A AAAAAA AA A AAAAAAAAAAA. A: AAA-AAA. AAA AAA AAAAAAAN AA AA AAA E‘ZAAAEA {AA {AA Sagas A? AA. 3.. .A’AEC'A “‘33.! ”iii $3; .53 IQJG‘? JA§§=1§uii1e AAA-AA This is to certiig that the thesis entitled "A Study of a Volumetric Method for the Determination of Thorium". presented In] Robert M. Paine has been accepted towards fulfillment of the requirements for M.S. degree inJhfilDifim (Analytical) I :5.” imp/V Majur III‘OfC‘SEj, 1m“? January 2, 1950 A STUDY OF A VOLUMETRIC METHOD FOR THE DETERMINATION OF THORIUM BY Robert M. Paine \ A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry l9h9 Appreciation is gratefully extended to Doctor Elmer Leininger under whose kind and efficient supervison this work was effected. 959399131. INTRODUCTION With the deve10pment of a relatively simple method for the preparation of disodium.dihydrogen hypOphosphate (l) and the simul- taneous develOpment of an excellent, rapid, volumetric method for its determination (2), it was thought that the use of hypOphosphate might be adaptable to a volumetric determination of thorium. This element forms an insoluble precipitate with hyp0phosphate, which is even rela- tively insoluble in strongly acid solutions. Two possible procedures were studied: (1) the precipitation of thorium with a standard solu- tion of disodium.dihydr0gen hypOphosphate in a volumetric flask with the subsequent determination of the filtered excess reagent after bringing to a standard volume in the flask; and (2) the direct oxida- tion of the washed precipitate of thorium hypOphosphate according to Chulski's method for hypOphosphate alone. Interest was mainly centered around the nature of the precipitate, its composition, filterability, solubility in strong hydrochloric acid solutions, and its ability to be oxidized by hot potassium dichromate-sulfuric acid solution. The need for such a method and its possible advantages will become apparent in the next few paragraphs. Moeller, Schweitzer, and Starr (6) have published a good re- view of the analytical aspects of thorium chemistry in general. It contains references to most of the gravimetric and volumetric methods for the determination of thorium, along with references to much other valuable work up to that time (19in). The titrimetric iodate procedure of Mbeller and Fritz (A), one of the more recent volumetric methods reported for thorium, is none -2- too accurate and excessive washing of the precipitate causes partial hydrolysis and dissolution of the appreciably soluble precipitate. The ordinary method requires about eight hours, but a double precipita- tion is necessary in the presence of the rare earth elements. Moeller and Schweitzer (3) have recently published a method for thorium using radioactive pyroPhOSphate. The authors obtained excellent results for small quantities of pure thorium, but the presence of very much of the yttrium earths caused high results, and a preliminary separation of the rare earths is recommended. Banks and Diehl (g) have modified the older molybdate pro- cedure by precipitating thorium as the normal molybdate with subse- quent reduction and titration of the molybdenum.that is combined with the thorium. Very large amounts of calcium cause considerable error, but small amounts do no harm. They report that thorium can'be separated ‘ from.an equal quantity of uranium by this procedure, but results are not too good. It is not a separation from the rare earths. It is also quite a lengthy procedure. All in all, few titrimetric methods for thorium have been reported. Those are either involved, highly indirect, or inaccurate. The need for an accurate and simple procedure adaptable in the presence of other elements, particularly the rare earths, should require no further emphasis. The use of hypophosphate as a precipitating reagent for thorium extends back to at least the turn of the century. Kauffmann (7), in 1899, prepared the salt, Th(P206? - ll HéO, and suggested the use of it as a means of separating thorium.from other elements. Wirth (8) used hypOphosphoric acid in the preparation of pure thorium compounds -3- from.monazite sand. He states that thorium hypoPhOSphate is amorphous, insoluble in water, and difficultly soluble in acids and alkalies; and that the hypOphosphates of the trivalent elements are readily soluble in water. Later Wirth (9) suggested the use of hypOphoSphoric acid by production directly in the solution by anodic oxidation of cupric phos- phide. According to Wirth, the use of alkali hypOphosphate (apparently in sulfuric acid medium) has the disadvantage that difficulty soluble double sulfates of the cerium.earths may be formed. He also states that a separation of thorium.from.the cerium earths can be made in al- kaline solutions with alkali hyp0pho8phate by precipitation in ammonia- cal tartrate or oxalate solutions. Similarly, Koss (10) found that thorium could.be quantitatively separated from cerium and the rare earths by means of disodium.dihydr0gen hypOphosphate by precipitation in 6% hydrochloric acid. .As little as 0.1 mg per ml of solution was determined by this method. The only interfering elements were titanium and zirconium, according to Kbss. Rosenheim.(ll) had used disodium.dihydrogen hypOphosphate for years for the determination of thorium in monazite sand. His method was as follows: "with 5% sand a 100 gram sample is taken, the sulfates obtained in the ordinary way, an aliquot portion taken, and precipitated with disodium.dihydrogen hypOphosphate. The precipitate is dissolved in sulfuric acid, changed to the chloride, and precipitated with oxalic acid. 