.1 L. _— _— __ —_ — _— —— — —— —— —— .— —_ —— _— _— THE TiTRATION OF TI-iALLOUS EON WfiT-H POTASSIUM CHROMATE Thai: for tho Dogm of M. i. MlCHIGAN STATE COLLEGE Edwin F. C. Cain E952 "'5va l w 1. Y ; w R H W; ,» S .MM . m A... .1 r h .1 gr! q )ix 1!“ e w I; 7.0 V v I l1 :. mix} .1 v 1n n w r: ‘4 a .X .U n. - n L mm“. P. W \ W“ ‘ _ \\\'\\§\m\m\m\m\1 \ “23 Ml 0119 93 3 12 .1 x (K. . C THE THRATION OF THAIJDUS ION WITH POTASSIUM CHROMATE By Edwin F. C. Cain A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture tad Applied Science in putitl fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry 1952 / 2/21 1/5/51 ( 5g) ACKNWLEDCHENT Grateful appreciation in expreued to Doctor Elmer Leininger under whose kind and efficient direction this work was carried out and to the Central Scientific Compuw for their financial aid in the form of the Central Scientific Company Graduate Scholar'- ehip for the year 1951-52. “mm “it?" if“ i} 1 4 N "('g ma..- e '5. 7.3M.) TABLE OF C ONTEEITS PAGE INTRODUCTION....................................................... QUALITATIVE SPUDY.................................................. PRELIMINARY QUANTITATIVE STUDI..................................... Experiments to Determine the Nature of the End-point.... QUANTITATIVE METHODS............................................... THE EF'FET 0F romm SALTS........................................ INTWERE‘NCES...................................................... COMPARISON Him HEHROTRA'S HEEEDD.................................. MLTHIL RED A3 A POSSIBLE INDICATOR................................. SEPARATION OF THALLIUH....... DISJUSSION AND CONCLUSION.......................................... LITEATURE CITEDOOOOOOOOOOOOOOOOOIICO0.0.0.0000...OOOOIOOOOOOOOOOIO 22 31 33 36 ho 1:6 LIST OF TABLES TABLE PAGE I . Smmary of Titration: of Thalloue Nitrate with Potaeeimn Iodide and Poteeeiun Bromide Using Various Adsorption Imic.wr..OCOOOOOICOCOUO0.....0......OCOCOOOOOOOIUOOOOIOI.O. II . Sumary of Reeulte with Various Substituted Indophenols . . . . . . III. Standardieetion of 20 .(D M1. of Stock Thalloue Nitrate alutionIOOOOOOOOO......OOCOOIOOOO......OOOOOOOIOOOOOOOOO0.0. IV. Titratione to Determine Experimental Conditions . . . . . . . . . . . . . . V. Titratione of Varying Amounts of Thallium. Initial pH Adjusted to 3.85-3.90 by Beckmen Model G pH Meter. . .. . . . . . . . . VI. Titration of Varioue Amounts of Thalliun with roteeeiun Chronete Using Ne 2,6—Dichlorobenzenoneindophenol as Indic.‘or00000000000.00.00.000....OOIOOOOOOOOOOOOOOOOOOOOI... VII . Summary of Titration: with Na 2 ,6-Dichlorobenzenoneindophenol VIII. The Effect of Foreign Selte Upon the Titration. . . . . . . . . . . . . . . IX. Titration in the Preaence of Ferrous Iron.............. ...... I. Standardieetion of Potaeeiun Iodide Solution. . . . . . . . . . . . . . . . . XI. Deteninetion of Thalliun byhehrotra Method................., h c“ ‘O h) VI 27 29 .30 32 3h 37 39 INTRDHICTION INTRODUCTION The gravinetric methods for the determination.of thallium have been somewhat limited. The thallous chromate (6) or thallic oxide method (6) have been used for accurate work; but the precipitation of thallous iodide (6) has been used industrially because of its rapidity even though the results were not as accurate. Thionalide (2), thicglyoollic amino- naphthalide, is specific for thallium in the presence of sodium hydroxide and.potassiul cyanide. The complex precipitate may be dried and weighed or titrated iodometrically. The field of volumetric analysis has been explored by various work- ers in an attempt to find a rapid and accurate method. Several methods which depend upon a change in the oxidation state of thalliun.have been proposed (5,8). However, these methods have not proved entirely satis- factory; In two recently published articles, Hehrotra (10,11) attempted to solve the problem of an accurate volumetric method for thallium by proposing the use of hromphenol blue as an adsorption indicator in the precipitation.of thallous iodide. 0n the basis of these two papers, it was decided to investigate the use of other indicators which.might function as adsorption indicators in the precipitation of thallium. The experimental problem was arranged to give special emphasis to the following: (1) sulfonphthalein indicators other than bromphenol blue, (2) adsorption indicators listed by Pagans (13) as suitable for the titration ong+ and 1' ions, (3) chrysoidin derivatives as proposed by Schulek and coworkers (11:45), and (h) conga red, which had been pro- posed by Hehrotra (12) as an adsorption indicator. Tricks and Sonnet (’4) used sodium 2,6odichlorobenzenoneindophenol as an adsorption indicator in the titration of lead with chromate. Since thallium forms an insoluble chronate, the use of various sodium bencenone- indophenol indicators was investigated as adsorption indicators in the precipitation of thallous clu'onate. Sodiun 2,6—dichlorobensenoneind0phenol was used successfully as an indicator in a volumetric precipitation of thallium From the data obtained in these experiments, an accurate volu- metric method for the determination of thallius was evolved. QUALITATIVE Sl‘UDY QUALITATIVE STUDY A. Adsorption Indicators The qualitative study of the various adsorption indicators chosen was carried out as described below. A stock solution of thsllous nitrate, approximately 0.01m, was prepared by dissolving 53 g. of c .P. grade thallous nitrate in 5 liters of redistilled water. The water, which was used in the preparation of all stock solutions, was