An Investigation on the Production of Sodium Thiosulfate from Trona Citation Ritchie, Charles Fisher (1922) An Investigation on the Production of Sodium Thiosulfate from Trona. Bachelor's thesis, California Institute of Technology. doi:10.7907/Y01W-T284. https://resolver.caltech.edu/CaltechETD:etd-06242004-143142 Abstract No Abstract. Item Type: Thesis (Bachelor's thesis) Subject Keywords: Chemical Engineering Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemical Engineering Thesis Availability: Public (worldwide access) Research Advisor(s): Unknown, Unknown Thesis Committee: None, None Defense Date: 1 January 1922 Record Number: CaltechETD:etd-06242004-143142 Persistent URL: https://resolver.caltech.edu/CaltechETD:etd-06242004-143142 DOI: 10.7907/Y01W-T284 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 2712 Collection: CaltechTHESIS Deposited By: Imported from ETD-db Deposited On: 28 Jun 2004 Last Modified: 03 Oct 2019 23:20 Thesis Files Preview PDF - Final Version See Usage Policy. 1MB Repository Staff Only: item control page

An Inve5tigation on the Production of SodiurA Thiosu.lfate from Trona Thesis by Charles Fisher Ritchie In partial fulfillment of the requ.iremente for the degree of ]aohelor of Science in Chemical Engineering. C.ALIFORi.UA nrsTITC'TE OF TECEHOLOGY Pasadena, California 1922. l. An Investigation on the Production of Sodium Thiosulf ate frotj Trona. The manufact"Ure of motion pict"Ures ie one of the la.r~est industries in California. Tv.ruout the fabrication of the entire film n:a.ny chemicals are employed. Ha.terials used by the photographer ha,ve heretofore been r.ianufa.ctured. almost exclusi vel~r iu tho ea.st. Tlie large :firms there have develcped the pruduction of fine photo~~a~hic chemicals to such a degree as to u.ake successful con;peti ticn b~· yoi.mg firms on the Pt:i.cific coast virtually inq;io$si'ble. There exists, however, a. good market for the chea:1)er mater- iali> whose ini.tial ooate are low as com:pa.red with their freight chargeB from New York to Los ~.ngelee. The mo&t outstanding ma.terial in this class is sodium thiosulfate or hypo. Vast qum1titieo of this chemiaal are con$Ul'i'.ed here a.nni.ielly. An examination of the availability of the raw n~•terials on this coast and in New York i~ interesting. T".ne two raw materials are sulfur and soda ash. Sulfur, which is produced in Louisiam~ and Texas,shou.ld be shipped by 11..ater as cheaply ·to this coast as to the New York :r.arket. We cnay obtain crude so~ ash in the f'orr11 of Trona. f'ron1 OV'..ens Lake a.t a. l".ll" iee comparable to eoda. ash on the east ooa.st. Hence, we find tl1e raw uieteriale market in a very favorablG condition for the production of heavy chemicals in direct co~wetition ~~th the e~stern n-ianufai.}turere. This work wo.s undertaken in view of deV'eloping a. pro- cess for the production of h.y't.-o from t:rona thru the labor1::1.tory ., ..,. and semi-works scale and to apply the knowledge so gained to the design of a plant. The fund.a.mental reactions of the prooe$s as e:ic:em.;.)lified by any text on. inoi•ga.nio chemistry are: is produced. by the sob.r eva:poration of the saline waters of 01;.