'The oxalate is ignited to the oxide for weighing. ’In rough.work the thorium precipitate with hypOphosphate may be ignited to thorium pyrqphosphate, but low results are Obtained." This is contrary to the work of later investigators who obtained high results upon igniting thorium hypOphOSPhate. Hecht and Korner (13) found that the precipitate -A- did not appear to be changed completely into the perphosphate upon ignition, as high results were Obtained. They furthermore assumed that this was apparently due to occlusion of uranium and other ele- ments. Later Hecht (lb) determined thorium accurately after precipi- tation as the hypophosphate by fusing the precipitate with sodium potassium carbonate, leaching with water, filtering and igniting the insoluble residue, fusing once more with potassium.pyrosulfate, dis- solving this melt in dilute hydrochloric acid, precipitating the thorium as oxalate, and finally igniting to the oxide. According to Hecht, thorium.is precipitated completely in a solution containing 10% hydrochloric acid by volume by means of hyp0phosphate. The pre- cipitate is practically insoluble in water, acids, and alkali hydroxide solutions. The elements most likely to precipitate with the thorium are zirconium, hafnium, cerium, and titanium. It is shown that there is little danger with reSpect to cerium.because hot hydrochloric acid reduces ceric to cerous. It was not yet clear whether there was any interference by zirconium or hafnium. Hecht stated that thorium hypo- phosphate is difficultly ignited to the perphosphate, hard or impos- sible to ignite to constant weight, and results are high even after prolonged ignition. Spitzin (12) studied the solubility of the insoluble and difficultly soluble thorium compounds. He found the solubility of ThP2O6 ' ll H20 to be as follows: in H2301+ - 53 mg of Tho2 per liter at 25° c l N l N HCl 2h 1 N l N HNO - l2 K2033 250 -5- With these things in mind it was thought that a possible rapid volumetric method could be worked out based on the precipitate of thorium.hyp0phosphate. It appeared that thorium.could be precipi- tated in the presence of many of the commonly interfering ions. -6- EXPERIMENTAL The basic method for the determination of hypOphosphate by oxidation with standard potassium.dichromate solution (according to Chulski (2) was checked as follows. Preparation of HypophOSphate: Disodium dihydrogen hypOphos- phate was prepared by the oxidation of red phosphorus with sodium chlorite solution according to Chulski (1). An accurately weighed amount of the air-dried precipitate (Na232P206 ° 6 H20) was dissolved in water and the solution made up to a liter in order that the final solution be 0.05flwith reSpect to the oxidation of the hypOphosphate to orthOphOSphate (7.85h0 grams per liter). Preparation of Standard Potassium Dichromate solution: Primary standard potassium dichromate was ground to a fine crystalline powder in a mortar. The crystals were dried in an oven at 11+0o C for about four hours, allowed to cool for 30 minutes in a desiéZtor, and used immediately thereafter in preparing a standard 0.15 N (for redox) solution of K20r207. Preparation of Standard Ferrous Sulfate solution: A 0.15 N solution of ferrous sulfate was prepared by dissolving approximately the correct amount of ferrous ammonium sulfate in three liters of water. The solution was made 1 M with respect to sulfuric acid to make possible fast, complete reduction in the Duke Reductor (15). This device was used to afford a constant, standard supply of ferrous sulfate solution. The solution was standardized against the standard dichromate solution. -7- Procedure: Fifty ml of a. 0.0500 N (0.0250 M) solution of disodium dihydrogen hypophosphate and 25 ml of 0.1500 N potassium di- chromate solution are pipetted into a 500 ml Erlenmeyer flask. Thirty- five m1 of concentrated sulfuric acid are added to make the solution 12 N, and the flask is heated in a boiling water bath for forty minutes. .After heating, the flask is cooled and the contents diluted to approxi- mately 250 m1. Three ml of 85% ortho phosphoric acid are added for each.50 ml of volume (or enough to equal half the volume of concentrated sulfuric acid present, whichever is greater). An excess of the standard ferrous sulfate solution (exactly 25ml) is added, and the excess back titrated with standard potassium.dichromate solution using diphenyl- amine sodium sulfonate as a redox indicator. An indicator blank of 0.05 ml is subtracted from.the amount of dichromate used. Results are shown in Table I. Table I Aliquot m1 0.157 1 N Grams NaafiéP 0 - Grams N 0 Error KéCr207 6 320 Presgn 6 Héo Nouns 6 Grams, 1 15.78 ' 0.3893 0.3893 1 0.0000 2 15.81 0.3900 + 0.0007 3 15.83 0.3905 + 0.0012 h 15.78 0.3893 * 0.0000 5 15-77 0.3891 - 0.0002 Average 0.3895 + 0-0003 Although the final average is very slightly high, the results of the method are in good agreement with one another. The precipitate, W206 - 6 320, is difficult to dry to an exact stage of hydration, 'and is therefore not intended for use mainly as a primary standard, but -8- in this case it suffices to prove the method. Preparation of Other Solutions Standard Thorium solution: An approximately 0.05 M thorium nitrate solution was prepared from.Baker's C.P. Thorium Nitrate crystals and distilled water. The resulting