled up with sheets of glass to a
length of 25 feet. The sheets lie in a horizontal position, and are
kept a little apart by strips of glass, to permit the passage of the
gases. Ward believed that upon these sheets (in lieu of which
tubes might be used) nitrous vitriol would condense and afPord a
large surface to sulphurous acid. His plan does not appear to
have been carried out in practice, or if it was it must have been
abandoned again, probably because his erection possessed too
little stability or was too easily stopped up. The horizontal
arrangement is also unfavourable to a systematic action of the
gaseous and liquid agents, for which streams in opposite directions
(up and down) are preferable, as we shall see below ; and the
total lack of a cooling arrangement would make the whole principle
of reaction on the solid surfaces incomplete, as will be proved
later on.
Mactear (Jouru. Soc. Clvem. Ind. 1884, p. 228) has carried out
some experiments showing the importance of surface condensation.
A tray, placed in a vitriol-chamber, one square foot area, was
found to give 708 grams H2SO4 in 24 hours. By placing in
the tray 12 pieces of glass, 12 in. by 6 in. each, in a vertical
position, the amount of acid obtained in 24 hours rose to 1644 gr.,
or 2*3 times as much, and by placing the glass slips horizontally,
the same distance apart as before, the acid rose to 3226 gr., or
4*5 times more than without the glass. Other experiments made
with " surface condensers '^ within the chambers showed that in
the case of flat vertically placed sheets the side facing the gaseous
current condensed more acid than the opposite side, in the propor-
tion of 100 : 78. When the same plates were placed horizontally,
with their edges facing the current of the gases, the amount
obtained from the double surface was 172, against 178 in the
former case.
The principle of surface condensation is also employed in
de Hemptinne's chamber-system, which will be mentioned in the
9th Chapter, in connection with his system of concentration.
At some places, e. g* at Uetikon near Zurich, there existed for
a time a peculiar kind of chambers. Each set consisted of only
474
CONSTRUCTION OF THE LEAD CHAMBERS.
one large chamber, 330 feet long ; ivithin this, however, there
were two partitions, dividing it really into three chambers. The
partitions are made as shown in
fig. 181. A row of perpendicular ^^S- ^^^'
iron gas-tubes of 1-inch bore,
covered with lead, a, is placed
across the chamber^ carried
through its top, b, and hung
from one of the joists c. At
vertical distances of 2 feet there
are lead hooks, dd, attached to the
tubes^ on the opposite side other
hooks^ d^d^, a little lower. These
hooks must not be made of sheet-
lead, because they bend too easily,
but they must be cast. On these
hooks sheets of glass 2 feet + 2
feet 6 in. are placed loosely,
leaving chinks of about 1 inch
width for the passage of the gases,
in order to mix them better.
These partitions do not seem to
offer any guarantee of durability;
and in fact, both at Uetikou and
at other works, formerly possess-
ing similar glass partitions, they
have been removed again ;
they are said to have sometimes
suddenly collapsed and cut
through the chamber-bottom.
The simplest kind of internal
partition, a lead one, is not prac-
ticable, because the lead, exposed
to the heat and the gas on both sides, is very quickly worn away.
This entirely disposes of the suggestion of Borntrager (Chem. Ind.
1885, p. 386) to make nearly horizontal (rather slanting) partitions
in the chambers, in order to multiply contact-surfaces. Both in
this case and in the Uetikon plan the internal contact-surface,
not being cooled, does very little service (comp. later on).
Walter and Boeing (Germ. pat. 71,908) employ hollow partitions.
DIMINISHING THE CHAMBER-SPACE. 475
made of acid-proof material^ arranged across the whole width of
the chambers. Double walls are constructed of such a form that
the principal gaseous current enters through large holes near the
bottom^ rises upwards in the space between the walls^ and issues at
the top ; at the same time the gases are allowed to penetrate into
the inner space by numerous small openings^ and to issue in the
same way on the other side^ so as to produce a good mixture.
Buttresses and binders produce sufficient stability without inter-
fering with the draughty which is also procured by making the
sectional area of the openings and joints much larger than that of
ordinary connecting-pipes. [This system aims at attaining the
same object as the previously introduced plate-columns in a simpler
way and without^interfering with the draught. But, in consequence
of the many outlets offered to the gases^ it is doubtful whether they
will travel exactly in the desired path. The stability of such inner
walls^ even when made of the best material^ is very doubtful indeed,
and a collapse must produce most disastrous results^ as has been
found in practice, wherefore the use of these partitions has been
abandoned. They remind one of the plan shown, p. 474.]
Brulfer (French pat. 220,402) also employs hollow brick par-
titions within the chambers ; he adds dividing apparatus, made of
lead tubes with air-cooling, fixed behind each partition. When
the gases have passed through a cooled divider, they again pass
through a hollow brick divider.
A similar principle, in which, however, the idea of mixing the
gases was the chief aim of the inventors, is involved in the
proposal made by Gossage and many others, and frequently
carried out in practice, of filling the chambers partially or entirely
with coke, or of erecting special coke-towers at the end of the set,
not as 6ay-Lussac towers, but to be merely moistened by water
or steam. In practice it has been found that even as a matter of
construction this plan gave much trouble, because the great weight
pressing upon the outside layers of the coke makes it bulge out or
even cut through the chamber-sides. But apart from this it was
found that th» yield of acid for a given chamber-space' was not
increased, that more nitre was used, and that the acid was rendered
impure by the coke. Everywhere, therefore, the coke has been
removed again from the chambers themselves, and has been
relegated to its legitimate place in the Gay*Lu8sac tower. The
cause of this failure is probably two-fold : firstly, the injurious
476 CONSTRUCTION OF THE LEAD CHAMBERS.
action of the coke on the nitrous gases, which would thereby
be reduced with formation of carbonic acid (comp. Chap. \T^-)*i
secondly, the lack of any cooling, just as in Ward^s case.
The same objections hold good for the apparatus of Verstraet
(Bull. Soc. d'encourag. 1865, p. 531), which was worked in Paris
for some time, but had to be abandoned as impracticable. It
consisted of a number of stoneware jars without a bottom, covered
430 square feet of ground, cost only £280, and was to supply
daily a ton of sulphuric acid of 106° Tw. There were twelve per-
pendicular stacks of five jars each, filled with coke and traversed by
the burner-gas ; nitric acid ran down over one of them, meeting
the sulphur dioxide ; and the resulting acid was run over the other
stacks in regular rotation.
The apparatus of Lardani and Susini (Bull. Soc. Chim. viii.
p. 295) is founded on the same principle. Its peculiarity is a
^^ reaction-apparatus,^' whose lower part is filled with sulphuric
acid, on the top of which a thick layer of nitric acid is floating ;
the upper part, divided from the lower by a perforated partition,
is filled with pumice ; the nitre-gas is regenerated to nitric acid
by an excess of air and water in a system of pipes filled with
pumice or coke.
That mixing the gases alone is not sufiScient is proved by the
small success of Bichter's apparatus (G. P. 15,252), consisting of
a steain-injector on the top of the chamber, which aspirates the
gases from the lower part of the chamber and re-introduces them
at the top. It is true that by this apparatus probably only a
small portion of the gases would be set into circulation ; other-
wise the draught would have been interfered with in an intolerable
way. At all events no great advantage has been obtained
by using it in all the factories visited by me; but at some
places a certain improvement is said to have been produced bv
this means.
The same proposal has been made in a somewhat modified shape
by N. P. Pratt (E. P. 4856, 1895). He places a fan or steam-
injector in front of the chamber, and a tower, fed with weak
sulphuric acid, at the end of the chamber, the gases issuing from
the top of this tower being re-injected into the chamber by means of
the fan. Baffling columns may also be placed within the chamber.
Modifications of this system by the same inventor are contained
in the U.S. P. 652,687 to 652,690 and E. P. 10,757, 1899.
MIXING- AND COOLING-TOWERS. 477
The Meyer's " tangential chambers'' also belong to this class
of apparatus (comp. p. 455).
Dr. Burgemeister '(private communication) proposed both to
mix and to cool the gases by arranging a number of lead pipes^ 15
to 18 inches wide, vertically between top and bottom of the first
chamber, and cooling these by air passed through. This plan is
hardly practicable, as the immense extension of joints, especially
at the bottom, will cause too many interruptions for repairs, but
it can be more easily carried out, according to Hartmann (Chera.
Zeit. 1897, p. 877), by constructing these inner pipes or shaft* in'
the same manner as an ordinary chamber-bottom, that is with a
hydraulic lute, formed by turning up the chamber-bottom round
the bottom of the pipes. Hartmann employed a number of such
shafts, 5x6 feet wide, from which he found an increase in the
production amounting to 20 per cent., viz. 0*9 to 10 cub. metre
chamber-space per I'O kil. sulphur burned. [This is not parti-
cularly high.]
F. Blau (G. P. 95,083) injects a spray of cooled sulphuric acid
into the first hot part of the lead chambers, in order to keep
down the temperature, and a spray of hot acid into the last
part of the chambers where the reaction is sluggish and is to be
revived in this manner. If the exit-gases thereby become too
hot, they are cooled by a spray of cold acid before entering the
Gav-Lussac tower.
A fourth way for increasing the production of acid is the
employment of special mixing and cooling towers and columns
between the chambers, even to the exclusion of all chambei's
except a first and perhaps a last small chamber. These ^^ inter-
mediate " or '^ reaction '' towers have had the greatest success in
diminishing the space for the production of sulphuric acid.
One of the first attempts in this direction was the plan of Thyss
(G. P. 30,211), of which I have myself given a detailed description
in Zsch. f. angew. Ch. 1889, p. 265, abstracted in our second
edition, pp. 378 & 379. This system having completely broken
down after a short trial, it may suffice to say that Thyss employed
lead towers, provided with a number of perforated lead shelves
over which the gas had to take a zigzag course. These towers
were not fed with any liquid, and consequently they would become
very hot and could exert no cooling action ; the draught was very
much impeded and the lead quickly corroded. Moreover, these
478 CONSTRUCTION OF THE LEAD CHAMBERS.
towers cost twice as much as a chamber producing the same
amount of acid. Still, although the Thyss columns were both an
economic and a technical failure, they proved that even in that
imperfect form an intimate mixture of the gases and their contact
with solid surfaces considerably hasten the reaction.
Much better elaborated was the plan of Sorel (French patent
of 1886; comp. his ^Fabrication d'Acide sulfurique/ p. 398, and
Zsch. angew. Ch. 1889, p. 279). He proposed to start with a
small chamber ; from this the gases were to pass through cooling-
pipes and then through two or three towers^ where steam is also
injected, while acid of 142° to 150° Tw. is running down, the out-
flowing acid not to fall below 130° Tw. Although he was then
connected with one of the largest chemical works in the world, his
proposal has never been tried on the large scale, probably owing to
constructive difficulties ; moreover, his idea of employing strong
acid for feeding the towers would rob the process of most of the
advantages of the principle.
It is recognized on all sides that the object in question was
first accomplished by my ^' plate^towers *' (Zeitsch. f. angew.
Cliemie, 1889, p. 385), in which I have endeavoured to combine all
the principles hitherto recognized as paramount in the manufacture
of sulphuric acid. In the 7th Chapter, when treating of the Theory
of the Chamber Process, we shall see that I formulate this theory
as follows: — ^Nitrous acid (or anhydride), or in the first part of the
chambers nitric oxide, acts as carrier of atmospheric oxygen and
water upon sulphur dioxide, by which action nitrososulphuric acid
is formed. This acid, which for the most part at once dissolves
in the sulphuric acid already present, like this floats about in the
chamber in the shape of a fine mist. When coming into contact
with water, or, which is probably the usual case, with dilute sul-
phuric acid, a decomposition takes place by which sulphuric acid is
formed, and all the nitrous acid is returned into the atmosphere
of the chamber to recommence the above-described action.
It is evident that all these reactions require in the first instance
a most intimate and constantly renewed mixture of all the gases^
vapours, and misty particles. In the ordinary large chambers a long
course, a vast space, and a correspondingly long time are needed
till the reactions are practically complete j that is, till nearly aU
the SO2 has been removed from the gases. If it were only the
question of a mixture of gases and vapours, probably very much
THEORY OF THE PLATE-TOWERS. 479
less time and space would be required ; but as both the nitroso-
sulphuric acid and the dilute sulphuric acid, which are to act upou
each other, are in the state of mist, that is, of minute liquid drops,
they may travel for some distance side by side without coming into
actual contact and, reacting as thev are intended to do. In manv
similar cases it has been found that simply mixing up the atmo-
sphere in question is nothing like so efficient as presenting large
solid (or liquid) surfaces against which the gaseous current must
strike in its progress. By the shock against these surfaces, and
the loss of velocity thereby incurred, and undoubtedly also by
surface attraction, the misty particles which would otherwise
float about for hours in the same state are condensed on those
surfaces in larger drops or films, and then the mutual reaction
above described, leading to the splitting up of nitrososulphuric
acid, will take place at once. From this we infer that we ought to
arrange a number of large solid surfaces in the path oi the gaseous
current, but so arrange them that this current must continually
strike against them and be constantly broken up into small
parts and mixed up again. (In this respect Ward's glass sheets,
p. 473, running parallel with the gaseous current, were not
properly disposed.)
There is, however, another condition to be realized for a proper
working of the chamber process. As we shall see further on, it
is indispensable that the temperature of the chamber is kept suffi-
ciently low to condense the requisite quantity of aqueous vapour
into liquid water or dilute acid, sufficient for decomposing the
nitrososulphuric acid. As the reactions in progress within the
chambers produce a large quantity of heat, the process cannot go
on without a portion of that heat being abstracted again, which iu
the ordinary system is done by radiation from the chamber-sides.
The separation of the whole chamber-space into several smaller
chambers acts favourably in this respect, as the ends of the
chambers and the connecting-pipes act as cooling surfaces; and
Sorel (comp. above) actually proposed increasing this by arranging
a set of cooling-pipes, which, however, would be nothing likfr
sufficient for the purpose. My own plan is, however, different
from anything hitherto proposed. I eflect the necessary lowering
of the temperature, not by radiation or convection to the outer
air, but from within by a shower of water or very dilute sulphuric
acid. Thus several objects are attained at the same time. The
CONSTRUCTION OF THB LEAD CHAUIIEBS.
PLATE -TO WEB
Fit'. 18-'-
482 CONSTRUCTION OF THE LEAD CHAMBEHS.
temperature of the chamber atmosphere is reduced to a proper
level, parts of its heat heing spent in heating and vaporizing water ;
but this water is just vliat is required for carrying on the chamber
process itself, and thus s saving is effected in the raising of steam
for the purpose of supplying the vitriol-chambers j we also supply
this water in a finely divided form, and exactly where it is needed
for meeting and decomposing the nitrososulphuric acid condensing
on the solid surfaces ; and by this cooling we protect the apparatus
employed against rapid deterioration, such as occurred iu the
Thyssplan (p. 477).
The apparatus employed is the " plate-column " or " plate-
tower," invented by myself and patented with Ludwig Rohrmaiin
(E.P. 10,355 of 1886 ; 10,037, 1887; 6989, 1889). It origin aUy
coQsiated of a column of large stoneware cylinders, filled with
the plates forming the peculiarity of the invention; and this is
the construction still employed for nitric and hydrochloric acid
(p. 125) ; but for the purpose of sulphuric-acid manufacture it is
constructed with a leaden shell («) of either round or angular
PLATE -TOWERS.
section, and stoneware plates {b b), as shown in tigs. 182 & 183.
The plates are supported by bearers, c c, in such a way that each
plate is independent of the others, and presses ouly upon the
hon^ontal ledge of its own bearer, whilst the pressure of the
superposed plates and bearers is sustained by the vertical part of
the bearers. The latter are easily arranged so as to protect the
whole inner surface of the lead against the attack of the chemicals
and the high temperature ruling within. We do not here notice
sucb parts as the feeding-arrangemeiitB, inlet- and outlet-pipes,
and the like, which require no special explanation ; the feeding-
arrangement will be described in the 6th Chapter, in connection
with the Gay-Lussac tower. A special CKplanation is only necessary
for the plates b b. Fig. 184 shows a small portion of their surface
as seen from the top, fig. 185 the same as seen from the bottom,
fig. 1 86 a section of pieces of two superposed plates.
Each of these is covered with a network of small ledges, d d,
and in each of the squares thus formed there is a perforation, e e,
with a somewhat raised margin. The height of this margin is
not quite so great as that of the ledges, hence there is always a
layer of liquid about J inch deep in each of the squares ; and as
V\g. 186.
there is always more liquid dropping in, the excess is forced out
through the perforations drop by drop. Tlie plates arc not
identical in shape, but differ as to the position of the holes. To
each perforation in any one plate there corresponds the point of
union of the ledges in the plates above and below (see fig. 186) .
Hence the liquid cannot drop straight through the holes in the
following plates, but strikes the solid portion of the next plate, U
3i2
484 CONSTRUCTION OF THE LEAD CHAMBERS.
scattered about^ and is divided among the adjoining squares. TliU
action is repeated from plate to plate. Thus the thin layer of
liquid resting upon the plates and clinging to the holes is con-
stantly renewed; and by the scattering about of the liquid another
absorbing surface is created.
The gases and vapours rising within the tower pass through
the numerous holes of the lowermost plate and are thus divided
into a great number of fine jets. Immediately on issuing through
the holes of this plate^ they strike against the solid places in the
next plate above, which correspond to the holes, and are thus
divided and again mixed; aud this process is repeated as many
times as there are plates provided. Whilst the gases and vapours
thus travel upwards in continuously renewed mixtures they come
into the most intimate contact with the absorbing-liquid, which
they meet within the narrow holes on the plates and scattered all
over in fine drops. By the incessant changes in the direction of
the current; aud the equally incessant renewal of the surface of
the liquid; the most favourable conditions are produced fur a
mutual action of the gaseous and liquid substances. Owing to
the principle of the apparatus, no false channels can exist in which
the gases or liquids would travel separately without coming into
proper contact with each other.
This circumstance partly accounts for the enormous difiereucc
in condensing-power between the '' plate-tower ^' and a perfectly
well-constructed and packed coke-tower, or any similar apparatus^
fitted with pieces of pottery and the like. The liquid within a
coke-tower is never quite evenly distributed ; there are always
many places where it drops down a considerable height without
meeting a piece of cuke, aud where, on the other hand, the gases
find channels in which they can ascend without for some time
getting mixed and coming into contact with liquid. Moreover,
the individual gas-channels are too wide, and the inner portion of
the gaseous current does not enter into reaction with the absorbing-
liquid. This is unavoidable, because the interstices between the
pieces of coke are quite irregular, and therefore the section of the
tower must be made wide enough and the pieces of coke large
enough to secure a sufficiency of draught for the worst case. Nor,
as experience has demonstrated, have any arrangements of cylinder?,
pipes, or other pieces of pottery hitherto had a better effect than
coke. Hence coke-towers must be made very wide and high, thus
PLATE-TOWERS* 485
offering a long time and corresponding opportunities for mixing
the gases and enabling them to come into contact with the liquid ;
and in this way the reaction is certainly very complete at the end.
But this enormous enlargement of space can be avoided by the
systematic way in which, in the plate-tower, the gaseous current
is split up into upwards of a thousand. very thin and exactly equal
jets, which must continually alter their direction, and must there-
fore become thoroughly mixed each time they pass through a new
plate. On their way they come into the most intimate contact
with constantly and systematically renewed thin layers of liquid.
The network of ledges prevents any unequal downward passage of
the liquid, unlike tbe action of coke-towers or of any other hitherto
known form of similar apparatus. Perhaps a still more important
difference is the following : — There is a very thin and constantly
renewed layer of liquid spread over each plate^ and the gases, in
passing through the perforations of the plate, must frequently
break through the drops of liquid. This seems to produce an
action somewhat similar to the Coffey still or other '^ rectifying '^
apparatus, and it may to a great extent explain why such an
intense action takes place in so small a space.
Owing to these advantages a plate-tower, in comparison with
a coke-tower, does from ten to twenty times as much work in
the same cubic space. It can therefore be made not merely much
smaller in section, but also much lower in height, and the feeding-
liquid requires correspondingly less pumping. A column of 40
plates would be only 18 feet high. The above is a comparison
between plate-towers and coke-towers ; the difference between the
former and empty chamber-space is much greater, as we shall
see.
In our present case it is of special importance that the injurious
action exerted by the reducing power of coke upon the nitrous
gases (p. 231) is avoided, the stoneware plates being absolutely
stable in the chamber atmosphere if manufactured of proper
quality. The plates, therefore, last for ever; even if cracked
they may still continue in use.
When a plate-column is partly obstructed by muddy deposits,
it is very easily cleaned out by a thorough flushing with water,
or, in bad cases, by removing the cover and lifting out one plate
after another.
Apart from the great constructive difference between the plate-
486 CONSTRUCTION OF THE LEAD CHAMBERS.
towers and all previously proposed apparatus^ there is an equally
great difference in their mode of application. If the tower
were left to itself, like Ward's or Thyss's apparatus, the very
completeness of the mixture produced therein would produce an
intense chemical reaction, and, consequently, a very injurious
development of heat. This is, however, entirely avoided by
feeding the towers with a stream of water or dilute sulphuric
acid, at such a rate that, by the vaporization of water, the tem-
perature does not rise above 70° or 80°. The intimate contact
between gaseous and liquid particles within the plate-tower must
bring out the cooling action of the evaporation of water to its
fullest extent, and at the same time the water required for the
chemical reactions of the acid-making process is supplied here
without any previous production of steam or spray ; the superfluous
steam passes over into the next chamber and does its work there.
The acid running off at the bottom is either used up as it is, or is
nm into one of the chambers, or it is employed for feeding the
Gay-Lussac tower.
