J‘.‘ '9 \\ ‘23. hi 0) § 921‘ .V c. 0. U. 0.3% \. ‘Ql. 3:234 :1) I!" \d ”R *4 ‘° ‘i l. x 3.4 mmw if: ‘{ a " ‘2‘: I 9’ a. ~& 9 0b “3‘ a. " 29' '.'\L!‘:3.' «A. :3 ‘:\5‘ 538 ”.‘.:‘°\ 31:56 o .‘ 7Q? i’!‘ a. 3% i «\3 a 3 4. -s: C d “1 \3! 9V a '.\" ’ “N - NHL 3 3.. Iva. ; . 4 “we I. t 3.1.. 4MB.“ g in 01.. . J v . . . xh «K 64m” ~3\ W0. A w. u . k. 8‘ u . :1 .. . u .. s. S “I! . — ”mil 5 ‘3‘ .\. ‘ m. . 9. w. — lo‘d «~06; 3 a!" o ' < C.‘ 0.. 4.4. \3\: 1"" 3.: i, l j, / 1., :1: _, ’: f IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII ; 12 93 00572 4871 I LIBRARY Michigan State University MICHIGAN STATE UNIVERSITY EAST LANSING, MICHSGAN BISUt—LI’K’ZU Liu/ “3).01")O III» Lg? XXVI) A3 A yuIGul.G F'DEJ F 0R 1H3 CLIL“‘IK.TI‘T OP h-LEJCquX By flenry J. Hoenee, Jr. A IEEHSIS Submitted to the School of Science and Arts of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of EASTER OF oCIEKCE Department of Chemistry 1356 ACKN OW LED Gil-LII T The author wishes to express his appreciation to Dr. Kenneth G. Stone for his invaluable guidance and assistance which made this work possible. II. III. IV. TABLE OF CON flNTS PAGE IIEIRODTJC'I‘IGIEIOOOOOIO....0...OOOOOOOOOOOOOOOOIO A.‘Historical.....................o...o....... l. Knowles' method and modifications....... 2. Molybdenum species present in acid SOIUtionoeeoeooeeeeoeeeoooooo.oooooooooo Bo Statement of Problem.......o..............o 1. Attempted volumetric method............. 2. Proposed modification of gravimetric method..0.0I.OOOOOOOOOOOOCOOOOOOOOOOO... mpiLRIFVZZ-41;TALOO‘COOOICOOOOOOOOOOOOODOOOOOOOOOO. A. Determination of Composition of Complex.... 1. DetGPMination of % EOooooooooeoooooooooo 2. Postulation of a structure for complex.. B. DeveIOpment of Gravimetric Hethod.......... 1. Solubilities in acetone and ethanol..... 2. Precipitation with 10% excess reagent... 3. Effects of varying acetone concentration 4. Effects of varyin rea«ent excess....... 5. Effects of Fe(III§, Cr VI) and V(V)..... C. Application of Method to Steel Samples..... 1. General procedure....................... 2. Determinations on N.B.S. samples........ v‘5glv Eatilfllflitfriro o e o e o o o o o e o o e o o o o o o e o o e o o o o e o o o o o o o o o o LILERAI‘URE CIWOOOOOOOOIOOOOOOOOOCOOOOOOOOCI. CD 0H3 01 FJP‘ P4 10 10 10 15 17 17 18 22 25 29 32 55 36 I o INTRODUC TI ON A. Historical l. Knowles' method and modifications While working with cx-benzoin oxime to determine the interferences in its use as a precipitant for copper, Knowles (10) discovered that in acid solution the reagent would quantitatively precipitate hexavalent molybdenum. From this he developed a gravimetric method for the determination of molybdenum. The structure of the analytical reagent dfibensoin oxime is as follows: ¥ __“ “<1 >‘?“fi“*<. > x J— 0H NOH\ — Its molecular weight is 227.3 and it melts at l49-51°C. The cxime itself is insoluble in water, but soluble in alcohol, acetone and ether. The general reagent solution is composed of a one to two per cent solution in 95% ethanol. It selectively precipitates copper fran an ammoniacal solution, and molybdate and tungstate from.acid solution. Knowles' method for molybdenum involves the precipitation of the complen from acid solution followed by ignition to molybdenum trioxide in a muffle furnace at SOC-525°C. He states that the best acidity is approximately 5% sulfuric acid by volume, but up to 20% acid concentration is permissible. Hydrochloric or nitric acid may also be used when the use of sulfuric acid is inadvisable. The method also calls for precipitation from a cold solution and addition of enough bromine to tinge the solution yellow. Both of these steps are to avoid reduction of hexavalent molybdenum.by the reagent. Two to five times the theoretical amount of reagent is required for complete precipitation in this method. The theoretical amount according to Knowles is deter- mined by the relationship of one molybdenum.ion to three d-bensoin oxime molecules. However, no data or theoretical evidence is given to substantiate the postulation of this three to one complex. Knowles also discovered that the precipitate can be filtered immediately or allowed to stand for ten minutes before filtration. If the precipitate is allowed to stand in contact with the solution for one-half hour or longer, low results are obtained. The most favorable wash solution is reported to be a 1% sulfuric acid solution containing a small amount of «~bensoin oxime. Elements which cause difficulty in this precipi- tation from acid solution are columbium, silicon, palladium, tantalum and tungsten. Vanadium(V) and chromium(Vl) also interfere if not reduced with sulfurous acid before precipitation (9). Ferrous ammonium sulfate has also been suggested to reduce the vanadic and chromic acids (11). Yagoda