Spectrophotometric determination of tungsten with thiocyanate

760

SHORT COMMUNICATIONS

Summary--Two new, simple, rapid, and accurate iodometric amplification methods are described for the micro and submicro determination of hydrazine. The first depends on oxidation with a chloroform solution of iodine and removal of its excess, oxidation of the resulting iodide with bromine, and iodometric titration of the liberated iodate. The second method is based on oxidation with periodate at pH 8, masking of the excess of periodate with molybdate at pH 3, and iodometric titration of the iodate. The order of amplification involved in the two methods is 6- and 3-fold, respectively. Micro amounts of hydrazine sulphate and dihydrochloride were determined satisfactorily by both methods, the average recoveries being 98"6 and 99"4~o.

Talanta, Vol. 22, pp. 760-762. Pergamon Press, 1975. Printed in Great Britain

SPECTROPHOTOMETRIC DETERMINATION OF T U N G S T E N WITH THIOCYANATE V. YATIRAJAM a n d SUDERSHAN DHAMIJA Department of Chemistry, Kurukshetra University, Kurukshetra 132119, Haryana, India

(Received 21 August 1974. Revised 30 December 1974. Accepted 28 January 1975) Most spectrophotometric methods ta for tungsten are based on its complexes with organic reagents having hydroxy groups, are only moderately sensitive and are subject to interference from several elements, including molybdenum, vanadium, chromium, iron, nickel and cobalt. The dithiol method uses a very high acidity, is moderately sensitive with a short Beer's-law range, suffers interference from molybdenum and other alloying elements and is also timeconsuming. The most commonly used tungsten(V)--thiocyanate method 2 is much more sensitive, and free from interference from iron and from an equal amount of molybdenum, though not from other coloured ions. Stannous chloride is used as the reductant in highly acid medium and extraction with organic solvents is avoided to keep down interferences. Titanous chloride a has also been proposed as reductant, but with no decisive advantage. In the following method, mercury metal is used for reduction of W(VI) to W(V) in the presence of thiocyanate, followed by extraction of the yellow tungsten(V)-thiocyanate •complex with a tertiary amine, offering many advantages. EXPERIMENTAL

Reagents Tungsten solutions. Stock solutions (mg/ml level) were prepared by dissolving sodium tungstate and standardized by the oxine method. 4 Working solutions were made by suitable dilution. Tribenzylamine solution (TBA), 2 ~ w/v in distilled chloroform.

Potassium thiocyanate solution, 5M. Mercury. Purified with nitric acid. 5 Procedure A solution containing not more than 600/~g of tungsten was placed in a 100-ml separating funnel, and adjusted to be 0"2M in potassium thiocyanate and 4M in hydrochloric acid in a total volume of 25 ml. Then 2 ml of mercury were added and the funnel was shaken vigorously for 1 min. The yellow thiocyanate complex formed was extracted by shaking for 1 rain with 20 ml of TBA solution. The mercury was run off and the solvent layer was transferred to a 25-ml volumetric flask and made up to volume with TBA solution. The absorbance of the solution at 410 nm was measured in a l-cm cell.

Modification for vanadium and titanium. When vanadium and/or titanium were present, the yellow solvent layer was transferred to another separating funnel and scrubbed with 20 ml of 7M hydrochloric acid for 1 min. The organic phase was transferred to a 25-ml flask and the absorbance measured as before. Modification for iron. With up to 100 mg of iron present, the reduction step was extended to 3 rain of shaking. High-speed steel. The sample (ffl g) was dissolved in 10 ml of perchloric acid (1 + 1), with heating. Concentrated nitric acid was added dropwise till all carbides had dissolved. Concentrated hydrochloric acid (1 ml) was then added and the solution evaporated to a paste which was taken up in 50 ml of water containing 5 ml of concentrated hydrochloric acid and 2 g of tartaric acid. The solution was made up to 100 ml in a standard flask and 1 or 2 ml were used for determination of tungsten. Ferrotunasten. The sample (0.1 g) Was carefully fused with sodium peroxide (2 g) in a nickel crucible, tb The cold melt was transferred with hot water to a beaker. The nickel crucible was rinsed with 13"0 ml of hydrochloric acid (1 + 1). The rinsings and tartaric acid (2 g) were added to the beaker and boiled. Any black particles were filtered off. The filter paper was dried and ignited and the residue was dissolved in concentrated nitric acid (1 ml) with heating. The nitric acid was expelled by three successive evaporations each with 1 ml of concentrated hydrochloric acid, until a paste was left. The latter was dissolved in 20 ml of water containing 2 ml of concentrated hydrochloric acid and 1 g of tartaric acid. The solution was added to the main solution, which was finally accurately made up to 250 ml. Aliquots of i ml were taken for determination of tungsten. RESULTS A N D D I S C U S S I O N

The use of stannous chloride for reduction of W(VI) to W(V) requires very high acid and chloride concentrations to avoid formation of tungsten blue and needs a 20-min waiting period. Tungsten concentrations > 15 #g/ml cause turbidity? Titanous chloride can be used in media of moderate acidity with practically no advantage other than some improvement in the Beer's-law range. The colour of the excess of titanous chloride requires compensation in the blank. 3 In both cases, extraction increases the interference from other elements. Mercury reduces tungsten to the quinquevalent state in the presence of thiocyanate and

SHORT COMMUNICATIONS % TBA i ' 0

:~

'MKSCN f 0-4 0.6 I I

0.2 !