In plate-towers there will always be a great excess of nitrous
gas and of oxygen ; hence there is very little fear that even when
employing water for feeding them there will be the conditions
present for the formation of nitrous oxide, which would mean
a waste of nitre. This can be avoided in any case by feeding
the columns with sulphuric acid of 1'3 spec. grav. or upward?,
since I have already shown (Ber. d. deutsch. chem. Ges. 1881,
p. 2200 ; comp. supra, p. 218) that in this case no NjO whatever
is formed. In practice such dilute acid or chamber-acid is
employed for feeding the columns.
The principal advantage of this system is that, like the Glover
tower, it brings about the mutual action of the ingredients within
the smallest possible space. We shall see in Chapter VI. that
one cubic foot space in the Glover tower effects the formation
of as much add as at least 180 cubic feet of ordinary chamber-
space ; and a similar difference may be looked for between the latter
and the plate-towers to be interposed between the chambers.
. We will now consider the question whether the thermal effects
produced in that system are not excessively large or small. The
heat of forming HjSG^ from SO2 + O + H2O is 54,400 calories ;
to this must be added the heat produced in the formation of
ordinary chamber-acid, say, of 110" TwaddcU, or H2SO4, 3 H5Q
PLATE-TOWERS. 487
= 11,100 calories; altogether 65,500 calories. This is the heat
produced in the formation of a quantity of chamber-acid corre-
sponding to 98 grams of H2SO4, and it is very little more than
would be required for converting 98 grams of cold water into
steam. This quantity of water then would have to be evaporated
within the tower in order to absorb all the heat produced in
the acid-forming process, on condition that the acid must run
out cold at the bottom and that the tower would lose no heat
by radiation. But as the former condition is unnecessary^ and
the latter even impossible to maintain, the quantity of water,
evaporated will be less than the weight of monohydrate produced,
and will probably be very nearly equal to that required for the
chamber-process, viz. three quarters of that amount. Any
deficiency of water could, of course, be made up by steam,
probably best by means of an injector placed in the outlet tube
from the plate-tower.
As a practical way of applying the new system^ I proposed
from the first the following plan : — Considering that by far the
greatest portion of the acid is made in the first part of the
chamber, we cut off the back part altogether, and leave behind
the Glover tower a chamber of only about 50 feet length. Behind
this we place a plate-tower of sufficient section for the amount
of gas to pass through and 40 plates high (say 18 feet). Then
comes a small chamber, say 30 feet long, again a plate-^tower, a
last chamber for drying the gases, and in the end a plate-tower
serving as 6ay-Lussac tower.
The question might be raised why I did not propose to carry on
the whole of the sulphuric-acid-making process in plate-towers or
similar apparatus. But a glance at the curves shown in the 7th
Chapter will show that the first part of the first chamber is really
very efficient, and whilst the gases are of such concentration
a lead chamber is possibly the cheapest apparatus for making
sulphuric acid. As soon as the curves begin to bend .towards
the horizontal, that is^ when the reactions become sluggish, it is
time to liven them up by apparatus like the plate-towers. But
if such were attempted to be used from the firsts the heat would
become excessive, which would be very injurious both to the
material of the apparatus and to the process. For this reason
the process proposed by Hannay {E. P. 12,247, of 1886, comp.
Chap. XII.) is not likely to be practically successful.
488 CONSTRUCTION OF THE LEAD CHAMBERS.
The first factory which ventured to try my plate-towers (which
have become known as ^' Lunge towers/^ both in their application
as intermediate " reaction " towers for the vitriol-chambers, and
as replacing coke-towers for the recovery of nitrogen acids, for
condensing hydrochloric acid, and so forth) was the old-established
acid-works at Lukawetz, in Bohemia, soon followed by a factor}^
at Valencia, in Spain, both in 1891, and by othei's in various
countries. Of course here and there diflSculties were experienced,
principally caused by the obstruction of draught. Thus in Zsch. f.
angew. Ch. 1895, p. 407, P. W. Ilofmann alluded to a trial which
failed because the hole>$ in the plates^ 8 millm. bore, became filled
with liquid and thus stopped the draught. I myself (ibid. p. 409)
completely refuted this objection, mentioning that already about
200 plate-towers were then at work, most of them with even
smaller holes, and a large number with 8 millim. holes, in
sulphuric-acid works. These works had been very successful,
as was autfientically proved by the replies to interrogations put
to certain firms, which also show that, if the section of the tower
is sufficiently lai*ge, no trouble is caused by draught of the kind
mentioned by Hofmann. Since the principle of artificial draught
by means, of fan-blasts is becoming more aud more applied to
vitriol-chambers, the complaint against the plate-towers on the
above ground is practically meaningless.
The greatest development of the plate-tower system took place
when H. H. Niedenfiihr, chemical engineer of Berlin, took the
matter in hand. He has designed and started many chamber-
plants on that system, and studied all the conditions necessary
for success, so that I shall refer principally to him in the
following description.
The part played by the Lunge towers in the manufacture of
sulphuric acid has been discussed at length by Niedenfiihr, iu
' Chem. Zeit.^ 1896, p. 31. According to him, plate-towers are
not very well adapted for replacing the whole of the ordinary
vitriol-chambers ; the first pai't of the process is always best
carried out in a single lead chamber, as here the gases are still
sufficiently concentrated to react upon each other. [In this
view, as well as in all other essential points of Niedenfiihr's
paper, I fully concur.] Here also the flue-dust and the excessive
rise of temperature would act injuriously. Hence it is not advan-
tageous to place a Lunge tower between the Glover tower and the
APFLICi^TION OF PLATE-TOWERS. 489
first chamber^ but it should be placed in the central or back part
of the set of chambers. Even then the results obtained with
these towers do and must differ at different works^ according to
circumstances^ viz.^ the available chamber-space^ the draughty the
size of the burners and of the connecting-pipes, of the Gay-Lussac
tower, and so forth. In some cases the working capacity of the
tower is partly taken up for correcting some fault in the set of
chambers to which it has been applied. Niedenfiihr quotes the
following special instances of the work done by Lunge towers, as
personally observed by him in the cases stated.
In one case, a Lunge tower was placed between the two chambers
of a set, which were of equal size; the total length of the set
was 193 ft. 6 in. and its contents = 9^,640 cubic feet. The
quantity of Sicilian sulphur burnt previously did not exceed
30 cwt. per 24 hours ; but after putting up the tower it could be
raised to 46 cwt., evidently because the time and the length of
the path of the gases had been thereby increased to such an extent
that the set could be worked with stronger draught, and more
sulphur could be burnt accordingly. In another case (4 chambers,
total capacity 38,150 cubic feet, length 78 ft. 6 in.) the production
could be increased from 15 to 18 cwt. of Sicilian sulphur per day.
This means that some improvement had been effected, but prin-
cipally in this respect, that the previously observed fault, viz. the
carrying forward of the process into the Gay-Lussac, was now
avoided. In a similar case, the production rose from 16 to 19 cwt.
of sulphur ; in both cases less nitre was consumed than previously.
The towers thus spent part of their efficiency in correcting the
faults of the old plant, and this in many cases would be a very
desirable object; but the quantitative results thus obtained do
not represent the whole of the working capacity of the towers,
which in some more favourable cases has admitted of increasing
the production by 45 per cent. Nearly in every case observed by
Niedenfiihr this capacity of the towers had not l)een exhausted,
for the chamber following upon the towers had hardly any more
work to do ; more work could have been done by increasing the
number of the burners, by enlarging the connecting-pipes, or by
other suitable measures. The very low temperature of the back
chambers in this case, however, is useful in condensing part of
the nitrous gases and lightening the work of the Gay-Lussac
tower. Formerly the acid from the Lunge towers was sometimes
490 CONSTRUCTION OF THE LEAD CHAMBERS.
too nitrous, but this drawback has been avoided by feeding the
toM'crs with acid of from 38° to 42° Be. (spec. grav. 1-357 to
1-410).
The Lunge tower cannot be simply substituted for a OloTer
tower, as the holes of the plates would be too quickly stopped up
by flue-dust; and in washing this down they would easily crack.
Niedenfiihr, however, recommends placing a few tiers of plates
with ^-inch holes in the upper part of the tower. He quotes a
case where a Lunge tower was found specially useful in com-
pletely denitrating chamber-acid required to be entirely free from
nitrogen compounds. [Comp. another case of special utility in
the 6th Chapter.]
Especially good results have been obtained in a number of
cases, personally observed by NiedenfUhr, where plate-towers were
employed as auxiliaries to Gay-Lussac towers. They act uot
merely in promoting the absorption of the nitrous gases, but also
in rendering the chamber-work much jnore regular, especially in
places where the chambers are subjected to sudden changes ot
weather, gales, &c.
Niedenfiihr in the above-quoted paper makes certain proposals
for the erection of sulphuric-acid works, forming a combination of
chambers and plate-towers, which we do not quote here, because
they are rendered obsolete by recent experience, the result of
which will be noticed in the 10th Chapter, where a complete plan
for this purpose will be given. [Comp. also p. 496.]
LUty (Zsch. angew. Ch. 1897, p. 484) also discusses at length
the function of '^ Lunge towers *' as intermediate reaction-towers.
He quotes the results of ten different works employing such towers
between the chambers as means for reducing the chamber-space,
from England, Scotland, Spain, Russia, Denmark, showing a large
saving of space, without any increase in the consumption of
nitre. He also gives detailed estimates showing that, apart from
the saving of ground-space, a set of chambers provided with such
reaction-towers costs 35 per cent, less than ordinary chambers for
the same production.
More recently, Niedenfiihr (Chem. Zeit. 1897, no. 20) quotes
practical results from two English works. One of them replaced
the last chamber of a set, with a capacity of 38,390 cub. feet, by
a LuDge tower of 256 plates iti 16 layers, with the same make
as before. Here each plate actually made 10*6 kils. H9SO4 in
ERECTION UF PLATE-TOnERS. 491
Fig. 188. Fig. 189.
m
l^h —
^^^^i
p^r-^ — frfii
& n
,pi.
^':
<<
W
1 ■;
i.
...,. ^1; fl
/■:
•^
:y
492 CONSTRUCTION OF THE LEAD CHAMBERS.
24 hours, or 216*7 kils. per cubic metre of the space filled with
plates, e, g. about 100 times as much as ordinary lead chambers.
Another factory, with very poor gases, still made 88*3 kils, HoSO^
per cubic metre of plate-space. [Comp. also Zsch. angew.
Ch. 1900, p. 960.]
Niedenfiihr (private communication) gives the following detailed
instructions for the erection of Lunge towers, taking, for instance,
a tower of a sectional area of 20 plates, 2 feet square each^ placed
4x5.
A brick foundation is usuallv made, but in case of need the
tower can be placed on the flooring surrounding the chambers, if
properly supported. The leaden shell is erected, with its outer
frame of wood and a lead bottom. On this a dwarf wall of
9-inch fire-bricks is placed all round the circumference, and a
central supporting pillar. This wall supports an iron frame, con-
sisting of one or more (in this case of six) pieces. The middle
pillar supports the places where four of the six pieces meet. This
frame is covered with lead^ the joints being so arranged that the
upper, bearing surface remains free and smooth. On this frame
the stoneware bearers (figs. 187-192) are placed. There are corner
pieces, a, which carry one of the corners of the plates, the three
others being supported by T-pieces b and cross-pieces c. Between
these pieces the longitudinal bearers are placed along the lead
sides of the tower. In order to prevent any shifting of the pieces,
wedges are put in suitable places.
Upon the frames thus prepared the Lunge plates are put, and
after finishing off one layer the parts forming the next frame are
placed upon the corresponding cross- and angle-pieces &c. The
parts a, d, c are so chosen that there is a distance of 2^ inches
between the single layers of plates. If a greater distance is
desired, we place below the parts a, i, c other cross-, T-. and
angle-pieces, aj, i^, Ci, 3*14 inches high. According to the number
of layers of bearing-pieces we designate the distance from plate
to plate as single, double, treble, &c. The tower in question has
in its lower part 10 layers at single distances and 6 layers with
double distances.
The top of the tower is made of lead, and is provided with a
proper feeding-arrangement.
If plate-towers are to be combined with existing systems^ it is
of course necessary to consider the place and level where the
PLATE-TOWKRS.
493
towel's arc to be erected^ and their dimensions^ in connection with
the existing circumstances^ so that it is difficult to lay down
general rules. But where new plant is to be erected, long expe>
rience now admits of establishing certain rules.
It is possible (as Mr. Niedenfiihr now holds) to replace the lead
chambers altogether by Lunge towera, by taking care to erect the
first part with as little loss of draught as possible, and to remove
the heat of reaction to the necessary extent. This is done by
making the reaction towers, immediately following the Glover
towers, of a wider section, and providing them with a very
efficient feeding arrangement. The last towers must be made
narrower than corresponds to the volume of gases passing through
the first towers, so as to exert a certain pressure on the gases
which promotes the reaction. The plate distances in these final
towers must be kept smaller than in the first towers.
The horizontal sections of Lunge towers may be chosen
according to the following particulars : —
Sulphur to be
burnt per 24 hours.
Up to 1 ton S
From 1 to 2^ tons S.
From 2i to 5 tons S.
Number of plates to be employed per ton
of sulphur burnt in 24 hours to replace the
chambers in dififerent parts of the system: —
in front. in the middle.
12
12 to 10
10 to 6
10
10 to 8
8to5
in the rear.
6
6to5
5to4
The number of layers for a given production is calculated as
follows : — In the first third of a system, where the plates are at a
treble distance from each other, each plate may be assumed to
produce in 24 hours from 10 to 12 kils. H2SO4 ; in the second
third, with double distances, each plate = 6 to 8 kils. H2SO4, in
the last third, with single distances, 1*25 to 2 kils. HsSO^.
Just as it is irrational to make a chamber system of one chamber
only^ which favours the diffusion of the inactive residual gases
witli the active ones, it would be wrong to try manufacturing
with one tower only, instead of dividing the work over several
towers.
Mr. Niedenfiihr believes the following arrangement to be
494 CONSTRUCTION OP THE LEAR CHAMBEaS,
suitable for vorkiiig with towers aloue, without chambers
(figs. 198 and 194). The burner-gases pass through a into a
preliminary tower b, and through c into the fan.blast d, which
conveys them through e to the denitrator /. This apparatus is
fed with nitrous vitriol and warm water or dilute acid so as to
furnish acid of 54" B. ( = 119J°Tw.). It also receives the nitric
acid required for making up the losses. It is packed like a Glover
tower, preferably with dish-like packing. Part of (he acid here
Fig. 104.
made is employed for feeding tower d. This deoitrating tower is
constmcted on the same principle as Niedenfiihr's or<linary Glover
towers (comp. next Chapter), viz. with an interruption of the
packing intended to facilitate the cleaning of the bottom part, and
with sufficiently large o^ienings for the passage of the gas. In
towor b the acid with which it is fed is concentrated, the gases are
|)urified and partly cooled so that they may pass through the fan-
Wast without any trouble. They effect the denitration in / and
then pass through gi, g^, to the first plate-tower A; pipes gi, g~, as
well as the further pipes Ji, ij, are provided with steam-pi|ies.
Tower h is filled with 12 layers of 24 plates each in treble
PLATE-TOWKRS.
495
distances ; tower k with 18 layers of 20 plates each in double
distances. In h and k the gases enter at the bottom and issue at
the top ; but in the last reaction-tower m they enter at the top
and leave it at the bottom through n, which arrangement has been
found to give the best result. Tower m contains 30 layers of
12 plates each in single distances. The gases now pass into the
Fig. 195.
«s
i^^'M:^Mr^;.u!4^.!..|.-.'^
Fig. 190.
^^^^
0
* ■ • •
a
-il
^^
e
^ K
f
first 6ay-Lussac tower o, containing 16 layers of 9 plates each,
then into the second Gay-Lussac p, packed with coke^ and finally
through q into the open air.
The acid coming from the first Glover tower b is freed from
most of its impurities by means of an air-cooler.
The just described system would serve to treat the gases from
496 CONSTRUCTION OP THE LEAD CHAMBERS.
2500 kils. = 2^ tons sulphur per diem. The production of acid
would be divided as follows : —
Tower h, 288 plates at 10 kils. HsS04=3168 kils. HjSO*.
„ A, 360 „ „ 7 „ „ ^2520 „ „
„ m,360 „ „ 1-75 „ „ = 630 „ „
Glover and Gay-Lussac towers 882 „ „
Total 7200 „ „
If the same production is to be attained by means of a combi-
nation of chambers and plate-towers^ this can be done as shown
in figs. 195 & 196.
From the Glover tower, a, the gases pass into a lead chamber^ b,
containing 1064 cub. metres = about 38,000 cub, feet, provided
with air-cooling shafts, ci, Cg, Cz (comp. p. 477). In this case no
fan-blast need be employed, but the gases must have a Insufficient
upward draught from the burners to the Glover tower and from
this to the chamber b. Now come the two plate-towers d and e.
These, as well as tower a, may be placed at a lower level, which
is all the better for the work. The Lunge tower d receives 20
layers of 20 plates each in double distances, tower e 30 layers of
12 plates each in single distances. At e the gases are passed in
at the bottom and out at the top, and then through / into the
first Gay-Lussac tower (a plate-tower) g^ into the coke-packed
Gay-Lussac h^ and through i into the open air. The production
will be approximately divided as follows : —
Chamber A, 1054 cub. met. at 2*75 kils. = 2926 kils. HjSO^.
Tower rf, 400 plates at 7 kils 2800 „ „
„ e, 360 plates at 1-75 kils 630,, „
Glover and Gav-Lussac towers 844
}} 9f
Total 7200
In a similar way larger systems can be constructed with one
chamber and a suitable number of liunge towers. Mr. Nieden-
fUhr would not even for the largest systems propose more than
t\\ o chambers, placing between them a very wide plate-tower with
great distances, and behind the second chamber all the remaining
towers. In the 10th Chapter complete plans will be given for
such a combination.
In order to apply the Lunge towers to existing systems, for the
purpose of increasing the production, the circumstances of each
PLATE-TOWERS. 497
case must be critically considered iu the light of the large ex-
perience now gained. It will usually be best to place the towers
behind the last chamber. The horizontal section of the towers can
be calculated from the last column of the table given on p. 493 ;
in the case of yery large systems from the second and the last
column ; with good draught the plates should be kept at single or
double distances.
If, however, plate-towers are to be placed between existing
chambers, all the conditions of the system must be carefully
considered according to the experience gained therewith. Above
all, the towers in large systems should not be arranged just after
the front chamber, but between the back chambers. The hori-
zontal sections, the places in the system, &c* must be calculated
according to the general instructions given above. If we were,
for instance, to place a Lunge tower between No. I. and No. II.
in a system of four chambers, the last three chambers would do
very little work. It will be much better to place the tower
between Nos. III. and IV. or behind No. IV,
Mr. Niedenfiihr thinks that, in view of the considerable saving
of expense effected by these improvements of the ordinary chamber
process, the contact systems cannot compete with the former in
the production of acid up to 142° Tw., and show their superiority
only for acids above that strength.
A very instructive plant was erected by Mr. Niedenfiihr at the
" Lazyhiitte ^' works. The chamber system consisted originally of
four chambers, with a total capacity of 7110 cub. met. (=250,000
cub. feet), and three Glover towers, two of which were always
working at the same time. In 1900, working with zinc-blende
and " forced style,'' this set produced on an average 25,580 kils.
acid of 50° B. (106° Tw.) per 24 hours, with 303 nitric acid
36° B. per cent, sulphuric acid. In 1901 four Lunge towers were
erected behind the last chamber, viz. : No. 1. 12 layers at 20 plates,
double distances ; No. II. 16 layers at 20 plates, partly single,
partly double distances; No. III. 25 layers at 16 plates, single
distances ; No. IV. (acting as a first Gay-Lussac tower) 18 layers
at 25 plates, single distances. The production now rose to 36,450
kils. acid of 50° B., with 1-90 per cent, nitric acid 36° B. per cent,
sulphuric acid. Evidently the plant was not working up to its full
capacity, but as there was not enough draught, a fan-blast was
placed behind tower No. IV. The production of acid rose at
VOL. I. 2 k
498
CONSTRUCTION OF THE LEAO CHAMBERS.
once, but during a few days also the cousumption of nitric acid
(to 4-76 acid 36° B. per cent.). When^ by a number of tet^,
the draught had been properly regulated '(which is still not quite
perfect, since there is only a temporary source of power for the
fan-blast, to be replaced by a better engine at the first opportunity;,
the production rose permanently to 44,600 kils. acid of 50° B.,
with a consumption of only 1-7 per cent, nitric acid 36° B. This,
for Upper Silesian zinc-ores, is a most excellent result, both as to
production of acid and saving of nitre.
The following estimations of temperature and manometric
pressure, made with this system^ after the erection of the plate-
towers, at three specified times (viz., before and after the increase
of draught by means of the fan-blast), are quoted here for the
sake of indicating their connection with the above-mentioned
alterations, but we shall later on describe the regulation of tempe-
rature and draught of chamber systems.
iAit«r Hpplyin|( the fiut-bliMt (Sortab^
and I>eceinher>.
\ Before applying
the fan-blaft (Sept.
and Oct).
Place where the obMrrations were
made.
Freaaure op
mm. water.
aa8-pip6 between Cb. I. & II +1-0 90
„ „ Ch.II.&IIl +11 ' 69
„ Ch. III.&IV +0-7r) o4
„ Ch.IV. ATowerl. -f06 , ...
„ „ Tower I.&II -09 | ...
)i ff II 11. & 111.... — I'D
M III. & Gay-,
LugsacClV.).' -3-6 i 38
It ,1 „ IV. & Coke-
tower (Qaj-Lussao) . — 4r>
„ „ Coke Ghiy-Lu9eac&
chimney —9*4
Before iiroperly After tboron^J
regulating the refifnlatiag the
draoght. draught.
104
86
I Preuare ;
mm. water.
+ 10
+0-2 i
-10 !
-3-8
~,rO
-6-2 : .
+7-4
+4-8
-1-4
»C.
preasnre
mm. water.
