* 908
T H E J O URXAL OF IiYD L-STRIAL AA'D En'Gln'EERlhTG C H E M I S T R Y .
All clerical work and calculations should be checked. Systems of weighing, manipulation, etc., devised t o prevent errors, should be employed and rigitlly :i~lhc*rc~l to, (:vcn in the simplest determinations. COKCLUSION.
The forms of organization described abovc should ncvc.1- t)c allowed t o fall into the rut o f dull routinc, b u t thc pcmonnel of the men engaged in this work maintained a t thc very highest standard, antl all the necessary plans as described carried out with cnthusksm, earncstness, and thoroughness. The perS l J n a l coirpcration of the men of this department with rncmbcrs o f other branches of the organization woultl provc an important aid in maintaining interest in the wcirk, and be mutually educating. 'I'hcsc departments should lead the way so strikingly as to he apparent to all; if mediocrity were tol,cratctl, t h m this branch of the organization would be indcetl 1':ir from the ideal t h a t should be its govci-ning spirit. IAUKEI,
€ 1 1 1 , ~h?.Y .
THE ELECTRICAL PRECIPITATION O F SUSPENDED PARTICLES BY THE COTTRELL PROCESSES.' By
LIKXBRADLEY.
I n previous articles which havc becn publishctl in THISJ O U R N A L , I , 686, 2, 372 and 3, j 4 2 antl elsewhere, Ilr. Cottrell has explained the original work on these processes and has also described the general operation of several installations, more particularly in the metallurgical fields. The application of the same to the Portland Cement Industry forms the subject of another paper by Mr. Walter A. Schmidt ( T H I S J O U R N A L , 4, 719). The problem of removing suspended particles from gases becomes a serious one when the volume of gases is very large, or the temperatures high, or when the gases contain materials which exert a corrosive action upon flues, chambers and apparatus. The variety of problems and conditions is very large, b u t they may be divided into three general classes as follows: First.-Where the composition of the gases is such t h a t they are the cause of complaint on the ground ( ~ nuisance f and damage to property. Sczuzid.-Where the valucs containetl in the gases rc~itlcrtheir rccovery quite necessary on the grountl of econoniy, hut wherc thc gases do not cunstitutc :L nuisance. Thid.--Where the gascs may bc considcrctl sufficicnt cause f u r complaint, antl yet contain V:LILICS which should be rccovcrcd. I n t h c first class, antl :IS :I further illustration (of t,hc growing rnngc o f applicntic-Jn t J f thew prucesscs, may bc ~ncntionctl the black smokes, ashes ;itid trintlcrs coming f r - t i m t h c stacks of sonic of thc largc p o ~ p1ant.s whcrc t h c boilers :ire being forcctl. I n somc of thvsc, p1:tnts thc operating d c p r t m c n t s h a w c'n(1c:Lvnrctl in various mays to purify their gaws, so :is to avi-iitl any compl:-tints. Tn ; I small plant,, w i t h pi-opcr arrnngcmcsntJo f b o i l ( n , nrchrs, set t.ings :ind f l ~ ~ t ttsh, c Paper presented a t t h e Eighth Internativnnl Ci~ngress of Applied Chemistry, New York. September. 1912.