4! 0-8 I

Table 1. Influence of anions on absorbance of tungsten extract (W 600 pg/25 ml)

,C B 2.0

Salt

2.0

f o

lt

' e--C

-'2

/a,.zx--~ - - , ,

a

0

..~e

o ,,it

1.5

!

/ I

c o J~

I I l I I

Jm <[

--I'O I 0

I I 30 60

I

2 I 120

I [ 4 6 IMHCl I I 150 240 :500 ,,tee

A

I D

Fig. 1. Dependence of W(V)-thiocyanate complex formation on various parameters. (Curves and scales indicated by the same letters). Tungsten concentration 24 #g/ml. A--[HC1], B - - [ K S C N ] , C - - [ a m i n e ] , D - - t i m e of shaking. acid very rapidly and no waiting time is required. As the reductant is not present in solution, the absorbance of the complex in aqueous solution decreases with time, but very slowly. The influence of various parameters on the absorbance of tungsten(V) thiocyanate is shown in Fig. 1. The reduction starts at an acidity of 2M hydrochloric acid and increases to a maximum at 4M., then decreases very slightly above 5M (curve A). The colour extracted at up to 3M acidity is unstable (shown by dotted line). Reduction increases with concentration of potassium thiocyanate up to 390 I

2"0

None Disodium salt of EDTA Sodium chloride Sodium sulphate Sodium acetate Tartaric acid Trisodium phosphate Potassium citrate Sodium oxalate

-1.0 2.0? 2-0t 3-0 2'0 2"0¢ 3.0 2.0

Absorbance 1-79 1.79 1.79 1.78 1.75 1.75 1.65 1.64 1-25

0'2M, remaining nearly constant thereafter (curve B). Extraction increases with the concentration of TBA up to 2~o and is found to remain constant up to 4~o (curve C). It also increases with the length of the reduction step, remains constant for 1-3 min shaking-time, but decreases on longer shaking (curve D). Only oxygenated solvents6-8 have been proposed for the extraction of the tungsten(V)--thiocyanate complex. In the present system, the complex can be extracted into isoamyl alcohol or methyl isobutyl ketone but is not stable, probably owing to the absence of reductant in the solvent phase. However, we have found that tertiary amines (Fig. 2) also extract the complex, which is stable in them even though the reductant is not extracted. The complex has ~a~ at 400 nm in aqueous solution and also in tri-nbutyl, n-hexyl and iso-octyl amines, and at 410 nm in TBA and n-octylamine. The colour in tri-n-butylamine is unstable, but stable in the other amines for 2 hr and probably longer. The absorbance increases with the number of carbon atoms (up to six) in the amine and then remains constant. The branched-chain amines give higher absorbances than the unbranched ones. The absorbance in TBA is about 50~ higher than that in aqueous solution, slightly better than that in tetraphenylarsonium chloride and 5 ~

420 E-G I

Table 2. Extraction of other elements under conditions of the method

A

Concn. in aqueous phase,

1"5

~ I'0 /x

0'5

0

Amount*

* g/25 ml of aqueous solution, added after forming the complex. "~added before forming the complex.

¢ ~'0

761

I

350

I

400

450 Wavelength, nm

500 A-4

Fig. 2. Absorption spectra of W(V)-thiocyanate complex in various solvents. (Curves and scales indicated by same letters). Tungsten concentration = 24 / ~ m l . A - - a q u e o u s phase, B - - 2 ~ w/v TBA, C--tetraphenylarsonium chloride, D - - l ~ o v/v n-hexylamine, E - - l ~ o v/v tri-n-butylamine, F - - 1~ v/v tri-iso-octylamine, G - - 1~ v/v tri-n-octylamine.

Element

/ag/ml

None Sb(III) Co(II) Ni(II) Ce(IV) AI(III) Pb(II) Pd(II) Mn(II) U(VI) Fe(III) Sn(II) Bi(III) Cr(III) Zr(IV) Ti(IV) V(V) Pt(IV) Mo(VI)

-400 400 400 400 400 400 240 400 400 400 400 400 400 400 400 400 200 40

Absorbance* Absorbancet - 0.004 0.000 0.002 0.002 0.004 -0.001 - 0.002 -0.002 - 0-004 - 0.005 - 0.006 - 0.008 -0.012 -0'017 -0.025 0.068 0.072 0'057 0"055

* Measured against 29/0 TBA in CHCl3. t After a single scrub with 7M HCI.

