Analyst, July, 1971, Vol. 96, pp. 475479 475 The Spectrophotometric Determination of Tungsten with Thiocyanate Part 111. Moderating the Interference by Molybdenum in the Determination of Tungsten in ‘Steel BY A. G. FOGG, T. J. JARVIS, D. R. MARRIOTT AND D. THORBURN BURNS (Department of Chemistry, Loughborough University of Technology, Loughborougk, Leicesterskire) Molybdenum in excess of 0.3 per cent. w/w interferes in the determination of tungsten in low tungsten steels (W < 0-25 per cent. w/w) with thiocyanate because of a molybdenum - iron inter-element effect. Two procedures are described which overcome this limitation. In the first procedure tungsten (VI) and molybdenum(V1) are extracted into chloroform as benzoin cc-oxime com- plexes prior to the determination of tungsten.In the second procedure iron is extracted from the tungsten and molybdenum into isobutyl methyl ketone. Up to 4 and 25 per cent. w/w, respectively, of molybdenum in the steel can be tolerated when these procedures are used. IN Part 111 a procedure in which tetraphenylarsonium thiocyanatotungstate(V) is extracted into chloroform was recommended for the determination of tungsten in steel. Precise results were obtained with this procedure in determining tungsten in a wide range of steels when either hydrochloric acid or orthophosphoric acid was used with nitric acid to dissolve the steel samples. The use of orthophosphoric acid is preferred by steel analysts, however, as its complexing action makes the hydrolysis of the tungsten less likely. One of the major advantages that the new procedure has over the existing British Standards’ procedure2 is that vanadium interference is eliminated completely.The tolerance to molybdenum, how- ever, is the same as in the British Standards’ procedure. For steel samples containing more than 0.25 per cent. of tungsten, molybdenum does not seriously interfere, as at least 3 per cent. of molybdenum in the steel can be tolerated. The extent of interference by molybdenum is largely a function of the iron-to-molybdenum ratio in the sample, and interference occurs when the sample weight is increased to obtain an acceptable sensitivity in the tungsten - thiocyanate reaction at low tungsten levels. For steel samples containing less than 0.25 per cent. of tungsten, when a large weight of sample had to be taken, only about 0.3 per cent.of molybdenum could be tolerated in the direct procedure . The present paper describes the development of modified procedures in which inter- ference by molybdenum in the determination of tungsten in low tungsten steels is greatly reduced. Several approaches to the elimination of molybdenum interference are possible, including the prior precipitation of molybdenum as its sulphide, as has h e n suggested by Bush and Higgs3 The present work has been concerned with methods that reduce the concentration of iron in the solution from which the tungsten - thiocyanate complex is extracted, vix., prior solvent extraction of the tungsten and molybdenum from the iron and prior solvent extraction of the iron from the tungsten and molybdenum.Benzoin cc-oxime has been suggested previously4 v5 as a reagent for extracting tungsten(V1) and molybdenum(V1) from iron, and this procedure, with certain modifications, has been applied satisfactorily in the present work. Luke,6 on the other hand, extracted iron(II1) from concentrated hydrochloric acid - water solution (1 + 1 v/v) into isobutyl methyl ketone in the determination of tungsten with thiocyanate; this procedure has also been adapted in the present work. 0 SAC and the authors.476 FOGG et al. : SPECTROPHOTOMETRIC DETERMINATION [Analyst, Vol. 96 METHOD 1. EXTRACTION OF TUNGSTEX AND MOLYBDENUM WITH BENZOIN ~-OXIME AS PART OF THE PROCEDURE FOR THE DETERMINATION OF TUNGSTEN I N STEEL The extraction of tungsten(V1) and molybdenum(V1) with benzoin a-oxime into chloro- form has been used previously, in conjunction with thiocyanate in the determination of tungsten in stee1.798 In these procedures the solid tungsten and molybdenum - benzoin a-oxime complexes were wet oxidised prior to the determination of tungsten.In the present investigation, however, satisfactory results were obtained without wet oxidising the benzoin cc-oxime complexes. After evaporating the chloroform extract nearly to dryness, the residue was treated directly with concentrated hydrochloric acid and tin (11) chloride. Attempts to back-extract the tungsten quantitatively from the chloroform solution into a 2 per cent. w/v solution of tin( 11) chloride in concentrated hydrochloric acid were unsuccessful. Orthophosphoric acid interfered with the extraction of the tungsten - benzoin a-oxime complex, and this precluded its use in the dissolution of steel samples.Steel samples were dissolved in a mixture of hydrochloric and nitric acids, as indicated previously (Dissolution of steels (i) in Part 11). In the procedure described below the tolerance to molybdenum is increased to 4 per cent. w/w in the steel. When more molybdenum is present the amount of benzoin a-oxime used is insufficient to extract all of the molybdenum and tungsten. This difficulty cannot be overcome by adding more reagent, as the increased amount of reagent interferes with the extraction of the tungsten(V) - thiocyanate complex. REAGENTS- Hydrochloric acid, concentrated, sp.gr. 1.16 to 1-18. Nitric acid, sp.gr. 1.42.Orthophosphoric acid, sp.gr. 1-75. Chloroform. The above reagents should be of analytical-reagent grade. Standard sodium tungstate solution, 25 pg ml-l, in concentrated hydrochloric acid-Dissolve 2.2432 g of analytical-reagent grade sodium tungstate dihydrate, Na2W0,.2H,O, in water, and dilute the resulting solution to 500 ml with water in a calibrated flask. Dilute 5.00 ml of this solution to 500 ml in a calibrated flask with concentrated hydrochloric acid. Benzoin a-oxime solution, 0.15 per cent. w / v , in chlorojorm-Dissolve 1.5 g of benzoin a-oxime in ethanol and dilute to 100ml with ethanol. Dilute 10ml of this solution to 100 ml with chloroform as required. Ammonium i r o n ( I I ) sulphate solution, 20 per cent. w/v-Dissolve 20 g of ammonium iron(I1) sulphate in water and dilute the resulting solution to 100ml with water.Prepare fresh daily. Tin(II) chloride solution, 10 per cent. w/v, in concentrated hydrochloric acid-Dissolve 20 g of tin(I1) chloride dihydrate in 180 ml of concentrated hydrochloric acid. This solution should be prepared fresh weekly. Sodium thiocyanate solution, 2 M-Dissolve 117 g of sodium thiocyanate dihydrate in water, and dilute to 500 ml with water. This solution is stable for about 3 weeks, but should be discarded if a pink colour develops. Tetraph,enylarsonium chloride solution, 0.025 m--Dissolve 1.05 g of tetraphenylarsonium chloride in 100 ml of water. This solution should be prepared fresh every 2 or 3 weeks. Chloroform containing 0.08 per cent. w/v of quinoL-Prepare a 1 per cent. w/v solution of quinol in absolute ethanol, and dilute 20ml of this solution to 250ml with chloroform.The ethanol solution is stable for about 1 week when stored in the dark, but should be dis- carded if a pink colour develops. The chloroform solution should be prepared fresh daily. Ammonium hydrogen dijuoride. Hydrochloric acid, approximately 6 M-Dhte 258 ml of concentrated hydrochloric acid to 500 ml with water. DISSOLUTION OF THE STEEL- Dissolve 0.5 g of the steel in 25 ml of concentrated hydrochloric acid and the minimum amount of nitric acid (5 to 10 drops). For more rapid dissolution warm the mixture gently. When the steel is dissolved, boil the solution for 5 minutes to remove oxides of nitrogen. Dilute the solution to 50ml in a calibrated flask with concentrated hydrochloric acid.July, 19711 OF TUNGSTEN WITH THIOCYANATE. PART I11 477 No difficulties were experienced with this dissolution procedure.If a small residue remains it should be filtered off and treated with a small amount of an orthophosphoric acid - nitric acid mixture (2 + 1, v/v). The resulting solution should then be returned to the main sample solution immediately after the solvent-extraction stage has been completed. For the routine analysis of a particular steel, the residue insoluble in hydrochloric acid - nitric acid should be examined spectrographically for tungsten. If absent, the residue in these instances can be discarded, with consequent saving in time. PROCEDURE- With a safety pipette, add 10ml of the steel solution to 35ml of water in a 100-ml separating funnel.Add 5 ml of ammonium iron(1J) sulphate solution and mix thoroughly. Extract the tungsten and molybdenum with four 10-ml portions of benzoin a-oxime solution in chloroform, then combine the extracts in a 100-rnl conical flask and evaporate the chloro- form solution nearly to dryness. Add 16 ml of concentrated hydrochloric acid and 4 ml of tin(I1) chloride solution. Simmer the solution just below its boiling-point for 5 minutes.* Cool the solution, preferably in an ice-bath. Transfer the solution to a 100-ml separating funnel fitted with a Teflon tap and use two 10-ml portions of 6 M hydrochloric acid solution to wash the solution remaining in the conical flask into the separating funnel. With a safety pipette, add 1 ml of tetraphenyl- arsonium chloride solution, swirl the solution, add 3 ml of sodium thiocyanate solution from a safety pipette, swirl the solution again, and extract the tetraphenylarsonium tungsten(V) - thiocyanate complex successively with 9, 8 and 7-ml amounts of chloroform containing quinol, adding 10 drops of tetraphenylarsonium chloride solution before each of the last two extractions.Transfer the extracts to a second 100-ml separating funnel fitted with a Teflon tap. Add 10 ml of