+ 1-9
+0i>
-0 5
-4-3
-71
91
7.3:1
61.1
+7-4
+40
-50
30
Other Apparatus on the principle of Plaie'towers.-^Atter the
success of the principle embodied in the '' plate-towers " had been
thoroughly established, it was only natural that other inventors
should try, more or less successfully, to attain the same end bv
other means, not 'coming under the Lunge- Rohrmann patents.
[is was all the more likely, as the price of the " plates '' and the
i
APPARATUS SIMILAR TO PLAT£-TOWER8. 499
fittings belonging to them was at first rather high^ owing to the
technical difficulties of their manufacture. Since these difficulties
have been entirely overcome^ the price of the plates &c. has been
so very much reduced that most of the imitations have lost even
any economical advantage^ apart from their inferior efficiency. I
shall, however, quote all the more important of these imitations.
One of the objects aimed at by some of the inventions concerned
is to avoid any impediment to the draughty for which the plate-
towers are sometimes blamed. As we have seen (p. 498), this
objection is of no account in view of the ease with which the
chamber-draught can be regulated by fan-blasts^ and it should be
borne in mind that the efficiency of an apparatus of this kind
is practically proportionate precisely to its draught-impeding
capacity.
Hacker and Gilchrist (Engl, patent 15,895, 1893) employ the
same principle that I have adopted in the " plate- towers,^^ to
which they expressly refer. Instead of my geometrically con-
structed stoneware plates for dividing the gases, the liquid acid,
and the acid vapour, they use a number of horizontal lead tubes,
running from one side of the tower to the other aud alternating in
position. These towers, which they call ^' pipe-towers,'^ are fed
with water or sulphuric acid ; cold air is drawn or blown through
the pipes. A paper in Journ. Soc. Chem. Ind. 1894, p. 1142,
contains a detailed account of this system, in discussing which
several speakers threw great doubt upon its efficiency. This is
hardly just, as the introduction of these *' pipe- towers " into many
American factories seems to show. They cannot possibly do as
much work as ^^ plate-columns " of the same size, but undoubtedly
they are of some use. Another paper on these towers was pub-
lished in Journ. Soc. Chem. Ind. 1899, p. 461, containing some
improvements in details. The success of this system as inter-
mediate towers between the chambers is the best proof of the
correctness of the principles which led me to the construction and
application of the " plate-columns," of which the " pipe- towers "
are an imperfect imitation, made of a material liable to corrosion
instead of indestructible stoneware. A similar plan is that of
Winsloe & Hart (B. P. 20,142, of 1901), who employ perpendiculap
air-cooling pipes in a shaft connecting two chambers.
Benker (French pat. 238,872) places between the chambers
leaden towers, 5 feet wide and 20 or 25 feet high, filled with
2k2
500 CONSTKUCTIOX OF THE LEAD CHAMBERS.
earthenware cylinders 4 inches vide and 3 to 4 inches high. These
are fed with nitrous vitriol from the Gay-Lussac towers at the top,
and with a steam-jet at the bottom. The strength of the out-
llowing acid is maintained between 113° and ISS'^Tw. ; it is kept a
little nitrons, to avoid the reduction going too far. Such a tower
is placed behind the first chamber, which is made lai^ enough to
consume all the steam coming from the Glover tower; another
tower is placed between the second and third chambers. — This is,
of course^ a simple imitation of Lunge towers by cheap, but im-
perfect means.
Gutlmann (G. P. 91,815) recommends as a "packing" for re-
action-towers perforated globular bodies made of earthenware,
glass, or metal. As shown in Hg. 197, the perforations are con-
tinued into short pipes turning into the inside of the globe.
Fi^'. 197.
These bodies may also he undulated inside and outside, to increase
the acting surface. They can be put into the tower without any
special care in packing. The liquid runs down both inside and
outside, and the gases are well mixed.
Niedenfuhr objects to hollow balls on the system of Guttmanu
and others, because, firstly, most of the perforations get closed up
when filling the tower ; secondly, the gases are sure to take the
easier way round the balls, instead of forcing their way with
increased friction into the interior, and whatever does enter the
balls will remain there for an indefinite time without taking part
in the reactions.
Another kind of reaction-apparatus consists o£ the " KegcU
thiirme " (cone-towers) of the Bettenhausen ceramic works.
They are filled with slightly conical bodies, open at the bottom
and provided with a shallow basin at the top. Niedenf iihr (Chem.
Zeit. 1897, No. 20) says that they are much less active in dividing
CHAMBER- FITTINGS. 501
the gas than Luuge plates (only from 14 to 138 times per square
metre^ against 2400 times in the case of Lunge plates)^ that they
contain less than half the acid-covered surface^ and that the hollow
space is altogether a mistake^ since the gases will stagnate in them.
Liity {eod. loc.) also criticizes the Bettenhausen cones adversely.
We shall refer to them again in the next and also in the 12th
Chapter.
The final result of reducing the chamber-space would be the
complete abolition of lead chambers in the ordinary sense. On
principle^ such a step cannot be regarded as irrational or impos-
sible, and it has several times been proposed or attempted, both
in former times (comp. p. 476) and recently (comp. p. 493), since
the success of the plate-towers had again shown that the vast
space of ordinary vitriol-chambers is not indispensable for carrying
on the process. We shall discuss some of these proposals in
Chap. XII. ; here I will only refer to what I have said on p. 487,
viz., that for the commencement of the process (that is, behind
the Glover tower) I still hold an ordinary, comparatively small,
lead chamber to be the cheapest kind of apparatus for the purpose,
followed by a number of reaction-towers, perhaps alternating
with one or two small lead chambers ; but the last word has not
yet been said in that direction, aud it may be that the ordinary
lead chambers can be entirely substituted by reaction-towers,
although hitherto this has not been economically accomplished,
as will be seen in Chap. XII.
Chamber^fit tings.
Every set of chaml)ers must contain a number of auxiliary appa-
ratus, which in part are absolutely necessary for the process, and
in part serve to check it chemically and technically : the former
are essentially those for introducing the nitre, the steam, and the
air ; the latter, smaller apparatus will be described first.
Drawing-off the add is never done by cocks attached to the
chambers. Such taps might be made of hard lead (4 to 5 Pb to
1 Sb) ; but they would soon get stopped up with sulphate of lead,
and could not very easily be repaired when leaking. It is best to
place beside a chamber a round or square lead^box, open at the top,
of the same height as the upstand of the chamber-bottom, and
connected at or near the bottom by a wide pipe with the chamber ;
or a suitable piece may be burnt on to the chamber, as shown in
oOJ! CO.VSTBDCTION OF THE LEAD CHAUBERS.
fig. 198, and the connection made by a slit. Tbe )k>x may be
provided with a stopcock ; but more usually, asshowain tbe Rgure,
Fig. 199.
it carries in its bottom a valve-seat a of regulus metal, into which
fits a conical plug b of the same metal provided with an iron handle
covered with lead. The ruiiuing-off pipe d
is either burnt to the valve-seat or joined to
it by an open funnel ; the latter permits the
running-ofT to be more easily observed, but
is apt to occasion running over, by getting
stopped up. Or, as shown in fig. 199, a
lead siphon may be employed, with a cup
attached at each end, a a, which keep it
iilways filled, so that it begins to act as
EooQ as one of its limbs has been put into the acid. The second
limb then enters into a large funnel of the mnning-off pipe b.
The simplest plan, which does not work at all badly, is this : to
bum a short piece of thick lead pipe to the chamber-side, and make
the joint very strong by casting lead round it. 'I'hb pipe ends
over the funnel of the runuing-off pipe, and is closed by a good
india-rubber cork. The men take this out and put it in by hand,
8IPHUN3. 503
having a bucket of water staadiag by to wash the acid off their
hands.
The arraDgemeut shown in fig. 200 is very good indeed. The
siphon b is firmly attached to tlie box c, or within the npstand of
the chamber. A cylinder d, surrounding the outer limb of the
siphon, is so suspended that it can be drawn up or down fay means
of the chain e and the pulley /, and fixed in any position by the
hook ff. The cylinder d forms a continuation of the outer limb of b ;
Fig. 200.
when it is quite drawn up, so that its overflow A is at a higher level
than the acid in c, it will cease to run ; but when h getsjbelow this
level, the siphon will at once begin to act, and the more quickly the
more d is lowered. Thus the acid can be run oflF with more or less
speed and with the utmost cleanliness.
Fig. 301 represents a siphon suitable for hot acids in any part
of the works. To the top of the siphon a a there is joined by a bent
tube a closed lead vessel b, which by an elastic tube is connected
with the open vessel c. The latter is filled with acid and lifted
into the dotted position^ whereupon b and then the siphon a are
filled ; c is then lowered, upon which the siphon begins to act,
some acid running back from & to c and thus producing a partiali
504
CONSTRUCTION OF THE LEAD CHAMBERS.
We shall in this place also mention the best arrangements for
inserting siphons into glass carboys or other vessels for carrying
corrosive liquids. The simplest and most efficient plan is that
shown in fig. 202. A glass or lead siphon, a, is inserted into a good
india-rubber cork^ made strongly conical so as to fit bottles with
various sized necks; another short tube, b, passes through the
Fig. 201.
[ i
same cork. The siphon, a, may or may not be provided with a
tap, c. It 'will be seen without explanation that the siphon can
be started by blowing into b. The flow of liquid may be stopped
either by closing the tap c, or, if there is no tap^ by lifting out the
cork, or by opening a third hole provided in the cork for this
purpose. In the (very frequent) case of the mouth of the carboys
being too irregular in shape for the cork to fit air-tight, the re-
maining air-channels are stopped up with damp clay ; and in an
emergency a lump of damp clay may replace the india-rubber cork
entirely.
Bode aud Wimpf s siphon (G. P. 28,794 ; Chem. Zeit. 1885,
p. 907; J. Soc. Chem. Ind. 1885, p. 484; further improvements
in Zeitsch. f. angew. Ch. 1889, p. 522) rests on a very similar
principle, with addition of a ball-valve for stopping and starting
the siphon. It seems to be specially adapted for nitric acid.
Fig. d02.
506 CONSTRUCTION OF THE LEAD CBAUBERS.
Alisch (G. p. 9133), Landel (G. P. 9307), J. P. y More (G. P.
28,721), Oplander (G. P. 30,662), and others hare constructed
different kinds of siphons.
Dc Hemptinne has written a pamphlet in vhicb he traces a
great many forms of siphons to their authors ; but there is nothing
specially new in it.
Pratt's carboy -emptier, sold by J. J. Griffin & Sons, London, is
showu in fig. 203.
J. Cortin, of Newcastle-upon-Tyne, makes non-rotative acid
valves of a special mixture of regulus metal, the plug rising or
falling into its seat out of a mixed setting without turning round,
so that it is free from friction in working, and the wear and tear
are reduced to a minimum.
Acid-dishes (drips, tell-tales) are placed inside the chambers, in
oitler to examine the process by ascertaining the quantity, strength,
and nitrosity of the condensing acid. They are made in very
different ways— for instance, that shown in fig. 204. A is a lead
Fig. 204.
ACID-DRIPS. - 507
vessel^ burnt inside against the chamber-side about 3 feet above
the bottom. The acid caught here runs by the tube a through the
chamber-side into the lead cylinder B, containing a hydrometer.
B is provided near its bottom with a side-branchy b, reaching above
its top^ and ending in a funnel for receiving the acid, which thus
constantly runs in at the bottom and out at the top of B into a
vessel C^ from which it is carried back by a small pipe into the
chamber. The greater the length of A, the quicker the acid will
be renewed in B, and the more reliable are the indications.
Many manufacturers place S-shaped drip-tubes in the connection
between ihe chambers, for a similar purpose. Others do not trust
to the collectors burnt to the chamber-sides, but place leaden
or stoneware dishes at some distance from the side within the
chamber. These rest on feet made of lead tubes^ or upon a stand
of stoneware, so as to be elevated above the level of the acid ; and
they have an outlet leading outside the chamber. In some works,,
both kinds of drips are fixed side by side ; and it is noticed that
those fixed to the sides always yield acid of 6° to 10° Twaddell
less than the inner drips, evidently because more aqueous vapour
is condensed on the sides along with the sulphuric acid.
Generally the cylinders of acid-drips are made far too large, so
that they show the changes in the process much too slowly. It
is therefore preferable in all respects to make the cylinders very
small, say holding about 20 cubic centimetres, with a side tube
and funnel, into which the fresh drips fall, as shown in fig. 204,
whilst the cylinder itself keeps overflowing, and thus its contents
are renewed about once in every ten minutes. Special small hydro-
meters, having only a range of, say, about 20 degrees Twaddell, are
made for the purpose of showing the strength of acid in these small
drips.
For taking samples of the bottom acid itself a recess is usually
made in some part of the chamber by dressing back the lower part
of the side. Some, in order to be quite sure, always take the
sample out of the chamber itself through a special small man-hole
luted with moist clay ; in this case there is a slight loss of gas, but
no danger of getting stagnant acid. Such a man-hole is shown in
fig. 205 in section. The large man-holes may be made in exactly
the same way ; or else their lids may fit into a groove luted with
damp clay, as shown in fig. 206. Large chambers are fitted with
several acid-drips, man-holes, &c.
508 CONSTRUCTION OF THE LEAD CHAUBBRS.
For taking the samples themselves a dipper of leatl or glai
emplovedj which is lowered slowly, so as to
get all layers of the acid into it." There is ^'?- ^^- ^f- ^***-
often a great difference between the top and
bottom acid.
In German works tkermomelers are fixed
at re^Iar intervals of 30 to 50 feet in the
length of a chamber, whose mercury- vessel
is inside, and whose scale is outside the
chamber. This means of observing the tem-
perature is undoubtedly infinitely better than
the rough one formerly in general use in England, by touching
with the hand.
For chambers not exceeding 100 feet in length, one set of drips,
thermometers, &c. is generally thought sufficient. For longer
chambers this is not the case ; at the German works there is
generally a special set of these fittings for about every 60 feet
length of chamber.
The pressure inside the chambers might be indicated by any of
the anemometers to be described further on; but usually simpler
means are employed, such as ordinary glass pressure-gauges.
Sometimes stoneware plugs are put into holes made in the
chamber-sides, in order to indicate the pressure inside the chamber.
The tension of the gas is also seen by lifting the lids of the small
man-holes {fig, 207), which are al-
waj's made on the top of the I'ig- 207.
chambers with hydraulic lutes, and
which generally cousistof glass jars,
so as to give light for observation
through the side-windows (see
p. 510).
A very sensitive pressure-gauge
has been described by Vogt (Joum.
f. prakt. Chem. xiv. p. 284). The
pressure is observed by the movement of a small air-bubble
playing in a horizontal glass tube of 4 or 5 millimetres diameter.
The glass tube, besides this bubble, is filled with water or another
liquid, and is connected on each side with a bottle tubulated near
the bottom. One of these bottles is 15 to 16 centims., the other
6 to 8 centims. wide ; the liquid stands at the same level in each.
FBESSUHE-OAUaES. 509
The pressure within the lead chamber \» made to act upoD the
surface of the liquid iu ooe of the bottles, and its amount measured
by tiie position of the air-bubble. The apparatus is all the more
sensitive the greater the difference between the diameter of the
tube and that of the bottles. There is a contrivance for admitting
a bubble of air previous to using the apparatus, and for again
equalizing the levels after use.
A lery simple pressure-gauge, sufficiently sensitive for ordinary
purposes, is shown in fig. 208 (from Sorel, ' Industries Chimiques,'
p. 142). The tube a has an inchnation from the level in the pro-
portion of 1 : 10 ; it is connected with a reservoir i, 1 ^ or 2 inches
wide, upon which the pressure is brought to act by the clastic
Fi;-. 208.
tube c (if there is suction, the vessel to be tested must be con-
nected with the bulb d) , The gauge is filled with a mixture of
water and spirit of wine coloured with magenta or otherwise.
As the movement of the liquid in the bulb b can be neglected,
any movement of the liquid in the tube a, as measured on the
scale e, corresponds to one-tenth of its extent in real height. If,.
for instance, each degree on the scale is = jL inch, it indicates a
real pressure of -^^ inch. It is best to cause the liquid to move
before each observation, in order to counteract the effect of friction
within the tube.
For gavffinff the height of the odd there are cither stationary lead
510
CONSTRUCTION OF THE LEAD CHAMBERS.
Fig/ 209.
gauges (which, however, are difficult to read-off exactly), or ac-
curately divided copper rods, which are
dipped in every time, but always in
the same place, since the chambers arc
never absolutely level, or glass floats like
that shown in fig. 209, the stem, a, of
which slides in a small lead frame, i,
whose upper edge serves as an index
for reading-off. The float will sink
more or less in the acid according
to any alterations in its specific
gravity. To make this cause of inac-
curacy less sensible, the ball of the float
is made pretty large. These floats are
the most convenient for reading-off.
A very great assistance in judging
of the chamber-process is afforded by
glass ivindows or sights^ which permit
the colour inside the chambers to be
Whoever has once got used to
seen.
these windows will never do without them. They are 8 or 9 inches
square, and are placed, at a convenient height for looking through,
in those places in the chamber-side which lie in a line with the
glass man-hole lids in the chamber-tops ; thus they are sufficiently
lighted. Where the chambers are roofed in, light must be procured
in some other way (for instance, by two opposite windows corre-
:sponding with a window in the chamber-shed, &c.) . The chamber-
glasses are put into small lead rabbets, and luted with white lead
:and boiled oil. The assertion is occasionally made that the colour
of the gaseous mixture, looked at across the width of the chamber,
-or in the diagonal line from the side to the man-hole lid in the top,
is too deep, and that " sights '^ in the connecting-tubes are pre-
ferable; but just the opposite is the case, since the observations
are evidently far more accurate, and any alterations of colour
much more easily perceived, in the former than in the latter case.
-Only in the first part is the gaseous mixture, through copious
condensation of acid, too opaque for observing its colour; but
just there it is quite unnecessary, for only in the back parts of
the set is it important to have always an excess of red vapours.
At some works they prefer to the ordinary side-windows, which
APPARATUS POR INTRODUCING NITRIC ACID. 511
are rather difficult to keep clean^ glass jars^ similar to those shown
in fig. 207, p. 508, but placed on special short, wide branch-tubes,
burnt in the sides of the chambers at convenient places. These jars
when dirty can be exchanged in a moment for clean ones, and they
are supposed to show all the changes in the chamber-atmosphere
quite as well as the glass panes fixed in the lead walls themselves ;
but my experience is decidedly to the contrary, as sometimes the
side jars are quite yellow while the chambers are already pale,
and vice versd.
Apparatus for introducing/ Nitric Acid into the Chambers.
These are divided into two classes. In the first class the nitric
acid enters the chambers in a state of vapour, mixed with the
burner-gas, whose heat evolves it from a mixture of sodium nitrate
and sulphuric acid contained in an iron pot. This is styled
^^ potting.^^ In the second class the nitric acid is made in a
liquid form in special apparatus, and introduced as such into the
chambers. Opinions still differ as to which of the two plans is
best. The plan of introducing acid in the state of vapour^ which
is quite general in England but only rarely used on the Continent
(in America both processes are found, but the English plan more
frequently than the other), has the advantage of greater simplicity
and of saving labour and fuel. The advantage sometimes claimed
for it^ that there is less loss than by making and employing liquid
nitric acid, is hardly a real one ; for in the first plan some nitric
acid is easily condensed during the conveyance of the gas to the
chambers^ and may corrode brickwork, iron, &c., whilst liquid
nitric acid is always introduced exactly in the place where it is
needed. It is necessary to employ much more sulphuric acid for
the decomposition of the nitrate of soda in 'Spotting'' than in
the regular manufacture of nitric acid. The presence of nitric
acid in the burner-gas will also induce a premature formation of
sulphuric acid, especially if it be 'much cooled ; but the Glover
tower obviates any inconvenience arising from this. On the other
hand, some are afriEtid that in " potting " the nitre-ovens may get
so hot that a portion of the nitric acid will be decomposed down
to NO or even to N ; but the men generally employ so much
sulphuric acid for decomposing the nitre, that this cannot easily
happen, nor are NO and N formed so very readily as was formerly
supposed (see later on).
512 CONSTRUCTION OF THE LEAD CHAHBEES.
The advantages of introducing nitric acid in the liquid form are
the following : — avoiding the entrance of false air into the chambers
and the escape of burner-gas into the atmosphere^ both of which
occur in many (not all) systems of employment of gaseous acid;
the possibility of employing as much nitric acid and as quickly
as desired, whilst in the other case this depends on the heat
of the burner-gas, which during a bad process, just when most
nitric acid is needed, proves insufficient ; lastly, and most of all,
the exact regulation possible with liquid nitric acid and its con-
tinuous supply, whilst gaseous acid is always given off from the
nitre-mixture very unequally. These advantages have induced
the great majority of continental manufacturers to adopt liquid
nitric acid. Muspratt (' Dictionary of Chemistry,^ ii. p. 1029)
reports that a continental manufacturer, who previously worked
with liquid nitric acid, after having seen the use of solid nitre in
England, had saved one third of his nitre by introducing the
English plan. This simply proves that that manufacturer had
not been very careful before, and is no guide whatever. The
opposite experience has been much more frequent. Liquid nitric
acid^ however, will do harm if the apparatus for introducing it
is not constructed in such a way as to completely volatilize it or
convert it into gaseous products before it reaches the chamber-
bottom, since it will act upon this. Accordingly, sulphur dioxide
and aqueous vapour, which decompose the nitric acid, must be
brought into as complete contact with it as possible.
