Extraction of gold, palladium and platinum from chloride, bromide and iodide solution with di-n-octyl sulphide (DOS) in cyclohexane

Talanta, Vol. 25, pp

163165

Pergamon

Press, 1978. Prtnted in Great Britain

EXTRACTION OF GOLD, PALLADIUM AND PLATINUM FROM CHLORIDE, BROMIDE AND IODIDE SOLUTION WITH DI-n-OCTYL SULPHIDE (DOS) IN CYCLOHEXANE M. MOJSKI Department of Analytical Chemistry, Warsaw Technical University, Noakowskiego 3, 00-664 Warsaw, Poland (Received 31 May 1977. Accepted 15 October 1977) Summary-The extraction of gold, palladium and platinum from hydrochloric acid, hydrobromic acid and iodide media by solutions of di-n-octyl sulphide in cyclohexane was examined. From distribution data it was concluded that the monosolvates AuXs.DOS and disolvates PdX,.2DOS are extracted. Extraction of platinum was efficient only from iodide solutions; a disolvate PtI,.2DOS was formed. The possibility of separation of gold and palladium from platinum by extraction from bromide or chloride solutions and simultaneous extraction of palladium and platinum from an iodide medium was demonstrated.

Alkyl sulphides are among the most promising extractants for the separation of noble metals. Complexes of noble metals with organic sulphides have been known for a long time, but it is only in the last decade that these reagents have been used as extractants. Nikolaev and co-workers1-3 have shown that gold(III) and palladium(I1) may be extracted from nitric and hydrochloric acid solutions with alkyl sulphides, R,S (where R is an alkyl group of the series C4H9-CiOHZl), in benzene and other organic solvents. Silver and mercury(H) are extracted effectively from a nitric acid medium. Lo* Extraction of platinum(IV) and with dialkyl sulphides is poor and that of iridium@) and palladium (IV) very poor. Other metals are not extracted by dialkyl sulphides. The production of organic sulphides from crude oil has facilitated their technological application as extractants and stimulated interest in this group of reagents.3.6-9 It is the purpose of the present investigation to obtain data concerning the extraction of noble metals with di-noctyl sulphide (DOS) in cyclohexane from bromide (or iodide) solutions and to compare them with those for chloride solutions.

mic acid (Pt) after removal of iodide by successive evaporation with concentrated nitric and hydrochloric acids. Other reagents used throughout this work were of analytical purity grade. Procedure

Equal volumes of the aqueous and organic solutions were shaken for 15 min for extraction of gold and 60 min for extraction of platinum and palladium. A 10m3M metal solution in the acid of known concentration was used as the aqueous phase, while a DOS solution of known concentration in cyclohexane was used as the organic phase. The phases were separated and the concentration of the metal was determined in one or both of them by spectrophotometric methods i” after evaporation and mineralization (if necessary). Gold was determined by the bromide or the Brilliant Green method, and palladium by the iodide or a-furildioxime method. Platinum was determined with stannous chloride or p-nitrosodimethylaniline. On the basis of the results obtained the distribution coefficients of the metals, defined in terms of the ratio of total metal concentration in the organic phase to total metal concentration in the aqueous phase, were calculated. All experiments were carried out at 20 + 2”.

EXPERIMENTAL

Reagents RESULTS AND DISCUSSION Di-n-octyl sulphide was prepared from n-octyl bromide and dehydrated sodium sulphide as described previously.’ The results for the extraction of gold(III), palladium(H) The crude product was distilled under reduced pressure and platinum(IV) from hydrochloric acid with a O.lM (b.p. 137-139”/0.3 mm-g, the distillate having a refractive DOS solution in cyclohexane are shown in Fig. 1. The index of 1.463 at 20” and giving an infrared spectrum extraction of gold is more efficient than that of palladium. which matched that of a standard sample. Platinum is extracted from hydrochloric acid to a negligible extent. Variation of the concentration of hydrochloric Standard lo-*M gold(II1) and platinum(IV) solutions in 4M hydrochloric acid (or 4M hydrobromic acid) were acid in the range l-8M has little effect on the efficiency obtained by dissolving the pure metals in aqua regia. The of the extraction. The distribution coefficients obtained for the extraction of palladium with O.lM DOS in cyclohexane solutions were evaporated several times with concentrated hydrochloric acid (hydrobromic acid) and the residue was from hydrobromic acid are ten times those for hydrodissolved in 4M hydrochloric acid (4M hydrobromic acid). chloric acid media but the extraction of platinum and gold Palladium(I1) solutions (lo-‘M in 4M) hydrochloric acid is similar in both media (Fig. 2.). The extraction of gold (or 4M hydrobromic acid were prepared from suitable salts diminishes with hydrobromic acid concentration. The disby dissolving them in the appropriate acid, evaporating tribution coefficients are approximately 82 and 13 for 1M to dryness and redissolving in the appropriate 4M acid. and 6M hydrobromic acid respectively. The extraction of The solutions were standardized gravimetrically with platinum from hydrobromic acid is negligible. dimethylglyoxime. It can be seen from the dependence of the distribution Palladium(H) and nlatinnm(IV) solutions (lo-‘M) in 1M coefficients of palladium and platinum on the sulphuric potassium iodide-O.iM sulphurid acid were‘prepared from acid concentration, at a constant concentration of potassuitable salts. The solutions were standardized gravimetrisium iodide of 1M (Fig. 3), that the effectiveness of the tally with dimethylglyoxime (Pd) and reduction with forextraction increases in this medium, particularly in the case 163