Dee., 1 g i 2
stack gases seldom give rise to complaint. Undoubtedly the proper method of attacking this problem is t o endeavor t o obtain proper combustion. However, there are a great many instances where complete combustion is not desired, and in such rases these processes may prove t o be of value in overcoming the nuisance due t o black smoke or other suspended particles. Among its other activities the Research Corporation is at present engaged in demonstrating the removal of such suspended particles by the electrical precipitation processes at the New k'ork Edison Company's Waterside No. 2 Station, one o f the largest power plants in the city of New York. I n this plant there are nearly one hundred boilers, connected t o four smoke stacks, approximately three hundred feet high b y twenty-two feet in diameter. The average velocity of the gases through these stacks is estimated at thirty-eight linear feet per second. During the evening, when the peak load is on the station, i t is necessary t o force the boilers, so t h a t they are delivering perhaps 100-150 per cent. in excess of their rated capacity. Automatic stokers are used, but even with the greatest care, considerable ashes and cinders are caught up by the draft and carried toward the stacks. The cross flues have been enlarged in order to reduce the velocity of the gases, but although some of the heavier particles are deposited in these chambers, the exit gases carry a fairly large amount of cinders and ashes. Experiments upon this particular material indicate t h a t the greater portion of the large particles can be caught if the velocity in the flue is reduced sufficiently and ample time is allowed for the particles t o settle out of the draft by gravity; but flues large enough t o produce this result would be prohibitive. Some particles are so light t h a t a slight movement of the gas easily keeps them in suspension. If we endeavor to increase their weight by water sprays or similar means, we interfere with the draft, and may not find our efforts successful, because some particles resist moisture and are extremely difficult t o remove b y this method. If we are able t o construct a n apparatus, which will not be prohibitive in size, and does not seriously interfere with the draft, whereby we can overcome the buoyancy of the particle in the gas and rcmove it from the influence of the draft, we shall succeed in cleaning the gas of the suspended particles. The actual net amount of energy necessary t o withdraw these particles from the moving gas is very small indeed. The problem is t o design and construct such an apparatus, then t o place the electrical charge upon the particle and introduce sufficient energy to compel the particle to migrate t o some point away from the rapidly moving gas. This may be accomplished by the use of high-tension direct current, r either intermittent or continuous. Discharge electrodes and collecting electrodes are placed in a chamber and the gas containing the particles passed between them. The gas is ionized and some of the ions collect tipon the particles, thereby charging thcm electrically. In;ismuch ;is the chargtd particles are within a strong electric field o f constant polarity, they are driven t o
the collecting clcctrodes. The construction 0 1 these clectrodes may be varied considerably, to suit the special conditions of each particular problem. I n most instances, the collecting electrodes are grounded. To meet different conditions, various forms of electrodes are used. The voltages used range from I S , O O O up t o as high as j 0 , o O o or more, depending upon the size of installation and conditions of the gases to be treated. I t will be realized t h a t i t is desirable to operate on gases at a high velocity antl during a short period of time, in order t h a t the installation may be made as small a s possible. Results a t present in hand indicate the feasibility of removing the cinders, ashes and black smoke from these power-house gases a t a reasonably high velocity and without any serious interference with the draft. Much more might be said about this problem, b u t i t may be more t o the point, in this genci-a1 survey, t o mention some phases of the work bearing upon other industries. Shortly after the organization of the Research Corporation, last February, investigations \vcrc undertaken for the Raritan Copper Works, Perth Amboy. N. j . , on the gases arising from the smelting of the slimes obtained in electrolytic copper refining. The copper, as received a t the electrolytic refinery, generally contains some gold. silver and other precious metals, as well as arsenic, selenium, tellurium ancl many other elements, in small quantities. I h r i n g the purification of the copper by electrolysis, most of these impurities settle out from the ckctrolytc. This sediment, callcd electrolytic slimes, is then treated in another department, primarily for the recovery of the gold and silver. I t is common p r x t i c c in some refineries to dry this slimc and then give it ;t furnace treatment. The volatile elements ;Lre tiriven off in the gaseous state and subscqucntly