redistilled from a Pyrex glass still. A potassium iodide solution, approximately 0.01m , was prepared by dissolving 6.7 g. potassium iodide, C.P. grade, in redistilled water and diluting to one liter. a potassiuw bromide solution of similar strength was prepared by dissolving h.8 g. of OJ. grade potassium bromide in water and diluting to one liter. Aqueous solutions of each indicator were prepared whenever possible, but in those cases where the indicator was not soluble in water, a 50% ethyl alcohol solution was used. The concentration of each indicator solution was 0.1 molar. A sample of 20 ll. of stock thallous solution was titrated with potassiul iodide or potassiun bromide in the presence of each indicator listed in Table I. The direct and reverse titrations were attempted. The pH of the thallous solutions were adjusted to give the desired color for. of the indicator. All possible color toms of each indicator were used. The results of these experiments are listed in Table I. mamas» hoHoo on canons nodes on vex nozm :No. codenamemoconm sun-no uoHoo on ego hence on nod new ounce nudge .330 8H8 a- ne own-so .58 an: .95 umo. .— .5885— eunwu end: ewneno nodes on here eunsne.nenoo Queens vex scam nmo. o o end-soon: engage peace on sundae peace on ewnsAe hence on sundae peace on sundae uoaoo on sundae peace on sundae uoHoo on ownino hoHoo on emu-no peace on sundae peace enudam smudge hedge on ewndno hofloo cu emnwno nodes on emqsno hence on sundae woaoo on sundae peace on euasno nodes on sundae nodes on sundae nodes on smudge peace on engage peace on emnwno peace on sundae woman on sundae hefloo on encode nodes on emssno uoaoo on euneno hcaoo on sundae node» on eaneao hoaoe o- ewneno hence on .8er .58 Ema encode nodes on sundae began on engine Lease on sundae oneo anuaam sundae peace unuuam engage hoHoe e- euaeno woaoe on ewnlno ooaoo on canine onoo on sun-no nodes on engage peace ea smudge nodes on ommune peace on casino nodes on sundae peace on manage onoc on sundae nodes on casino nodes on ensenu weHoo on canine hoaeo e- emneno hoaoe on canons woaoo on sundae nodes on sun-ac nodes on 85.: ...-38...: can!» .58 o- ..31» €83. undone Medea on soaflehwneaneeho. «one: 33.» bag» can 32.x 3. 33 fig .255 season-u ednhnm gonad.» one fined-u com .5331» 8% :QADOHHGH . 3 inane» 555 icon.“ 33 union.” £095 .333." 33 qqqqquqqq«qeonqnqeqqqquqqqqo r-r-uwesucexp-ear-enr4c~oau>aouxucr-aaunr-e-r-GNVDFor-owe- Sachem 58.37538 Second.» 55.37938 .m 53859 . .m 33850 33850 awesome 538.65 Passed 33 Saint: soda: auto .33; 33a: eon agonofioflo eanhnm Hoeonoue cannon Hoeeho1l com H83.— eem H23:— 3 Honcho ecu 28.8 enameaclbna 33 H85. 38 ~82... ennusm HoeeoQIOhm eanusmedoeoueloum 3:. Honorsfim SS Agosto g Hal’s; GOO-5 Hosea-um 33 again 3.8 assign 58m 508:8: 312? .3: a8 m .. g... eu.0 .33“: 5:38 «o 3:38 O . noeeodnnm mududounzH IOHHmmomndgmaondb GIHM=.HQHSQQI qumn, . 0 c1 “0&4“ <66} " 0 "nab-l" d - 0 Cl Br . 0 Br N30 n.0—2H:}n-< >- o - Br B 7.3: “‘0’ ‘3. ......— .0 Br N80 ._._oz 0.... uum} 2.0. or. .2: P... (5.3526 _z_.:>_u DI <>rcmm m .m..// m.o.. mo. v0.2... n: m.9\ mo 1 bu . 0.0 .. won K k f ab . o w news 0' OJ‘ mu. mo um . so ea 2:: on conuuz x~oxoi Graph 1 showed a definite pH change occurring during the titration. It also showed that the point of color change for a pH indicator could be shifted by varying the initi pH of the solution. Thus the amount of standard chromate necessary to cause a sharp change in the pH of the solution was dependent upon the initial pH of the solution. Similar curves were developed for titrations in which the indicator was present. These titrations were carried out in the same manner as those described above except that 10 drops of a 0.2% indicator solution were added. Both indicators, sodium 2,6~dichlorobenzenoneindophenol and sodims 2,6—dibromobenzemneind0phenol, change color from red to blue at e.pfi of 5.7. On the basis of Graph 1, an initial pH of 3.8 was chosen because the pH at which the indicators change color was located approxi- Iately at the equivalence point on the curve ‘with an initial pH of 3.8. The results of these titration: are plotted on Graph 2. Graph 2 showed that the pH curve of the titration was unchanged in the presence of either indicator. It also showed that the indicator night be functioning only as a pH indicator. Therefore a titration was ' lads upon 30.00 ml. of redistilled water containing an amount of’potassiun nitrate equivalent to that formed during the titration of 189.? ng. of thallous ions Titrations were carried out in the absence of either indicator and also in the presence of 2 drops of a.0.2$ solution.of sodima 2,6-dichlorobensenonsindophenol or 2 drops of a 0.2% solution of sodiul.2,6odibronobenzenoneindophenol. Both indicators changed from red to blue exactly as when thallium was present. The results of these titrations are plotted in Graph 3. l5 OEDUI 20. N O... w~m.¢§m.0w .2. >... .2243: _uI Om Wm 3.235202 20 _Zo_0>._.om wwmmmzae ZO.N .0 omovm O.No\o 2» N.m-O_OIPOmOmm2Nmzozm_ZUOVINZO... boomo X 0cm No.0.nxrowomNZNmzoszoou vzmzor X m.ml VI 0.01 \ who; 9? e. 99 \+ e.ol.\,..\.\r Anne; m u a u m u m o .o 233 Poeuuz x» 9. 0+ 17 The pH of the 0.0h35‘N potassium dichromete solution was 8.10. T113 was due to the hydrolysis of the potassium chromete according to equations (3) and (h). Kfiro‘ + up—zucw.