164

SHORT

COMMUNICATIONS

3

2 02

Ol 0"

$ 1

0

0

-1

-1

-2

I

I

1 2

I6

a CHCI.M

Fig. 1. Dependence of the distribution coefficients of gold (O), palladium (0) and platinum (A) on hydrochloric acid concentration for O.lM DOS in cyclohexane. of platinum. The distribution coefficients of platinum from an iodide medium are higher by a factor of 100 than those from chloride and bromide solutions. However, in this system, platinum can also be extracted as a sulphate complex (because of the mixed iodidesulphate medium). The dependence of the distribution coefficients of gold and platinum on the DOS equilibrium concentration in the organic phase was determined, the concentration of hydrochloric or hydrobromic acids in the aqueous phase being 2M . As mineral acids are practically non-extractable with DOS, the equilibrium concentration of the reagent was calculated from its initial concentration, taking into account the fraction bound to the metal. In both systems the slope of the plot of the logarithm of the distribution coefficient us. the logarithm of the DOS equilibrium concentration was approximately 1 and 2 for aold and palladium, respectively, at DOS- concentrations higher-than O.OlM. In the case of the iodide medium IlM iodide-O.% sulphuric acid) the slopes of the plots of log D vs. log [DOS] were approximately 2. It has been found that at a constant DOS concentration in the organic phase (0.1&f) the distribution coefficients of gold, palladium and platinum are nearly constant for initial metal concentrations in the range 5 x 10v31 x 10-4‘J4.

::_-_ 0

O-1

I

-

-2-U

I

I

I

2

4

6

1

2 CH,SO~,M

Fig. 3. Dependence of the distribution coefficients of palladium (0), and platinum (A) on acid concentration at a constant potassium iodide concentration of 1M for O.lM DOS in cyclohexane. These results suggest that gold and palladium are extracted from hydrochloric and hydrobromic acid media as the monosolvate AuX,.DOS, and disolvate PdX2.2DOS, respectively. This is in agreement with earlier data on the extraction of gold and palladium from hydrochloric acid with DOS in benzene. The results also show that palladium and platinum are extracted from an iodide medium as the disolvates. Pd12.2DOS and PtI,.2DOS. In DOS and other alkyl sulphides the bivalent sulphur atom has unoccupied d-orbitals. According to Pearson” they have the properties of a soft base. Such ligands form strong complexes with metal ions (or their complexes) which are soft acids, capable of donating d- or p-electrons with the formation of x-bonds. Hence dialkyl sulphides form stable complexes with Pd(II), Pt(II), Au(III), Cu(II), Ag(I), Au(I) and Hg(II), which are considered to be soft acids. Attempts to isolate alkyl sulphide complexes of platinum(W), which exhibits intermediate hardness, failed.” There are no data available on the interaction of hydrogen ions with dialkyl sulphides. Hence it can be concluded that the extraction of protonated complexes of Au(II1) and Pd(I1) does not take place. and that the extracted species are compounds containing metal-sulphur bonds. As in halogen acid media AuX; and PdXj- occur, the distribution coefficients should diminish with increasing acid concentration when AuXa.DOS and PdX,.ZDOS are extracted. This can be clearly observed only in the case of the extraction of gold from hydrobromic acid medium. This is probably due to the high stability constant, I& = 10s5-of the gold bromide complex,i3 compared with the value of K,-.-d = 1O5.56of the nalladium comdex.‘4 The extraction of gold from bromide solutions is, however, less efficient than that from chloride solutions. The dependence of the distribution coefficients on the acid concentration (Figs. l-3) indicates that in the extraction system hydrochloric acid-DOS in cyclohexane, the separation of gold from platinum is more effective than the separation of palladium from platinum. Separation of gold and palladium from platinum can be done more effectively from hydrobrotnic acid medium. Platinum cannot be separated from palladium by extraction with DOS from iodide media.

I

CHBr. M

Fig. 2. Dependence of the distribution coefficients of gold (Cl), palladium (0) and platinum (A) on hydrobromic acid concentration for O.lM DOS in cyclohexane

REFERENCES

1. A. V. Nikolaev, V. G. Torgov, E. N. Gil’bert, V. A. Mikhailov, V. A. Pronin, L. G. Stadnikova and I. L. Kotlyarevskii, Izv. Sib. Otd. Akad. Nauk SSSR, Ser. Khim. Nauk, 1967, (6), 120.