solution was slightly cloudy and was filtered. The thorium content of the final solution was determined by the gravimetric potassium iodate method of Bonardi (16) with a few slight changes. Procedure: Twenty-five m1 aliquots of the 0.05 M thorium nitrate solution were diluted to 75 ml and 25 ml of concentrated nitric acid were added to each. These solutions were made up in 200 m1 centri- fuge bottles. Sulfur dioxide gas was passed through the solution for a few minutes to reduce any ceric ion which might have been present. To the cooled solution a cold solution of 15 g potassium.iodate in 50 ml concentrated nitric acid and 30 ml water was added. The precipitate was allowed to stand one-half hour with frequent stirring. The solutions were centrifuged and decanted through Whatman #2 filter paper. The precipitate was washed in the bottle with 100 m1 of wash solution (2g potassium iodate in 50 ml dilute nitric acid (sp. gr. 1.2) and 200 ml water). The precipitate was again centrifuged and decanted. This cycle was completed once more with an additional 100 m1 of wash solu- tion. Finally the small amount of precipitate was washed from the filter paper with hot water into the bottle containing the main portion of pre- cipitate. The precipitate of thorium iodate was dissolved by heating with hydrochloric acid while a stream of sulfur dioxide gas passed through the solution. Thorium, zirconium, and titanium were precipi- tated with ammonium hydroxide, filtered, and washed with hot water until -9- free of iodides. The hydroxides were redissolved in 20 m1 of 6 N hydrochloric acid and diluted to an acidity of about 0.5 N. The diluted solution was brought nearly to boiling and 60 m1 of a hot, 10% oxalic acid solution was added slowly with stir- ring. The precipitate of thorium oxalate was allowed to stand at least twelve hours. It was then filtered on ashless paper and washed a few times with a solution containing ho ml concentrated hydrochloric acid and 25 grams oxalic acid per liter. The precipitate and paper were ignited in a previously weighed platinum crucible to thorium oxide. The final ignition was carried out under the full heat of a Meker burner. The results of one set of three samples were 0.30h7, 0.30u7, and 0.30h3 grams Th02 in 2h.92 m1 (calibrated 25 ml pipette) of thorium solution. This gave an average of 0.30h6 grams. In order to somewhat check the purity of the reagent used in making up this solution, samples were taken as before and precipi- tated with oxalic acid directly without any preliminary separations. Results obtained were 0.3052, 0.3066, 0.3062, and 0.3058 grams Th02 in 2h.92 ml of thorium solution. The average was 0.3060 grams. It is apparent that the standard thorium solution was comparatively free from other elements such as the rare earths. Asbestos suspension: Asbestos used in the following deter- minations was grade A long fiber, washed and ignited, asbestos treated as follows. Heavy strands were broken up as much as possible and enough water was added to make a fairly thick suspension. After shak- ing for some time, this was filtered through a Buchner funnel, washed successively with concentrated hydrochloric acid, water, concentrated nitric acid, and more water. This was then digested with 12 N sulfuric -10- acid and potassium dichromate solution for twelve hours on a steam bath. After filtering and washing thoroughly, the asbestos was finally made up into a rather heavy suspension with distilled water. A blank under the particular set of conditions applicable was obtained on each new asbestos preparation. This blank was substantial, but re- markably constant in most cases. This blank, oddly enough, did not increase in proportion to the amount of asbestos taken. Agreement between different lots of asbestos suspension was good. The following results are typical (Table II): the asbestos was digested three hours with standard dichromate solution in 12 N sulfuric acid, excess ferrous sulfate added, and the excess back titrated with more standard di- chromate. The amount of potassium dichromate consumed constitutes the blank. Table II Lot ml of asbestos Blank - ml of standard Dichromate suspension 0.1630 N 0.1553 I 20 0.15 I 20 0.15 II 20 0.16 II #0 0.22 III 30 0.1h III 30 0.1h IV 30 0.17 IV 50 0.19 IV 50 0.21 It was thought that perhaps the asbestos delayed the end point or obscured it in some way, but added asbestos without digestion gave no blank. Also, it should be pointed out that the different lots -11- of asbestos suspension did not necessarily contain the same weights of asbestos per unit volume. Many preliminary determinations were made before any satis- factory results were Obtained. The main misleading factor was that the precipitate of thorium with hypOphosphate did not contain a phos- phorus:thorium ratio of 2:1. The precipitate contains considerably more phosphorus, when precipitated under strongly acid conditions, than is represented by the formula, ThP206 ° 11 H20. Because the method depends upon the amount of phosphorus (present as hypophosphate), the first results obtained appeared too high. They were also very incon- sistent. This inconsistency was later found to be due to incomplete oxidation of the precipitate. Attempts were made to check many of the variables during these preliminary determinations. These included digestion of the precipitate, acid concentration of the wash solution, amount of wash, and conditions of precipitation, such