‘ + 011‘ + 2x” (3) Hero; + Hack-:znacm. + on“ (1.) Several titrations were eade in which the pH of the chromate solution was varied by the addition of nitric acid or potassium hydroxide. The pH curves of these titrations varied pestly tron those in which the pH of the chromate solution was a function only of the concentration. These curves are reproduced in Graph h. In the case of the acidified chromate solution, the precipitate formed during the titration was orenge instead of yellow. However, when potassiun hydroxide was added to the chronate solution, the pH of the sample increased steadily. It was obvious, that in both instances, a marked variation fro- the other curves had occurred. In summary, aw theory in regards to the nature of the end-point lust take into consideration the following: (1) The pH of the solution remained fairly. constant until the eQuivalenoe point was approached. (2) The color change of the indicator was independent of the presence of thalliun and was dependent upon the pH of the solution. (3) The amount of standard chromate which must be added to cause the color change or the indicator was dependent upon the amount of thallitn and the initial pH of the solution. (1;) The addition of acid or base to the standard chronate solution caused wide variations in the subsequent pH curves. PH GRAPH NO.4 ’8 TITRATION OF 379.4 MG TL WITH 0.0463 N KICRO4OF VARYING PH CURVE I, PH OF STD Kg“: 6.60 e CURVE 2, PH 0F STD KZCRO4 6.95 x CURVE 3, PH OF STD «5.0,, ms . 9.0. EQ. POINT ‘ l/A 8.5 / so, ' ,/ 7.54 / 7o, 6 5 " . . . ,/ / ./° 6.0.. . x / / I" 5 5. X ‘ ‘ I // 4.5. 4.04 X 3 ,5- "\ 3 .0. \'*" 2 .5; 2.0-. o 5 IO I5 20 25 30 35 4045 50 ML OF 0.0453N K2094 WITH VARIOUS PH VALUES 19 The above sentioned experimental facts can be explained by consider- ing that the color change of the indicator was due to a change in pH caused by hydrolysis of the excess potassium chromate. The nearly con- stant pH of the solution prior to the color change was caused by the removal of essentially all chronate ion from solution. In order to verify the above statuents, a single drop of 0.01051! potassiul chromate was added to 15 ll. of stock thallous nitrate solution containing 23 .7 ng. of thallium. A precipitate of yellow thallous chromate was formed and did not dissolve upon standing for five hours. The formation of the precipitate was considered likely even though the solubility product were not exceeded. However, calculation of the solubility product from the solubility of thallous chronate given in the literature (16) , subsequently showed that the solubility product of thallous chrosate was exceeded. The required calculations are reproduced below. The solubility of Tl.CrO‘ given was 0.0M? g. TlfirO‘ per liter of saturated solution. Therefore the molar concentration was 8 .1h x 10—5 . Imp. - {11“1'. [oz-of] mam-o. I. p - (16.3 x 10"]. . [8.11I J: 10"] ' ”£1,010. Imp. - 1.75 x 10'“ run-o. Fifteen milliliters of solution contained 23.7 mg. of thallium or 7.7 x 10" lillilolcl/ml. Pros the solubility product of thallous chronate calculated above, the concentration of chronote ion necessary to cause precipitation was calculated as shown below. . £7.7x10"]‘ . [Cr04'] - 1.75 x 10"“ [Cr-0"] - 2.93 x 10" millimoles/nl. Neglecting the small change in volume, the molar concentration of chmnate ion was calculated when one drop of 0.03435N POtassium chromate was added to 15 ml. of solution. 1 drop . 0.05 ml. n1 3: R . milliecLuivalents of Kfi’fle I§ ml. 1 nilliequivalent “Grog :- 1/2 millim01eé CF04: Therefore the concentration is 7.25 x 10" nillimoles of 610,71311 According to. the above calculation, the solubility product of thallous chromate was exceeded by the addition of one drop of 0.0h35N potassium chromate solution to 15 ml. of a thallous nitrate solution which contained 23.7 mg. of thalliun. These calculations substantiate the theory that in the initial phase of. the titration, essentially all chroaate was effectively removed from solution by the formation of thallous chronate according to the reaction below. 2T1NO, + 11.620, T135101 + ZERO. The potassiun nitrate, which was the other product formed, did not affect the pH of the solution. Thus, the increase in pH at the equivalence point was caused by the hydrolysis of potassius or thallous chronate. 21 Note should be nade at this point that during further experiments- tion, it was found that some adsorption of the indicator probably occurred. In attempting to establish optimum conditions for carrying out the titration, attmpts were made to match the blue color in the supernatant liquid with a color blank. The color blank was prepared by adding the indicator being used to a buffer solution with a pH of 5.7. The buffer was composed of potassium acid phthalate and sodium hydroxide in quantities which were specified by Clark (3). The volmne of buffer solution was approximately that of the ample after titration and identi- cal amounts of indicator were used in each instance. However, it was noted that the color blank was nore highly colored than the supernatant liquid of the sanple after titration. It was seemed from this data that some adsorption of the indicator had occurred. However, when thallous chromate was filtered from a sample which had been previously titrated in the presence of indicator, no difference in color was noted when compared with thallous chronate which was precipitated in the ab- sence of indicator. This fact say be due to the large amount of bright yellow thallous chronate and the small mount of indicator which was probably adsorbed. QUANTITATIVE KETHODS 22 QUART IT ATI V}; I‘IETHODS After establishing the color change of sodium 2,6-dichlorobenzenone- indophenol and sodium 2,6—dibromobenzenoneindophenol to be that of a pH