-er~r_; Lake. :Besides :r::razco3 and Ma.HOO ~here are a.loo present small amouitts of J3orates, chlorides, sulfates, and silica. The materials of vital interest to the process Analysis of Ti·ona. A representative sample of crude trona was crush- ed and ground and placed in a.it air tight container for future use. It was found tha.t the ~.aterial, containing about 15% of water, could be very satisfactorily ground in a disc pulverizer. Total AlY..a.li. A weighed (4. gran.is) portion of the ground trone. was dissolved a.nu. rm.de up to 250 cc. A 25 cc. i;or- tion was titr~ted with standard RCl, using methol orange as ait indicator. The O.UT HCl was standardized ar~ainst analytica.l Ha.2 co3 (I">.ahlbau.m) using methol orangej the liberated co2 being boiled off just previous to reaching the end point. 3. "Borates. To a 25 oc. portion of the ci.bove soluticrl of trona the exact a.mount of standard HCl found in the above to be reqµired to neutralize was added. The solution wa.s boiled tmcler a re:f'l11x condenser far sever;.l minutes to ex_vel the co • 2 The vapors are conde1i.sed in order to prevent the lO!!'.·S of boric a.aid which is volatile with etean. dovm and 35 oc. of glycerine 1 neutral to pnenolphtha.lein 1 \"J'.:l.S The resu.ltine solution was titrated to colorless with added. standard Na.OH solution, using phenolphthalein as an indioa.tor. From the net a:noui1t of standard NaOH found necessary to neutralize the final glycerine solution, the a.mount of Hazt.34 07 w;;;,.s caloub.ted.. - The lfaOH used was prepared from J3a!;;:er 1 ·:. stick NaeH. The 053 wa.s removed with :BaCl 2 • . with Ua.,S0 ,., 4 ; The excess Ba.-f"twas then removed c.p. ohemioal being used thruout. Upon standardiz- ing the resulting l'fa.OH solution against the standard HCl solution) using phenolphthalein and. methol orange• two distinctly different ratios were obtained. l.iethol oranee was titrated to a. yellow and phenolphthalein tc colorleso. Tests showed that the Nazso4 used did ntit cause the diacrepency. The time bsing shoi·t, it wac deoided to recognize these ratios in the case of the two indic~tors and to forego further investiciation on the subject. :Bicarbonate. very unsatisfactory. .Analyses for bicarbonate were found to be The spot plate method of the Solvay Process Cc;t".,a:.>any was tried, but was not applicable in O.lN solutions. It was thought that the following method should prove satisfactory. To a. 25 cc. portion of trona. aolution add a ..... t• iooaau:r·ed excei:is of sta.n.da.rd NaOH, an exue~:l of neutral J3aCl? sol~1tion, ..., a."ld 35 co. of i-1e·u.tral glycarine. Titr~te the solu'tiein with RCl, using phenolphthalein <:\.fl '6n indic61.tor. Then, frcir. the l'J'aOH fow1d to have been neutralized, the 1JaE.C03 content of the trona may 'be ca.laulated. Sever- al analyses rrere ma.de on this basis, but such Nidely varying results were obt~ined that th~ method 'i'/~$ a.bandonad. It wa~ fou11d. froff; trial anal~.r- sei:J 011 a mixture of ~Ta.HC03 ~d Na2 co3 that the n1ethod dl.'u;c1·ibed in Treadwell-Hall "Analytical Chemistry" (whiah is es::ientiei..lly the abo·,.ro ~"l.'""lysio without the use of gly~erine) e;ave very oo:nsi.stcnt results. If this at1£!.lysie ie erc:.!Jloyed in the presence of sodiinn borf::.te, Ma.IIC03 will be re!?orted low due to the forme~tion th~ of IIBO,,. <;, decided to titrate the trona ~:3,m1>les 1)y this ,;:>rocodure en1 to t::;..:pply a. correction for the small amount of Na.zl3407 present. Assuming that the trona consiats of 1._z% U'aJ3Q,,;;,; <lnd that wa hs..vc a. .7 grrun sun1ple in 100 ~c. of final sol1.ltion, the eo:ncentratio11 