and Fales (17) applied the method to the determination of tungsten and molybdenum. Due to the difficulty of filtration of the complex formed, they unsuccessfully sought a substitute for debenzoin oxime. They separated the molybdenum and tungsten by means of hydrogen sulfide. Since in the case of samples having a low molybdenum concentration the weight of the oxide was so small, they found a more suitable weighing form was thallous molybdate or thallous tungstate. Taylor-Austin (16) used the method with a few modifications for the determination of molybdenum in cast iron. He states that three times the theoretical amount of reagent is used as determined by the relation that one molybdenum ion is equivalent to three cx-benzoin oxime molecules, but also gives no basis for the formulation of such a complex. However this work shows that complex formation is incomplete in the presence of tartaric acid, and also that acetone may be successfully substituted as the reagent solvent. S. W. Cravens et.al. (4) have developed the method to be used for the determination of molybdenum in steels in the presence of tungsten. After the precipi- tation of the molybdenum and tungsten with x-benzoin 4 oxime the molybdenum is determined by the formation of the thiomolybdate in alkaline solution, followed by decomposition with dilute acid to the sulfide, and final ignition to the oxide. Sterling and Spuhr (15) have also devised a method for molybdenum with certain modifications. The precipitated molybdenumpcr-benzoin oxime complex is dissolved in ammonium hydroxide and hydrogen peroxide. The molybdenum is then reprecipitated from a lead acetate buffer mixture as lead molybdate. It has the advantage of providing a heavier weighing form than the oxide. Arrington and Rice (1) ignite the complex to the oxide, dissolve the oxide in sulfuric acid, reduce the molybdenum in a Jones reductor, and titrate with standard oxidant. A good review of most of these methods may be found in Flagg's, Organic Reagents, (6) and Blanco (3) has also published a review article on the determi- nations of molybdenum with.q%benzoin oxime. In all of these determinations an excess of two to five times the theoretical amount of (1-benzoin oxime is added on the basis of a three to one complex. However, there has been no work done to establish that the complex is actually three to one, and no theo- retical basis for the postulation of such a complex has been offered. 2. Molybdenum species present i2_§gid_solution. It is also interesting to note that it has not been definitely established what molybdenum species is present in 5 to 20% acid solution. Both molybdenum and tungsten are reported to form isopoly acids of various complexities depending upon the pH of the solution (12). However molybdenyl sulfate, M002334, can be prepared by dissolving molybdenum trioxide in hot concentrated sulfuric acid and cooling the solution until the sulfate precipitates (8,14). The sulfate is very hygroscOpic and is reduced by dust in the air to form a blue colored material, a molybdenum blue. Molybdenyl sulfate is reduced to a deep blue color when heated with pure metallic molybdenum. hicholls,.Saenger, and Wardlaw (13) have also shown the existence of molybdenyl sulfate in sulfuric acid solution. The precipitate formed by molybdenum with B-quinolinol in slightly acid solution has been definitely established and is employed as a suitable weighing form. The complex is bis(8-quinolinolo)- dioxomolybdenum(VI).(2). This complex of molybdenum with B-quinolinol is dried at 130°C. and weighed, while all the work done on the txhbenzoin oxime complex indicates that it is incapable of being dried and weighed as such. In fact 6 Dupuis and qual (5) ran thermal studies on the complex and found that it formed no stable material until the temperature reached about 550°C. where the complex was all converted to the oxide. This they determined to be stable to approximately 780° when sublimation began. They prepared the complex for these determinations exactly as outlined by Knowles. B. Statement of Problem 1. Attempted volumetric method The method of Knowles (10) for the determination of molybdenum with. debenzoin oxime is a gravimetric method involving ignition to the oxide which has an unfavorable gravimetric factor. The develOpment of a volumetric method using the same reagent seemed quite desirable, and an attempt was made to develop an amperometric method using a small applied potential across a pair of platinum electrodes. In acid media, from which molybdenum is quanti- tatively precipitated by cx-benzoin oxime, the oxime was found to be quite active at a pair of platinum electrodes at a potential of 400 millivolts. This and point seemed to present a definite possibility for the development of a