conclcnsc~tl in the flue. During the furnacc treatmerit. consiclerable silver is vaporized, some o f which condenses quitts near the furnace exit. At this point in the flue, dust may be found assaying very high in silver. Considerable gold may also be fvuntl a t this point, probably having been carrictl over mc~chanically. Further along tht: flue, arsenic. antimony, selc.niuni and other elements, or their compounds, contlcnsc. thus fprming a thick chemical fume. I t is interesting tu note the tendencies of the various compoumls to deposit. For instance, some o f the silver conipouii(1s or metallic particles are very light and finctly tliviclc(1, and it is extremely d cult to effcct a satisfactory recovery by settling chambers, centrifugal (lust apparatus or by water sprays and scrubbers. Golil has been found a long distance from the furnace, b u t as a rule it is the first t o deposit. The influencc of one or several volatile metals upon the volatility of other metals, especially at various tcnipcraturcs, has not been studied very extensively, but most authorities state t h a t the loss of silver is increased in the presence of lead or zinc. Probably arsenic or antimony also exert a n influence upon silver. It would be interesting and of considerable value t o be able t o consult curves of vapor tensions of these various
clciiicnts utit1c:t- thc varying c i ~ i t i l i t i i ~ i i(11 s llicsc IT. Samples of this kin(l 0 1 (lust froin v:triokis refineries show it t o have a range of a t l c m t froin L O O 2000 oz. of silver per tun. T h e gold content is 1 l I J t a p t t o vary so greatly. Fume has been collcctctl after passing through water scrubbers (so arranged t h a t the gas was saturated and carried water particles i i i suspension) and found to still contain large xmount s of precious metals. The character of the fume will depend, of course, upon the composition of thc c s i i p p c ' r received a t the refinery, as well as upon the operations of the refining furnaces. It is almost impossible to filter this gas, because of its corrosive action u 1 m i the filtering medium, and its tendency tc-i clog t h < , samc. Results obtained upon these gases lcad to t h t , conclusion tha't practically all of thc values can l x readily recovered by the electrical precipitatioii 1)r(I ccsses. thercby insuring a greater net profit in t h ( w departments. The problem of recovering thc vu1iic.s from the dust collected is receiving consiclcrablc ;ittcntion. If all thc collected dust is fed hack i i i t o the furnace. the highly volatile clcmcnts \vi11 agiiiti tic driven oft', subsequently conclensctl anil, in tui-11. collected. This mill permit the irnpuritics tcI c o i i centrate and incrcctse to such a point as to 1)e c ~ l ) j ~ , c tionable. Espcrimcnts arc n o w tx'ing concluctccl with ;I vie\\- t o ixcilitating the rcccivctry 0 1 the. precious metals in ;I pure state frcirii this prccipitati.il funic. T h e priiposc'(1 method may open ~ i p:in (113provcments in thcs presciit practice, clectrolyt ic slimes. ha\.c d s o been matic to i i ~ s t a l l t IIC. processes upon thc gasc's Uroin thc ci J1q)C'r c o n v e r t c,i.!; :it thc, Garfielti ( C t : t h ) Sirivlter o f the Atncric:in SiricAtiiik; lcad. ;mil Refining C o . T h e tunic cont;tins cc~nsii1c~r:tblc and small amounts o f silver a n t l cithc-i-clcnicrits. A ] ) [ J ; L t-atus has bceii iiistak>tl](.I treat the gasc's I'roin O I l C ~ ( I I I vcrtcr, m t l this \\-ill be placed in opcrxticJn in the vcry ncxr future. 'cs 1i:ivc- I ) c ~ i i i l l '1'h ct clcc t r i e;11 1) rev i 1) i t :L t i ( > n 1) r(i stcatly u s e for t h c rollc.ction o f siil1)liuric' ;iciil tiiist froin Imrting kettles i i i o n e ' silvc'r 1.c~liric~i.yt i i i . ;I I I L I I I I IKT o f yc~;trs,an(l h;ivc. givcsii c~scc~llc.nt rc,sults. A i i o l t i c ~ i ~ n t l y tlcvelol)c~(l is the ccluiLiiiic,iit ( ~ t ' w i t 1 conccntration stills, so ;is to rcxcovc'i- tliv siii;i11 amount of sulphuric x i i 1 lwing carric>(la\v:iy I J t liv ~ gas. The installation is v c ~ ycotnlxict, thc. I J O \ V V ~ consumption small, ant1 vc1.y littlc attenticin is rcquirc,tl. For installations o f this charactcsr q)cc.i:tl clc,vic,cas t i 11. o vc rc( imi ng insu 1: t t i ()TI I i ili c: i i 1tic s i I1i i 1 1)1.1 i t 1.1i i i i 1 i6 t 1i i ionization I)y the tlischargcs 11;tvc hac1 to 1)c \vcirlii,ili l l i t . 'Thc: puhcscrnt cslectrodcs ;ire oiic' ( i f the* iiic.jst i i i i p o r t m t features in instalktticms o f t h i s t y p c . The ( ) h i 0 ani1 Colorarlo Snii~lting x n i l Rcliiiiiig Compan y ;L t S ;ili , c d I J K t l i 0 , ollc' I i i t h e 1;ti.gC lc:tli smclters in the L\ tern portion [ i f the IJnitc>tl S t x t v s , is installing the ~ ~ C J C C S S C Sc ~ n their roaster s t a r k . Godfrey machines and Dwight-Lloyd sintering m : t chines are employcd, delivering their gases to a coinmon flue. This flue has a cross section of such sizc t h a t the average velocity of the gases is about j o o linear feet per minute. A large amount o f the very
910
T H E J O U R N A L OF I N D U S T R I A L ' 4 N D E N G I N E E R I N G C H E M I S T R Y .