of no2 - woul1 be 0,0013 rnol61. At change of color of phenol:phth~tlein (Ir") =3 x io-9 : K =2 x 10 .. -0 (IrBO,.,) ('j ( 3 x io- 9) ( l.3 x io-3 - x) then = (X) solving; JC : (HB0 2 ) == 0.00078 molal. Hence the concentration of EBO-; at the chan~e of color ~ of 'flhenol.t;hthalein v-vill be .00078 ::nolal. tra.tfori. of B02 - Since ~ha original concen- v1a:~ sce:n to be 0.001:3 molr..1, i t h evident tliat at the 5. HB0 2 • In vi.:;.'t 1Jf the nature of the a.ilalysis and of the relative- ly small arro:i:- to be introd·.:i.ced by ;$UCh aszurqpt:tons, this proce.iu.re W<:.$ a.d'.)pted. The oorrection to be adJ.ad to ·the a\J.1.0U.."lt of 0.10917 Na.OH used to n~~tralize the NaTIC0 3 wao 0.12 oc. Wt:.ter. heating a wei[~hed ea:d.ple on a aa.nd bath for one to tv:o hours st 250-3000 C. and determining the loss ir• v1ei3ht. Thh lo~s rr.rust repi·esant the H;f0 3 fl.•0;11 the 1'ra.IIC03 , (the amount of which may be c;;..1- culated )<-::i..'1.:1 the ::noisture. The analys6a showed the following for the compoiaition of the trona: Ua.-;,C03 HaHC0 3 Na.z.13407 --------- 44.2% --------··-- 37 .1% ------··---- 6.9~b Water -·----------15 •.2% NaCl , Ha2so,1 insoluble, etc. ----- 2.6% by difference --1 100.vi;' In the ;process, as O"J.tlineU., we shall aaiO.ify a eol;.t- tion of trona. with SO.,>;,, • ·The two most vital points concerning the nia.taril::i.l are (l) its Na.2 0 content, and (2) the amount of wa.ter to be introduced by its use. This water 0011sistc cf the rJoisture as report- ed. ir. the above analysis a..."ld the water :;.:iroduced by the deco~josition of We may report the above a.1;.a.lysie c:~s: Ha.zo ---·------- 39 .s·;:b Availabl~------ 10 2 ~ ;rater -:J. .- • /J .,.,, £ore 1g;i1 ----~--- •.i::"l .i.;·<'1 ~ 1'~etl:ocl o:f Analy-eb 6. of the Prod.uct. hypo forn~ed'un:ier different orisr2.ting con.4itione1 1 it is necess.'lry that we hl'-ve chloridot, otc. A~.,, a ret';ul t of a 21J.rvey of work done on this su11ject, ''' the followin:; \V&s accepted au a. po~sible rr.-ethod of analy&in: The fil tere~t _product s- with HgC1 2 , in acid eohi.tion. fro~i• D. trial :r...J.n i:;; te~1ted for J...r. e.liqu.ot portion of the fi.1 trate, con- - taininc about lC grams of h~~Ja, ix removed and, if s- is absent, is nade up to 250 ec. If - s- is :present, it is ma.de slightly basic with 6 Iv lfa.OI:I, em excess of B12 (C03 ) 3 suspension added a..."'1:1 digested. for 15 rtdnutes. The :mixture is fil tared a11d. washed thoroughly and the filtrate and wash Praters r.nade up to 250 cc. To a 10 oc. portion of' thb solution, 10 cc. o:f frosh- ly prrJpa.red Ne'~C?J solution is added, the vessel covered ai1d heatei en a wat·· er ~ath for an hour. The lfo.CH used shot,_ld correspor.d to an axcese of five times the amount req_uired au ohown "oy the equation The solution ie diluted to 200 cc. and the undecor'.iJ"O:Sed. Na.CU titri;:,ted with ntandard Ag!T0 3 by the method of Liebig. From the amount of Na.C?r found to h~we been converted, the a.mount of N'a..zs2 o3 in the orieir.al The N'e.CN" solution employed should be prarared Jt.1.$t, prior to ih use, from the pijreot material obtainable. Its strenJth may be * Z. Anal. Chem. 46 (1907) 485, abe. J.Chem. Soc. 92 II (1907) 812 Che1u. Zte;. 30 (1906) 58.Abs. J. Chem. Soc.. 9Cl II (1906) 196. 