indicator, it was necessary to detemine the initial pi! of the solution which was required in order that the indicator change color at the equivalence point. From Graph 1 and Graph 2, it was apparent that the initial pH should be somewhere be'tween 3.7 and 11.2. It was also desirable at this point to determine the amount of indicator which must be added to obtain the sharpest bend-point. Since the indicator used was the sodium salt of a weak acid, it had basic properties upon lvdrolysls. Therefore, if it were added at any point in the titration, the pH of the solution would be affected. By experimentation, it was found that the addition of two drops of a 0.2% aqueous solution of the indicator to 30.00 ml. of solution was sufficient to give a good color change. Titrations were carried out in which the initial pH of the solution was varied between 3.7 and h.2. The initial.pfl was adjusted by using 0.013 and‘3.1l nitric acid and ammonium hydroxide solutions in conjunction with the Beckman Model H-2 pH meter which was standardized as described previously. It was found that if 30.00 ml. of thallous nitrate solution was adjusted to a.pH of 3.85 to 3.90 and two drops of a.O.2% indicator solution added, the color change could be used to establish the equiva- lence point. Attempts were made to titrate the solution to a color blank which has been described previously, but it was found.more satisfactory to titrate to a pure blue color in the supernatant liquid. 23 In order to make the method more widely applicable, it was con- sidered desirable to adjust the original pH of the solution by means of a second indicator. Methyl orange was chosen because it changed color at the desired pH and also because the yellow color of the basic fern would not interfere with the color change of the indophenol. It was determined that two drops of a solution containing 0.2 g. of the sodium salt of netlvl orange per liter was sufficient. The color was matched with a freshly prepared bleak which was made by adding two drops of the methyl orange solution to 30.00 :11. or a solution which was buffered to a 9‘1 of 3.854 .90. The 3.85 buffer was prepared in the following nanner (3). a 0 .1 solar solution of potassius acid phthalate was prepared by dissolving 20.132 g. of C.P. anhydrous potassiun acid phthalate in redistilled water and diluting to one liter. An 1101 seluo tion which was standardised in the usual manner was also used. The buffer was prepared by nixing 50 ml. or the 0.1 molar potassium acid pbthalate solution with 0.233 nilliequivalente of 301. The solution was then diluted to 100 ml. Ira the abeve data, conditions were tentatively. established for carrying out the titration in order to deternine the effective concentra- tion range of thallium over which the method would work. Titrations were nsde with varying amounts of thallium present and it was found that the concentration of the thallimn must be taken into consideration. The count or thalliun present cid‘not affect the action or the indicator; but , if the comentretion of thallim was too low, the precipitated thallous ehrmate did not coagulate and the end-point was obscm‘ed. 214 Thus the initial value of solution to be titrated must be regulated. Ithis was also true of the amount of indOphenol indicator added. It was found that if the thallium concentration were 914 to 190 mg, an initial volume of 30 ml.and two drops of the indophenol indicator was satisfactory. However, 1: the amount of thallium being titrated were 23 to 91; mg., the initial volxme should be 15 ml. and one drOp of the indophenol indicator should be used. The limits given above in regards to initial volume are not highly critical, but must be observed within reason. It was found that it was not necessary to reduce the two drops of metlvl orange to one drop when titrating only 15 ml. of solution. The yellow color of the basic form of methyl orange did not interfere even when two drape were used in adjusting the initial pH of 15 ml. of thallous solution. During these experiments , the use of sodium 2,6-dibromobenzenoneindo- phenol was unsatisfactory. II‘he failure of this indicator was apparently due to adsorption of the indicator by the precipitate. Thus at the equivalence point, very little indicator was left in solution and the color change was poor. Therefore all further work employed only sodim 2 ,6—dichlorobensenoneind0phenol as indicator. The use of a pH meter to adjust the initial pH of the solution was also investigated. The initial pH of the solution was adjusted by loans of a Bechsn Model G pH leter which was standardized in the same manner as described previously for the Bookman Model 3-2 pH neter. Although experimental results show that a pH meter may be used successfully, some difficulty was encountered in keeping the voltme of solution within set TABLE IV TITRATIONS T0 DQ’EfiL‘iIHE EEERDIHTAL CONDITIONS Titration Mg Tl file. of Drops of Drops of Hg Tl Error lumber Present Soln. H .0 . Indophenol Cale , pp [1000 1 189 .7 30 2 2 189 .11 -2 .5 2 189.7 30 2 2 189.29 .1.5 3 189 .7 30 2 2 189 .38 -1 .0 h 9h.85 3o 2 2 9h.69 -1.0 5 9h .85 30 2 2 95 .13 +3 .3 6 9h.85 30 2 2 9h.h2 -3.8 7 h7.h2 30 2 2 h7.7 +5.0 8 h7.h2 30 2 2 h7.9 +10.0 9 h7.h2 30 2 2 h6.h «20.0 10 h7.h2 15 2 2 h7.2 ~heo 11 h7.h2 15 2 2 h7.6 +3.0 12 h7.h2 15 2 1 h7.h 0.0 13 h7.h2 15 2 1 h7.h 0.0 lb 23 .71 15 2 1 23 .h ~15 .0 15 23.71 15 2 1 23.7 0.0 16 23 .71 15 2 l 23 .h -15 .0 17 9h.85 15 2 1 9h.7o -2.0 18 9h.85 15 2 1 9h.39 ~5.0 26 Units due to the quantitative transfers between the titrating