volumetric method using the "dead—stop" technique. At this point, a review of the literature revealed that no actual determination of the compo- sition of the complex had been carried out. Therefore Knowles original assumption of a three to one complex was followed, and solutions of molybdenum in 5% sulfuric acid, and (x-benzoin oxime in 50¢ ethanol were made up accordingly. That is, a 0.0052 M molybdenum solution and a 0.0156 M tx-benzoin oxime solution were made so that one m1. of each solution would be equivalent. than twenty m1. of the molybdenum solution were diluted with 5% sulfuric acid and the cx-benzoin oxime solution was added dropwise from a burst the results were very erratic. The first drop or two gave a large galvanometer deflection. The results after this were dependent upon the rate of addition of reagent. When the reagent was added rapidly the deflection increased, and with slow addition it decreased gradually. An attempt to estimate the end-point by controlled. rate of addition of the reagent to twenty ml. of the molybdenum solution was unsuccessful. The results of plotting ml. added against the galvanometer deflection (Figure 1) indicated a break at about fourteen m1. At the time, the results were inconclusive; however, they indicated the complex might be two to one instead of three to one as Knowles originally assumed. GALVANOMETER DEFLECTION eoo #- soo - 400- f soar 200“ IOO " o ; Kl) I; 2:) 2'5 3'0 ole ML. (X-BENZOIN OXIME SOLN. F I G U R E I. l 40 2. Proposed modification g£_gravimctric method At this point, he most logical approach appeared to be a study of the composition of the complex formed. As no basis could be found for the postulation of a three to one complex it was hoped that the determi- nation of the actual c'mposition would lead to a simplified gravinetric method. A method in which the complex could be dried and weighed as such would also have a much more favorable gravimetric factor than the oxide which is approximately two-thirds molybdenum. A. Determination of Composition of Complex 1. Determination 93. 55 £3 In order to revise the existing gravimetric method for the determination of molybdenum it was first necessary to establish the composition of the complex. The method which was first attempted involved the precipitation of the complex according to tinowles method, drying and weighing the complex and then igniting the complex to molybdenum trioxide and reweighing. The first attempts at drying the complex in 1 105°C. oven resulted in failure as the white precipitate turned to a black tarry mass. A melting point determination on some of the dried material yielded the following results. At 85°C. it turned yellow; at 95°, blue; at 100°, it melted to a pale yellow liquid; and at 104°C. it decomposed into a black tarry mass. Since upon reduction Mo(VI) is reported (8) to form a material, molybdenum blue, of partially reduced molybdenum, it appeared that one of the steps in the melting process involved a reduction. As Knowles reported that the reagent will reduce some of the Mo(Vl) at room temperature when no bromine is added, the preceding method was modified. In order to avoid reduction by excess «bbenzoin oxime, only one-half as much reagent was added as was ll theoretically necessary for complete precipitation. This could be done since the main purpose was recovery of pure complex and not quantitative precipitation. Also, the method of drying the complex had to be altered, so it was dried in a vacuum desiccator over concentrated sulfuric acid. The following procedure was used. Forty ml. of 0.0052 M molybdenum solution in 5% sulfuric acid were cooled in beakers in an ice bath and then 20 ml. of 0.0156 H qkbenzoin oxime solution were added dropwise to the beaker: with constant stirring. The precip- itates were allowed to stand for one hour in solution and then filtered through filter paper and washed with water. Some of the precipitate was then transferred to previously weighed silica crucibles. The precip- itates were next dried to constant weight in a vacuum desitcator over concentrated sulfuric acid and finally ignited to the oxide and weighed. See Table I-A. In all three cases the precipitates turned a spotty blue color during the drying process, and in sample 2 the blue coloration was quite intense. Six more samples were precipitated exactly as before, the only difference being the fact that the precipitates were first allowed to dry overnight in the air before drying over concentrated sulfuric acid. Also, samples 4, 5 and 6 were treated with enough TAELE I ATTEEPTS TO ESTABLISH COMPOSITION OF COK?L3X Sample G. of Complex G. of M003 % M003 A. Original attempts I .0712 .0146 20.51 2 .0658 .0092 13.98 3 .0771 .0158 20.49 8. Effects of air drying and bromine addition 1 .0602 .0135 