'i'. determine~ by titrating a 10 ~c. aarnple ~gainst standard f~JN03 solu- tion of a concentration of 0.1 normal 01· less. KI i~ added to the solution to :facilitate the reaognition of the first permanent tm•bidity • .A solution of li.;ypo was made up (about 0.51~) and stand.e.rdized against the !ry solutions of the quantitative analysis ola$s, as well as ~gainst ""' a standard soli.ition,I;Ja..ie up from. a weighed. amount of analytical (Kahlbs.i.u,1) the two checking wt thin 0.5%. Portions of this solution were rur1. according to the proposed analysis. The results obtained for the iodine; concentration of the thiosulfate solution were about 45~ higher than thb:se found with the I 2 solution. Comparison runs were made with and without present. The amounts of thiosul:t'ate found in the oases where s= present we1~e seen to be a.bout 4% lower than in the absence of s-. v:Ja.s This effect corqpensates the tendency of the analysis to give high re$ults. A more thorough study of this a:tlalysie vilO'l.1ld. 'be :t-.,rcfita'ole. Since the available time was short and since at this stage of the process we were interested. rr.ainly in com1Ja.rati ve results, it was decided to employ this Effects of Reagents upon Ma.teria.ls of Construction. HYIJo on steel. A piece of :polished steel bar was :ple;.ced in boiling hypo and allowed to stand for several days. The $teel i:rr..1.'ned.is_tely turned. black upon contact with hot hypo a.."1d the air. No pitting or oorro~ioi-1 wa;;i noticed when the s1.i.rfs.oe wa~ again polished. A tes·t on the hypo after the treatment of the steel failed to show the concentrated. $Olution of trona. A discoloration of the :;oli!liied surface Wi.3..s noted., bnt no pitting couL:. r)e cl.i:;iervecl u:11on revolish:i.ng. 'J:est'.3 slwi'!ad small amomits of iron }:.resent. &..ilfi ta on cite el. A piece of fJolished steel b0.r ".'ras heated for several hours in a solution of 1Ta.2S03 • v1ere apparent; the su.rfi:::.ce ren:ainir.g brit:;ht. The mat'3ria..l was al- lowed to remain in the solution for several days, producil1g a growth of su.lfite crystals m;1 the metal surface above the liquid. Up::in vleaning the metal, a very decided pitting wa~ noted,where there had been combtned. a.at ion of air and. neutral aulfi te. '.!.lo one. on e:na:ne 1. tre~ted by Os.:t1i)le A solution of trona was conoen- intermittent boil inc; tover a. period of sevor:.ll days.1 in a. vessel of Elyria Enameled iron. Ho action ttI:lon the sin•face of the en.arr.el could oe sean. Acid r.ml:f.'ite on er.a.mel. A concentrated. ~olution. of fJ,cid S'tllf'i te was r;lace1 in the en~:t::l.eled. d.ish for twenty-four ho\U"S. The acid sulfite wac neutralized \•Jith trona, au.lf·o.r adc~ed u:nd. the mixture heated for seve.ral b .-ru.rs. lfo injury to the enam.aled trurtu.ce co-;.1.ld b9 noted. Consideraticn1 of Working Conditions. For a process :i.n v.Jhich a solid i~o; to he o·fJtained fro-.cJ aolu.tion, it is desirable that the amount of wi:~ter entering i::J.to the 9. The other operations in this case are the 'liatu.rating of the material wi.th so2 , stirring, cooking, filtering and As one extrerne or'. the we..t::ir problem, we have the possibility of producing sodium thiosulfate in such concentr~tion that upon completion of the cooking operations, the ratio cf anhydrous Na.,,S?O,., ,., ..., .::> to water shall be 158 :90. fact that lTazS2 03. 