flask and the pH neter, and the necessity of rinsing the electrodes. The results are tabulated in Table V. The final conditions for carrying out the volumetric determination of thalliu- by precipitation of thallous chromate were established to be as follows. A. Titration of 9b to 190 mg. of thallium. The sample containing the thalliun in the thallous state was placed in a 125 nl. Erlenmeyer flask. The total volume of the solution was adjusted to approxinately 30 ml. and two draps of methyl orange solution, 0.2 g. of the sodiu- salt per liter, were added. The pH of the thallous solution was adjusted to 3.85-3 .90 by using 0.11 and 0 .OlR solutions of nitric acid and moniun hydroxide. The color of the solution was snatched with a color blank which was prepared weekly by adding two drops of methyl orange to 30.00 :31. of the 3.85.3 .90 buffer previously discussed. After adjusting the initial pH as described above, 2 drape of a 0.2% aqueous solution of sodium 2 ,6-dichlorobensenoneindophenol were added. it this point the solution was scarlet. Standard potassiu- chronate was added slowly while the flask was rotated vigorously. II'he bright yellow thallous chromate precipitated causing the suspension to have an orange color. The flask was rotated continually during the titration to coagulate the thallous ohronate. After the addition of several milliliters of potassiul clu'onate, the titration was stapped and the flask swirled for about 60 seconds. At the end-point, the thallous chromate should be TITRATIONS or VARIING AMOUNTS or THALLIUH. 1'0 3.85-3.90 BI BECKEJJJI MODEL 0 pH MEIER TABLE V 11mm. pH ADJUSTED 27 Titration 141. Kg T1 Mg Tl Error Number Solution Present Calculated pp/lQOO 1 30 189.70 189.1; -1 .5 2 30 189 .70 189 .3 '2 .0 3 30 189 .70 189.3 -2 ,0 h 30 189 .70 189 .5 -1 .0 . 5 30 189 .70 189 .h -l .5 6 30 9h .85 9h .51 -2 .5 7 30 9h .85 9h .h2 -3 .5 8 30 914.85 91.33 ~h.5 9 30 911 .85 9h.33 44.5 10 30 9b .85 9h .33 ~24 .5 11 15 h7 .h2 h? .3 -2 .0 12 15 1.7 .h2 h? .2 ~14 .0 13 15 in .t2 1:? .3 -2 .0 1h 15 1‘? ehz h? e1 06.0 15' 15 23 .71 23 .h -5 .0 16 15 23 .71 23 .6 -5 .0 l? 15 23 .71 23 .3 -8 .0 18 15 23 .71 23 .2 ~10 .0 28 alnost completely coagulated. Near the end-point a green color appeared where each drop of cm'onate entered, but disappeared upon mixing. Upon further addition of potassium chronate, the entire solution became green- ish, but reverted to orange won shaking for several seconds. At the end- point, the suspension renained green when it was swirled vigorously. The solution was allowed to stand for several seconds. The yellow precipitate settled and the color of the supernatant liquid was observed. The end- point was reached when the supernatant liquid exhibited a true blue color. If a purple color existed, the end-point had not been reached and if a green color existed, the end-point had been overstepped. B. Titration of 22 to 2y :3. of thallium. The titration was carried out in exactly the some manner as above, enept that the initial volume was 15 ml. and one drop of the indephenol indicator was used. A 15 :1. methyl orange color blank was also used, but two drops of nethyl orange solution were added . The results of the titrations carried out according to the above procedures are smarised in Tables 71 and VII. TABLE VI TITRATION OF VARIOUS AMOUNTS 01" THALIJUH WITH POTASSIUM CHROMATE USING Ila 2,6-DICHLOROBENZFJONEINDOPHENOL A3 INDICATOR Titration Ml . Ml . N . M T1 Error Nunber Soln. no.6. new. M pp/looo 1 30 21 .26 0.011350 189.70 189 .0 -3 .5 2 30 21.28 0 31.350 189 .70 189 .2 -2 .5 3 30 21.35 0 .08350 189.70 189.8 +0 .5 h 30 21 .27 0 .0h350 189 .70 189 .1 -3 .0 5 30 21 .29 0 014350 189 .70 189 .3 ~2 .0 6 30 21 .30 0 014350 189.70 189 .h -1.5 7 30 21 .32 0 .011350 189 .70 189 .6 ~0 .5 8 3o 21 .27 0 .08350 189 .70 189 .1 -3 .0 9 3o 21 .29 0 108350 189 .70 189 .3 -2 .0 10 30 21 .30 o .0h350 189 .70 189 .h -1 .5 11 30 10 .61 0 .0h350 911.85 91.: .33 -6 .0 12 30 10 .61 0 .0 14350 9h .85 9h .33 -6 .0 13 30 10 .70 0 .Oh350 9h .85 95 .13 +3 .0 1h 30 10 .68 0 .011350 9h .85 9h .95 +1 .0 15 30 10.65 0.017350 911.85 911.69 -2 .0 16 30 10 .63 O .0h350 9h . 85 9h .51 ~11 .0 17 30 10 .66 0 011350 98.85 9h .78 -0 .5 18 30 10 .65 0 .0h350 9h .85 911.69 -2 .0 19 30 10 .70 0 .0h350 9h .85 95 .13 +3 .0 20 30 , 1o .62 0 .0h350 9b.85 98 .112 -h .5 21 15 5.32 0.01.350 87.82 117.3 —2.0 22 15 5 .26 0 .014350 177 .12 L6. - 2 .0 23 15 5 .32 0 .08350 h7 la la .3 -2 .0 2h 15 5 .32 0 .011350 117 .112 h? .3 -2 .0 25 15 5 e31 0 eOhBSo h? e142 h? e2 .111 so 26 15 5 .28 0 .0h350 h? .112 h? .0 -8 .0 27 15 5 .30 0 .0L350 1:7 .h2 117 .l -6 .0 2 8 15 S .33 0 .014350 h7 JR 1:? .h 0 .0 29 15 5 .311 0 .0h350 L7 .82 h? .h 0 .0 30 15 2 .66 0.011350 23 .71 23 .7 0 .0 31 15 2 .59 0 .0h350 23 .71 23 .0 ~35 .0 32 15 2 .611 0 .08350 23 .71 23 .6 -5 .0 33 15 2 .614 0 .0h350 23 .71 23 .6 -5 .0 3h 15 2.63 0.01.350 23.71 23 .1: ~15.0 35 15 2 .634 0 .0 11350 23 .71 23 .6 -5 .0 36 15 2 .65 0 .0h350 23 .71 23 .7 0 .0 37 15 2 . 63 0 .0h350 23 .71 23 .h ~15 .0 38 15 2 .66 0 .0h350 23 .71 23 .7 0 .0 TABLE VII SUMMARI 0F TITRATIONS UI‘I'I-I Na 2,6-DICHLOROBENZENONEINDOPHENOL lunber of H Thalliun Av. Error Han. Dev. Titrations fiesent Found pp/lOOO DPflOOO 10 189 .7 189 .30 -2 h .0 10 911.85 98.147 41 8.5 9 h? .112 h? .21 -h 12 .0 8 (Omit #22) 117.112 h? .31 ~3 8.0 10 23 .71 23 .5 ~10 35 .0 9 (Chit #31) 23.71 23.6 -8 15 .0 THE EFFECT OF FOREIQI SALTS 3.1 THE EFFET OF FOREICRJ SALTS The effect of foreign salts upon the titration was investigated. Titrations were carried out with the addition of varying amounts of foreign salts, which were nest likely to be present. The pH of each ti- tration was adjusted to pH 3.85-3.90 after the addition of the foreign salt but prior to the titration. The results of these titrations are given in Table VIII. The addition of sodium or ammonium nitrate did not affect the deter- lination. 