as temperature of solutions, rate of addition of hypOphosphate solution, and amount of excess precipitant. In some cases where an estimated 10% excess of hypOphosphate was present, actually no excess was present. This was due to the fact that the precipitate of thorium hypOphosphate was actually composed of more phosphorus than even the formula, ThP206 ° 11 H20, plus 10%, represents. A 10% hydrochloric acid solution was used in most cases as the medium in which precipitation was made, according to the suggestion of Hecht (15). This method was studied chiefly as a method of deter- mining thorium without troublesome separations. Precipitation in strongly acid solution appeared to be the only answer. Not until the dried precipitate was finally analyzed for thorium and phosphorus did -12- a solution to the pr0blem become apparent. At the same time these first experiments were being carried out on this direct oxidation method, work was also being done on the indirect method which involved back titration of the excess hyp0phos- phate. The indirect method was later set aside because the direct method, although materially longer, appeared more promising. In the former method inconsistent results could not be laid to incomplete oxidation. The determination of the excess precipitant was carried out exactly according to the basic method for the determination of hyp0phosphate (see page 6). In this case precipitation had to be made in sulfuric acid medium. Chloride could not be present due to the subsequent oxidation by dichromate. The fact that thorium hypo- phosphate is more soluble in sulfuric acid than in any other acid (13) mdght be one reason for the variable results obtained. Another reason for abandoning this method in favor of the other was because of the possible danger of the c0precipitation of double alkali metal sulfates of the rare earth elements when a sulfate medium is used (9). The general method used was to precipitate the thorium hypo- phosphate directly in a volumetric flask with a known amount of standard disodium dihydrOgen hyp0phosphate. This was then cooled, brought up to the mark, shaken, and set aside for a few minutes until the precipi- tate settled. About 60 ml of the supernatent liquid was decanted through a filter paper. Exactly 50 ml of this was used in the deter- mination of hyp0phosphate. The sulfuric acid concentration of the final volume was varied between 5 and 10“. Results were very incon- sistent. Some of these are given as typical examples (Table III). -13- Table III Determination . Acid concentration grams Th02 grams Th0 of final solution Present Calculated v01. % H280g 0.1219 0.1230 0.1285 0.1331 0.12u7 0.1213 0.1327 0.1371 1/2 0.1277 -1/2 0.1318 2 0.13h3 0.1092 0.11u3 FBI-4 [\3 O\OCD\]O\\J1-P'UUI\)I-' O ONNNU‘IU‘IWWU‘IWW F’PJ The results are calculated on the basis of a phosphorus - thorium ratio of 2 to l. The precipitate is undoubtedly quite soluble in 10% sulfuric acid. This would account for the exceptionally low results obtained in the last two determinations. In general, the results using this method were high, but not as high as the results obtained when using the other, longer method. The inconsistency here might be attributed to adsorption of excess hyp0phosphate. Then too, precipitation in sulfuric acid medium may cause a difference in composi- tion of precipitate, or precipitation may be incomplete. Analysis of the Dried Precipitate: Due to the erroneous results Obtained in the preliminary work and to the fact that reviewing previous work by other authors indicated that perhaps the high values obtained upon the direct ignition of thorium hyp0phosphate were due to a precipitate containing a phosphorus - thorium ratio greater than two to one, it was decided to analyze portions of the dried precipitate for thorium and phosphorus. -1h- The precipitate was obtained under the same conditions as had been used in the previous analyses. A 10 to 20% excess, based on the formula, ThP206 ° 11 H20, of cold 0.05 M sodium dihydrogen hypo- phosphate was added drOpwise to a boiling hot solution of thorium nitrate (25 ml of 0.05 M Th(NO3)u in 170 ml of solution) containing 10% concentrated hydrochloric acid by volume and a little hydrogen peroxide. The precipitate was allowed to digest for a very few minutes and then set aside to be centrifuged with the other samples. The pre- cipitation was carried out in 200 ml centrifuge bottles which were contained in a boiling water bath. AlthOugh these bottles are pyrex, the walls are thick and breakage is very high when heated over a direct flame. The precipitates were alternately centrifuged, decanted, and washed until free of excess hyp0phosphate and hydrochloric acid. A 5% (by volume) hydrochloric acid solution was used as a wash. Due to the gelatinous nature of the precipitate, it was necessary to break up the lumps with a stirring rOd each time in order to wash thoroughly. Upon drying, horny lumps were formed which were ground to a powder in an agate mortar. The actual analysis of the thorium hyp0phosphate was carried out by dissolution of the precipitate in concentrated sulfuric and fuming nitric aCids, dilution, and precipitation of the hydroxide. This precipitate was dissolved in hydrochloric acid, and finally pre- cipitated as the ignitable thorium oxalate. Spencer (17) cites the method but does not give details. The procedure used here was Obtained from two references on the decomposition of thorium