water from a measuring cylinder and 1.5 to 2 g of ammonium hydrogen di- fluoride, and shake the mixture. Pass all of the chloroform layer through a No. 1 Whatman filter-paper directly into a 25-ml calibrated flask. Wash the aqueous fluoride layer with 1 ml of chloroform containing quinol, and pass the chloroform through the filter-paper into the flask.Wash the filter-paper with a small amount of chloroform containing quinol, and dilute the chloroform extracts to 25 ml with the same chloroform. Mix the solution thoroughly and measure the absorbance at 402 nm against chloroform by using 1-cm cells. METHOD 2. USE OF ISOBUTYL METHYL KETONE TO EXTRACT IRON It was considered that a procedure by which iron could be extracted from tungsten and molybdenum would be simpler than the benzoin a-oxime procedure. Luke6 used isobutyl methyl ketone for the extraction of iron(II1) from hydrochloric acid - water (1 + 1, vjv) solu- tion in the determination of tungsten with thiocyanate. The procedure described below was developed in the present work. By dosing a hydrochloric acid solution of British Chemical Standards steel No.322 with sodium molybdate, it was found that molybdenum in concen- trations up to 25 per cent. w/w in the steel could be tolerated. When molybdenum was present in greater concentrations the red colour of the molybdenum(V) - thiocyanate complex was observed in the measured solutions and the absorbance values obtained were high. As in the benzoin a-oxime procedure, orthophosphoric acid cannot be used to dissolve the steel in this case, because of its complexing action on the iron(III), which prevents the extraction of the iron into isobutyl methyl ketone. PROCEDURE- Prepare a concentrated hydrochloric acid solution of the steel as described in the benzoin or-oxirne method above. With a safety pipette, transfer 10 ml of this solution into a 100-ml separating funnel containing lOml of water and 20ml of isobutyl methyl ketone.Immediately shake the mixture vigorously to extract the iron and run the lower aqueous layer into a 100-ml conical flask. Extract the isobutyl methyl ketone layer with two 4-ml portions of 6 M hydrochloric acid solution, adding the aqueous layer on each occasion to the conical flask. Evaporate the aqueous layer nearly to dryness and immediately add carefully 16 ml of concentrated hydrochloric acid and 4 ml of tin(I1) chloride solution. Complete the analysis as described in the benzoin cc-oxime method from “Simmer the solution just below its boiling-point for 5 minutes. . . .” * The procedure described hereafter is also used in Method 2 below.478 FOGG et d.: SPECTROPHOTOMETRIC DETERMINATION [Analyst, VOl. 96 RESULTS The extent of molybdenum interference in the benzoin a-oxime procedure, described in Method 1, was investigated by using standard solutions of sodium tungstate and sodium molybdate; the iron was added as iron(II1) chloride. The results given in Table I were obtained under conditions simulating those for the analysis of a low-tungsten sample and indicate that up to 4 per cent. w/w of molybdenum could be tolerated in such a steel sample. Serious molybdenum interference, when it occurred, could be detected by visual examination of the chloroform extract ; when tungsten alone was present the chloroform solution was straw-coloured, whereas with a trace amount of the molybdenum(V) - thiocyanate complex present it was tinged with red.TABLE I MOLYBDENUM TOLERANCE OF THE BENZOIN CC-OXIME PROCEDURE Molybdenum, per cent. w/w, in the steel equivalent to molybdate added . . 0 1 2 3 4 5 6 * 150 pg of tungsten level with 50 mg of iron. Absorbance* . . . . . . . . 0.487 0-466 0.473 0-472 0.465 0.428 0.262 The extent of molybdenum interference in the isobutyl methyl ketone procedure, des- cribed in Method 2, was investigated by determining the tungsten in British Chemical Standards steel No. 322, by dosing the sample aliquots with various amounts of sodium molybdate. The results obtained with this procedure are compared in Table I1 with those obtained with the direct procedure given in Part 11. With the isobutyl methyl ketone procedure up to 25 per cent. w/w of molybdenum in the steel can be tolerated, whereas the direct procedure clearly gives high results with molybdenum contents above 0.5 per cent.w/w. TABLE I1 EFFECT OF ADDED MOLYBDATE ON THE DETERMINATION OF TUNGSTEN IN B.C.S. 322 STEEL BY USING THE DIRECT AND ISOBUTYL METHYL KETONE PROCEDURES Molybdenum, per cent. w/w, Apparent tungsten content of B.C.S. 322, per cent. w/w in the steel equivalent to f A \ molybdate added* Direct procedure1 Isobutyl methyl ketone procedure 0 0.5 5 10 15 20 25 30 40 50 100 0.049 0.074 0.114 0.122 0.049 0.050 0-05 1 0.052 0-053 0.062 0-053 0.054 0.057 0.062 0.070 * B.C.S. 322 contains 0.045 per cent. w/w of molybdenum. The amount of iron present remained constant; in practice, the