It is claimed as an advantage for introducing the nitre by
^^ potting,^' that the chambers are not exposed to the damage possible
with incautious handling of nitric acid, whilst, on the other hand,
the irregular evolution of gas from the nitre-mixture is equalized
by employing several decomposing apparatus^ and charging them
in turns, say, once every hour, just after a fresh pyrites-burner has
been charged; thus the stronger evolution of nitre-gas runs parallel
with that of sulphur dioxide. Some prefer a contrivance for
supplying gradually, and not all at once, the sulphuric acid serving
for decomposing the niti*e. It is contended that the best English
works, all of which employ soUd nitre, work with as small a con-
sumption of it and as good a yield of vitriol as the best of the
continental works employing liquid nitric acid ; also on the Con-
tinent some manufacturers work quite as well with solid nitre as
their neighbours with nitric acid ; but it is extremely difficult ta
INTRODUCTION OF NITRIC ACID. 513
check such statements^ as few manufacturers divulge their real
working results to outsiders, and, moreover, very many of them do
not even know these results themselves with that degree of
accuracy which is required to decide this question.
There is no doubt whatever that the chamber-process can be
worked more regularly by the continuous supply of nitric acid in
the liquid form (comp. Eschellmann^s experiments at Widne3,
infra) \ and the just objection to this, formerly existing, that
there was a risk of damaging the first chamber in case of a collapse
of the ''cascades^' (see p. 521) has been entirely removed, in the
first place by the almost general plan of introducing the nitric acid
into the Glover tower, and in the second instance by spray-pro-
ducers &c. The labour of, making nitric acid in large apparatus
and condensing it in receivers is not much greater than that of
frequent '^ potting ^' on the English plan ; the waste of sulphuric
acid for decomposing the nitrate of soda is much less in the former
than in the latter case, in spite of utilizing the i\itre-cake, which
pays for the coal consumed in manufacturing nitric acid. The
nitric-acid retorts are even sometimes heated by pyrites-burner
gases. These reasons explain why the majority of continental
manufacturers prefer the employment of liquid nitric acid for
the chamber-process, in spite of the somewhat greater *' trouble ''
involved, which, however, is more apparent than real. If all the
trouble caused by the English potting-process, in producing a
nuisance, running over of the mixture into the kilns, occasional
bad decomposition of the nitre, frequent introduction oE an
excess of air into the chambers, &c., were summed up, it would
greatly exceed that involved in making and supplying liquid nitric
acid. There remains hardly anything in favour of the English
plan except the force of habit and the fear of having some
trouble in the transition from one mode of working to another.
That the imperfection of the present plan is felt even in England
is proved by the various attempts at feeding the chambers with
a solution of sodium nitrate, which are irrational in principle
and have necessarily failed (see later on).
We shall now describe both ways of introducing nitric acid, and
begin with the
Introduction of Nitric Acid as Vapour ('' Potting ''),
The apparatus serving for this has been partially described in a
VOL. I. 2l
514 CUXSTRVCTIOX OP THK LEAD CIIAMUEK3.
former chapter (pp. 270, 279, AU), along with the BulphuT' aud
pyrites-burnerB. The drawbacks huve heeo pointed out which
attend placing the iiitre-pota within tiie burners, or, generally,
in such a way that the acid emlphate boiling over can run iutu
the burners. Recently, however, there has always been a special
I'ifT -JIO.
nitre-oven constructed by enlarging a suitable place in the gas-flue.
It is situated either above or, preferably, just behind the .burner-,
and provided with the necessary working-doors and a cast-iron
saucer for collecting what boils over. The nitre-pots ibemselvcN
have various shapes — for instance, that shown in tig.[210,; at «'«
ledges are cast ou the bottom, which facilitate pushing tbe pots
backwards and forwards. Thev hold from 8 to 12 lb. of nitre.
FOTTIXO THE NITKE. 515
TLe "pottiug" with these pots, which, strauge to say, are atill
(or have been uutil quite recently) met with in some English
works otherwiBe abreast of the times, is very troublesome and
imperfect. During the emptying and refilling of the pots the
doors of the nitre-oven are wide open, which does even more
harm than in the case of the burners. The heavy, pots, along
with their melted contents, have to be taken up with long fork-
Fig. 212.
shaped touls and emptied, wliicli requires great strength and skill.
The pots, freshly charged with nitre, are placed just within
the door of the oren ; ilie necessary acid is poured in from a jug
by superficial estimation, and the pot pushed into its place; not
till then eau the door be closed. If the draught is good, a great
deal of air must enter, meeting not even the same resistance as in
the burners; if it is not very strong, which will more frequently
be the case, so much gas escapes that it can be smelt for some
2l2
516 CONSTRUCTION OF THE LEAD CHAMBERS.
distance. Special clampers would partly obviate this, but are
rarely met with. (Such dampers are mentioned in the official
Belgian Report of 1855, p. 23.) The pots are quickly worn out,
and must be replaced, especially if chamber-acid is employed in
them. They last much longer if acid of 140° Tw. is used.
A much more perfect plan is that of decomposing the nitre in a
fixed apparatus, and running off the acid sulphate (nitre-cake) in
a liquid form. This consists of a seraicylinder of cast-iron^ a
(figs. 211 & 212), with a cast-on tube i, bored somewhat conicallr.
The latter projects out of the nitre-oven, and during the working is
closed by a ground-in iron plug with a long handle. Outside there
is a cast-iron saucer for holding the nitre-cake, which at once soli-
difies. The internal saucer, c, catches the boilings-over. The nitre
is introduced by the hopper, rf, which is provided with a damper:
and after putting in the damper again, it is made gas-tight by filling
up with the next charge of nitre. An S-shaped tube (not shown
in the diagram) serves for running in the sulphuric acid, for which
it is best to provide a small tank with a siphon or stopcock. The
acid should be run out of this tank by a pipe with a very fine
point into the S-shaped tube, so that the running shall take a long
time, and the nitre be only gradually decomposed. Sometimes au
iron scraper with a long handle (passing through one of the ends
of the oven) serves for stirring up the mixture in the pan. The
hole for this must be kept air-tight with clay. An apparatus of
the size drawn here holds 56 lb. of nitre, which can be easilr
decomposed in two hours. In any case there should be two or
more of these apparatus, so as to make the current of nitre-gas
more regular by charging them in turns.
Even preferable to the arrangement here shown is that of placing
the nitre-trough in such a way that the burner-gas can play round
the bottom as well. The saucer for the boiling-over stuff, which
forms the bottom of the nitre-oven, must then be placed some-
what lower.
At Oker the potting is effected in cast-iron retorts set in the
liigh kilns used there (p. 295) ; the gas-delivery pipe opens into
the gas-flue belonging to the kilns.
Rice (Engl. pat. 16,757, 1892) patents a contrivance which has
been in operation in several places for many years past, viz.
putting the nitre-oven between the burnei-s and the Glover
tower in such a way that, l)y means of valves, the burner-gas can
INTRODUCTION OF LIQUID NITRIC ACID. 517
be made to travel either through the oven or directly into the
chambers.
A. P. O'Brien (U.S. P. 694,024) describes a very complicated
nitre-oven, the practical results of which are unknown to me.
An intermediate process between the '' potting '' system and the
application of liquid nitric acid is the generation of nitric acid in the
ordinary retorts, fired with coals (p. 103 et seq.), but without con-
densing the vapours to liquid nitric acid by passing them straight
into the chambers. This system, which was followed in several
English works about 1880 (Jurisch, ' Schwefelsaurefabrikation,'
p. 130), seems to have neither the simplicity and (apparent)
cheapness of potting in the nitre-oven, nor the exact regularity of
supply by liquid nitric acid. It costs as much coal and very nearly
as much labour as the latter; and the passage of the vapours from
the nitric-acid retorts to the chambers presents great difficulties
on account of the unavoidable comdensation of liquid acid, which
is not entirely overcome or rendered harmless by lining the cast-
iron pipes with stoneware pipes, with an asphalt or asbestos cement
between them. If an acid-maker once emancipates himself (as he
ought !) from the old system of ** potting," he should proceed to
the thorough reform of making and using liquid nitric acid.
Introduction of Liquid Nitric Acid
Any of the apparatus described (p. 103 et seq.) may be employed
for the manufacture of nitric acid intended for use in vitriol-
chambers, but no special precautions are needed for obtaining
the acid in a concentrated state or free from lower nitrogen
oxides. On the contrary, any low-strength and impure acid ob-
tained in the manufacture of commercial nitric acid may be
turned to use in the vitriol- chambers.
It is, however, of the greatest importance for the process to
supply the acid in an even, continuous way, and to regulate the
supply to a nicety. This can be most simply done by a Mariotte's
vessel, as shown in fig. 213 on a scale of 'f^y The stoneware
vessel. A, containing the nitric acid, is closed by a caoutchouc cork,
a, holding a glass tube, b. The latter is the only channel for the air
which must take the place of any acid running out through the
cock, c. As the liquid above the level h h\ down to which the
tube h reaches, is supjiorted by atmospheric pressure, only the
'I
CONSTRUCTION OP THE LKAD CtlAMRKKS.
IMTBODUtTlON OF LIQUID NITRIC ACIU. 519
Iieight of acid below this can iaflueuce the speed of outflow ; and
this i-emains coustant till the level of the acid has suok below
this point. The glas-4 watcr-gau^, d, and the lead scale, e, admit
of observing the height of liquid within the vessel. The latter is
tilled up through the tube b, which ends in a funnel at the top.
During this either the cork must be raised, or it must be provided
with a separate open glass tube which at other times is kept
closed. The funnel / carries the acid into the glass or stone-
ware pipes conveying it into the chambers.
At some works there are two tanks which are filled up in turn,
one of them every \i liours, or both every 24 hours. The acid
is coDtinually running out of each tank. When one of them
is half empty, the other is just full ; and thus the vanatiou of
pressure is compensated to a certain extent ; but this plan cannot
at all vie in regularity with a Mariotte's bottle.
The Mariottc bottles sometimes become stopped up by grains of
saud &c. getting inio the >ligbtly opeiicd stopcock. Bode (Diugl.
Jovrn. vol. 220. p. 538) avoids this
by opening the cock full bore, i-jp. ^14,
stopping the neck of the bottle
tightly by a caoutchouc cork {as
shown in fig, 214), through which
a tube, a, goes down to the desired
depth, and is connected by an
clustic tube b with a metal or
glass cock, c, of 5 inch bore,
which serves for regtilaiing the
supply. further improvements
in this apparatus have been made
hy Liebig (Post's Z<-it8chr. f, d.
chem. Grossgew. 1878, part 2).
Formerly the nitric acid used
to be introduced in one or two
"tambours" (that is, small lead
chambers) about a2x lOx 12 feet,
or cylinders of 10 to 13 feet
diameter and 12 feet height, placed between the burners and the
main chamber. The second of these contained the "cascades"
or other sprcading-apparatus; it stood at a higher level than the
first tambour, into which it emptied its acid, and which only
520
C OXSTKUCTlOiV OF THE LEAD CHAMBERS.
Fig. 215.
served for furthler exposing it to sulphur dioxide and completely
driving oflP • the nitrogen oxides. This first chamber received
enough steam to prevent the formation of chamber-crystals, or to
decompose them if they arrived in solution from the second
chamber ; the acid collecting in the first small chamber ran away
into the main chamber. The first tambour is unnecessary ; with
proper regulation the nitric acid can be completely decomposed in
the first apparatus by means of SO2 and steam ; but the latter
ought to be supplied to such an extent that the sulphuric acid
formed contains rather more than four molecules of water to each
molecule of acid. At some French works the above-described
faulty arrangement of the cascades caused the acid in the second
tambour to contain a good deal of iiilric acid.
The tamboui*s have been mostly abolished^ and the process is
carried on in the main chamber; where the nitric acid, as is now
usual, is introduced through the Glover tower, the tambours are
entirely unnecessary.
Some manufacturers do not prefer to run the acid continuously
in a very small jet, but intermit-
tently in larger quantities. For this
a siphon arrangement is mostly
employed (fig. 215). a conveys
nitric acid into the stoneware vessel
b ; through its bottom passes a tube ?
reaching about three-fourths up
its height, and open at both ends.
This is covered by the wider tube d,
which is closed at the top and open
at the bottom, so that the acid fills
up the space between the inner tube ; \
and d. As soon as it has got to [
the top of the former, this, along
with fl?, forms a siphon which almost directly empties the contents
of d, whereupon this is slowly filled till the acid has again risen
to the top of til e inner tube, and so forth.
If the nitric acid were simply run into the chambers it would
cause very great mischief. It would dissolve in the chamber-acid
and quickly destroy the chamber-bottom ; moreover, much of it
would find its way outside with the chamber-acid without doing
its duty within the chambers. It is therefore necessary that no
CASCARES run NITKIU ACIU. 521
nitric acid should arrive as such at the bottom of the chambers,
but tliat, before reachiDg there, it should be decomposed into
gaseous osides of nitrt^en. This is doue by exposing it to the
action of sulphur dioxide, that is by the chamber-gases them-
selves. Before the introduction of the Glover tower, and even
long after, no other means were known for this purpose than
ri-. :
spreading the nitric acid out over a large surface so that it w&i
thoroughly exposed to the chamber-gases, and nothing could reach
the bottom in an undecomposed state. This was done by means
of stoneware or glass "cascades," of which there existed many
descriptions, which are fully explaineil and illustrated in the first
edition of this work, pp. 308 to 318. Since these cascades have
been almost entirely superseded by the Glover towers, we will here
522 CONSTRUCTION OF THE LEAD CllAMBER-S.
show only one of the best descriptions of cascades, that made by
Fikentscher, of Zwickau, and shown in fig. 216. The acid ran into
its top is spread over a large surface before reaching the bottom.
A fault inherent to all such systems is this : that there is no
really practical way of knowing whether the nitric acid has been
entirely decomposed before the chamber-bottom is reached. The
means adopted for this end at some works leave much to be
desired.
By far the simplest method of feeding, which dispenses with all
cascades, tambours, &c., is that otrunninff the nitric acid along with
the nitrous vitriol through the Glover tower » Few manufacturers
ventured to do this at first, because a loss of nitre was apprehended
with this plan ; but at most of the best-manacred works it has now
been done for many years without involving any extra consumption
of nitre, and it may be safely asserted that wherever a Glover
tower in proper working order exists, no other apparatus is re-
quired for feeding the chambers with nitric acid.
This is a principle universally accepted, and borne out not
merely by the practice of nearly all sulphuric-acid works where
liquid nitric acid is used, and where a Glover tower forms part of
the plant, but also by the practice of all the works following the
universal English plan of " potting^' nitre between the burners and
the Glover tower. It is all the stranger that in a modern treatise
on the manufacture of sulphuric acid (Jurisch, ^ Schwefelsaure-
fabrikation,' pp. 135 & 153) the old story is repeated of a great L)ss
of nitre in the Glover tower by reduction to NjO or elementary N,
on the strength of some absolutely inconclusive experiments made
by Vorster, and refuted 25 years ago by me (comp. Chap. VI.),
and of some alleged '^experiences'^ made in the North of Prance,
without any proof by positive data. Wherever figures are given,
e,g, Hurter's experiments of 1877, Jurisch is compelled to concede
that with careful work no more nitre was used after introducing
the Glover tower than before.
But the reduclio ad absurdum becomes even more complete by
the fact that Jurisch, when speaking of the deuitration of nitrous
vitriol, states that he decidedly prefers the Glover tower to all
other descriptions of apparatus ! Seeing that far more nitre passes
through the Glover tower in the shape of nitrous vitriol than in
that of nitric acid, and that the alleged destructive action of the
tower must be precisely the same in both cases, the above is a
INTRODUCTION OF LIQUID NITRIC ACID. 523
case of straining at a gnat and swallowing a camel. And this is
further illustrated by his remark (p. 156) that the amoiint of
nitre might be to some extent reduced by employing the Glover
tower only for treating the chamber-acid and itetiitrating the Oay-
Lu»!>ac acid in cascades, but that the slight gain in nitre would not
pay for the expenie of working l/ie cascades [which is almost nil, so
that the "gain in nitre" would amount to the same thing] !
The strangesc oversight committed by that obstinate adherence
to the exploded idea of the destruction of nitre in the Glover
tower is this: whenever the " potting" process is employed, t. e.
everywhere in England, in the majority of American and in many
continental works, the whole of the nitric acid is evolved in the
presence of the greatest possible excess of sulphur dioxide
and at the highest possible temperature ; hence Voster, Hurter,
Kuhlmann, Jurisch, and all those who believe in the destructive
action of the Glover tower on nitrous vapours, ought to wonder
why the greater part of the nitric acid is not reduced to N^O or
N in the ordinary nitre-oven t
Sorel (' Fabrication de I'Acide sulfurique,' 1887, p. SOi) states,
as the positive ri-sult of the experience of the St. Qobaiu works,
that there is no destruction of nitric acid in the Glover tower.
Since in exceptional cases a Glover tower is not available for
the introduction of nitric acid, we shall describe a very efficient
spray-apparatus, constructed for this purpose by M. Liebig
Mi CONSTRUCTION OF THE LEAD CHAMBERS.
(Xeitsch. des Vereiiis deiitscher Ingenieure, ]879, p. 111). Ii
consists of a lead steam-pipe li (fig. 217) with a platinum noule.
parallel to which ruos a glass pipe m, for coDvejing the nitric acid,
bent up in front and drawn out into a fine point, Tlie steam mafa-
ing past this causes a vacuum in the glass tube, and sucks acid
INJECTORS FOR NITRIC ACID. 525
through the latter from a stock-bottle^ a glass cock h regulating
the supply. The acid is divided into a fine mist, and none of it
arrives at the bottom undecomposed.
An apparatus for the same purpose, constructed by Mr. Stroof,
of the Griesheim Avorks, has been made known to me by that
gentleman, and is illustrated by figs. 218 & 219.
Fig. 218 shows the general disposition, fig. 219 the details of
the injector b. The nitric acid runs from a Mariotte^s bottle A
into a Woulfe's bottle B, standing in a glass dish, provided with an
overflow-pipe c, which conveys the acid, in case of the injector
breaking down, on to the cascade C. From the bottle B the acid
is sucked away by the glass injector b, whose steam-jet is connected
with the steam-pipe a by a stuffing-box. Such injectors are best
made of well-annealed water-gauge pipes, drawn out to a point.
The point projects but loosely into the suction-pipe, so that a little
air is sucked in as well, and no breakage can take place by expan-
sion. At a pressure of 1^ atm. the injector can carry away 16 cwt.
of nitric acid in 24 hours in the form of spray, along with a little
air. The mouth-piece of the injector must be contracted and
widened out again, like that of a fire-engine, to prevent any larger
drops forming at that place. The acid is thus completely converted
into a mist, and a sensible saving effected in compai*isou with
cascades.
Another glass injector for nitric acid has been described by
Burgemeister (Fischer^s Jahresb. 1880, p. 228). He employs a
platinum nozzle {not soldered with gold !) and a steam-jet placed
just below, the latter consisting of a platinum nozzle, about ^ inch
wide, inside the chamber, continued outside into a copper tube.
Both tubes pass through a lead pipe burnt into the chamber-side,
and are fastened in this with glycerine-lead cement.
At the Freiberg works, where, in consequence of the complete
cooling of the roasting-gases, they do not employ a Glover tower,
formerly the nitric acid was introduced by cascades. These have
been replaced by glass injectors, constructed by Wolf, as shown in
fig. 220 & 221. A glass tube, a^, 28 mm. wide, is sealed to a
narrower central tube, a*, at the place a. At b there are three
glass knobs for steadying the inner tube a^, and at c four small
glass knobs for steadying the platinum capillary d. Tube a^ is
provided with a funnel e for running in the nitric acid. The
tube d, made of platinum-iridium, is connected with a steam-pipe
526
CONSTRUCTION OP THE LEAD CHAMBERS.
and is held in the centre of pipe a* by an india-rubber cork /
and the knobs c, c. The outlet of a' is 6 or 7 mm. wide. The
nitric acid flows through e into the aimular space between a' and
a" and is sprayed into the chamber by the steam issuing througli
the platinum capillary. The whole is inserted into the chanaber-
side by means of an india-rubber joint at g^ so that it is easily
taken out and cleaned.
•-7
Potut (G. P. 122,920) introduces the nitric acid (or nitrate of
soda solution, com p. later on) by a steam injector into the pipe
leading from the Glover tower to the first chamber, with the
ridiculous assertion that thus two-thirds of the nitre are saved in
comparison with running the nitric acid down the Glover tower or
straight into the first chamber.
Other manufacturers inject nitric acid into the last chamber
(com p. next Chapter).
The simplest way of introducing the nitric acid through tlit
Glover tower is to run a suitable quantity of it into the uitfois-
vitriol tank at the bottom of the tower, and pump up the mixture
in the usual manner. It is sometimes preferred to carry the nitiic
I.NTROIICCIKO SOLUTION OF NITRATE OF SOKA. 53?
acid to tlie top of the chambers or the tower, and so run it into
one of the lutes of the toirer as required.
The slorivg of nitric acid on the top of the chambers or of the
(ilover tower is geuerally effected iu large stonewure receivers, or
else iu a number of smaller stoneware jars or ordinary glass
carboys, ail of which are connected by glass siphons, so that the
niDiiing olf by means of a tap-sipliOD need take place only from
the last vessel of the set (Hg. 222), Vessels proof against nitric
rij;. 23S. Fig. 223.
acid may also be composed of single pieces of stone joined togetlicr
by a cement made of finely-ground asbestos and a dilute solution
of silicate of soda, kneaded into a putty and preferably mixed with
ground sulphate of baryta.
E. Pohl (0. P. 30,188) employs irou vessels lined inside with
asbestos cloth soaked in paraffin. The riveting of the iron
shell is effected in the manner shown iu fig. 223, so that the acid
nowhere comes into contact with the iron.
Introducing Nilre as an Aqueous Solution of Nitrate of Soda.