0n the other hand, the amount of sulfate present affected the end-point. The presence of a large sulfate concentration caused high results. The addition of even a mall amount of phosphate or acetate caused very high results. These high results were probably due to the ' buffering action of the phosphate and acetate ions. The results obtained in the presence of 500 ng. of perchloric acid were excellent. Thus perchlorio acid in large amounts does not interfere. However, the addi- tion of even a small amount of potassim fluoride decolorized the indi- oator completely . TABLE VIII THE.EFFfiDT OF FOREIGN SALTSLUPOR THE TITRATION Salt Mg His . g Mg L T; A _ Error Added Added Soln. Present Calculated—' pp/IOOO New. 100 30 189.7 189 .9 +1.0 Haaso‘ 200 30 189.7 191.7 +10.0 17.350. 500 30 189 .7 193 .2 +18 .0 (101.; .30. 100 30 189 .7 189 .9 +1 .0 (NH. 2304 200 30 189 .7 190 .7 +5 .0 mm. 100 30 189 .7 189.9 +1 .0 mo. 200 30 189.7 189.5 -1.0 III-IO. 500 30 189 .7 169 .2 -2 .5 1711.240, 100 30 189 .7 189 .8 +0 .5 1111.100, 200 30 189 .7 189 .6 -0 .5 mm, 500 30 189 .7 189 .2 -2 .5 la.P0. 10 30 189 .7 190 .5 44.0 lafio‘ 20 30 189.7 193 .1; +18 .5 Ila-.130. SO 30 189 .7 197 .6 +39 .5 let: ,a,o , 10 30 189 .7 191 .1 +8.5 7111.010. 585 30 189 .7 189 .8 +0 .5 I! 100 30 189 .7 decolorized indicator completelqz .«3 ed Immenw INTERF EICiNC ES Several different interferences were noted during the experimental work and the literature survey. One of the most troublesome inter- ferences is iron. Iron is commonly found with thallium and constitutes an interference when present in either the ferrous or ferric state. fitrations were made in which ferrous iron was added as ferrous ammonium sulfate. The initial pH was adjusted after the ferrous iron was intro- duced. The results of these titratione were high and the precision was very poor. The results are reproduced in Table II. In order to study the interference of iron in the ferric state, one milliliter of a ferric nitrate solution, which contained one milligram of iron per 'milliliter, was added to several samples of thallous nitrate . During the subseQuent determination, ferric hydronde precipitated. The freshly precipitated hydroxide immediately removed all indicator from solution. Apparently the removal of the indicator by the ferric hydroxide was due to adsorption. It might be expected that alminiun and chromium (111) would act in a similar manner. The buffering effect of phosphate and acetate has been mentioned previously. Therefore it was assuned that am ion which will establish a buffered system ainilar to that established by phosphate and acetate would constitute ion interference. Mention has also been made that sul- fate ion acts as an interference. Several cations and anions are obvious interferences. Thallous iodide and broside were precipitated in the initial phase of this work and thallous 3h TABLE III TITRATION IN THE PRESENCE OF FERROUS IRON W fl Titration Hg . Ye" M Tl Error lumber Added recent oun pp/lOOO 1 10 189.7 1914.3 +23.o 2 10 189 .7 192 .8 +15 .5 3 10 189 .7 193 .7 +20 .0 h 10 189 .7 193 .9 +21 .0 35 chloride is insoluble. Therefore, Cl", Br- and I" would constitute interferences. The commonly used hydrogen sulfide qualitative scheme employs the precipitation of lead chromate and barium chronate as con- fimtory tests while insoluble silver chromate is used as an indicator in the Hohr method for the determination of chloride ion. Therefore it is readily apparent that bari‘, lead and silver constitute interferences in this method . CCMPARISCE WITH mans METfiOD 36 COMPARIWN WITH MEHROTRA'S METHOD Hehrotra (10,11) proposed the determination of thallium by precipi- tation of thallous iodide using bronphenol blue as an adsorption indicator. Since this research problem was initially based upon Hehrotra's work, a comparison between the two methods was desirable. The comparison was carried out in the following manner. A standard [I solution was prepared by dissolving 8.35 g. of C.P. potassium iodide in redistilled water and diluting to one liter. The solution was standardized (7) by measuring 20.00 ml. of the potassium iodide into a 1100 nl. beaker. The solution was diluted to 200 ml. and reds alkaline by the addition of a dilute moniun hydroxide solution which was prepared from 2 part of concentrated mania hydroxide and one psrt water. Silver nitrate, 0 .05K , was added slowly with stirring until the precipitation was complete . The precipitation was carried out in diffuse light. Nitric acid was added in 11 by volume excess. The pre- cipitate was filtered and washed with 11 by volume nitric acid. The final washing was nade with water to remove all nitric acid. water was sparingly used because it tended to render the iodide colloidal. The precipitate was dried at 100410" 0. for one hour and then at 130-150" C. for one hour. The results of the standardization are given in Table I. llehrctra stated that the end-point was reversible, but that it was tech sharper if iodide ion was titrated with a standard thallous solution. Titrations were attempted in both directions, but in titrating thallous ion 3? TABLE X STANDARDIZATION OF POTASSIUM IDDIDE SJLUTION Sample Mg of XI Normality of Number Hg £31 per 20.00 :11. II Solution 1 29h.h 208.1 0 .0626? 