heavy particles are deposited in the first zoo feet of flue, but the fume is being discharged from the stack, after having traveled a much greater distance. Preliminary tests were made on these gases a t a point about 1 5 0 feet from the stack base. The temperatures varied from 100' C. to 150' C. During these preliminary tests, the ore charges upon the sintering machines were varied and checked against the assays of the dust recovered electrically, during the day, in order to ascertain how the different charges influenced the loss of lead. Assays of recovered dust varied from 20 per cent. t o 45 per cent. lead during this period. Even a t this distance from the roasters, considerable silver and an appreciable amount of gold were found in each sample. The dust recovered was dry and could easily be handled, as i t was so dense and compact that a little draft did not blow it away. The smeltcr is now proceeding with a n installation on a largcr scale and it is expected t h a t considerable values will be recovered. Furthermore, this installation may make i t possible to push the roaster department to a greater tonnage, due to the ability to collect any materials carried in suspension by the gases. In some roasters, a fairly large percentage of sulphur trioxide is produced. Apparatus can easily be designed t o handle this character of gas successfully, without undue deterioration. Another recent application of these processes is in the removal of tar and suspended carbon from illuminating gas. Mr. Walter A. Schmidt has directed the work upon the gas manufactured from crude petroleum a t one of the gas plants near San Francisco. A complete clearance of all visible particles was obtained. I n the eastern part of the United States, the work of Prof. A. H. White, of the University of Michigan, should be mentioned. Under the auspices of the Michigan Gas Association, Prof. White has conducted extensive investigations, primarily upon the removal of tar from gas manufactured from soft coal. The results were so gratifying t h a t negotiations have been instituted for a larger installation, for this purpose. The tar is collected upon the collecting electrodes and runs off into sealed chambers. This field promises t o be a very wide one. Several months ago, prior to the organization of the Research Corporation, Dr. F. G. Cottrell made arrangements with Prof. R . K. Duncan, of the University of Pittsburgh, whereby the latter arranged to use the Cottrell processes for conducting investigations upon the smoke problem of the city of Pittsburgh. Dr. Strong, Dr. Benner and others have been of assistance in this important problem. Many other problems of a nature similar to some of those mentioned above have been brought before us and an effort will be made to adapt the processes to them a t the earliest possible date. There are many places adjacent to New York City wherein considerable values may be recovered and a t the same time, improve the atmosphere, a t least in so far as suspended particles and obnoxious fumes are concerned. Much more of the details of the processes themselves or their application to specific problems might
Dec.,
1912
be given, but i t would prolong this paper unnecessarily. I sincerely trust t h a t the above general survey may bc of value as well as interest to the chemists and cngineers. NoTE.-Mr. Bradley explained in detail the construction of the small apparatus prepared for demonstrating the Cottrell processes. The first test was upon the precipitation of cement dust, which was blown through a pipe 6 inches in diameter. The effect of the high tension current was to collect all of the dust within the pipe. A test was next made upon somc chcmical fumes. Compressed air was blown through bottles containing concentrated ammonia and concentrated hydrochloric acid, separately, and thc chemically laden gases mixed in a chamber just prior to being introduced into the electric treater. I3cnsc fumes of ammonium chloride were thus obtained and the effect of turning the current on and off was demonstrated. Samples of the various matcrials, which were collected by the processes in large scale opcrations, were upon the desk for inspection. THE CONTROL OF TEMPERATURE IN THE OPERATIONS OF ANALYTICAL CHEMISTRY.' BY THEODORE W. RICHARDS.
The control of temperature is a very important question in the work of the analytical chemist. The reason is a t least three-fold. I n t h e first place, temperature affects greatly the speed of all chemical reactions which are generally accelerated to extents varying perhaps seven t o twelve per cent. by each degree's rise in temperature. I n the second place, temperature affects the final equilibrium attained by many reacting systems and therefore influences both the yield and the composition of the products dealt with by the analyst. In the third place, accurate physical measurements, t o which the quantitative experimenter must frequently resort-such as weighing the measurement of the volumes of gases and liquids, and the determinations of calorimetric or electrical magnitudes-demand considerable control of temperature if any accuracy is sought. Clearly, the subject is too large for the brief ten minutes to be devoted t o it; but a few words may be able t o point out the more vital features. Let us begin with the control of temperatures near t h a t of the room. I n the first place, it is clear t h a t every chemical laboratory may advantageously have a thermostat attachment t o its heating arrangements. For years I have used a commercial contrivance which, when operating properly, has kept my laboratory a t 2 0 O C. within half a degree, greatly to my satisfaction. I n order t o attain any such constancy, the air of the room must be efficiently agitated by means of a n electric fan; just as any other form of thermostat should be adequately stirred. Of course, when a n operation affected by currents of air is undertaken, the fan must be temporarily stopped. Entirely within this room, surrounded by glass walls and without any outside windows, is built a
'
Paper presented at the Eighth International Congress of Applied Chemistry, New York, September, 1912.