5Hz0 rool h Taking advant'::·,ge of the in itz water of crystallization at 48° C., we might filter this concentrated solution hot, s:n•aad it onto a cold re vol vin::; iron dru1'!1 and sera,i:ie off the solid as it cools. We CO"\.lld thus produce a sodium thioflulfate in annll flaky chip~ which .woul1 be fm.- more easily d.iszi:il ved them the commo11 cryr:italir1e form. Eve..porators and their operating costs could be di~~ensed with in favor of this sirrrpler £1.nd less expend ve process. Th.e most a::..'l?arent difficulty to be e11co1..mterad. in au.ch a scheme is ·the handling of a hot 1 concentrated syr11p in filter ~Tesses. line sodium thiosulf'ate rr,i;.iy be heated e.s hicr,h as 110° C•. B.i.'lii rriay even oool oonsideraoly below 48° C, without crystallization taking place. Even if the treatment cf such a conce:ntri::..t.ed ::•Jh:tion should :prove :feasible with oo:.:imercial a.}Jpu.rt:itus, there h always the possibility of having to chi:f· hypo out of c~:;pare,tus or to melt H 01.i.t one. of pi:J?e 1 inee when the plant h suddenly shi..1.t d.o·wn. The The extent to which the soluble unconverted matter and ir.'.[Jurities of the i·ei.vv rr.ateria.h: wou11 t)e carried into the final :product can only be dets:rm:.:ned. by experim.el'1t. It seer.~2s re~isoriable Vlhether these iwJ;urities would 'be detriuental to the utility or ssila.bilitJq 10. of the 11roduot can only be conjectm•ed at this time. Experiments showed that the least possible ratio between trona and water during its saturation with SO~ was l : 1.5. (;, This ratio v:;a.$ only :possible. when :part of the v1Jater vvas added in success! ve portions as the gas wa.s beint! introduced. The rat:i.o of l : 2 ·wae found to be the rriinimutll. where a batch :process Vvd.S employed. of the inni:'lediate sat'UI'ation of the wetted trona. If the materi;;:tl is allowed to form. the d.eca-hydrate, the mass becomes qp.ite u.n- workable at the$e proportions. If this method of working ~1th slu.dgea is adopted, the satl.U'ation with so2 would. have to be effectecl in Jr..ettles. This would be more corilJ:•licated and costly than the use of wooden saturatins towers which could be eu~loyed in the case cf true liquids. l,fu,ch diffioul ty i<.> experienced. with frothing in concentrated solutions, neceoe<itating the nee of oversize a-,pparatu;;i. Upon the addition of sulfur and subseqµent ooo~ing, the rn.a!:'ls 'Jl.rt::;.s f'our1d to becorr.e very vi2ccn1s, necessitating the ..f'urther add.i tio11 of water to perm.it the use of a, r•o•ver stirre1•. lillij :reactior1 between xuateriall!l in two d.iff'erent phases req)J.iros intimato mixing. Ra;pid stirring is es;;:;e:nti~l'.l :i.n this :process. '.l:he high power ~nu attendant installation costs to be dealt with in ths handlin€; of concentrri..ted materials will of:f'eiet to a ls.re;e extent the advantages to be gair.ied.,by such a. process. A consideration of the above ha.e lad to the point of e11{Ploying at lear;t three pe.rts of. r:ater to one l'Jart of trona. This is ~i.bm:i.t ti ve tin10B the figure ori;zim.,lly considered. 11. which 11.as been seen to be c1uite irrr.practic:.;:.ble. A u~re practicable plan ie to tre~t a solution of trona with so2 in saturating tower~, to boil the sulfite litllJ,ors so fornwd with sulfur in open kettles$ filter, evaa_.:iorate a1"l.d crystallize. to prevent the precipitation of NaHC03 in the procese of saturating Frora a. considerution of tho ar.no·o.nt o.f l\fa.HC03 the solution with S0;3. to be formed from the Na2o content of the trona, it was shown that the r.r.axinrJ.rtl concentration was 18.5 grams of trona :per 100 pat•te of water a.t 20° C. solid Such a. solution was prepared and. sat.ura.ted with so2 • When 24 grams of trona was dissolved NaJICO,..o,J appeared.. in 100 cc. of water und SOz passed in, a large i:ireci11itate of !fa.HOO;; separated. 