2 29h.5 208.2 0.06273 3 29h.) 208.1 0.06267 1: 295.3 208.8 ' 0.06286 Average of lies. 1,2 and 3 . 0.06269 38 with potassium iodide, the end-point was so very poor that it was unusable. However, in titrating iodide ion with thallous nitrate a fair end-v-point was found. The titrations were carried out by pipeting varying amounts of standardized potassium iodide solution into a 125 ml. flask and adding 1; drops of 0.1% alcoholic solution of bromphenol blue. The pH of the solution was 5.0. The total volume of the solution was adjusted to I) .0 ml. with distilled water. Standard thallous nitrate solution was added with vigorous agitation. The titration was continued until the precipi- tate turned a dark green. The titration had to be carried out in very diffuse light because strong light turns the precipitated mum iodide from a yellow to a green color. The results of these titrations are given in Table I. ' In comparing nethode, the following facts should be taken into con- sideration. The proposed chronate method is a direct titration of thallium while the Hehrotra nethod involves a reverse titration. The standardisa- tion of the chrmate solution was carried out volunetrically while the standardization of the potassium iodide solution was done gravinetrieally. However, the iodide solution night have been standardised voltmetrically using an adsorption indicator as proposed by Teams (13) . The end-point of the chronate nethod was sharper and clearer than that of the Hehrotra nethod. In view of the above cited facts, it is believed um. the proposed chronate nethod is superior to the Hehro tra method. TABLEXI DEIEMINATION or THALIJUH 81' wow METHOD ' W 39 3:52: “.21“ 3’35. ”111%?“ $15.15.. We:- 1.1%.» 1 20.0 30 1.251: ' 27.08 256.3 256.7 +2 2 20 .0 30 1 .251: 27 .13 256 .3 257 .1 +h 3 20.0 30 1.2514 27.08 256.3 257.1 +2 h - 10 .0 30 0 .6269 13 .60 128.2 128.8 +6 5 10 .0 30 0 .6269 13 .58 128 .2 128.6 +h 6 10.0 30 0 .6269 13 .63 128.2 129 .3 +11 7 5 .0 30 0 .3135 6.81; 6h .1 65 .8 +28 8 5.0 30 o .3135 6.85 6h .1 66.0 +32 9 S .0 30 0 .3135 6 .82 6h .1 65 .0 +15 T METHYL RED AS A POSSIBLE INDICATOR 2:0 METHYL RED AS A POSSIBLE INDICATOR Methyl red was tried as an indicator because of the relative un- availability of sodium 2,6-dichlorobenzemneindophenol and because it changed color over the desired pH range . Several titrations were made upon 20 ll. of standard thallous nitrate solution which was diluted to 30 ml. The solution contained 189.? mg. of thallium. The pH of the solu- tion was adjusted to 3.89-3.90 with methyl orange in the acne manner as described previously. Four drops of a prepared methyl red solution were added. Standard potassium chronate was added with vigorous swirling as described under the directions for carrying out the titration using sodium 2,6-dichlorobensenoneindophenol. A color change in the supernatant liQuid from red to yellow took place. The color change was fairly sharp. However, some of the methyl red underwent adsorption because the pre- cipitated yellow thallous chromate had no orange out. This caused the suspension to appear orange upon swirling, but if the precipitate was allowed to settle, the supernatant liquid underwent a fairly sharp color change tron red to yellow. The quantitative results of this experinent were extremely poor . However, if the preper experimental conditions were determined, the methyl red color change probably could be used to mark the equivalence point in the precipitation of thallous ion with potassiun chromate . SEPARRTION OF TEXALLIUM SEFARATION OF TZIALLIUM Wade (1?) proposed that thallic bromide could be separated from the salts of all other metals except gold by shaking a hydrobromic acid solution with other. McBryde and Too (9) used the other extraction from hydrobronic acid to separate gold. According to heBryde and Toe the percent of extraction of iron was dependent upon the approximate molarity of the hydrobronic acid and the other used. The smallest amount of iron was removed if the molarity of the hydrobmmic acid was maintained at 2-2.5 molar and isopropyl other was used. In view of the two papers cited above, an attempt was made to separate thallium from iron by extraction with ieopropyl ether. Eastinan Kodak technical grade isoPropyl ether was purified in the following manner. The ether was shaken in a large separatory funnel with acidified ferrous ammonium sulfate solution to remove the peroxides which might be present in the ether. The ether layer was separated and stored over sodiun to remove all alcohols and water. After drying for several days, the ether was redistilled from capper turnings and placed in a dry bottle whose stopper contained a calciun chloride tube filled with anhy- drous magnesiun perchlorate. The following procedures were used in attupting to separate thallium fron iron by extraction with ieoprOpyl ether. 1 . With continuous extraction apparatus. A continuous extraction apparatus as described by Ashley and Hurry (l) r... [0 was used. A 5 n1. ample of stock thallous nitrate solution which con- tained h7.h1.