perphosphate. Cartledge (18) decomposed thorium perphosphate with concentrated sul- furic and fuming nitric acids. Carney and Campbell (19) used concentrated -15- sulfuric acid in the presence of ammonium perchlorate. Instead of precipitating the oxalate directly from the sulfate solution as is suggested by Spencer, they found that to obtain an oxalate entirely free of phosphorus, it was best to first change the sulfate to hydrox- ide, dissolve in hydrochloric acid, and precipitate the thorium as oxalate. The combined filtrates were evaporated to a reasonable volume and the phosphorus determined by precipitation first as am- monium.molybdiphosphate; then with subsequent precipitation as mag- nesium.ammonium.phosphate and ignition to perphosphate (20). Pre- cipitation directly as magnesium.ammonium.phosphate could not be made because of the high concentration of sodium sulfate present. Procedure: About 0.3 g (weighed exactly) of the dry pre- cipitate was placed in a 250 ml round-bottom.flask. Ten ml of con- centrated sulfuric acid and 5 to 10 ml of fuming nitric acid were added and the mixture was heated gently for about two hours with fre- quent shaking. The excess nitric acid was removed by boiling for two or three minutes after fumes of sulfuric acid began to appear. The flask was cooled in cold water and 75 m1 of diStilled water was added slowly. This mixture was set aside until completely clear. It was then poured slowly into a solution containing 15 grams sodium hydroxide in 125 ml of water. The flask was rinsed with distilled water several times with the washings added to the hydroxide solution. The solution was boiled and stirred for several minutes to convert thorium to hydroxide. The hydroxide was filtered on paper (Whatman 5912), using slight suction, and washed thoroughly with hot water. The precipitate and paper were transferred back to the original beaker and 5 ml of -16- concentrated hydrochloric acid were added. After a minute of stir- ring, the contents were diluted, boiled, filtered, and washed thoroughly with hot water. The filtrate was diluted to approximately 2h0 ml and made 0.5 N with reSpect to hydrochloric acid. The solution was heated to boiling and 60 m1 of a hot, 10% solution of oxalic acid were added slowly and with stirring. .After standing overnight, the solution was filtered and washed with a warm solution containing ho ml of concentrated hydrochloric acid and 25 grams of oxalic acid per liter. Precipitate and paper were ignited, and weighed as Th02. The combined filtrates were evaporated nearly to dryness in the presence of nitric acid in order to oxidize the excess oxalate. The solution was diluted and neutralized with ammonium hydroxide. This gave the necessary concentration of ammonium nitrate. The solution con- tained about 10% ammonium nitrate at this time. Ammonium molybdiphos- phate was precipitated from a hot (ho - 50° C) solution by adding 300 m1 of ammonium molybdate reagent (5 - 6% M003, 10% NHLLN03, 20% 111103). This was allowed to stand overnight. The precipitate was then filtered, washed once with cold, dilute nitric acid (1:100), and twice with cold 5% ammonium nitrate. This precipitate was dissolved in 20 ml of 1:1 ammonium hydroxide and the filter paper washed successively with 1:20 ammonium hydroxide, hot water, and 1:20 hydrochloric acid. The final volume was kept between 100 - 150 ml and was rendered acid with hydro- chloric. To the slightly acid solution 10 ml of cold magnesia mixture were added. Ammonium hydroxide was added slowly drOp by dr0p with stirring until precipitation started. After standing ten minutes, enough ammonium hydroxide was added to make the solution 10% with respect to concentrated ammonium hydroxide. The precipitate was -17- allowed to settle at least four hours. It was then filtered and washed with dilute ammonium hydroxide (1:20). The washed precipitate was dissolved with a few ml 1:9 hydrochloric acid, and magnesium ammonium phosphate was reprecipitated as before, except that only one ml of magnesia mixture was added and only enough ammonium hydroxide to make the solution 5% with respect to that reagent. The final pre- cipitate was filtered through a weighed porcelain filtering crucible. The precipitate was gradually brought to the full heat of a Meker burner and ignited to constant weight as magnesium.pyr0phosphate. Results are given in Tables IV and V. Table IV Sample % Th % P P/Th (Atomic Ratio) 1 h3.h2 13.02 2.238 ) 2 h3.76 12.87 2.203 ) 3 u3.5u 12.99 2.235 ) 2'228 h h3.75 12.98 2.228 ) Table V Sample grams Th grams P P/Th (Atomic Ratio) 1 0.1h12 ‘ 0.0m 2.195 ) 2 0.1h92 0.0h57 2.285 ) 2 2 1 3 0.1616 0.0h9h 2.286 ) ' 7 h 0.133h 0.0u13 2.320 ) Table IV shows the results obtained when all four samples were taken from a composite batch of precipitate obtained by grinding tOgether thoroughly the separate precipitates of thorium hyp0phosphate. -18- This was a good check on the method of analysis used. Because the pre- cipitate of thorium hyp0phosphate loses water continually upon exposure to the atmosphere, results obtained from separate samples (Table V) did not agree when compared by percentages. The percentages Obtained did not correSpond in either case to eleven molecules of water of hydration, but rather,closer to five or six. This does not mean, however, that the water of hydration is not equal to eleven molecules at sometime during the drying. Here the interest lies mainly in the ratio of