SHORT

2. V. G. Torgov, V. N. Andrievskii, E. N. Gil’bert, I. L. Kotlyarevskii, V. A. Mikhailov, A. V. Nikolaev, V. N. Pronin and D. D. Trotsenko, ibid., 1969, (5), 148. 3. A. V. Nikolaev, V. G. Torgov, V. A. Mikhailov, V. N. Andrievskii, K. A. Bakovets, M. F. Bondarenko, E. N. Gilbert, I. L. Kotlyarevskii, G. A. Mardezhova and S. S. Shatskava. ibid.. 1970. (4). 54. V. A. Mikha%v, S. .S. Shatskaya, D. D. Bogdanova and V. G. Torgov, ibid., 1976, (3), 34. V. A. Mikhailov. N. A. Korol and D. D. Bogdanova, ibid., 1975, (6), 29. 0. S. Shelovnikova, A. V. Nikolaev and R. J. Novoselov, ibid., 1976, (3), 44. M. F. Bondarenko, N. K. Lyapina, Yu. E. Nikitin, A. Kh. Sharipov, L. M. Zagryatskaya, V. S. Nikitina and M. A. Pais, ibid., 1973, (3), 16.

Tafanta, Vol. 25, pp. 161-166.

Pergamon

EXTRACTION

Press, 1978. Pnnted

165

COMMUNICATIONS

8. D. M. Chizhikov, B. P. Kreingauz and G. M. Denisova, ibid., 1970, (4), 120. 9. P. A. Lewis, D. F. C. Morris, E. L. Short and D. N. Waters, J. Less-Common Metals, 1976, 45, 193. 10. Z. Marczenko, Spectrophotometric Determination of Elements, Horwood, Chichester, 1976. 11. R. G. Pearson, J. Am. Chem. Sot., 1963, 85, 3533. 12. S. E. Livingstone, Quart. Rev., 1965, 19, 386. 13. A. G. Sillen and A. E. Martell, Stability Constants of Metal-Ion Complexes, The Chemical Society, London, 1964 14. L. V. Companys, X. T. Morer and A. T. Carbonell, Ajnidad 1974, 31, 559.

in Great Britain

AND SPECTROPHOTOMETRIC DETERMINATION COPPER WITH THIOBENZOYLACETONE (3-MERCAPTO-l-PHENYLBUT-2-EN-l-ONE)

OF

M. V. R. MURTI and S. M. KHOPKAR Department of Chemistry, Indian Institute of Technology, Bombay-400 076, India (Received 11 March 1977. Revised 4 October 1977. Accepted 20 October 1977)

Summary-The copper(H)-thiobenzoylacetone complex, extracted into benzene, is used for photometric determination of copper at 410nm after excess of reagent has been removed by scrubbing with a buffer at pH12. The extraction is quantitative over a broad pH range (3.5-9.5). The method is highly sensitive for the determination of copper.

The spectrophotometric determination of copper with l,l,l-trifluoro-3-(2-thenoylacetone) has been recently reported by Akaiwa et al.’ They claim advantages of this method over previous methods for solvent extraction of copper with various B-diketonesz4 It is now shown that the solvent extraction of copper with thiobenzoylacetone (HSBA) followed by removal of the excess of reagent by scrubbing with a buffer solution of pH 12.0, without decomposing the Cu(SBA), complex in benzene as a diluent, provides a more selective and fairly sensitive method.

General procedure

Transfer a sample solution containing l-20 pg of copper to a separatory funnel. Add 10ml of pH 6.0 buffer, dilute with distilled water to 25ml and shake the mixture with 10ml of 10e3M thiobenzoylacetone in benzene for about 10min. After settling, run off the aqueous phase, then shake the greenish yellow organic phase twice (for 2 min) with 10ml of pH 12 buffer to remove excess of reagent. Measure the absorbance of the copper complex at 410nm against a reagent blank prepared similarly. Compute the amount of copper from the calibration curve.

EXPERIMENTAL

A stock solution of copper sulphate was prepared by dissolving 3.935 g of the analytical grade reagent in 500 ml of water containing 1% sulphuric acid, and standardized complexometrically with EDTA, PAN being used as indicator. Solutions of lower concentration were prepared by appropriate dilution. Acetate buffer of pH 6.0 was prepared by dissolving 16g of sodium acetate in 1 litre of distilled water and acidifying with acetic acid to pH 6.0. For scrubbing, buffer solution of pH 12.0 was prepared by mixing 10 ml of 0.05M borax with 12.6 ml of O.lM sodium hydroxide. Thiobenzoylacetone was prepared from benzoylacetone as described earlier.’ A 10-3M solution in benzene was used. The reagent is preferably preserved in the refrigerator.

RESULTS

AND

DISCUSSION

E@ct of organic solvent

Various organic solvents such as benzene, toluene, xylene, chloroform, carbon tetrachloride, n-butanol, amyl alcohol and amyl acetate were tried. The absorbance of the complex was highest when benzene was used as the solvent. Absorption characteristics

Figure 1 shows the absorption spectrum of the copperthiobenzoylacetone complex measured against the reagent blank obtained by the procedure described, showing almost complete removal of residual HSBA from the organic phase. The spectrum of the reagent blank measured against benzene is also shown. The complex has