75 grams of trona was placed in 250 co. of water and hes.ted on a -qater be.th for several hours. A. small a.mou..."1.t of trona Fi..U'ther experiments showed the solubility of rerr.ained indissolved. trona, so trcated,to be 27 grr:uns per 100 cc. of water. found t'h.at, when this sa.t'\.trated solution was treated vJith so2 at 20° C., no NaJIC0 3 separated out. A consideration of the c1'mposition of trona and its rea.ot ion 'iiVi th Tl..l SO,.,. will aer ve to ex-f' lain thi ft phenomenon. (;, When an excess of Na..,f0 3 "' 1.a acic:'.ti:f'ied ·with so2 , the folloYti:nc; reaction takes i1lace: ]'rorri this rea.ction it is seen tha.t one mol. of Na..,CO~ :prod.\wes one IV c...~ mol. of lfaHC03 • In the e.r.1;;;.lysis of trona, we su.w that it ,~ontained 41;'~ If 100 $I'aTi'lS of trona be acicUfied. 1 then vve 12. have 3? grarns of WaRco , plus 35 grarns :produced. from the Na. co 3 or ei total of 72 grams of Na.HC03. 2 3 , Howa1rer • if the trona solution be first boiled., the 'bicarbonate itl\ de~ompoeed; 4 WaHco3 -t ha&.t : 2 na. co:s t n 2o ,. COz 2 2 1Ta.2C03 + H.,zS03 = 2 H.::1EC03 t lTazS03 Hence, it is seen that the "oic&r1:onate to be encounte1~ed is reduced by an amount equal to o:ne-half the Na.IICID3 content of the original trona.. The NaIICOz to be :produced in a 0olution of 100 grazxs of boiled tron.a. will be that coming from the Na.zco content, 35 gra.rt.s, a.nd. one-lia.l:t' 3 the amount originally present, 18.5 grams, or a total of 54 grwr.s, as compared with 72 grams to be produced from a solution of 11 raw11 trona~ From the above it is seen tr.at the saving in water to be evaporated, by the use of boiled trona solution, is about 25%. In the design of. a plant the heat from the Gulfur burner ~ight be utilized in treating the trona solutions. The added cost of a hee,t ex- changer and wooden storage tanks is more than offset by the increased CaJ>~city of the plant to be gained. in handling a 25% more concentru.t1~d liY)ld>r, as well as by the saving on evapor!:',,ticm clu:;r.ges. The quea- tion was thus raised as to the econow.y to be gained by utilizing waste he;;.t in cookin3 the trona s:::lution instead of in eva:po1•atb1g the 255~ excess water at some later step. A considere.tion of the heats of' :formation, as ~iven by the Landolt-<Bornstein tables, of the substances involved should settle this point. 2 NaHC0 3 t 20,800 oalories : Na2 co3 t HzO t COz. From the above it is seen that the heat necessary to decompose one mol, of l\l"aFIC03 is 10,400 calories. The heat reqµired ution of cooked trona (100 J,rams of trona in 375 oo. of w0,ter) cc1J.ld be sa.tu.r:~ted v1i th so2 wi thcru.t :precipi ta.tins bicarr.)011~1.ts. approxi1kately 50,000 c~lories. Hence, it i~~ evident t:r...a..t a l:::.rze A substc.ntial reducti011 in ~V'iter to be eva._pora.ted to acid sulfite and s~.:iJ:isequemt neu.tralizat3.on of the resn.ltil:lg solution vd th a calc".lated amount of' trona. &dvantage of this syctem lies in the fact tl'leht acid 6ulfites are more in.jv.riou.s to material~ of tower oonstruotion, either wood or iron, than are neutrel sul:f'i tes. The procedure also produces a ra.l,lid evolution of co2 , Which tend.s to cat1se the me.tarials to froth out of their containers. In trial e>~eriments stirred into the acid sulfite liquors, showed a very great tenT'h.ia ca.J..