phosphorus atoms to thorium atoms in the precipitate. A check on the concentration of the hyp0phosphate solution used in the precipitation of the thorium hyp0phosphate disclosed that the precipitation was not made with an excess of hyp0ph03phate in the first determinations (Table IV). In fact, there was an excess of thorium, based on the phosphorus:thorium.ratio Obtained. This situa- tion was corrected in the second set of determinations. There is, therefore, a difference in the two values obtained, 2.228 and 2.271. At any rate, it suffices to show that there is an excess of phosphorus in the precipitate when thorium.hyp0phosphate is precipitated under these conditions. Whether this is due to adsorption, to a variation in species (e.g., Th(HP206)h), or to some other phenomenon is not known. From.the facts of the foregoing work it appeared possible that consistent results could be Obtained, but prObably the precipi- tate had not been completely oxidized. There were very few results in the preliminary work which were high enough to correspond to the phosphoruszthorium ratio Obtained. Aware of the fact that an empirical method was the most promising at this time, a definite standard -19- procedure was set up in order to determine whether consistent results could be obtained. Acid concentrations, conditions of precipitation, volumes, etc., were maintained as closely as possible. On a set of four samples run simultaneously excellent results were obtained. Uh- fortunately, these values did not check with other sets of samplea.al- though within each set, good results were Obtained. The underlying cause for these inconsistent results was finally laid to the solubility of the precipitate (thorium hyp0phosphate) in dilute sulfuric acid, which was being used as a wash. The figures for this conclusion are given in Tables VI and VII. The basic standard method used in the following determinations is given. Variations of this method were applied from time to time. Procedure: To a hot solution which contains: 25 m1 standard thorium solution (0.05 M) 5 ml 3% hydrogen peroxide 100 ml water 25 m1 thick asbestos suspension 18 ml concentrated hydrochloric acid a 10% excess (on the basis of P:Th : 2.3) of 0.05 M sodium dihydrogen hyp0phosphate is added drOpwise from a burette. It was found that the use of asbestos suspension greatly facilitated oxidation of the precipi- tate, apparently by aiding dispersion. After settling a few minutes, the solution is centrifuged and decanted through a gooch crucible which has been fitted with an asbestos mat. The precipitate is washed with 75-100 ml portions of a 5% sulfuric acid solution, centrifuged, and decanted. The washing is continued until the wash liquor is free of hyp0phosphate and chloride. The filtrate should not give a precipi- tate with a dilute silver nitrate solution or with a solution of thorium nitrate. At this point the asbestos mat is carefully removed -20- from the crucible by means of a fine nickel spatula. The bottom of the mat is used to wipe the sides of the crucible. The mat is placed in the original bottle with the main portion of the precipitate and the inside of the crucible is rinsed twice with hot water. The total volume of solution should now be around 50 m1. Exactly 25 m1 of 0.15 N potassium dichromate solution are added, and enough concentrated sul- furic acid is added to make the solution 12 N with respect to sulfuric acid (35 ml). The mixture is heated on a boiling water bath for at least two hours with constant stirring. Individual electric stirrers (glass) were used, the Speed of which could be controlled by rheostats. Each time the precipitate is washed and before digestion the lumps are broken up as much as possible by means of a glass stirring rod. This same rod is used as an aid in decanting the supernatent liquid. A centrifuge bottle with a pouring lip would be very handy for this Operation. After the precipitate has been completely oxidized, the contents are cooled, transferred to a 500 ml Erlenmeyer flask, diluted to about 250 ml, and enough 85% ortho phosphoric acid is added to equal half the volume of concentrated sulfuric acid present. An excess of standard ferrous sulfate is added (exactly 25 ml). The excess ferrous ion is back titrated with the standard dichromate solution. Diphenyl- amine sodium sulfonate is used as an indicator. An asbestos blank, determined on a sample of asbestos, and 0.05 ml for an indicator blank are subtracted from the number of m1 of dichromate used. -21- Table VI Set Sample cone. of No. of other actual g. g. Th02 Apparent wash soln. wash- varia- Th02 calcu- Meq. wt. (HQSOA) ings tions present lated Th0 (1*) I l 5% h None 0.30h6 0.3373 0.1193 2 0-3373 0-1193 3 0.3366 0.1195 h 0-3373 0-1193 II 5 5% h (1) 0.30u6 0.3375 0.1192 6 0.3368 0.1195 III 7 5% 5 (2) 0.30h6 0.33h3 0.120h 8 0.33h9 0.1202 9 0.3356 0.1199 10 0.3363 0.1196 IV 11 5% 5 (3) 0.30h6 0.335h 0.1200 12 0-3358 0-1198 13 0.3361 0.1197 14 0.3363 0.1196 v 15 1% h None 0.30u6 0.3121 0.1176 16 0.3h10 0.1180 VI 17 1% 6 None 0.30h6 0.3376 0.1192 18 0.3376 0.1192 (1) 8% hydrochloric acid used as precipitating medium. (2) 50% excess hyp0phOSphate used in precipitation. (3) 20% excess hyp0phosphate used in precipitation. (h) Calculated on basis of phosphorus:thorium ratio of 2:1. From the data in Table VI it can'be seen that there is a tendency towards lower results as the number of washings for a deter- mination is increased. This could be