There is yet a third way of introducing the nitre. Many yeara
ugo several w oiks ran their nitre as a solution in water into the
chambers. This has long since been discontinued, both because
sodium sulphate gets into the acid, which is not allowable for
nian\ purposes, and because the lead always wears away very
quickly at the point where the solution enters. The same method
was patented by Burnard {14th Aug., 1S75). The solution of
nitre was to be injected into the chamber in a thin jet, or, better
still, at once mixed with sulphuric acid by means of a steam,
jet, exactly similar to Sprengel's water-spray (videinfr&). The
528 CONSTRUCTION OF THB LKAD CHAMBERS.
chief object sought to be attaiucd in this process was an imaginary
saving of nitric acid, which in the decomposition of nitre by the
burner-gas was supposed to be I'educed to NjO and N. It h&^
been shown on p. 522, and will again be referred to in Chapter VI.,
that no sensible decomposition of this kind takes place at all ;
and any advantage accruing therefrom would be far more than
-counterbalanced by the difficulty of keeping the nitre solution
long enough in suspension to completely decompose it and to
prevent liquid nitric acid from getting at the chamber-bottom*
The process also takes so much steam that the acid in the first
chamber gets too weak. This is certainly contradicted by tlie
patentees (Chem. News, xxxvii. p. 203) ; but no independent
favourable testimony has yet been published^ and a saving of nitre
appears out of the question. It would, however, seem feasible to
tun a solution of nitre through the Glover tower along with
chamber-acid and nitrous vitriol, so that the nitre would he decom-
posed in the tower itself, sodium sulphate and nitrous vapours
being formed. Of course this plan, as well as that mentioned
before, is restricted to the case of all the sulphuric acid heins:
intended for decomposing salt, or for the manufacture of super-
phosphates, &c.
The last-mentioned plan is undoubtedly the simplest imagin-
able for introducing the nitre, uniting the advantages of both
solid nitre and nitric acid — easy regulation, introduction of anv
quantity at a time, dispensing with all apparatus for introducing
the nitre or manufacturing nitric acid, saving of labour and coals
(in the case of nitric acid), avoiding the handling of nitric-
acid carboys or of fluxed nitre-cake, the latter being an article
difficult to utilize to any extent. Unfortunately these advantages
are counterbalanced by a drawback, which has induced most manu-
facturers who have tried this process to give it up : it is found that
sodium sulphate crystallizes in the towers, tanks^ and connecting-
pipes and causes obstructions. It would be necessary to have two
Glover towers for each set, and to run the nitre solution down oulv
one of these, whose acid would not be used for the G ay-Lussac tower,
but for the salt-cake pans or for superphosphate only. This would
be verv inconvenient, and for smaller works not at all feasible.
Blinkhorn (E. P. 1084, 1878) runs a solution of sodium nitrate,
of spec. grav. 1*35, in a regular jet upon sulphuric acid contained
in a pot heated by the burner-gas, and draws off the solution of
sodium sulphate from time to time. This will hardly decompose
all the nitrate !
NITROUS OASES OBTAINED AS BY-PRODUCTS. 529
Potut (B. P. 7710, 1900) injects a solution of sodium nitrate
between the Glover tower and the first chamber (comp, p. 526).
Feeding the Chambers with Nitrous Gases obtained as
By-products,
Several proposals partly carried out in practice have had no
lasting success. Thus, for instance, the attempt has been made
in France to obtain oxalic acid as a by-product in evolving the
nitre gas by heating molasses with nitric acid and conducting the
vapours into the chambers. The yield of oxalic acid, however, was
not large enough to compete with its manufacture from sawdust
by fusing caustic potash. Not more successful was a proposal of
Laing and Cossins, to heat sodium nitrate with arsenious acid or
chromium oxide, in order to obtain arseniates or chromates along
with nitrous acid for the chamber-process (Wagner's Jahresb.
186 "i, p. 207). It is also quite feasible to convey any nitrous gas
generated in making arsenic, antimonic, phthalic acid, &c. by
means of nitric acid, which formerly used to be lost, into the lead
chambers; but this process, which (like all similar ones) can
hardly be so conducted as to give a sufficiently regular supply of
nitre to the chambers, is no longer called for, since the respective
works now regularly regenerate nearly the whole of the nitric acid
by mere contact of the vapours with an excess of air and water in
^'plate-towers^' or similar apparatus (comp. p. 125 et seq.).
An ingenious process, invented by Dunlop, was for many years
carried out at St. RoUox, but it has not been used at the new works
at Hebburn belonging to the same firm. A mixture of common
salt, nitrate of soda, and sulphuric acid is heated in large iron
cylinders ; the principal reaction setting in is as follows : —
3SO4H8 + 2NaN03 + 4NaCl = SNajSO^ + N2O3 + 4C1 + SU^O ;
but the further deoxidation of nitric acid will only be prevented
by keeping within certain limits of temperature. Thus there re-
mains a soluble residue of sodium sulphate, whilst chlorine and
nitrous acid are given off in a gaseous state. The two gases are
separated by passing them through a series of leaden Woulfe's
bottles filled with sulphuric acid of 1*75 spec, grav., which retains
the nitre-gas, being converted into " nitrous vitriol ^' and used as
such {vide infi'a) ; the chlorine passes through without absorption
and is utilized for bleaching-powder. The advantage of this
VOL. I. 2 m
530 CONSTRUCTION OF THE LEAD CHAMBERS.
process is^ that chlorine is obtained direct from salt without
making any hydrochloric acid and without wasting manganese.
The drawbacks are : — that the nitre-gas has to be evolvcsd again
from the nitrous vitriol^ which at that time could only be done by
diluting with hot water^ necessitating a reconcentration of the
vitriol ; that there is a danger of losing nitrogen compounds ; and
that the apparatus is very complicated. This process consequently
did not obtain permanent success.
At the Uetikon works^ near Zurich^ nearly all the nitre re-
quired for the chambers is obtained in the manufacture of iron
mordant for dyeing purposes. This is made by treating ferrous
sulphate with nitric acid^ and thus oxidizing the ferrous to ferric
sulphate. The nitric acid is thereby reduced mostly to lower
oxides of nitrogen; these gases are conveyed into the vitriol-
chambers^ and there do exactly the same duty as if the nitric acid
had been directly supplied to the chambers. Recently the manu-
facture ofcupric sulphate from metallic copper^ sulphuric and nitric
acid has been introduced at the same works^ equally carrying all
the nitrous vapours into the vitriol-chambers.
Supply of Water as Steam or Spray.
The water required to produce UsS04 and to dilute this to the
point required for the practical working of the chambers must be
presented to the gases in as fine a state of division as possible.
This was formerly in all cases and is still generally effected by
injecting into the chamber a certain quantity of steam, which
rushes forward and on its way is condensed to a mist of very fine
particles of liquid water. At many works^ however, water is now
injected in the shape of a mechanically produced spray*
The Steam
is always generated in an ordinary steam-boiler, since boilers
placed above the sulphur-burners have been given up everywhere.
The boilers are constructed in the usual manner, but are mostly
made for low pressure, rarely working above two atmospheres^
more frequently only at one or one and a half atmosphere; in the
south of France they work at three or three and a half atmo^
spheres. A high pressure has no object so long as the liquid is
spread over the whole chamber-space ; for even low-pressure steam
fulfils this requirement and sufficiently assists the draught. Low^
SUPPLY OF STEAM. 531
pressure steam is more easily kept at a uniform tension tlian high-
pressure : without this no regulation of the supply of steam to the
chambers by the attendant is of any avail. High-pressure steam
certainly condenses less readily than low-pressure steam ; but this
is a doubtful advantage^ so long as the steam possesses enough
'^ carrying-power" to convey the minute globules of water right to
the other end of the chamber. Experience has shown that this is
the case even with low-pressure steam; at most English works they
employ only a single jet at one end of each chamber^ and consider
this quite sufficient to supply the whole chamber with moisture^
but I do not like this practice (see p. 534).
Of course low-pressure steam may be obtained from a boiler
working at high pressure by means of a reducing-valve. Thus at
the Oker works the steam-pressure in the boilers is 8 to 3*5
atmospheres^ and is reduced to 1 to 1*5 atmosphere for feeding
the chambers. At small works the same boiler may be utilized
for supplying the chambers and for driving the machinery of
stone-breakers^ air-pumps, and so forth.
It is also almost a matter of course that the chambers may be fed
with the exhaust-steam of engines^ if these are worked in such a
way as to leave some pressure in the exhaust. The utilization of
the waste steam of the Gay-Lussac air-pump for this purpose had
been practised by myself for many years, as described in the first
edition of this work (1879), vol. i. pp. 393 & 565. A proposal
not essentially differing from this was patented much later by
Sprengel (No. 10,798, 1886).
At some large works, in order to control the uniform tension
of the steam, so important for the regularity of the chamber
process, registering steam-gauges are employed, which show the
tension during the whole day on a sheet of paper wrapped round
a drum making one revolution in twenty-four hours. Such a
gauge, made by Schaeffer and Budenberg, of Magdeburg, is de-
scribed in 'Dingler's Journal,' ccxxvii. p. 519.
The conveyance of the steam to the chambers usually takes place
in cast-iron pipes, with one or more branches for each chamber^
The main-pipes in any case, and, if possible, also the branch-pipes^
considering their great length, should be surrounded by bad con*
ductors of heat to restrict radiation as much as possible, and avoid
a considerable loss by condensation of water.
The pipes must always be laid with a slight fall towards the
2m2
532
CONSTRUCTION OF THE LEAD CHAMBERS.
boiler^ so that the condensed water may run back. Where^ from
local circumstances^ this cannot be done^ automatic apparatus for
removing the water should be fixed at the lowest points.
Of course the size of the main-pipes must correspond to the
number and size of the chambers. When more than one steam*
boiler is required^ they are placed together and their main-pipes
connected so as to equalize the pressure. The branches for each
chamber need not be above 1 inch wide^ even for large chambers
(up to 70,000 cubic feet) supplied by one jet. They are made of
wrought-iron tubes, a (fig. 224), sometimes of copper, up to a short
Fig. 224.
distance from the chamber, where they end in a cock or valve, b,
to which a lead pipe, c, equal in width to a, is attached and pro-
jects into the chamber itself. It is not, however, burnt to the
chamber-side, c; but a short wider tube, rf, is burnt to this,
and c is loosely put into it, the joint being made tight with tar,
cement, &c. Sometimes in lieu of this an india-rubber cork is
employed, but this does not last long. In the latter case, if the
outlet is stopped up by lead sulphate, the pipe c can easily be drawn
out and cleaned and no platinum nozzles are required (as had
been proposed by Scheurer-Kestner).
SUPPLY OF STEAM. 533
The same figure shows another commendable contrivance^ viz. a
simple mercurial pressure-gauge y consisting of a bent glass tube, f,
with a scale^ g, connected by means of a caoutchouc bung with a
branchy h, of the lead pipe c. Thus the pressure behind the regu-
lating.cock can be observed at any time ; and the chamber«manager
has thus a means of very accurately regulating the supply of steam.
Any water condensed in the gauge can be easily allow^ for.
A good steam-cock is preferable to a wheel-valve, because the
wheel does not show how far the valve is opened, whilst the handle
of the cock can be fitted with a graduated arc so that its position
can be fixed with precision.
Automatic steam-regulators, if reliable^ save a great deal of
trouble^ but do not dispense with constant supervision on the part
of the attendant, as they are somewhat liable to get out of order.
In England the usual way is, or formerly was, this : to employ
only one jet of steam for each chamber^ mostly beside, above,
below, or even within the pipe conveying the gas from the burners,
the Glover tower, or the preceding chamber. Some introduce the
steam quite near the top, others in the centre of the chamber^end.
A single steam-jet sufiices, if the length of the chamber does not
exceed about 130 feet ; in longer chambers it would not carry
right through.
It is maintained by English practical men that a single steam,
jet from a 1-inch pipe is quite snfiicient for feeding chambers up to
130 feet length, and also that* the distribution of moisture through
the chambers is thus properly efibcted. By arranging a single
8team-jet, the cost of cocks, branches, &c. is saved, and the
regulation of the supply of steam is much simpler and easier than
if, for instance, four cocks were to be opened a quarter as much
as the cock of a single jet. It is also true that in this way the
front part of each chamber, which makes most acid and evidently
requires the greatest supply of moisture, actually receives it ; but
this does not hold good of the first chamber, which obtains a
considerable portion of its steam from the Glover tower, so that
a steam-jet placed in the just described way is certainly not in the
right position. The steam-jet should enter the chamber near the
top, or at least in the upper part of the side. Experience has
shown that it is not advisable to send the steam into the lower
portion of the chamber.
Most experienced managers, however, now agree that the single
534 CONSTRUCTION OP THE LEAD CHAMBERS.
steam-jet for each chamber is a faulty appliance. The chamber
should not be left to haphazard supply of its different parts witli
the necessary amount of moisture^ but each part should receive
just what it needs.
On the Continent^ indeed^ most manufacturers have alwa^^s
preferred employing a number of steam-jets for each chamber, so
as to make themselves independent of any casualties in the proper
distribution of steam by a single jet. These branch jets are intro-
duced at right angles to the direction of the gaseous current either
in the long chamber-sides^ not far from the top, or, which is
most usual, through the roof of the chamber, so that the single
jets can be regulated by a man walking over the top. Thu9^ for
instance, at the Oker works there is a steam-pipe extending above
the chambers, from which, at intervals of 17 feet, branches of
I -inch bore enter the latter; from these the steam issues, by
several small openings immediately below the top^ in several
directions (BiiLuning, Preuss. Zeitsch. 1877, p. 137). A similar
arrangement exists at Aussig and elsewhere. In all these cases
each branch-cock must be regulated separately.
Where a Glover tower is in use, the first steam-jet should not be
in front of the first or ^Meading '^ chamber, as this part receives
enough steam from the Glover tower ; the first steam-jet should
be 20 or 80 feet, or even farther, from the front side.
An apparatus by which steam can be introduced at many places
and yet regulated at a single point has been described by Scheurer-
Kestner (Wurtz, Diet, de Chimie, iii. p. 149) ; it is shown in
fig. 225 : a is the copper main-pipe running in the centre of the
chamber-top, and held fast by the joist £ 6j, as well as the
branches c. The latter are arranged alternately on the left and
right hand at distances of 16^ feet ; they are made of lead, pa^
through the chamber-top, and are burnt into it. The arm rf,
covered with straw rope, serves for making the communication
between a and c. Both pipes have hydraulic lutes, so that only a
very low pressure can be employed. The main-pipe, a, is provided
with a cock, and the supply of steam regulated by this. The
steam, entering the pipe at the front end of the chamber, will
principally escape through the first branches, where it is most
needed, because in the beginning a large quantity of unchanged
sulphur dioxide is present. The pipe a has suflScient fall for
emptying the condensed water. (This arrangement seems to offer
SUPPLY OF STEAM.
535
no advantage over simple branches on a main-pipe^ and has the
great drawback that only a very low pressure can be employed,
as the water is easily thrown out of the hydraulic joints a and c.
Moreover, it disregards the necessity of supplying the various
parts of the chamber with different amounts of steam.)
Perfectly absurd is the arrangement given in every edition of
Payen's ' Precis/ even up to the last one (1877), and copied from
it into many other treatises. Here the steam-jets are shown partly
in the chamber-bottom, coming through the chamber-acid. No
practical man can imagine that this plan, if it has been actually
carried out anywhere, has not been discontinued at the first oppor*
tunity; for the shaking by the steam must gradually cause a
leakage at the joint, which cannot be got at, owing to the chamber-
floor, nor can it be repaired till the chamber has been entirely
emptied.
The total quantity of steam required for a set of chambers, which
should be known approximately in order to fix upon the boiler-
space and the size of the main-pipes, of course depends, first,
upon the quantity of sulphur to be burnt, secondly upon the
existence of a Glover tower, and thirdly upon the strength to
which the acid is brought in the chambers. A general rule,
therefore, cannot be laid down. The two latter conditions are
536 CONSTRUCTION OF THE LEAD CHAMBERS.
partly reciprocal ; the stronger the acid is made in the chambers,
the less water is evaporated in the Glover tower, and vice versd. If
we assumCj adopting a proportion very usual in England, that all
the chamber-acid is brought up to 124° Tw., and that it is con-
centrated in the Glover tower up to 148° Tw.^the amount of steam
required will be as follows: —
Each pound of sulphur burnt requires,
Ist, for forming - SO4H2, -^ water =0-5623 lb.
2nd, for diluting it down to 124^ Tw.
( = 70percent. SO,H2),^2^|| = 1-3125 ,.
1-8750 „
Of this nothing is lost with the escaping gas,
as this passes in the Gay-Lussac tower through
strong vitriol ; on the contrary, the Glover
tower saves the steam corresponding to a con-
centration from
124° (=70 percent.) to l48°Tw. (=80 per
cent.), viz.^^^ -^ =0-43/0 lb.
Leaving 1*4375 lb.
which must be supplied to the chambers. To this must be added
a certain quantity for water condensing in the steam-pipes ; but
this cannot be estimated generally^ since here everything depends
upon the length of the pipes, their thickness, surroundings, &c.
On the Continent the chamber-acid is kept more dilute and cor-
respondingly more steam is used. It is safe to say that the steam
to be generated in the boiler, without a Glover tower, amounts to
about 2^ times — with it, to about twice the weight of sulphur
burnt.
Employment of Water in the form of Spray,
Instead of feeding the chambers with steam, Sprengel (patent
of October 1st, 1873) proposed liquid water in the form of a
INJECTION OF WATER AS SPRAY.
537
fine spray. His reasons are these : — that the steam increases the
volume of the gases by its heat, and consequently more chamber-
space and nitre are required, which can be avoided by introducing
the water in a liquid form, suflSciently divided ; and that the
cost of evaporation can be saved in this way. The water is
made into a spray by the employment of steam, a steam-jet of
301b. tension escaping through a platinum nozzle in the centre
of a water-jet, as shown in fig. 226 (where a is the steam-pipe,
b the water-pipe) ; 20 lb. of steam is sufficient for converting
80 lb. of water into a mist. Such jets are arranged in the chamber-
sides, at distances of 40 feet apart, and supplied with water from
a tank fixed at some height above. Sprengel assumed that
FiK. 226.
two-thirds of the coal can be saved in this way, instancing the
M'orks at Barking Creek, where at the same time a saving of
6| per cent, pyrites and of 14| per cent, nitre is said to have been
effected. At those works there was no Gay-Lussac or Glover
tower. In the case of factories working with a Glover tower,
Sprengel estimated the saving in coal at a third less (Chem. News,
xxxii. p. 150). Of course the water- and steam-cocks must be
exactly regulated, and the two nozzles must have a particular
shape, so that only a fine mist and no lai^er drops shall be
formed, which would at once fall to the bottom and only dilute the
chamber-acid.
A different way of producing a spray or mist of water instead
of a steam-jet for feeding vitriol-chambers is employed at the
Griesheim works, and has thence been introduced with great
538 CONSTBUCTION OF THE LEAD CHAMBERS.
success into several other factories. The spray is here not produced
by the injection of steam^ but by allowing the water, at a pressure of
two atmospheres, to issue from a small platinum jet against a small
platinum disc. Two rows of such water- jets are introduced through
tubes in the chamber-top, each tube about 20 feet away from the
other. Thus the whole chamber is uniformly filled with a fine
mist, which, together with the steam comiug from the Glover
tower, supplies all the water required for the chamber-prcicess.
The water must be carefully filtered, as otherwise the jets would
soon be stopped up ; but this trouble is far more than compensated
by the considerable saving in fuel caused by doing away with the
chamber boilers. The fear formerly entertained, that the intro-
duction of the moisture in the shape of liquid water would reduce
the temperature of the chambers below that most favourable for
the acid-making process, is entirely groundless. At Griesheim it
was noticed that the temperature of the gases, arriving from the
Glover tower sometimes at only 35^ C, quickly rose within the
chambers to 50° C. Similar observations have been firequently
made, most extensively by Lunge and Naef (comp. Chap. VII.) .
This is explained by the fact that the evolution of heat, con-
sequent upon the chemical reactions going on within the chamber,
is far more important than the heat brought in by the steam,
and that, in fact, the local cooling produced by the water
being supplied in the liquid form is actually beneficial in most
cases.
It might be objected to the introduction of the water in the
shape of a spray, that steam is preferable on account of being only
gradually condensed in its onward course within the chamber, and
that the moisture would thus be more uniformly distributed
through the chamber. But this objection is not at all valid, and
would not be so even if no sulphuric acid were present in the
chamber. Calculation shows that the gas introduced for each
kilogram of sulphur, whose volume at 50° C. and 760 millim.
pressure amounts to 83 4*5 litres, can contain only 0*6868 kilogram
of aqueous vapour, whilst the total amount of water is nearly four
times as much, and three-fourths of the steam entering into the
chamber must therefore be at once condensed into water. This
calculation, given in our 1st edition, pp. 348 & 349, is not repeated
here, since it docs not take into account the fact that the tension
of aqueous vapour within the chamber is very much reduced by
INJECTION OF VTATEVL AS SPRAY. 539
the presence of sulphuric acid^ and it is hence useless for our
purpose. Hurter (J. Soc. Chem. Ind. 1882, p. 51) somewhat
more correctly applies to our case Regnault's table for the
tension of aqueous vapour in sulphuric acid of various strengths,
and he there gives a diagram which allows of finding this tension
for any intermediate concentration of acid. But this is in-
correct for the principal working part of the vitriol-chamber ; for
Regnault's determinations only go as far as 35^ C, that is much
below the ordinary chamber-temperature, and it is not admissible
to calculate tensions at 60°, 80°, or even higher temperatures by
simply applying Regnault's table or Hurter's diagram to them.
This gap has been filled by a set of elaborate observations made by
Sorel, and first rendered accessible to the public by me (Zeitsch.
f. angew. Ch. 1889, p. 272). Sorel's table extends to acids
from 44 to 82 per cent. H2SO4, and to temperatures from 10°
to 95°. We have given it in the third Chapter, p. 196, where
the specific gravities corresponding to the acid percentages have
been added for the reader's convenience. At the close of this
Chapter we shall give a table for reducing volumes of gases to
the conditions of the vitriol-chamber atmosphere, which equally
takes into account the aqueous-vapour tensions of sulphuric acids
of various strengths.