-:.ed material redis~ol ved very slowly. denay to cake. There is also a considerably decreased efficiency of a'1-;;so1~rltion of SOr. when the solu'l;ion of trona is :t"l:tn to the - t:j acill s12.l.fite instead of to the neutral aulfite. intake tube for SOP', ~va~ ir1·trod11ced u.nd ·t:he ~ain in ,;:, ":J:reight notod. on S02 • The availe:~'ble time bei11.;; short 1 i t was decided to treat a concentratsd. (cooked) ool\ttiori of ti·ona with SO?.» heat with an w excess of rml:f'u.r, and examine the yields produced. Time of heatii:lg GI.ltd the effect of the acidity d.uction of hypo. Jrn indica.to:t• vra.s desired which 'WOu.li give a :f'a-.i:rly following conaiderations served to define the cnaracteristics desired 100 e;rS?J:1s trona in 375 cc. of wata:r v..&len convei·,ted to !\fa,,.,SO,,, produce& a. solution containing 200 grams of . ;;;, .:;i 2 3 per liter (l.6 molal). l~a i:io Let us assume th3 ionization of such a solution to 1)a 50%. (ll+) (SO~) ... - 1C =5 x 10-6 If we asS"l»lle the indieator to change (a) whan 0.1% and ("o) when 0.5% of the sulfite has bo;ian converted to the acid sul:f'i te, we have: 15. (a) tp+\ \""".J.. I' ( .8) ... 5 x l"'-;'3 ".,/ ( .0016) (If*) (b) <rr+J (.8) - io- 8 " 5 x 10-o ( .008) + -- 5 x 10-8 ( H) .An examination of the characteristios of the more common indicators showed litrPU.r:: and rosali:~ ~cid. changing colo1• at (II+) :: io- 7 which is 2,s nG:ll" the de9ired (11'1') ~~ is :pcssi'ble to Re~J.l ts of Trial Rm1.!l. It w:is decided to :make trial run~ 1m1.er consta."'."1t conditions of dil11tion and varyi:nz C·:mditioni.3 of acidity and. to note the yields prodticed v.>i th vs.ryi:n;;i; time of he1I:.ting. 1\ one liter clistillil1g; fla$k 'Nas fitted 'Nith The side neck of the flask wo:;..s 'bent upward tlll'u. so 0 and. attached to a reflux condenser. of trona was dissolved in 375 oc. water a.nd the solu.ti021 satui'ated with so2 V..'i'ltil a 10 cc. :portion (boiled to expel the co ) ohowed. only 2 a al:l.ght :pink with rofu1,lic acid. This solution was show11 to contain The $Olu.tion was placed in the d.ietilling fla.sl:: with 35 grart1s o:f su.1.- The mixture was he<.i.ted on a water bath and };led. with a. rheostat to control it$ speed.. ia. from th~ flask, suction fil tared, anc'l. the volUL1e and te:t4ie1•a.tu.re of the £11 trate noted. A portio11 o:f thiri :filtrate W'il.S takel'l fo1~ analyzing. The remai:rider, together with the sulfur, was replaced. and lle;J.ted for another hour; when another a<:.:i.m;ple vr.a.s remov~d. taken. These sartiples were arul.lyzed. as follows: of :Bi2 S3 thorou.ghly washed. f 01' one 1101..i.r. 'Ihe 10 cc. of KCN solution Wt:H>o added and heated The excess KCH was determined with st:::.ndl-ird AgH0 3 according T11e washings from to the method. of Liebig, using KI as an indica.tor. This difficulty waz ovei·oome 'by removing the s- frow. the original S:oiX\'lple and ma.king it up to 250 cc. with the wash This analysis is given on page 6. Time did not permit :t\'trther trial rwts on "basic solutions. It may be stated, however, that the reaction betw~en Na2 so3 and S lk~d taken place to a considerable extent at the end of the firzt hour of heating, as evidenced by the fact that the sul:f\:lr residue filA s irni lar ru..11 wa.s made on a s~le cf trcna Which h..