amount of iron present would decrease as the amount of molybdenum increased and the tolerance would be somewhat greater.The results of analyses of three steels with tungsten contents of less than 0.25 per cent., and molybdenum contents of greater than 0.3 per cent., are given in Table 111. Molybdenum was expected to interfere in the determination of tungsten in these three steels when using the direct thiocyanate procedure, but this was found not to be so because of the high con- centration of other alloying elements present in these steels, which effectively decreases the amount of iron present in the aliquot taken. Nevertheless, these three steels were also analysed by using the benzoin a-oxime and the isobutyl methyl ketone procedures. The results agree well with those obtained with the direct thiocyanate procedure and with the dithiol method.July, 19711 OF TUNGSTEN WITH THIOCYANATE.PART I11 TABLE 111 COMPARISON OF THE DIRECT AND MODIFIED PROCEDURES IN THE DETERMINATION OF TUNGSTEN IN STEEL SAMPLES 479 Sample Spectrographic values, Dithiol* Direct Benzoin or-oxime Isobutyl methyl No. per cent. w/w* method procedure procedure ketone procedure I A \ Cr Ni Ti Mo W 1 1 8 8 - 2.6 t 0 . 1 0.055 0.052 0.054 0.053 2 18 8 0.5 0.5 <0*1 0.05 0-046 0.044 0.044 3 1 8 8 - 0.6 0.15 0.12 0.121 0.115 0.121 * Spectrographic and dithiol values from Firth-Brown Limited. DISCUSSION The benzoin a-oxime and isobutyl methyl ketone extraction modifications to the direct colorimetric procedure for determining tungsten with thiocyanate allow its extension to the determination of tungsten in steel samples containing amounts of molybdenum that would normally interfere.Unfortunately orthophosphoric acid interferes in both of these modi- fications to the procedure, and therefore this acid cannot be used in the dissolution of the steel. This disadvantage will detract from the use of these modifications by steel analysts who are conscious of the possibility that tungsten(V1) will hydrolyse in the absence of ortho- phosphoric acid. Orthophosphoric acid can be used to treat residues insoluble in hydro- chloric acid; the resulting solution is added after the extraction stage. Difficulties arising from the hydrolysis of tungsten(V1) were experienced in the present work only in the isobutyl methyl ketone extraction procedure when the tungsten was kept for longer than 5 minutes in concentrated hydrochloric acid - water (1 + 1 v/v) solution.Delay at the iron extraction stage should be avoided; once the concentrated hydrochloric acid is diluted in the separating funnel, the extraction of the iron and the transfer of the aqueous layer to the conical flask should be made as rapidly as possible. In many instances the direct procedure involving orthophosphoric acid dissolution can be extended to low tungsten steels containing relatively high molybdenum contents by taking a smaller sample weight than that recommended and either accepting a lower absorbance read- ing or using cells with a longer path length, or both. For example, if the recommended sample weight and aliquot volumes were taken in determining tungsten in a steel containing 0.1 per cent.of that element, then the final chloroform extract would contain 1OOpg of tungsten. An absorbance value of 0.32 would be obtained, and at least 0.3 per cent. of molybdenum could be tolerated. If one fifth of this sample weight were taken the absorbance reading would be 0.060, and 12 per cent. of molybdenum could be to1erated.l The use of cells with a longer path length would give a more acceptable absorbance value without reducing the molybdenum tolerance. The exact amount that could be tolerated would depend on the percentage of other alloying metals present in the steel, as has been shown in this paper. The presence of other alloying metals reduces the amount of iron present in the sample weight taken, and reduces the interference of molybdenum. The authors thank the British Iron and Steel Research Association for financial support and, in particular, acknowledge gratefully the experienced guidance of Mr. P. H. Scholes. They also thank Mr. B. Bagshawe of Firth-Brown Limited for helpful discussion, and for providing analysed samples. 1. 2. 3. 4. 5. 6. 7. 8. REFERENCES Fogg, A. G., Marriott, D. R., and Burns, D. T., Analyst, 1970, 95, 854. British Standard 1121 : Part 32 (1954); Amendment No. PD2916 (1957). Bush, G. H., and Higgs, 14. G., Analyst, 1955, 80, 536. Jeffery, P. G., Ibid., 1956, 81, 104. Pfeiffer, V., and Hecht, F., 2. analyt. Chem., 1960, 177, 175. Luke, C. L., Analyt. Chem., 1964, 36, 1327. Got6, H., and Kakita, Y., J . Japan Inst. Metals, Sendai, 1961, 25, 184; Analyt. Abstr., 1962, Peng, Y., and Sandell, E. B., Analytica Chim. Acta, 1963, 29, 325. NOTE-Reference 1 is to Part I1 of this series. 9, 1918. Received August 5th, 1970 Accepted March lst, 1971