-g. or thallium, and 1 ml. or a ferric nitrate solution which contained 10 mg. of iron were placed in the apparatus. Bromine water was added in excess to oxidize the thallium from the thallous to the thallic state. The nolarity of the solution with respect to hydro- bromic acid was adjusted by the addition or 18.5 ml. of 3M 2132-. The solution was diluted to 50 :1. The isoprOpyl ether was added and the extraction carried out for eight hours. After extraction, the ether solution was evaporated on a steam bath to 5 ml. and transferred to a 50 n1. Erlemeyer flask. The solution was diluted to 10 I1. and 0.15 ll. of 70-72% perchlorio acid was added. The solution was evaporated to dryness on a steam bath and then taken to {has or perehlorie acid to remove any bromine. TheresiduewastakenupinIO. n1. otwater andlnl. ofazoi sodiu- sultite solution was added to reduce the thalliu- to the thallous state. The solutions were boiled for one ninute and then cooled to room teaperature. The initial pH was adjusted against the color buffer and one drop of a 0.23 solution of sodium 2,6-dichlorobensenoneindophenol was added. The solution was titrated with standard chronate as described before. The results are smarised below. Sample Mg. 1'1 Mg 1’1 No . Present Found 1 h7.h1 h8.0 2 h7.h1 h9.h 1&3 B.‘With.nechanical shaker. A 5 nl. sample of thallous nitrate solution‘which contained h7.bl mg. of thallium was pipetted into a 300 m1. iodine flask. Iron was introduced by the addition.ot 1 ml. of a ferric nitrate solution which contained 10 ng. of iron per milliliter. Bromine water was added until the color of free bromine was observed. After the oxidation, 18.5 ml. of a 3&3 hydrobromic acid solution was added and the volume of the solution was made up to 50 ml. The extraction was carried out by the addition of 50 ml. or purified isopropyl ether to the water solution. The flask was tightly stoppered and shaken fer 30 minutes by means of a mechanical shaker. The flask was then removed from the shaker and the contents placed in a 250 ml. separatory funnel. The two layers were separated. The water layer was replaced in the iodine flask and three more extractions were made upon that particular solution. The ether ex- tracts were combined and evaporated.over a steam bath to 5 ml. of solu- tion. The sasples were then treated identically'as those which underc want continuous extractions . The results of these extractions are given bolas . Sarnple Mg. '1‘]. Mg. Tl No . Present Found 1 h7 .h1 I49 .1 2 In .hl 50 .5 PM the above results, it can easily be seen that this method was not successful. The high results are probably caused by iron being extracted along with the thallium . DISCUSSION AND COLCHJSION DISCUSSION AND CONCLUSION The titration of thallous ion with potassium chromate proved success- ful.when sodium 2,6—dichlorobenzenoneindophenol.was used as indicator. The error was less than 8 parts per thousand when titrating cpproximately 20 to 190 mg. of thallium. .A comparison of the proposed chromate method with the Hehrotra (10,11) method was made and the chromate method was found to be superior. Contrary to the proposal of Fricke and Sonnet (h) which stated that sodium 2,6-dichlorobensenoneindophenol functioned as a,pH and an adsorpo tion indicator in the titration of 1ead.sith.potassium chromate, sodium 2,6-dichlorobenzenoneindophenol functioned only as a.pH indicator in the titration of thallous ion.with potassium chromate. Several other substi- tuted indophenols were also tried, but they proved unsuccessful. Experiments were carried out in which.sethyl red was substituted as the indicator instead of sodium 2,6—dichlorobensenoneindophenol. A color change was Observed, but it did not occur at the equivalence point. How- ever, it is believed that methyl red could be used as the indicator’by adjusting the initial conditions. Common.interferences of the preposed.method were noted and studied. An attempt was made to separate thallium from iron'by extraction from a hydrcbromic acid solution with isOpropyl ether. The method used was not satisfactory. However, further investigation of this separation might yield a successful procedure. 145 Several additional experiments were carried out using various indicators and dyes as adsorption indicators in the titration of thallous ion with potassium iodide or potassium bromide. The results of these experiments were completely negative. LITERATURE CITED Lumen}; omen 1. Ashley, 8. E. Q., and Murray, W. M. Jr., Ind. Eng. Chem., Anal. Ed., 19,, 367. (1938). 2. Berg, 3., and raru-enkmp, E. 3., z. anal. Chem., 9:92, 305, (1937).' 3. Clark, W. M. , 'The Determination of Hydrogen Ions ," Williams and Wilkins Company, Baltimore, Md., 1920, p. 75. 1.. Fricke, 3., and Sammet, 3., z. anal. Chem., £9, 13, (19M). 5.1mm, 1;. 1.. J. Am. Chem. 309.. 22, 300. (1907). 6. Hillebrand, W. F., and Lundell, G. E. F., ”Applied Inorganic Analysis ," John Wiley and Sons, Inc., New Iork, 1929, p. 371-8. 7. 22.19." p. 588. 591. a. Marshall, 8.. J. Soc. Chem. Ind., 2.2, 991;, (1900). 9. McBryde, w. a. 22., and Ice, J. 3., Anal. Chem, 39, 109s,(191.8) . 10. Hehrotra, R. 0., Anal. Chim. Acta, 2, 73-7 (1910). 11. 139.. pp. 78~82. 12. Hehrotra, a. c., Anal. Chim. Acta, 3;, 36, (191.8). 13. Oesper, R. 12.,- "Newer Methods of Volumetric Chemical Analysis ," D. Van Nostrand Co., Inc., New Iork, 1938, p. 203. 111. Schulek, E., and Roses, P., 2. anal. Chem., 112, 185, (1939). 15. Schulek, E., and Somogi, 2., Z. anal. Chem, l_2_8_, 398, (19118). 16. Seidsll, 1., I'S‘aolubilities of Inorganic and Metal Organic Compounde," D. Van Nostrand Co., Inc., New Iork, 3rd Edition, Vol. I, p. 1552. 17. Nada, 1., and Ishii, 11., Bull. Inst. Phys. Chem. Research (Tokyo). 3:2: 26h.7l‘3 (1931‘). CO A.) fl, 3331‘, (1932-1). 18. Willard, H. H., and Diehl, 11., “Advanced Quantitative Analysis," D. Van Hostrand Co., Inc., New Iork, 19311, p. 2117. ._ MICHIGQN STQTE UNIV. LIBRQRIES 1 1 1‘ 41293011080383 ll Hlllllii 11111111111 11 1M 1 I