attributed to a solubility loss or to a de-sorption of excess hyp0ph05phate. The former conclusion seems more feasible when note is made of the effect of a 1% sulfuric -22- acid wash. Further experiments, which will be discussed shortly, also indicate a solubility loss. It is apparent from the figures in Table VI that consistent results are possible under certain controlled conditions. In the two instances where a greater excess of hyp0phos- phate was used (Sets III and IV) it was necessary to wash the precipi- tate more than usual, causing lower results. One might expect a greater excess to cause higher results instead of lower. In Set II, where precipitation was made in 8% hydrochloric acid, results checked with the preceding set. It appeared at this time that determinations, in which the same concentration of wash solution was used the same number of times, were in fairly good agree- ment. The last two sets of results shown in Table VI were Obtained to bear out this fact. Using a 1% sulfuric acid wash caused higher results, presumably due to less dissolution of precipitate. Washing six times with 1% acid definitely points to an error due to the solu- bility of the precipitate. In order to verify this conclusion a set of determinationswas made in which each sample was washed progressively more than the one before it. Results were as eXpected, prOgressively lower as the number of washings increased. When a determination was attempted on a sample containing one-fifth the quantity of thorium dioxide in the same amount of solution (5% sulfuric acid wash), results were very low. This was due no doubt to the magnifying effect small quantities of thorium would have on the solubility error. Portions of the dried precipitate of thorium hyp0phosphate, prepared as in the analysis of the precipitate, were weighed out and placed in 500 ml Erlenmeyer flasks. To two flasks #00 m1 of hot water -23- were added, to one #00 m1 of a hot, 5% solution of sulfuric acid, and to another h00 m1 of a hot, 1% solution of sulfuric acid. All were mixed thoroughly and allowed to stand. After cooling about two hours, each was filtered through an asbestos mat contained in a porcelain crucible. The thorium content of each sample of precipitate was deter- mined in the usual manner by oxidation with dichromate solution. Diffi- culty was experienced in getting complete oxidation of the precipitate; therefore, a longer digestion period was allowed. It was by no means intended that this experiment should represent the actual solubility loss involved in a determination, for the precipitate was completely dry before washing, a fact which could have tremendOus influence on its solubility under the conditions used. As a result of this simple experiment, however, thorium hypo- phosphate appeared to be about twice as soluble in 5% sulfuric acid, as in 1% sulfuric acid. The thorium content of the water-washed samples had to be used as a standard in this case, the real composi- tion of the precipitate being more or less unknown. Results are shown in Table VII. Table VII Sample Wt. of ppt. taken % Th02 wash Treatment 1 0.266h g 68.h2 0 2 0.1973 68.69 350 3 0.2hl3 65.55 5% Hesop u 0.2608 67.00 1% Hgsop -gh- Due to the apparent solubility of the precipitate even in 1% sulfuric acid, it was decided to control the amount of wash solu- tion, and to reduce its volume as much as possible. In order to keep the volume down the precipitate was washed directly in a gooch crucible without centrifugation of each wash portion. This washing requires about two hours, but is no longer than the previous procedure of centrifugation and decantation. The precipitate is very gelatinous and slow to filter in either case, although in the first instance not very much of the actual precipitate is transferred to the crucible. An asbestos suspension, as used, aids filtration. The volume of wash solution was set at 250 ml by experimentation. This was found to re- move the chloride and excess hyp0ph03phate sufficiently. Procedure: The thorium is precipitated in the usual manner, centrifuged, and decanted through a gooch crucible which has been fit- ted with an asbestos mat. The precipitate is then mixed with a small quantity of warm.wash solution (1% sulfuric acid by volume) and the entire precipitate transferred to the crucible. The original bottle is rinsed a few times with wash solution and set aside until it is used again in the digestion of the precipitate. The precipitate is washed with approximately 10 ml portions of wash solution until a total of 250 m1 of it have been used. This volume is indicated by a mark on the filter flask. After washing, the precipitate and mat are transferred back to the original bottle in a manner previously described. The remainder of the determination is exactly the same as before. Results are given in Table VIII. -25- Table VIII Sample grams Th02 ml 0.15h8 N g. Th02 on Apparent Averages Present K'QCrZO7 basis Meq. Wt. P:Th : 2.0 of Th02 1 0.30%6 16.63 0.3399 0.1183 ) 2 16.66 0.3h05 0.1182 ) 3 16.6L 0.3h01 0.1183 ) 0.118h h 16.61 0.3395 0.1185 ) 5 16.59 0.3391 0.1186 ) 6 0.2hh3 13.35 0.2728 0.1183 ) 7 13.3h 0.2726 0.118h ) 8 13.37 0.2732 0.1181 ) 0°1183 9 13.35 0.2728 0.1183 ) The quantity of thorium.present in the solution, within certain lhmits at least, appears to have little or no effect on the results Obtained. For extremely small amounts, however, low results could be expected. The results of this last