22.43 2 .0631 .0145 22.98 3 .0751 .0172 22.90 4 .0560 .0125 22.32 5 .0792 .0180 22.73 6 .0516 .0112 21.71 C. Further tests on the effects of air drying 1 .0587 .0126 21.39 2 .0662 .0154 23.35 3 .0550 .0130 23.64 4 .0480 .0109 22.71 5 .0662 .0158 24.17 6 .0766 .0185 25.56 13 bromine water to turn the solutions pale yellow. See Table 1-8. The precipitates after drying in air were still vlrite, however after drying over the concentrated sulfuric acid they turned pale blue. This blue color faded during weighing. In order to test the effect of air drying upon the results, six more samples were precipitated as before. Samples 1 and 2 were transferred immediately to the desiccator containing the concentrated sulfuric acid, samples 3 and 4 were first dried in the air for twelve hours, and samples 5 and 6 for seventeen hours before placing in the desiccator over sulfuric acid. They were all brought to constant weight and then ignited to the oxide. See Table 1-0. The results were so erratic on these determi- nations with each group yielding different results it began to appear that little could be done to obtain consistent results. The one remarkable feature being that in almost every case the precipitates acquired a blue coloration of varying intensity. This indicated the presence of excess reagent which was actively reducing the molybdenum. In order to remove all of the oxime which might be present in the complex, the precipitation was carried out as before, the precipitates were allowed to stand 14 for fifteen minutes, filtered and washed twice with five ml. portions of 95% ethanol. The precipitates were dried to constant weight in a 70°C. oven before ignition to the oxide. In this case there was no reduction of the molybdenum as no blue coloration appeared in the precipitate. See Table II-A. A new melting point on some of this complex from which all excess reagent had been removed revealed that the complex itself was quite stable and melted sharply at lee-9°C. Three more samples were run as above except that this time the complex was dried in a 105°C. oven to constant weight. See Table 11-8. ‘ABLE II DETERuIHATION 0F giro In COMPLEX G. of Complex G. of M005 % Mo A. Drying at 70°C. .0783 .0193 16.65 .0715 .0174 16.25 .0739 .0182 16.42 B. Drying at 105°C. .0764 .0188 16.40 .0710 .0176 16.52 .0799 .0195 16.45 15 2. Postulation gg‘g structure for complex Since the preceding results were reproducible and the complex formed appeared to be fairly pure, the original assumption of a three to one complex (10) appeared impossible. Therefore a new structure for the complex had to be postulated. There has been + + ion in evidence given for the existence of an moo2 acid solution (8,15,14) which is especially supported by the precipitation of molybdenyl sulfate from cold concentrated acid solution. The $002++ ion has also been definitely estab- lished as the species which precipitates with 8-quinolinol from acid solution (2). This lead to the theory that the complex formed with 0.23. 0.33. 1.25g. fl acetone 25 50 100 ci-benzoin oxime 0.2g. 1.9g. 13.5g. molybdenum complex .0043. .0063. .0145. 2. Precipitation with 10% excess reagent The following solutions were then prepared to be used in the development of a gravimetric method using he complex as the weighing form. (1) A 3.75075. sample of Merck Reagent Grade 3005, assay minimum 99.5335 2.1003, was. dissolved in a small amount of dilute sodium hydroxide solution. It was then neutralized with sulfuric acid and diluted with water to one liter. The resulting solution was 0.026l M, or contained 0.0500g. of molybdenum per 20 ml. of solution. (2) Crude c(-benzoin oxime from the Paragon Testing Laboratories was recrystallized twice from ethanol, and 5.92543. of this purified, dried material were dissolved in one liter of 50% acetone. The resulting solution was 0.0261 M, or two ml. of rx-benzoin oxime solution were equal to one ml. of molybdenum solution. 19 It was heped that by adding only a 10% excess of reagent and washing with small portions of 50% acetone, complete precipitation could be effected and any excess oxime could be successfully removed. This would permit a simple weighing of the complex after drying in a 105°C. oven. The major advantages which would result from this method would be (1) a large decrease in the amount of reagent required, 2} elimi- nation of the necessity of ignition to the oxide in a muffle furnace, and (3) a weighing form with a very favorable gravimetric factor, as molybdenum is only 16.51% of the total weight of the complex as compared to 66.65% when weighed as the oxide. Two samples containing 20 ml. of the standard molybdenum solution (50 mg. of M0) were pipetted into a beaker and one ml. of concentrated sulfuric acid was added to each. Forty-four ml. of the