~d been treated with SO? until slightly ~ acid. The a.na.lysh on page S was applied. to the product obtained. after three hours li.eo.ti1"l.g. .Three saraplss were analysed. The a'mo·lll1 ts of 17. the sta.nd.ard. Aglil'03 solution requ.i:red for es.ch of these three sa.nples va:i.•ied widely. A mee.n of these showed a 70.8% yisld of Na2s2o3 from the original trona. Experirtlents were tlla.de to determine the reasons for these varying results in ana.lysaa. Pure KC!IJ' waa titrated. against the standard .AgM03. usi11g varying &mounts of KI iniiicator, The results obtained showed that the an1om1.t of KI used had no effect upon 'Ghe end point. Tests were made to determine the approxi:nate defu-ree of reversibility to be expected as the excess cyanide is removed from the final solution by titra.tio1'l.. 20 co. of 0.6 molal Na.2 so3 , corresponding to six times the SlilO!J.nt to be formed :from a san~1le of hy-90 irpon analysis was a.d..ded to the 275 cc. of solution of the oort11:ileted analysis, conta.i."l.ing the slight t"UJ.•bid.i ty of AsI .. ;.\fter standu1g cold. for one hour, none of this precipitate had dissolved, showing & non-reversal, under these conditions, of the reaction, Na2Sz03 t KCN V~ien heated =Na.zS03 + KSCN. for an hour, the precipitate v:tdS seen to redissolve, requiring about 2% mo1•e AgN03 to restore it. Henoe,it \~~s conclud- ed thst,under the conditions of dilution a:nd tffi~perature of the w1alyda , the reversion of the reaction to lie encounter eel dtu·ing tit- ration would be negligible. A consideration of the left hand. i:;wmber of the above 18 .. equation shows that, in the case of iDOomplete conversion of the lV.fiO, the error introduced. upon titratiorJ. with .AgN03 is not only tl'l.at p:-i:>odi.ioad by the action o:f' the KCN, but also an effect :p:rod.u.aed by the dh$olving of the end point precipitates by the h;n?o. The method of analysis on page G had been previo~sly was made that it was applicable to the prod:l1cts of the trial runs. The excess of Na.Cl'!' sol'i..i.tion added is calculated on the basis of a eq,uilib:rium equation (so;) ( sm;r) (s2 o~) (mr) -- K Let us assume that the filtrate from the run contains only 5o% of the - calcnlated s203 - the remainder oonaistin~- of unconverted so:. ..;, From a consid.eration of the above equation, we find that the conoe;:itration of s2 o3 at eqp.ilibriuin in this oase rm.tat be less than in the oase when the filtrate contains the cal~alated (l0<1'~) amount of Na2 s2 o3 • This is aee11 to Vlndicate the a.ssu.i7fptio:n that the p1.. 01Jo&ed method. of analysis should be applicable to the analysis of reaction products. However, the facts of the c:~:le hava not, in the limited i'lumber of analyses ma.de, been in acc;ord. with this asst'U.!1ption. Results of the~e analyses lu~ve been widely varying. It wa.o concluded. that, to continue the study of the :process by mea11s of trial n1m1) the conditions £o11.nd in the f:Utr2,tes. 19. Summary. A study ·wa.s i::.::1.de on th~ producticn of sodh1m thi::i- su.l:f\;,te fr :ml trona 5.n view of it?. developm~nt thru the l:;;.borat!lry 1 of the :r;.l'ooess. Im:.i.'rove:men-1'., of tha a11alysio of t.he :c.iro.J.u.et was found n"!CeH~y. on raaterials of construction wa.e st'Udiee.• of this w·ork.