The importance of this table will be indicated at present by only
one example. In a special instance the temperature close to the
chamber-side was 80° ; the acid running down the side stood at
114° Tw.=66 per cent. H2SO4, and the aqueous-vapour tension at
this place was, therefore, =39 millim. Only 6 centim. (say 2^
inches) within the chamber the temperature was already 95° ; but
at this temperature an acid, whose aqueous-vapour tension is=39
millim., must have a strength of 128 J° Tw. = 72*33 per cent.
H2SO4, and this was found to be really the case. We shall see
the importance of this in Chapter VII.
Looking at the great reduction of the tension of aqueous vapour
by the presence of sulphuric acid, we must conclude that the
steam introduced into the chamber must be condensed almost
immediately into a liquid mist, and this must reduce the alleged
superiority of steam in " carrying power " to a properly com-
minuted spray of water, introduced at high pressure, to almost nil.
Of course the water mtist be properly comminuted ; otherwise,
that is when it drops from the jets in the shape of rain, it dilates
540 COSSTROCTION OF THE LEAD CMAMBEBS.
the chamber-acid to aa intolerable degree, and this is all the
more injurious as this dilute acid floats on the top of the stronger
chamber-acid, and is not noticed for a long time at the places
where the acid is drawn off, till it becomes too late to meet the
evil at once. This accident will happen whenever the jets are out
Fig. 229.
of order, and this has, very unnecessarily, caused several works to
abandon the plan of introducing the water in the shape of a liquid
spray.
A special platinum jet for converting water or acid liquids into
ft thin spray has been constructed by P. Benker, of Paris, aad is
shown in fig. 227. In this jet the distance between the noszle a
and the disc b is adjustable, the disc being moved backwards or
forwards by means of a rod, at the end of which is cut a 6ne
thread, which works in a similar nut c, placed on the top of the
IXJECTION OP WATEIt AS SFKAY. Oil
cylinder. The screw and nut are made _,
of en alloy of platinum and iridium, so
that there is no fear of their wearing
out. In this manner the best distance
betireeii a and b can be easily attained,
and by removing b altogether the
nozzle a is easily cleaned. More
I'ecently Benker has employed spray-
producera of the shape shovn in
fig. 228, with the platinum-iridium jet
cased in antimony ^lead, and without
an adjusting-screw.
Korting Brothers' spray-producer
(fig. 229) (Zeitscb. f. angw. Chem.
1888, p. 401) contains within the con-
tiacted part a metal spiral, which by
the pressure of the liquid is kept
tightly ill its place, whilst the liquid on
passing through the helicoidal channel
takes a rotating movement, so that,
on issuing, it is projected equally OD
all sides as a conical spray. Tlie
nozzle and spiral spring can be ar-
ranged for spray of any degree of
fineness. This apparatus was origin-
ally intended for damping the air in
cotton-mills &c. for precipitating dust,
for absorbing acid vapours, and so forth.
It has also been made of platinum, and
is in several places used for producing a
fine spray of water in vitriol-chambers.
These K or ting's spray-producers
(" Streudiisen ") were also made of
antimony-lead with a platinum lining,
but this did not stand the corrosion in
acid- chambers, and had to be replaced
by solid platinum nozzles. Even these
do not last eo long as nozzles made of
glass, as shown in fig, 230. The glass tube a is drawn out to a
capillary point at a', where it is cut oSf quite straight so that the
542 CONSTRUCTION OF THE LEAD CHAMBERS.
jet of water comes out centrallj^ not sideways ; in 24 hours 9UL)
or 1000 litres of water should be delivered from this. This glass
nozzle is fixed in the antimony-lead part d by means of a iIud
india-rubber washer b, the joint being made tight by the vater
pressure^ and both orifices being at the same level. Within a
the Korting spiral e, made of gun-metal (this metal can be employed,
as no acid penetrates here) , is fixed by the thick india-rubber tube c.
This apparatus is set in the chamljer-top in such a way thit
the jet comes out horizontally^ and that the whole can be re-
moved at will for the purpose of cleaning or inserting a nev
glass nozzle.
In practice the orifice of the glass tube a is about 0*75 mm. bore,
that of the hard-lead nozzle d is 4 mm. wide. In order to fix a iu
the hard-lead part d, the small india-rubber washer d is put od
and placed with its orifice at the top, and d is put on with a twisting
motion^ but without exercising much pressure or trying to centn*
a exactly in d, the water pressure afterwards accomplishing this
much better. With a pressure of 3 to 5 atmospheres and a
properly adjusted nozzle no droplets whatever are formed, but a
uniform mist travelling over a great distance. The Korting spray
is bought in the ordinary ^ay, and is fitted at the works with a
glass jet^ as shown here with every detail.
[Glass-nozzled spray-producers, precisely as here shown, have
been employed for many years at one of the best conducted
continental factories, and are still preferred there to every other
form of apparatus of this kind. Since their introduction all
previous trouble in feeding the chambers with liquid water has
disappeared. The orifice of the glass jet has | millim. bore, that
of the hard-lead nozzle 4 millim. bore.]
A special advantage of the introduction of liquid water in the
form of spray is that the temperature of the chambers is kept
lower than when employing steam, since the latent heat of the
latter, becoming free in its condensation, decB not come into play.
How useful such a cooling effect is, will be seen in Chapter VII.
The only drawback to the spray system is the liability of the
orifice of the spray-producer to get choked up by dirt, or else to
be widened by corrosion. The former is avoided by careful
filtration of the water, the latter by a suitable construction. Glass
nozzles, as shown above, are much better even than platinum. I
have found this system applied for many years with full succes:
=3h
I
CHAMBERS SUPPLIED WITH WATEll-SPRAYS.
543
in a number of the best managed works^ e.g. Griesheim and Aussig.
AVherever it has proved unsuitable^ this has been due to want of
attention.
During the warmer portion of the year the whole of the chambers
can be supplied with water as a spray ; but in the winter season
the back part of the system must sometimes get a little steam.
For all descriptions of spray-producers the water must be care-
fully filtered^ preferably by means of sponge-filters. Benker more-
over places wire-gauze in front of every spray-producer. He works
these at a pressure of from 2\ to 5 atmospheres, by means of a
small intermediate cylinder^ about 3 feet wide and 6 or 9 feet high.
For the first start a little high-pressure air is employed ; after-
wards the feed-pump gives sufficient pressure^ which can be reduced
by means of a tap if too much water is discharged.
This plan is very well illustrated by figs. 231 & 232 (folding
plate) ^ which at the same time show a chamber as erected by
Mr. Benker for the French ''high-pressure style." A is a small
Fig. 2.33.
air-pump^ B the water-pressure vessel, made ot an old steam-
boiler, with two safety-valves, a and b. One of these, destined
for the air, is very small ; but the other, for the water, must
be lai^e enough to discharge the whole of the water supplied by
pump D ; otherwise B might burst. The second valve is weighted
i atmosphere more than the air- valve. The level of the water in B
must be kept 12 inches below the top. Pump D is an ordinarv
541' CONSTRUCTION OF THE LEAD CHAMBERS.
feed-pump, kept continuously going; between B and D is interposed
the sponge- filter C. The water rises in the high-pressure pipeEE
to the top of the chamber, where there are taps c c at distances of
about 16^ feet. Behind each tap is placed a leaden filter (shown
on a larger scale in fig. 233), with a very fine wire-gauze sieve;
then comes a swan-neck pipe, which passes through the chamber
top^ and this ends in a spray-producer, fig. 228 [or else 230].
From time to time the air-tap on B is opened in order to replace
the air dissolved by the water under the high pressure.
ArrangemefiU for producing the Draught in Vitriol-Chambers.
The draught necessary for driving acid-chambers is produced bv
various agencies, the most important being the high temperature
with which the gases leave the burners and enter the chambers,
which counterbalances the greater density of the burner-gases
when compared with that of air in the cold state. We shall cal-
culate these factors for the various cases in question, employing
the following values for the density of gases and vapours at CP C.
.and 760 mm. mercurial pressure: —
grams.
1 litre of dry atmospheric air weighs . . . 1-2932
1 „ „ oxygen 1-4298
1 „ ,, nitrogen 1-2562
1 „ „ sulphur dioxide 3-8721
1 „ aqueous vapour 0*804343
We shall begin with burner^gases from brimstone, the normal
•composition of which has been calculated (p. 397 et seq,) =0*1123
SO3 + OO977 0 + 0-7900 N = 1 litre of gas. This must weigh at
O^ and 760 mm. pressure :
01 123 X 2-8731 +00977 X 1-4298 +07900 x 12562
= 1"4547 grams.
Taking the temperature in the outlet-pipe from the brimstone-
burner to average 100° C, (which is much below the actual
temperature), the above 1*4547 grams would occupy a space of
?Z^t^ = 1-3663 litres, or 1 litre of the burner-gas at 100° C
1-4547
and 760 mm. = ttqc^q = 1*0647 grams.
PRODUCTION OF DRAUGHT. 545
Atmospheric air at 0^ and 760 mm. pressure weighs per litre
1*2932 grams; at 20^ this quantity occupies the space of 1*0733
litres, at 35°=1'1282 litres, so that even at the highest summer
temperature, say 35° C, 1 litre of air weighs as much as rrrooo
= 1*1463 grams; it is therefore in any case heavier than the burner-
gas at 100° C.
The aqueous vapour always present in the air need not be taken
into account, since by its expansion in the heat of the burner it
can only*increase the difference between the weight of the gas and
that of the air.
Owing to the fact that the gaseous mixture in the vertical pipe
of the sulphur-burner is lighter than air, it must issue out of the
top of the pipe into the chamber with a speed corresponding to
the excess pressure of the atmosphere acting upon it from bdow.
It must therefore by itself exercise a pressure upon the gas in the
lead chamber. Its speed or the draught increases with the height
of the vertical pipe ; and the latter therefore ought to enter the
chamber-side as high up as possible. By thus providing more
than sufficient drawing-power the supply of air is assured in any
case ; and its excess can always be moderated by narrowing the
area of the inlets.
A second cause of draught is the formation of sulphuric acid
itself, as the space occupied by the consumed gas cannot remain
empty, and must at once be filled again. The condensation of
the gas to sulphuric acid thus acts as an aspirator.
A third cause of draught is the vertical pipe taking the gas away
from the last lead chamber or the chimney with which it is con-
nected. As the gas in these contains all the nitrogen introduced
into the chambers with only 5 per cent, of oxygen (a mixture of
95 vols. N-f 5 vols. O has the litre weight grav. 1*263; ordinary
atmospheric air 1"293), as it is saturated with aqueous vapour,
and as it is usually warmer and never can be colder than the
atmospheric air, it must necessarily be lighter than the latter ;
this is evident without any calculation.
If the nitrogen-acids are not recovered by a special process, to-
be explained later on, the gas certainly contains a little of those
acids and of sulphur-dioxide, by which its specific gravity is
somewhat increased. We shall, however, see that their eflFect
is very slight, and does not materially interfere with the causes
producing the draught.
VOL. I. 2 N
546 COXSTHUCTION OF THE LEAD CHAMBERS.
The draught produced by all the above-mentioned causes regu-
lates the quantity of air which can enter the apparatus by openings
of a certain size. We have already seen^ on p. '697, that we mu&t
not introduce the exact quantity of air required for transforming
the burnt sulphur into SOs, but a certain excess, which we have
calculated = 5*18 vols, of oxygen upon each 14 vols, of SO^.
With this, for each 14 vols, of SO^,
14 + 7 + 518=21+ 5-18=26-18 vols, oxygen,
and 79 + 19-50= 98-50 „ nitrogen,.
together 124'68 „ atmospheric air,
must be introduced into the chambers. From this it follows that for
121*68
each vol. SOg iX" =8*906 vols, of air are required. Now 1 litre
of SO2 at 0° and 760 millims. pressure weighs 2*8731 grams, and
SO3 consists of equal parts by weight of sulphur and oxygen.
Accordingly 1 litre of SOj at 0° and 760 millims. contains
'- — = 1*43655 gram sulphur,
and 1*43655 „ oxygen.
Thus for each 1-43655 gram of sulphur burnt 8*906 litres air at
0° and 760 millims. are required. Since
1*43655 : 1000 : : 8-906 : jc,
QQTift
each 1000 grams or 1 kilogram sulphur requires , .o^^.,==6199
litres air at 0° and 760 millims. pressure to be introduced into the
sulphur-burner, weighing 6199x1-2932=8017 grams or 8017
kilograms.
At 20^ C. this weight would occupy ^^U,-^-x 6199 = ^565:5
litres.
All these calculations refer to dri/ air. If the air is saturated
with moisture, its volume is increased by the vapour-tension e for
the temperature in question, according to the formula :
^n V X 760
V =- , - >
o—e
IMPORTANCE OF KEOULATINO THE DRAUGHT. 54?7
where b is the actual barometric pressure. At 20^ C, c is = 17*4min.,
and for b = 760 the above 6653 litres will be = 6809 litres, if
saturated with moisture.
The last increase of 156 litres is only fully realized in the ex.
ceptional case of air completely saturated wit^ moisture. As this
increase is only 2*34 per cent, of the volume of the necessary dry
air, whilst, according to the calculation on p. 398, nearly 25 per
cent, of the theoretical quantity of air (that is, more than ten times
as much) is introduced in excess, the changes in the moisture of
the air and the differences of volume resulting therefrom are of no
practical consequence. We shall therefore not enter into a calcu-
lation of the differences caused by the real percentage of moisture
in the air.
In the case of pyrites-kiln gases, it follows from the data given
on p. 399, that for each 100 pts. of sulphur employed as FeSo,
375 pts. oxygen must be supplied for oxidizing the Fe,
1000 ,, „ „ forming SOg,
500 „ ,, ,, oxidizing this to SO;,.
Since 1 litre of air at QP and 760 miilims. pressure weighs 1*42S8
gram^ at this temperature and pressure
375 grs. O give 262-3 lit. mixed in the air with 9867 lit. X.
1000 ., „ 699-A „ „ „ 26311 ,,
500 „ „ 349-7 ,, ,, ., 1315-5 ,,
1875 „ „ 1311-4 „ ., ,. 4933-3
Theoretically, then, for each kilogram of sulphur consumed
as FeSg, 131 1*4 + 4933-3 = 6244-7 litres air at 0° and 760 miilims.
pressure must be supplied.
The normal pyrites-burner gases, as calculated p. 400, contain
8-59 per cent, by vol. SOo, 9*87 O, and 81-54 N (not reckoning any
SOa present).
1 litre of this gas at 0^ and 760 miilims. weighs
0-0859 X 2-8731 +0-0987 x 1-4298 + 0-8154 x 1-2562=1-4122 grm.
whilst the 1 litre of the gas resulting from the combustion of
brimstone, according to our former calculation, weighs 1-4547.
o V *>
¥W •'I /W
548 CONSTRUCTION OF THE LEAD CHAMBERS.
I
The former being, under equal conditions, lighter than the latter, I
consequently gives stronger draught.
Of course the volume of air necessary for a certain consumptiou
of sulphur is also dependent upon the elevation of the site above
the level of the sea, which regulates the mean barometrical pressure.
Thus at Munich a quantity of air will occupy a space larger by 5*3
])er cent, than the same quantity at Widnes or New York.
It is easy to introduce the minimum of air required for proper
work. But this is not all ; an excess of air is just as hurtful a<
a deficiency, although not to the same extent. Air in excess cools
the gas, and thus may sometimes interfere with the process;
it fills a portion of the chamber-space and renders it inoperative ;
it dilutes the gas and weakens the energy of the chemical action.
The regulation of the supply of air must therefore be accurate,
and must be adapted to the frequent variations in the state of the
atmosphere. This must be done by great attention in enlarging
or diminishing the openings serving for introducing the air and for
taking away the gas. By either means the supply of air can l)e
diminished ; but it is not indifferent which of them is selected. Br
the latter the draught acting upon the contents of the chambers at
the end of the apparatus, by the former the pressure upon the
contents of the chambers at the beginning of the apparatus, is
lessened. With the latter method the pressure inside the chamber
is increased ; with the former it is diminished. Accordingly, if the
chimney-draught is too much cut off, the gas issues forcibly from
any openings in the chambers, &c., whilst the air may enter properly
by the holes in the front of the sulphur-burners. If, however,
these latter are stopped up too far, the chambers suck in air in any
places not completely closed against. the atmosphere.
The draught may also be increased in two different ways, viz., by
enlarging the opening in the exit-tube, or by increasing the inlet-
holes in the door of the burner. Then the chambers, if the exit-
tube is not suflBciently closed, suck in air; if, on the other hand,
the inlet-openings are too wide, gas is forced out from any leaks in
the chambers by the excess pressure. This is especiallv noticed
when the doors are opened for charging. Both can be avoided bv
arranging a certain proportion between the inlet and the outlet
openings. Usually the area of the latter is two-thirds that of
the former. For the changes of draught made necessary by the
variations in the state of the atmosphere no certain rules can be
PRODUCTION OP DRAUGHT. 549
given ; observation and practice must come into play here. In
well-arranged works, however, this is not left to chance, but the
supply of air is checked by regularly estimating the oxygen in the
escaping gas, as we shall see later on.
We have thus seen that the hot gaseous mixture in itself con-
tains the conditions for causing a draught, since it is much lighter
than the air, and will always have a tendency to rise from the
burners to the chambers. We must also point to the second source
of draught, viz. the formation of liquid sulphuric acid within the
chambers from the mLxture of the gases, which must necessarily
have an aspirating action, although not only from the burners, but
from all sides.
Along with these two sources of draught furnished by the peculiar
nature of the acid-making process itself, there must always be
another arrangement for causing further draught, especially because
otherwise the current of gas could not be turned into the required
direction. In the simplest case a plain outlel-pipe behind or above
the last chamber will suffice. The Belgian Commission of 1854 even
preferred this arrangement to a chimney, because the latter might
produce an excessive draught; and many factories work quite
well in this way. But it cannot be said that the excessive draught
of a chimney must lead to a loss of uncondensed gas and too quick
a passage through the chambers ; for it is always very easy to cut
off an excess of draught by a damper &c. in the outlet ; but it is
nothing like so easy to increase the draught in the outlet-pipe or
chimney if insufficient. For the latter object a steam-injector
placed in the outlet-pipe was formerly considered the most con-
venient apparatus. Sometimes, in lieu of a proper injector, a
simple steam- jet, turned in the direction of the draught, is used;
hut this is a very wasteful proceeding, and a proper Korting's
injector, made of regulus metal (lead and antimony), should always
be employed. Such injectors can be applied in various places.
Scheurer-Kestner (Bull. Soc. Chem. xliv. p. 98) describes his
experience in this direction. He employed a Korting^s injector
which produced a gaseous mixture of 7*9 per cent, steam and
92' 1 per cent. air. Thus a quantity of 1814 kilog. of water in the
shape of steam sufficed for aspirating the air required for burning
7000 kilog. of 45 per cent, pyrites. At first the injector was placed
in the pipe entering into the first chamber. This is the best place,
where there is no Glover tower, as the steam is then very service-
I
550 CONSTRUCTION OF THE LEAD CHAMBERS.
able for workiug the chambers^ and thus costs nothing ; but iu
case of a Glover tower this produces an excess of steam in the first
chamber. The regulus metal of the injector wore out pretty
quickly ; nor could it be replaced by porcelain, which cracked very
soon ; a thin casing of platinum^ howeyer^ was found sufficient for
protecting a regulus injector. It was tried to place the injector
between the first and second chambers^ but here also too much
steam was introduced into the chamber. This is avoided by
applying the injector at the exit from the Gay-Lussac tower ; but
then all the steam is lost and the process is thus made expensive.
In the case of seleniferous pyrites the injector between the burners
and the chambers is stopped up so quickly with a deposit of
selenium^ that two injectors must be employed side by side^ one of
which can be cleaned out while the other one is going.
Steam-injectors between the Glover tower and the chambers
are^ as we see^ impracticable. This holds good of any place in the
system, except in the exit-pipe or chimney itself. But in the
chimney where the steam cannot be utilized for the chamber
process, they cause considerable expense.
It must not be overlooked that with steam-injectors a regulation
is all the more called for, lest the draught should be too strong;
and in the end a cheap source of draught, viz. the chimney, has
been replaced by a dear one, without any gain as to constant
supervision and regulation. We should accordingly in ordinary
cases prefer a chimney to a steam-jet, all the more as the former
will always be necessary in any case for the steam-boilers. Of
course the chimney, to do its work, must be higher than the
chambers.
Where a chimney cannot be employed for one reason or another,
nothing remaius buc to have recourse to a mechanical contrivance
(fan-blast), as will be described below.
It answers much the same purpose as a chimney if the outlet-pipe
fixed to the last chamber has a considerable height — for instance,
50 feet (in the south of France). Where several sets of chambers
exist in the same works, it is preferable to carry them all into a
common chimney, providing the connecting-pipe of each set with a
contrivance for regulating the draught. It is not a good plan to
utilize for the chambers a chimney with which ordinary furnaces
are connected, as the draught will be of a very variable character in
this case, and the working of the chambers will not be easily kept
PRODUCTION OF DRAUGHT BY CHIMNEYS. 551
entirely regular. Stilly at some works this plan cannot be avoided^
and must be provided for by more careful regulation of the
draught. In such works more than anywhere else the auto-
matically acting dampers^ described below, are recommended to
be used.
The employment of a chimney is even more advisable if, as is
now the case in all well-appointed works, a Gay-Lussac tower is
fixed at the end of the set. In this case the draught must be
regulated with even greater care than otherwise ; but there must
he an excess of draught at disposal to begin with. It is also a great
improvement if the " sight ^' necessary for checking the work of
the tower (comp. Chap. VI.) can be arranged in the down-draught
near the ground-level, or at least the gangway round the chambers.
If there is no down-draught, but a direct top-draught out of the
tower, it is always necessary to mount to the top to observe the
'^ sight/' It is certainly quite possible to employ the tower itself
as a chimney, if it is built with its top a good deal higher than the
chambers ; and this is actually done at a good many works, but
probably in some cases only because there is no chimney available.