procedure, although not comprehen- sive, are excellent. This procedure shows definite promise as a com- paratively rapid control method for the determination of thorium. -26- CONCLUSIONS There is yet much to be done if this method of determining thorium is to be perfected. However, it does appear to have possi- bilities and certain advantages. More work needs to be done on the limiting quantities of thorium which can be present in a given volume of solution to give satisfactory results. In this work 0.3h and 0.2h gram quantities of thorium oxide in 200 ml of final volume were deter- mined by the last direct method. It would seem that one could deter- mine somewhat less than this, but below 0.15 gram the error might be considerable. The upper limit would be fixed by the amount of precipi- tate one could handle efficiently. The more precipitate the more difficult it would be to wash and filter. It would be highly desirable to Obtain a zero blank on the asbestos used. Anything, for that matter, which.might cause an error of even 0.05 ml of standard dichromate solution, should be carefully checked. The method does not have a particularly favorable factor (0.05 ml of 0.15 N K'2Cr207 5 0.0009 grams Th02). The acid concentration of the original solution should be satisfactory if kept between 8 - 10% hydrochloric acid by volume. Wider limits might do no harm, however. A high acid concentration is neces- sary if separations from troublesome elements are to be made. This brings up another prOblem, the question of separations using hypo- phosphate. It is hoped that cerium and titanium.will not interfere in the presence of hydrogen peroxide. Titanium is changed by hydrOgen peroxide to a peroxy form and ceric ion reduced to cerous. The tri- valent elements such as the rare earths do not form hyp0phosphates -27- which are insoluble in strong acid. Zirconium.and hafnium are pos- sible interferences. Although these are published facts, the method should be checked in the presence of these elements. The wash solution will most likely have to remain slightly acid. This means that there will be some solubility effect. If pure distilled water could be used, this effect would be reduced to prac- tically zero. Washing with pure water, therefore, should at least remain a possibility until tried. The rate of addition of hyp0phosphate was controlled to some extent in order to maintain a standard procedure while other variables were being checked. It is not known whether the rate of addition has any bearing on the results or not. The amount of excess hyp0phosphate should be kept at a mini- mum. This should not be too difficult, because the precipitate settles fairly rapidly when removed from the hot water bath; and a drOp of hyp0phosphate solution in a solution containing thorium forms a precipi- tate immediately. An empirical factor is obviously necessary in this method. Instead of using the normal equivalent weight of thorium oxide, 132.1, it is suggested that the apparent equivalent weight be used. From the nine determinations in the last procedure an average equivalent weight of 118.h is obtained. The apparent equivalent weight is ob- tained by dividing the grams of thorium oxide present by the number of equivalents of dichromate used in the oxidation. In the oxidation hyp0phosphate is oxidized to phOSphate, a change in oxidation number of one for each phosphorus atom. The formula, ThP206, would require an equivalent weight for thorium oxide of 132.08 (Th02/2). -28- The 1ast direct procedure is recommended for further study. While nof being a rapid method, as volumetric methods go, it is com- paratively so as far as methods for thorium are concerned. Its accuracy appears good and it has the added advantage of not requiring troublesome separations from interfering elements. (l) (2) (3) (h) (5) (6) (7) (8) (9) (10) (ll) (12) (13) (1h) (15) (l6) (17) (18) (19) (20) -29- LITERATURE CITED Leininger, E., and Chulski, T., J. Am. Chem. Soc., 7;, 2385 (19A9). Chulski, T., M. s. Thesis, Mich. State College, June, 19h7. Moeller, T., and Schweitzer, G., Anal. Chem., 20, 1201 (l9h8). Moeller, T., and Fritz, N. D., Anal. Chem., g9, 1055 (19h8). Banks, C. v., and Diehl, H., Anal. Chem., g9, 222 (19h7). Moeller, T., Schweitzer, G., and Starr, D. D., Chem. Rev., E2, 63 (19A8). Kauffman, 0., Zur Kenntnis eininger neuer Thoriumsalze. Rostock 1899. (From Mellor, J. W., ”Inorganic and Theoretical Chemistry,” Longmans, Green, & Co. Ltd. (1928). Wirth, F., 2. angew. Chem., g5, 1678-9 (1912); C. A., Q, 3379 (1912). Wirth, F., Chem. Ztg., 3 , 773-u (1913); C. A., 1, 3726 (1913). Koss, M., Chem. Ztg., 36, 686-7 (1912); C. A., I, 38 (1913). Rosenheim, A., Chem. Ztg., 3g, 821 (1912); C. A., 1, 38 (1913). Spitzin, V. 1., Russ. Phys. Chem. Soc., E2, 357-70 (1917); C- A-: ll: 3291 (1923)- Hecht, F., and Korner, E., Monateh., E2, h60—75 (1928); C. A., gg, 3861 (1928). Hecht, F., 2. Anal. Chem., 75, 28-39 (1928); C. A., gg, AA06. Duke, F. R., Ind. Eng. Chem., Anal. Ed., 11: 530 (19A5). Bonardi, J. P., U. S. Bur. Mines Bull., 212, 19 (1923). Spencer, J. F., "The Metals of the Rare Earths,” Longmans, Green, & Co. Ltd., London (1919), pp. 59-93. Cartledge, G. E., J. Am. Chem. Soc., E1, h9 (1919). Carney, R. J., and Campbell, E. D., J. Am. Chem. Soc., 36, 113k (191k). Scott, W. W., "Standard Methods of Chemical Analysis," Vol. I D. Van Nostrand Co., Inc., N. Y. (1939), p. 595. CHEMISTRY UBWY . T5451 $29054 P146 Paine 293’ 024(46 7/890 :