------------0.002 0'012 0.007 0.002 0.002 ---

762

SHORT C O M M U N I C A T I O N S

Table 3. Analysis of synthetic samples by the proposed method Sample composition* Fe(3400).Ni(350),Cr(1000),Mn(25) Fe(7600) Fe(3420),Co(2100),Cr(180) Fe( 1500);Cr(120).V(6) Fe(95),M n(72),Sn(9).Bi(1.7)

W added. g#

W found, /~0

200 400 300 360 500

199 402 301 358 498

* These samples are analogous to Midvale HR, tungsten steel, K.S. Magnet steel, high-speed steel, and Spanish wolframite respectively. Figures in brackets are the number of/zg of the element present in the aliquot analysed. less than in the most efficiently extracting amines. Therefore, TBA is chosen as it is also much cheaper and easily recoverable.9 About 99.5% of the tungsten is removed in a single extraction. Beer's law is obeyed up to 24 /zg of tungsten per ml in the final solution.

Effect of diverse ions Large amounts of chloride, sulphate, EDTA, tartrate and acetate do not decrease the absorhance, or do so only very slightly (Table 1). Phosphate, citrate and oxalate in large amounts decrease the absorbance in that order. Nitrate should not be present. Fluoride even in small amounts suppresses the extraction. Uranium, titanium, vanadium, chromium, iron, cobalt, nickel, manganese, aluminium, lead, tin, bismuth, palladium and antimony do not interfere in the method, if any necessary modifi~tions arc made, such as scrubbing with 7M hydrochloric acid, or compensation in the blank (Table 2). Platinum and molybdenum can be tolerated in amounts equal to that of tungsten, with errors of up to 0.4 and 2% respectively, but in larger amounts should be separated. Copper, in concentrations of more than a few #g/ml, is precipitated, but the interference can be avoided by filtration of the solvcnt layer. When arsenic is present, the mercury does not collect nearly together, and the solvent phase requires filtration.

Applications With better sensitivity and wider Beer's-law range than the existing thiocyanate methods, the present method takes less than 8 min for a single determination. With suitable cells and standard curves the method can be used for determination of a wide range of tungsten concentrations with an error of around 0,5% and good reproducibility. The usefulness and wide applicability of the method is shown by satisfactory analysis of several synthetic samples (Table 3) analogous to industrial products. Analysis of BCS 241/1 high-speed steel (19.61% W) gave 19.5 and 19"5%W, and analysis of two ferrotungsten samples gave 75.0 and 75"3% for 75'2%W, and 73"4 and 73'5% for 73'3%W.

Acknowledgements-- The authors wish to express their sincere thanks to Prof. S. M. Mukherji, Head of the Chemistry Department, for facilities. S. D. is grateful to Kurukshetra University for a research fellowship.

REFERENCES

l . W. T. Elwell and D. F: Wood, Analytical Chemistry of Molybdenum and Tunosten, (a) pp. 103-130; (b) p. 55. Pergamon, Oxford, 1971. 2. E. B. Sandell, Colorimetric Determination of Traces of Metals, 3rd Ed., p. 887. Intcrscience, New York, 1959. 3. D. N. Finkelshtein, Zavodsk. Lab., 1956, 22, 911. 4. K. Kodama, Methods of Quantitative Inoroanic Analysis, p. 424, Interscience, New York, 1963. 5. A. I. Vogel, A Text Book of Quantative Inoroanic Analysis, 3rd Ed., p. 1056. Longmans, London, 1961. 6. K. V. TroItskil, Primenenie Meehenykh Atomov v Anal.

Khim. Akad. Nauk SSSR Inst. Geokhim. i Anal. Khim, 1955, 133. 7. B. Ncef and H. G. Doge, Talanta, 1967, 14, 967. 8. H. Goto and Y. Kakita, J. Japan Inst. Metal, 1961, 25. 184. 9. V. Yatirajam and J. Ram, Anal. Chim. Acta, 1972, 59. 383.

Sumnmry--The yellow W(V) thiocyanate complex is formed by shaking sodium tungstate solution in 0`2-0.8M potassium thiocyanate and 4-5M hydrochloric acid, with mercury. It is extracted with 2% tribenzylamine solution in chloroform and measured at 410 nm. U, Ti, V, Cr, Fe, Co, Ni, Mn, AI, Pb, Sn, Bi, Pd, Sb and Cu do not interfere.Pt and M o in amount equal to that of tungsten give errors of up to 0.4 and 2% respectively.The sensitivityis 0,013 ~g/ml and Beer's law is obeyed up to 24 ~g/nd.