The drawbacks of this plan are well illustrated by the following
passage from the oflBcial Alkali Reports, No. 21 (for 1884), p. 74; — -
"No. 2256. The vitriol-exit from the plant in which the pyrites
smalls are burnt used to be at the top of the Gay-Lussac tower. I
found an exceedingly high escape from here on my first three visits.
The manager has since connected this exit to the main chimnev,
and now finds he can better regulate the draught in his chambers.
Since this has been done the tests have been invariably good.'^
The more recent sets of chambers at Oker, utilizing the configura-
tion of the ground, are arranged in such a way that the burners,
Glover towers, chambers, and Gay-Lussac towers rise one above
the other, terrace-wise. The outlet of the whole is at a height
of 62 feet above the level of the burner-grates. Drawings of this
arrangement are given by Brauning (Preuss. Zeitschr. f. Berg-,
Hiitten- u. Salinenwesen, 1877, table ii.). It is stated there that
formerly the draught could not be made suflScient, even by con-
necting the Gay-Lussac towers with the boiler-chimneys.
I have received the following notes concerning the Oker system
during 1902:— A set of chambers, started in October 1883, was
formerly connected by a 2-foot pipe from the Gay-Lussac towers
with the steam-boiler chimney. The draught was good, but the
552 CONSTRUCTION OF THE LEAD CHAMBERS.
chimney, made of common bricks, suffered very much. Since 1896,
when the chimuey became superfluous through the centralization
of the boiler plant, the gases have been passed straight out of the
Gay-Lussac tower in all five sets of chambers. The draught is
quite sufficient, without any artificial help; but this is easy to i
understand from the conditions described above. There is also
the convenience of running both the chamber-acid and the Gay-
Liissac acid into the Glover tower bv natural fall.
It has been noticed at Oker that a very long draught-pipe,
connecting the last chamber with the tower, has the advantage of
neutralizing to some extent the oscillations of the outer atmo-
spheric pressure, and thus facilitating the regulation. Se this as
it may, such a long pipe, although it causes some loss of draught
by friction, will always be very useful, by cooling the gas previously
to entering the absorbing-tower. The same has been found at the
Stolberg works (1902).
Very frequently one chimney has to serve two or more sets of
chambers. It is perfectly well understood, from innumerable
analogous cases in ordinary firing operations^ that, where several
apparatus are served by the same chimney, special care must be
taken that they receive the same amount of draught. Wherever
possible, the main flues are taken separately to the chimuey and
are introduced into the latter in such a way as not to interfere
one with another, which can be attained by erecting mid-feathers
within the chimney. Where it is necessary to connect several
sets of chambers with the same main flue, it must not be over-
looked that the draught is stronger in the part nearer than in the
part further removed from the chimney i by suitable arrangement
of the dimensions, by avoiding sharp angles in the places where
the branches form the main flue, by midrfeathers, and by regu-
lation by means of dampers, a proper equalisation can generally
be attained.
Sometimes none of the ordinary measures secure an equal
draught for two sets of chambers, even when the flues from these
meet about the same distance from the chimney. In such cases
it is best to make the individual flues end in a large chamber,
from which starts the main leading to the chimney, and to fill
this chamber loosely with bricks, ot course, not to such an extent
that the draught is too much restricted. This produces numerous
small and constantly changing currents, which prevent any odc
PJIODUCTION OF DRAUGHT BY CHIMNEYS. 553
of the large currents getting the advantage of the others, and thus
equalizes the draughts.
C. L. Vogt has patented (July 29th, 1875) a peculiar contri-
vance for producing draught in acid-chambers, which introduces
the air along with the necessary steam through a pipe with an
opening of ^ inch. The steam is under a pressure of 3 to 4 atmo-
spheres. Sorel ('Fabrication* &c. p. 291) states that this had
been practiced in France 20 years earlier. Such a contrivance
is only exceptionally called for ; but there are cases in which a
supply of air behind the burners seems desirable (7th Chapter).
At some factories they work in this way : the Glover tower
is packed very loosely, and itself acts as a chimney, so that the
burners have always very good draught and never blow out,
whilst it is quite possible at the same time to keep the exit draught
so low that there is some little outward pressure even in the last
chamber. In the next Chapter we shall describe an arrangement
by which this aim can be attained even more perfectly.
We have already said something about the principles according
to which the supply of air must be regulated ; and we shall have to
return to this in the next Chapter. Here we can only remark
that there must be in any case enough total draught behind the
chambers, but not too much j otherwise, even if the burners them-
selves are protected against excess of draught by diminishing the
air-holes below the grates, there is all the more tendency for air
to enter the chambers from all other sides through the finest
chinks and thus disturb the process. If the draught is excessive,
the incubus of the vitriol-maker, pale chambers, at once makes its
appearance.
Whether, therefore, the draught is produced by a chimney or by
ail open pipe, there must always be some contrivance /or regulating
it. At many works this is done by a simple damper, introduced
into the respective lead pipe by a slit, luted with clay or not at all.
The arrangement shown in fig. 234, partly in elevation, partly in
section, and in fig. 235, in cross section, is far more perfect. The
draught-pipe, a a, is widened out into a rectangular vessel sur-
rounded by a jacket, bb, forming a hydraulic joint; and the
damper, c, is surrounded on all sides by the jacket dd, dipping
into the water-lute at 6. The damper is raised and lowered by
the help of the chain^ pulley, and balance- weight, e,f,g.
In continental works the arrangement shown in fig. 236 is
aoi CON'STKUCTION OP THE LEAD CHAMBERS.
frequently met with. The draught-pipe, a a, is interrupted b; a
vider drum, b, divided iuto two parts by a horizontal diaphragm, c.
The latter is perforated by a number of holes whose total area is
somenhat larger than that of the pipe, a a. When, therefore, all
the holes are open, there is no obstacle whatever to the draught ;
but this can be produced at will by closing a certain number of the
holes with clay or lead plugs. For this purpose the space above
HEOULATION OP THE DRAUGHT. 555
the diaphragm is accessible by a small door, vhich may consist of
a pane of glass, d (fig. 236), to which another on the other side
corresponds, so that the whole at the same time serves as a
" sight."
Automatic regulation of the draught in the chambers. — Especially
in the case of cliambera not connected with a high chimney, where
changes of wind &c. produce great variations of draught, it is
advisable to adopt some automatic regulation along with the ordi-
Fig. 237.
nary dampers, &c. Such an automatic apparatus can be made by
putting on to the horizontal part of the exit-pipe a perpendicular
12-inch pipe, closed by a bell standing in an annular water-lute.
The bell hangs on one arm of a lever, the other arm of which is so
weighted that the bell can travel freely. When the draught is
just right, this second arm has a certain position, in which a throttle-
valve within the exit-pipe connected with it is half open. When
the draught increases, the bell descends, owing to the iucrease of
556 CONSTRUCTION OF THE LEAD CHAMBERS.
atmospheric pressure^ and partly shuts the throttle-valve; in the
opposite case of the draught decreasing, the throttle-valve is opened
wider. This apparatus^ as constructed by M. Delplace^ is shown
in fig. 237, where a is the entrance-pipe from the Gay-Lussac
tower, c the exit-pipe, b a conical valve, d the regulating bell,
ee the water-line of the hydraulic joint,/ the lever, g the balance-
weight.
Somewhat different in detail, and apparently very accurately
working, is the apparatus of Mr. W. G. Strype, of Wicklow, of
which the following is a description (patent No. 705, Feb. 21st,
1879) :—
The drawings illustrate two forms of the apparatus, fig. 238
being the most desirable, although somewhat more expensive in
construction than the arrangement shown in fig. 239. Refer-
ring to fig. 238, an inverted vessel or receiver A, open at its
lower end, dips into a tank A', containing water or other suitable
liquid acting as a hydraulic joint. The interior of A is placed in
communication, by means of the pipe or passage a b c, with a
receptacle R connected with the main flue from the Gay-Lussac
towers and chambers. This receptacle is also in communication
with a flue leading to the chimney, or other device for supporting
the draught, and is divided by a partition having apertures fitted
with valves or dampers D D, made of an alloy of lead and anti-
mony. The operation of opening and closing D D to ensure
uniformity of draught is regulated automatically by the action of
the suction itself in the following manner: — The dampers are con-
nected to a lever B, mounted and turning on a centre or fulcrum e.
Suspended from one end of the lever is the vessel A, whilst the
opposite end is loaded with a weight C, sufficient to preponderate
to the required extent over the load of A. Assuming that the
draught has an excess of '' pull '^ over that which is adjusted and
necessary for the proper working of the chambers, the dampers
being open, the suction within the vessel A, when accelerated,
will draw down that end of the lever and elevate the opposite or
weighted end, and so partially close the dampers. C is so calculated
that the weighted end of the lever can only be elevated when the
required draught is exceeded, and it will fall by gravitation as
soon as the draught is unduly diminished. It follows that thus
the desired uniform action is obtained.
The connection between the dampers and the lever B is by
HEGULATIQN OF THE DBAUOHT. 557
means of rods or links passing through vater-sealed stuffing-boxes,
0 G'; and to avoid friction these rods are suspended from knife-
edge centres d d'. The other centres are constructed with knife-
edges in like manner.
The vessel A becomes sensibly lighter when deeply immersed in
the liquid, owing to the thickness of the sides, and would thereby
constitute a source of disturbance to the proper action of the
Fig. 238.
apparatus. To counteract this, the lever B is provided with
a projecting arm E, carrying an adjustable weight F, arranged in
such a position that, as the arm partakes of the motion of the
lever, the centre of gravity of the system will be moved in the
direction and to the extent necessary to effect the required
correction.
Fig. 239 is a simpler, and in some applications a more convenient
form of the apparatus, the action being of course identical with
that described for the arrangement in fig. 238. Should the
558 CONSTRUCTION OF THE LEAD CHAMBERS,
draught fluctuate very much, the diaphragm shown dotted at H
{with an opening in its centre to conimuaicate with the vessel A)
can be interposed to prevent the movements of the regulator
being too sudden and rapid.
This apparatus has no wearing surfaces, is practically frictiouiess
in its working, and is balanced in all positions. By means of it
any disturbance to the steady and uniform flow of gases through
Fig. L'39.
the chambers caused by irregular chiniuey-draugbt is prevented,
the admiGHion of air to the burners is more uniformj the regulation
of the proper quantity and relation o£ the gases to each other
throughout the chambers is facilitated, and better working and
more economical results are obtained in the process with less
supervision and attention than hitherto required to carry on
successfully the manufacture of sulphuric acid.
MECHANICikL PRODUCTION OF DRAUGHT. 559
Mechanical Production of Draught in the Chamber System,
We have spoken above (p. 549) of the various drawbacks
connected with the application o£ injectors for this object : I have
indeed not found such apparatus at any of the works I have
recently visited. But several works have adopted the plan,
originally followed at Freiberg*, of promoting the draught by
fans made of lead alloyed with antimony, or wood or iron covered
with lead, fixed on iron axles, running in somewhat tightly fitting
lead journals without stuflSng-boxes. Such fans are arranged
either between the Glover tower and the first chamber, or between
the last chamber and the Gay-Lussac tower, or in both places.
These fans are worked at a trifling. expense, most conveniently by
electromotors, which avoid the necessity of shafting and gearing ;
and this process should be more frequently employed, not merely
in such extreme cases as at the Freiberg works, where the gases
must travel through flues of 330 feet length, but in ordinary
chambers, which are thus made independent of accidental variations
of pressure, of low chimneys, &c.
At the works of Messrs. Matthiesen and Hegcler at La Salle
(111.), where zinc-blende is roasted in a mechanical shelf-burner,
and the necessary draught for the chambers could not be obtained
by a chimney, an iron fan-blast, covered with an alloy of lead and
antimony, is placed between the Glover tower and the first
chamber, and another such apparatus between the Gay-Lussac
tower and the chimney. This arrangement had been working foe
several years when I visited the works in 1890.
The systematic production of draught by placing one fan
behind the Glover tower and another in front of the Guy-Lussac
has been especially worked out by F. J, Falding : comp. Min.
lud. vii. p. 672. In this way the draught of the burners is
rendered independent of the pressure in the chambers, where quite
different conditions prevail.
Falding's fans have a cast-iron casing lined with lead, and
a spindle and arms made of antimony-lead. They are very
carefully and substantially mounted, and work up to 700 revolu-
tions per minute.
Niedenfiihr (1902) considers that a fan-blast would be best
placed between the burners and the Glover tower, but he believes
♦ According to Miihlhauser (Zsch. angew. Ch, 1902, p. 072), this invention
is due to a mining engineer of the name of Ilageu, at the Ilalsbriicke works.
I
I
560 CONSTRUCTION OF THE LEAD CHAMBERS.
this to be impossible with ordinary fans on account of the high
temperature and the flue-dust. Directly behind the Glover tower
a fan would also act very weil^ but here antimony-lead is too
quickly corroded, and it is therefore generally preferred to put
the fan between the last chamber and the Gay-Lussac tower.
Recently the firm of March Sohne, at Charlottenburg (nov
"Vereinigte Thonwaarenfabriken''), have constructed very good
fans of stoneware, which probably will do perfectly well at
temperatures below 70° C, for instance, between the Glover tower
and the first chamber (comp. Chem. Zg. 1902, p. 1057). But
even in that case it is best to place another fan between the last
chamber and the Gay-Lussac tower. In very long sets with
small sections, such as are found when a small original plant is
gradually enlarged by adding more chambers, there is frequently
irregular work with large consumption of nitre, which is very
easily remedied by a rational use of fan-blasts.
Usually the faus are made of iron covered with lead, or
altogether of " hard lead " or " regulus '^ (antimonial lead), but of
course the axle or spiudle should be made of iron, and the jouma/s
must also consist of this metal. This is a weak pointy at least
where there is heat to contend with as well as the acids. In very
hot places, however, where lead is out of the question, cast-iron
may be employed, which is not acted upon so long as no acid
is condensed upon it. Here also the journals are the weak point,
but this has been overcome by A. P. O'Brien, at Richmond, Va. ;
in the following manner (Falding, * Min. Ind.* ix. p. 621) : — A
cast-iron fan is placed immediately behind the burners, before the
nitre-oven and Glover tower. It serves five Herreshoff furnaces,
consuming 30,000 lbs. 49 per cent. Rio Tinto fines per 24 hours.
The fan has 27 in. suction and discharge, and is cast-iron
throughout, including the spindle: it is covered with a 1-inch
coat of asbestos cement. The temperature inside the fan is about
540*^ C. The journals are not oiled at all, but flooded with water
from several g-iuch pipes. Water also surrounds the jacket of
each journal, and is admitted to the oil-ehamber in lieu of oil as a
lubricant. After 9 months' work it had not required a cent's worth
of repair or oil. No wrought-iron or steel is in contact with the
gas, only cast-iron ; otherwise there is nothing special in its
construction.
Figs. 240 to 242 show the construction of a hard-lead fan, as
HECHANICAi. PRODUCTION OF DBADOHT. 561
coBstructed by H. Lentpold, at the Dora-Lys works at Pont
Saint-Martin (commnnicated by him to the author). Fig. 240 is
a sectional elevation, fig, 241 a sectional plan, fig. 242 a per-
specfive view of the iron casing. It is clearly seen bow the iron
spindle is protected by a hard-lead casing.
Fig. 240,
Henker (1902) always employs fan-blasts (and water-sprays)
for the " high- pressure work" (oomp. pp. 468, 543, and Chap.
VII.), where 8 kil. acid of 116° Tw. is made per cubic metre in
24 hours. He places the fans preferably between two Oay-Lussac
towers ; if there is only one Gay-Lussac, the fan is placed behind
this, but is followed by a small tower fed with water in order to
VOL. I. 2 o
562 CONSTRUCTION OF THE LEAD CHAHBEHS.
condeaae the acid miat. For this purpose Beaker prefers plate-
towers to any other kind of apparatus ; he places one of these
towers at such a height that the weak acid contained therein
can be mu into the Glover tower. The object of the fan-blast
is to avoid the inequalities and temporary losses of draught
caused in the case of chimneys by wind, sunshine, &c., and to
Tig. 242.
produce a regular composition of the exit-^ases, 4'5 to 5 per cent,
oxygen for ordinary pyrites, up to 6"5 per cent, for cupreous
pyrites. As there is a difiereuce of temperature between day aud
night, the speed of the fan must be regulated at least twice a day.
Attempts at placing a fan between the Glover tower and the first
chamber were not successful, principally on account of the necessity
of frequent repairs, although even shells made of Volvic lava were
tried.
Anemomelera.
Although we have in a previous Chapter {p. 322) excluded the
use of a Combes' anemometer for regulating the access of air to
the burners, because it is too delicate an inBtrument, and because
it OD^ shows the draught in the place which it occupies, vre have
here to speak of an anemometer adapted for controlling the draught.
This is P^clet's differential anemometer as modified by Fletcher
and Swau. Fletcher's modification is described in the Third
ANEMOUETBKS. 563
Annual Report on the Alkali Act, 1863, by the Inspector, for 1866,
p. 54 el aeq. ; Swan'a in the ' Transactions of the Newcastle
Chemical Society/ Jan. 26, 1871. Peclet's anemometer is founded
upon the physical principle that a eurrent of air passing the open
CQcl of a tube causes a partial vacuum in the tube. If, therefore, a
straight tube is introduced through a hole iuto a chimney, or into
the draught-pipe taking away the chamber-gas, so that the gaseous
current passes the open end of the tube at a right angle, a partial
vacuum will be formed in the latter, proportionally to the velocity of
the current ; but the aspirating action of the chimney will be equally
communicated to this tube. We must here distinguish between
Fiv'. i>43.
these two actions. To do this, we must introduce two tubes into the
chimney, one of which euds straight, whilst the other is bent to
a right angle, so that the current of air blows into it. Both tubes
will now be affected by the aspirating action of the chimney; but in
the straight tube this is increased by the aspirating action of the cur-
rent crossing its open end, whilst in the bent tube it is dminished
by the air blowing into it. The drEerence between the aspirating
action of the two tubes is thus reducible to the action of the current
of air ; and by measuring it the speed of that current can be
ascertained. For this purpose the two tubes are connected with a
U-shaped glass tube containing water or another liquid ; this will
rise in one of the limbs to an extent corresponding to the difference
2o3
564 CONSTBUCTIO\ OF THE LEAD CHAUBEKS.
of suction. Since the sucking-action of the chimney acts upon botb
limbs, it is eliminated, iind the difference of level corresponds
merely to the different action exerted hy the current of air upon
the straight tube, which it crosses, and the bent one, into which it
bloH'S. This action rises and falls with the speed of the curreut;
and the latter accordingly can be de<Iuced from it. Water (used
by Peclet), on account of the friction exercised in the U-tobe, is
only adapted for currents of a greater speed than 5 feet per second.
Fletcher overcame this difficulty thus : — In order to lessen the
friction, he employed two cylinders, a a' (fig, 2i3), of 4 iuches
diameter, connected at the bottom by a narrow tube, i. This
arrangement is ten times as sensitive as a U-tube of 0'4 inch
width would be, since the area upon which the pressure acts is
increased 100-fold, but tiie circurafereuce upon which friction acts
only 10-fold. The rising and falliug of tlic liquid is observed by
means of metal floats, c c, upon which a very fine horizontal line is
marked by a lathe ; and the scale, d, provided with a vernier and a
very fine adjusting-screw, permits the difference of level, down to
one- thousandth part of an inch, to he read off. This is possible,
not with water, whose mobilitVj^owiug to its adhesion to the glass.
ANEMOMETERS. 565
is too slight, but with ether, whose adhesion is only one two-
«thousandth of that of water. The two glass tubes, e and /, are
inserted into the draught-pipe, k, by means of a cork, ^, at right
angles to the current of gas (so that it blows into the bent tube, /),
and are connected by elastic tubes, h i, with a a' .
The form of anemometer shown in fig. 243 has now been
replaced by the simpler one represented in fig. 244. Lenses have
also been added to make the readings more accurate; but these,
in my experience, give more trouble in making the observations
than is gained by the greater accuracy of reading the scales.
In the original communication by Mr. Fletcher, as well as in
the 6rst edition of this w^ork (pp. 333-335), we find the mathe-
matical evolution of the laws for ascertaining the relation of the
readings to the speed of the currents. We abstain from repeating
this reasoning here, and merely give the final formula found for
ascertaining the velocity of the gaseous current v from the height
of the column of ether (of 0*740 spec, gravity) =jo, for any tem-
perature t (in degrees Fahrenheit) and barometric pressure A (in
inches) : —
/•
/' 29-92 519 „-..
The Table given on p. 567 et seq,, for the speeds corresponding to
different readings of the anemometer, is computed from the formula
t;= Vj»x28'55 j
and another Table is added for correcting the variations in the
temperature of the current of gas. Tlie corrections for small
variations in the barometrical pressure are usually not consider-
able ; but they can be made by means of the above formulae —
,,_ / 29-92
t'''=Y/yrYx 28-55,
or
X 28-55.
^=\/^29^92
If the pressure is read off in millimetres the number 760 is
everywhere substituted for 2992 ; or if the readings are in milli-
metres and the speed in metres per second is required to be known.
566 CONSTRUCTION OF THE LEAD CHAMBERS.
the constant 28'55 is converted into another, accoi'ding ta the
formula
0-8048 ^^_ , .^
-7-= X 28-55 = 1-727;
V25-4
so that the formula for t/ and j/ in metrical measures will read
r'= 1-727 x/j?.
A correction for the expansion and contraction of the ether in
the instrument itself is mostly unnecessary, since it is only ex-
posed to the ordinary temperature ; it amounts to about 1 per
cent, of the speeds shown in the Table for each 10° F. ( = 5°-55 C.)
deviation from 60° P., — more for temperatures below, less for
temperatures above 60° F.
In order to make the readings more exact^ first the height of
ether in one of the limbs is noticed, then the current is reversed
by connecting the tube e with a and / witli a! (fig. 243) ; another
reading is made ; and thus twice the difference of pressure caused
by the suction at / is found. The number thus found is read off in
Table I. and corrected for temperature by Table II. To take an
instance^ let the first reading be ]'039^ and the second readings
after reversing the current, 0-861, the difference will be 0*178.
On referring to Table I., the speed 12*05 feet per second will be
found. This, however, is only true if the temperature of the air
is 60° F. Should it in the case in question be 520° F., Table II.
gives the correcting multiplier, 0*7280. This, multiplied by 12*05,
is 8*772, the true speed of the current if measured at the tempera-
ture of 60° F.
This instrument is not influenced by soot, heat, or corrosive
vapours ; it can be placed at some distance from the flue to be
tested, if longer elastic tubing be used ; and it can, of course, be
employed both for aspirating and for pressure currents (fan-blasts
&c.), and as a measure for the speed of atmospheric currents.
Of course, like every other anemometer, Fletcher's only indi-
cates the pressure at the place occupied by its receiving portion ;
and accordingly the tubes e and / must be introduced so far as to
reach into the air-current to the extent of about one sixth of the
diameter of the flue. The velocity at this place is assumed to be
nearly equal to the average ; but this is very doubtful, and there
are no means at present known of measuring the absolute quan-
tities of air passing through a flue of any considerable sectional
area with any degree of certainty.
ANEMOMETERS.
567
Table I. — Showing the Speed of Currents of Air as indicated by the
Ether Anemometer,
t;=2i/jox 28-55.
Temperature 60^ Fahr. Barometer 29*29 inches.
Manometer
readiDg.
Speed of air.
ft. per see.
0-903
Manometer
reading.
in.
Speed of air.
Manometer
reading.
1
Speed of air.
in.
ft. per sec.
in.
ft. per sec. \
0001
0047
6-189
0-093
8-707
0-002
1-277
0048
6-255
0-094
8-754
0003
1-564
0049
6-320
O095
8-800 i
0-004
1-806
0-050
6-384
0-096
8-846
0005
2019
0-051
6-448
0-097
8-892 ,
OOOG
2-212
o-a52
6-510
0O98
8-938
0007
2-389
o-a53
6572
0-099
8-988 :
0-008
2-554
0-054
6634
OlOO
9-028
0000
2-709
0-a55
6-695
0-102
9118
0010
2-855
0056
6-756
0104
9-207
O-Oll
2-994
0-057
6-816
0106
9-295
0012
3127
0-058
6876
0-108
9-383
0013
3-255
0059
6035
0-110
9-469
0014
3-378
0-060
6993
0-112
9-554
0-015
3-497
0-061
7051
0114
9-639
' 0016
3-612
0062
7-109
0-116
9-724
0017
3-723
0063
7-166
0118
9-808
0-018
3-830
0064
7-223
0-120
9-891
, ooio
3-935
0065
7-279
0-122
9972
0-020
4-038
O066
7-336
0-124
10O63
0-021
4-137
0067
7-390
O120
10-13
0022
4-235
0-068
7-445
0128
10-21
, 0023
4-330 '
0069
7-500
O130
10-29
0024
4-423
0-070
7-554 '
0132
10-87
I 0-025
4-514
1 0-071
7-608
0-134
10-45
0026
4-604
0-072
7-661
0-136
10-53
1 0027
4-691 !
0^3
7-713
0-188
10-60
1 0028
4-777
; 0074
7-766
O140
10-68
. 0-029
4-862
1 0-075
7-819
0-142
1076
0-030
4-1^5
' 0-076
7-871
0*144
10-88
: 0031
5-027
0-077
7-922
0-146
10-91
^ 0032
5-107
1 0078
7-974
0148
10-98
0-033
5187
0079
8-0-25 I
O150
11-06
0034
5-265
0080
8075 '
0-152
11-18
0-035
5-342
0-081
8125
0-154
11-20
a036
5-418
0-082
8175
0166
11-27
0-037
' 5-492 ;
0-083
8225
0-158
11-34
0038
5-565 1
0-084
8-275
0-160
11-42
0039
5-638
0-085
8324 1
0162
11-49
0-(M0
5-710
1 0-06*5
8-373 1
0164
11-56
0-04J
5-781
O087
8-421
0-166
1103
00i2
5-851
0K)88
8-469
0168
11-70
aoi3
5-921
' 0069
8-517
O170
11-77
0044
5-989
0-090
8-565
0-172
11-84
0045
6-056
O091
8-618
0174
11-91
0046
6-123
0-092
8-660
1
0176
1108
568
CONSTRUCTION OP THE LEAD CHAMBERS.
Table I. (continued).
' Manometer g ^ - . i Manometer « j * •
reading. ^P*^ ^^ *»'• ' reading. ^P^ *^^ ^*'"-
Manometer
reading.
Speed of air.
■
in.
ft. per sec.
in.
]
0178
1205
0-284
0180
i 12-11
0-286
0182
12-18
0-288
0184
12-25
0-290 1
0186
12-31
0-292 !
0-188
12-38
0-294
0190
12-45
0-296
0-192
12-61
0-298 1
0-194
1257
0300
0-196
1264
0-302
0-198
' 12-71
0-304
0-200
12-77
0-306
0-202
1283
0-308
0-204
12-90
0-310
0-206
1296
0-312
0-208
1302
0-314
0-210
13-08
. 0-316 1
0-212
1315
0-318
I 0-214
13-21
0-820
0-216
1327
0-322
0-218
13-33
0-324 !
0-220
13-39
0-326
0-222
13-45
0-328
0-224
13-51
0330
0-226
13-57
0-332
0-228
13-63
0-334
1 0-230
13-70
0-336
1 0-202
13-76
0-838
; 0-234
13-82
0-3-10 1
0-236
13 88
0-342
0-238
1394
0-344
0-240
13-99
0346
0-242
1405
0-348
0-^44
14-11
oaoo
0-246
14-17
0352
0-248
14-23
0-354
0-260
14-28
0-356
0-252
1
14-34
0-358
0-254
14-40
0360
0-266
14-45
0-362
; 0-258
14-50
0-364
0-260
14-56
0-366
0-262
14-62
0-368
0-264
14-68
0-370 ;
0-266
14-74
0372
1 0-268
14-79
0-374
0-270
14-84
0-376
0-272
14-90
0-378
0-274
14-96
0-380
0-276
15-01
0-382 ;
; 0-278 I
1506
0-384 !
. 0-280 i
15-11
0-386
0-282
15-17
1
0-388
ft. per sec.
in. :
ft. |3er aec. ,
15-23
0-390 .
17-83
15-28
0392 *
17-88 ;
15-33
0394 '
l7-4»
15-38
0396
17»8
15-44
0-398
18H)2-
15-49
0-400
18-06
15-54
0-402
18-11
15-59
0-404
1816
15-64
0-406
18-20
15-70
0-408
' 18-34
15-75
0410
18-28
15-80
0-412
18-33
15-85
0-414
183S
15-90
0-416
18-42
15-95
0-418
1846
16-00
0420
18o0
1605
0-422
18-5r>
1610
0424
18-60
1615
0-426
18-64
16-20
0-428
18-68
16-25
0430
18-72
16-30
0432
18-77
16-35
0-434
18-82
16-40
0-436
18-86
16-45
0438
18-90
16-50
0-440
18-94
16-55
0-442
18-99
16-60
0444
19-03
16-65
0-446
19-or
16-70
0-448
1911
1675
0-450
1915
1680
0-452
19-20
1685
0454
19-24
16-89
0-456
10-28 ;
1694
0-458
19-32
1699
0460
1936
1704
0-462
19-41
17-09
0464
19-45
17-13
0466
19-49
17-18
0468
19-53
17-23
0-470
19-57
17-28
0472
1962
17-33
0-474
19-66
17-37
0-476
19-70
17-42
, 0-478
1974 ;
17-47
0-480
1978 I
17-52
0-482
1982 1
i7-r>6
0-484
1986 '
17-(50
0-486
19-90
17-65
0-488
19-94 '
1770
0-490
1998 1
17-75
0-492
2002 1
17-79
0-494
20-06
ANEMOMETERS.
569
■Table I. (continued).
Manometer
reading.
1
1 Speed of air.
Manometer
reading.
Speed of air.
Manometer
1 reading.
in.
1
Speed of air.
in.
ft. per sec.
in.
ft. per sec.
ft. per sec.
0-49C>
20-10
0-590
21-94
0-700
23-89
0-498
2014
0-600
•.>*^12
0-750
•24-73
0-500
2018
0-610
22-30
' 0-800
25-^
0-510
2o-;«
0-620
22-48
0850
26-32
0-520
20-58
0-630
22 66
0-900
27-08
0-530
20-78
0-640
22-84
0-950
' 27-83
0-540
20-98
0650
23-02
1-000
28-55 1
0-550
2117
0-660
23-20
1-250
31-93
0-560
21-37
0-670
23-38
1-500
34-97
' 0-570
21-56
0680 ;
23-55
1-750
37-77
, 0-580 1
1
(
21-75
0-690
23-72
2000
40-37
/ 519
Table II. — Showing the Values ^/\/Tro • //^^
Values of
if'
om 0 /O ICK
\/ 459+^
}0; or Cor
rectionsfo)
■ Temperat
degrees
Fahrenheit.
ure.
' degrees
1 Fahrenheit. '
1 '
degrees
Fahrenheit. |
1
/~5i9~
\/ 459+ j5.
/ 519
\/ 459+ J'.
0
10634
130 '
0-9388
260
08497
5
l-a>77
135 :
0-9348
265
0-8467
10
10520
140
0 9309
270
0-8438
15
10464
145
0-9270-
275
08409
1>0
10409
150
0 92;^2
280
08380
25
1-0355
155
0-9194
285
0-8352
;jo
10302
160
0-9156
290 i
0-8324
a->
1-0250
165
0-9119
295
0-8296
40
10198
170
0-9083
300
0-8269
45
1-0148
175
0-9047
305
08242
50
1-0098
180
0-9012
310 1
0-8215
55
10049
185 ,
0-8977
315 '
0-8189
HO
1-0000
190
0-8943
320
0-8163
«J5
0-9952
195 ;
0-8909
325 1
0-8137
70
0-9905
200 1
0-8875 '
330
0-8111
75
0-9858
205
0-8841
335 1
0-8086
1 80
09812
210
0-8808
340 :
0-8060
'^
09767
215
0-8775
345
0-8035
90 '
0-9723
220
0-8743
350
0-8010
! 95
0-9679
225
0-8711
.355
0-7985
i 100
0-9636
230
0-8680
360
0-7960
105
0-9593
235 i
0-8649
3(>5
0-7936
110
0-9551
240 1
0-8618
370 1
0-7912
115
0-9509
245 !
0-8687
375
0-7888
120
0-9468
250 1
0-8557
380 '
0-7865
125
1
0-9428
255
0-8527
:J85 1
0-7842
570
CONSTRUCTION OF THE LEAD CHAMBERS.
Table II. (continued).
degrees
/ 519 '
, degrees
/ 519
degrees
Fahrenheit.
1
^ / -^'^
Fahrenheit.
\/ 469 -f^ .
' Fahrenheit.
1
\/ 459+/.
•
390
1
0-7819 1
595
0-7017
800
0-6420
395
0-7786
6U0
0-7000
805
0-6407
400
0-7763
605
0-6983
810
0-6395 1
405
0-7741 ,
610
0-6967
815
0-6382
410
07729
615
0-6951
820
0-6369
415
0-7707
620
0(>9a5
825
0-6357
420
0-7685 ,
625
0-6919
' 830
0-6345
425
07(M>3
630
0-6903 1
835
0-6333
430
0-7G41
635
0-6887
840
0-6321
435
07619
640
0-6871
845
0-6309
440
0-7598
645
0-(>85(i
850
0-6297
445
0-7577
650
0-(>841
855
0-6285
450
0-7556
655
0-6826
860
0-6273
455
0-7535
660
0-6811
865
0-6-261
4(30
0-7514
665
0-6796
870
0-6249
465
0-7494
670
0-6781
875
0-6237
470
0-7474
675
0(5766
880
0-6225
475
0-7464
($80
0-6751
885
0-6214
480
0-7434
6S5
0-6736 ,
890
0-6203 ,
485
0-7414
690
0-6721 1
895
0^192
490
0-7394
695
0-6706 '
900
0-6181
495
07375
700
0-6691
905
0-6169
500
0-7356
705
0-6676
910
0-6168
505
0-7337
710
0-6662
915
0-6147
510
07318
715
0-6648
920
0-6136
515
0-72VKJ
720
0-6634
925
0-6125
520
0-7280 1
725
0-6620
930
0*6114
525
07261 !
730
0-(«06
935
0-6103
530
07243 1
•735
0-6592
940
0-6092
535
0-7225
740
06578
945
06081
640
0-7207
745
0-(>565
1 950
0-6070
545
0-7189
750
06552
955
0-6059
550
0-7171
755
0-(5538
9(>0
0-6048
555
0-7153 '
760
0-6524
' 965
0-6037
560
0-7137
765
0-6511
970
0-6026
565
0-7119
770
0-6498
975
06015
570
0-7102
775
06486
980
06004
575
0-7085
780
06472
985
0-5994
580
07068
786
0-6459
990
0-59W
685
0-7051
790
06446
995
0-5974
590
0-7034
795
06433 i
1000
0-59G4
'■
_ — .. —
.
Fletcher^s anemometer has been improved by Swan in the
following way, practically returning to Peclet's original construction
(a similar plan has been independently proposed by P. Hart, Chem.
News, vol. xxi. p. 200). In lieu of the 4-inch cylinders he takes
a U-tube of \ inch diameter, narrowed in the bend to diminish
the oscillations. The tube is 10 inches long, and placed with an
ANEMOUETEKS. 571
inclination of 1 ia 10 ; each limb has a scale and vernier, the latter
partly made of glass and coverinx at the same time the scale and
the tube, so that it is easy to read off to ^^^ inch. The ends of
the tube are connected with a two-way cock, so that the current
can he reversed without opening any joint. Fig. 345 shows the
instrument as seen from above, so that its inclination to the
vertical line does not appear. It is fixed on a stand provided with
a spirit-level and adjusting-screws. It is employed just like
Fletcher's anemometer; hut, owing to the inclination of 1 in 10,
the column of ether in the tube occupies ten times the space
corresponding to its height, and the reading of j},^ inch gives thus
the same result as the very difficult one to iQti(i i^"^^ ''^ Fletcher's
instrument. The narrowness of the tubes does not matter in the
case of ether, as the friction may he entirely neglected with this
substance (the later form of Fletcher's anemometer, shown in
fig. 244, hears this out as well). Swan's anemometer must
always be placed exactly level in the direction of its length;
hut it need not he levelled across, if a reading be made in one
limb, the two-way cock turned, and the new reading in the same
limb subtracted from the first j thus it is unnecessary to read ofi
at both limbs, which would involve levelling across as well. The
speeds are found from Fletcher's table, dividing the readings
by 10. Quite similar to the above is the presaure-guage designed
by Sorel (comp. supra, p. 509).
Other instruments for measuring the draught are, for instance,
those of Kretz (Dingl. Journ. cxc. p. 16), of Ramshottom (ib.
c^xxx. p. SSi), of Scheurer-Kestner (ib. ccvi. p. 448 and ccxxi.
p. 427), none of which can vie with Fletcher's in sensitiveness.
^72 Construction op the lead cbajibers.
The very iagenious anemometer of Hurler (Dingl. Journ. ccxxis.
p. 160) is only adapted for laboratory use. Compare also Bour-
don's multiplying anemometer (Compt. Kend. vol. xcir. p. 5 ;
Joura. Soc. Chem. Ind. 1882, p. 60).
One of the most delicate anemometers is Fryer's, described in
the Inspector's Report on the Alkali Acts for 1877-78, p. 08. Its
Fig. 2«.
principle is to measure the difference of pressure on each side of
a watch-glass shaped copper plate connected vith a spiral spriug.
It will measure a pressure of ^j/^g of au inch.
Recently difi'erential anemometers on another principle have
come into use very largely, and seem to be preferable to nil others.
There are already a good many forms of this apparatus, one of the
best known being that of Professor Seger (ti. F. 19,426), slioivii
VOLUMES OF CHAMBER-GASES. 573
in fig. 246. The calibrated U-tube A is surmounted by two
cylindrical cups^ B and C, of equal width. The board on which it
is fastened also carries the sliding-scale D^ adjustable by slits a a
and screw-pins bb. The tube is filled with two not miscible
liquids^ for instance heavy paraffin oil and dilute^ coloured spirits
of wine, of nearly equal'specific gravity, to such an extent that the
zero-point of the scale D can be put exactly at the lin'e of contact
of both liquids at X. If an aspirating force is acting on the
surface of the liquid in C, which raises the level in that part of the
tube, the point X will be lowered at a multiplied ratio, corre-
sponding to the difference in the sectional area of the narrow part
of A and the enlargement in C. If, for instance, the ratio of the
sections is as I to 20, a difference of pressure of 1 millimetre will
be indicated on the scale by a sinking of X to the amount of 20
millimetres. The scale is graduated in such a way that it
indicates the pressure, expressed in millimetres of water. This
instrument is much cheaper and easier to handle than those
constructed on Peclet's principle and quite as accurate.
Calculation of the Volume of Chamber-gases according to
Temperature and Moisture.
In all calculations concerning chamber-gases it is not sufficient
to take into account the difference of temperature and barometric
pressure from the normal state of 0^ and 760 millim., but the
amount of moisture present in the chamber-atmosphere must be
equally brought into the calculations. It is evidently impossible
to do this on the assumption that the tension of aqueous vapour
within the chambers is that ordinarily existing for any given
temperature ; the presence of sulphuric acid, not merely at the
bottom but all over in the shape of mist, greatly changes the
aqueous-vapour tension according to the varying strength of the
acid. The tables of Regnault and Sorel, given on pp. 195 and 196^.
would admit of making the calculation in the proper manner; but
it will be more convenient to consult the following Table, p. 574
(calculated by Sorel), which immediately gives the volume occt^ied
by a cubic metre (or cubic foot) of air, originally at 0° C. and 760
miUtmetres pressure, after being brought into equilibrium of tempe-
rature and vapour-te^ision with dilute sulphuric acid of varying
strength and temperature, but without any change of pressure.
574
CONSTRUCTION OF THE LEAD CHAMBERS.
o
a
o
s s
cc i-* c;
I'- 55 o
^Mk ^^ **^
C'l
S g
(M •*♦< •*: lO Ol
St t^ CI -< '* C4 Ti »5 c:
.. ^ 59 r- Ct tr C5 »5 Tt C5
liTi Ob -r^ I- Ob 05 p CO i-i ct t^ c5 »'?: •*: ^
o
o
o
00
o
o
:rA
o
5
c^ c^
C? 00 CO C"* ?t
■« v^ >*-
<N
O'M^O':PQQC^0aG^l'«Q0Q'^C0l-*CjC
x^C'ic5icriQtr5'H*co«ci-*Q— 5
»r5.^ip'<|h'n'^^w«cpcpcoopopc*c:«
» 55 CO CO so 80 ^ o5 CM 5i CI Ti -Ji
X ca CO
1-* ^ <N
^1 •«(< «<ji
CO oi oa go X o <?i »j: Ji ri 01 00 a> -^ ^ 1-. **; -r
I- »': oi o X i? >o CO oi ^ o g> X 00 x It- I'- i-
^^^^^^0OS0nc0C0<MC4OIOI01C1tl
•»< CO CO CO CO CO CO CH Ol CI W 01 (N <M 0^ 2^1 01 Ol CI
»c -*• ci ^ Q cS I'- sp sp »r: '«r OT CO CO 04 ^1 oi 01 2»
rO CO to CO CO Ol CI C^ « Ol Ol Cl <N Ol CI 01 01 Cl Ol
01 r-^ O o 01 -i* i'^ C "^ o >o CI o s on i^ ^ i2
cixi'-S»0'^cococi»-i'-«»-«'— pcopw
CI 01 CI CI Ol Ol Ol CI (N « CI Ol CI CI ?l 01 5l 5l
:o'^c5i-i:ait-cio©3CQt::r2?:z'SS'^«'*<^
5pkO'n*'^cooici — ppooc»Cio35xxxx
Ol CI CI CI CI CI CI 01 CI CI CI »-• »-< »-i i-H ^H rH »-i — ,«
lOXCIr^CIX'^P'iPClXvi^COCIi-HOClXl^tC
~" — 7ICI»-^»-ti— (t-^i-H-^f^t-Hf-tT— li-^rHi-H.— ,— I
CI 01 01 CI
oix»*Qt^uocsi:£cooxt^?r»ococi^Qc:r:
CiXXXl'-l>-:OCO»CDi*0«0»Q»CiOk::»^'rr"^
3
o I- i-o CI o t-. '<*' CI Q X cp »3 Tt: 55 52 rt
^^LOO^OO-^-^Tf^COCOCOCOCOCOCO
«MS 9
co^pX»^CI^pCJXi>-:C'^
CO CO 01 CI 01 CI 01 CI 1-H p- I— 1 ^ i-i
■^ 00 CI i-i — c: w
o
o
01
o
8
o
CO CO Ol 04 ^H
^ tr '* :r 00 53 Q 5i X i^ — lO i-o ^2! :!? CO CO oi 04 ^H
1(0
'^coci»-^Qoxt*t>^cooio-*<"^'^'Tj»rtr:
CO CI d ^
I S 8 S
Oi Ci X X !>. i>. cp CO CD ?p i;t o ir: «.o
^r • • • ■ • « « * • • ^^
Ol CI 01 -^ r-i Q Q Ci Ci X X X X X X t- tr t^ i^ i-^
3^_^^»»cococ2«grtcortco9ocj?oco
OwOOOOOOOOOOOOOO^CO
'f',CXp01-U5DqCQ01^CPXOCl'»**«XCC1