Analyst, May, 1981, Vol. 106, pp. 529-536 529 Thiosulphate as a Complexing Agent in the Separation of Cations by Cation-exchange Chromatography Lalit C. T. Eusebius, Animesh K. Ghose and Arun K. Dey" Chemical Laboratories, University of Allahabad, Allahabad-211 002, India A batch equilibration technique has been employed using an aqueous and aqueous - alkaline thiosulphate - Dowex 5OW-XS system to separate metal ions. The formation of strong anionic complexes by Cu(II), Cd(II), Ag(1) and Pb(I1) lower their distribution coefficients ( K ) appreciably and they there- fore remain in solution while other cations are taken up by the cation exchanger. A critical analysis of the K values indicated the conditions under which a number of binary and ternary mixtures could be resolved by column chromatography.In 0.06-0.10 M sodium thiosulphate solutions, the following binary separations have been achieved: Cu(II), Cd(II), Ag(1) or Pb(I1) from Mn(II), Fe(II), Co(II), Ni(II), Zn(II), Mg(II), Ca(II), Sr(II), Ba(II), La(II1) or Ce(II1) ; and in 0.28 M sodium thiosulphate solution Sr(II), Ba(II), La(II1) or Ce(II1) were separated from other metal ions studied. The following ternary mixtures have also been resolved: Cu(II), Cd(II), Ag(1) or Pb(I1) from Mn(II), Fe(II), Co(II), Ni(II), Zn(II), Mg(I1) or Ca(I1) from Sr(II), Ba(II), La(II1) or Ce(II1). In all of the separations favourable column kinetics have been observed. The distribution coefficients, elution characteristics of metal ions, elution curves and statistical analyses of the results of the resolution of artificial binary and ternary mixtures are reported.Keywords : Thiosulphate ; cation complexes ; cation-exchange chromatography Anion-exchange studies involving the analytical application of sodium thiosulphate have been reported previous1y.l This paper reports an extension of the technique to cation- exchange studies. Materials Strongly acidic Dowex 50W-X8 (H+ form), 100-200 mesh, was treated in a column with an excess of approximately 0.2 M EDTA solution (pH 8-10) and then with water. The resin was converted into the Na+ form by passing a large excess of approximately 10% sodium chloride solution followed by washing with water until free of chloride. The capacity of the resin was 2.48 mequiv g-l. Metal ion solutions: Stock solutions of Mg(II), Ca(II), Sr(II), Ba(II), Mn(II), Co(II), Ni(II), Cu(II), Zn(II), Cd(II), Ag(I), Pb(I1) and La(II1) were prepared in water from their nitrates, and Ce(II1) from its chloride. Fe(I1) and Pd(I1) were freshly prepared by dissolving ammonium iron(I1) sulphate and palladium chloride, respectively, in sodium thiosulphate solution. All salts used were either of AnalaR grade or were obtained from Johnson Matthey and Co.Other chemicals were of laboratory-reagent grade. De-ionised distilled water was used for preparing the solutions. Sodium thiosulphate solution. A standard 1.0 M aqueous solution of AnalaR grade sodium thiosulphate (BDH) was used. Wash solution. EZuting agent. All experiments were performed at 30 -J= 2 "C. A batch equilibration method was employed for determining the distribution coefficients Resin (1 g) was added to a solution containing the metal ion under investigation and * To whom correspondence should be addressed.Experimental Cation-exchange resin. It was then air dried and stored in an air-tight bottle. 0.10 M sodium thiosulphate solution. 0.10 and 0.28 M sodium thiosulphate solutions. Procedure (K).530 different concentrations (0.02-0.28 M) of sodium thiosulphate. (5 ml, The solutions were shaken in 100-ml Pyrex flasks for 1 h. been previously described.l EUSEBIUS et al. : THIOSULPHATE AS A COMPLEXING Analyst, VoZ. 106 Sodium hydroxide solution M) was also added to the solution when studies were made in alkaline medium. Details of the procedure have Values of K in alkaline thiosulphate medium are shown in Table I.In achieving the various separations by column chromatography the general procedure previously describedl has been followed unless otherwise stated. Metal ion Mn(I1) . . Fe(I1) . . Ni(I1) . . Zn(I1) . . Cd(I1) . . MgfII) . . Ca(I1) . . Sr(I1) . . Ba(I1) . . La(II1) . . Ce(II1) . . Pb(I1) . . Pd(I1) . . Co(I1) . . CU(I1) . . M I ) * * .. .. .. . . .. .. .. 1 . .. .. 0.02 360 140 360 400 32 250 39 250 440 660 1100 1300 1100 0.2 P P 0.04 170 97 160 150 110 11 130 250 340 610 1100 500 P P 3.1 0.2 0.06 95 68 93 91 60 73 150 210 370 900 300 0.4 4.0 0.2 4.0 P 0.08 67 51 63 62 37 52 100 150 250 730 230 0.4 2.0 0.2 1.6 0.2 0.10 50 38 48 48 28 40 75 110 200 570 170 0.2 <0.1 0.2 0.4 0.8 0.20 16 19 15 14 0.3 5.0 0.4 14 26 41 69 120 58 0.2 <0.1 0.2 TABLE I DISTRIBUTION COEFFICIENTS OF METAL IONS IN ALKALINE SODIUM THIOSULPHATE SOLUTION Concentration of sodium thiosulphate/M* -7 0.28 9 11 9 8 0.3 2.0 0.2 8 15 30 42 64 32 0.2 <0.1 0.2 * p = precipitation.Results and Discussion With increasing concentrations of sodium thiosulphate there is a decrease in the K values of all the metal ions examined, though the extent of the decrease varies for the different metal ions. The presence of thiosulphate therefore reduces the affinity of all of the metal ions towards the cation exchanger, probably due to the reducing and complexing properties of the thiosulphate. At low concentrations of thiosulphate the decrease in the K values is due to the formation of species with zero and a small positive charge for bivalent and trivalent metal ions, respectively.On increasing the concentration of thiosulphate, anionic complex species are formed in some instances, while in others the concentrations of neutral and positively charged species increase, causing a further decrease in K values. The abrupt fall in K values with increase of thiosulphate concentration for Cu(II), Cd(II), Pb(I1) and Ag(1) is due to the formation of anionic complex species2-6 that have negative affinities towards the resin phase. Cu(I1) is reduced to Cu(I), which then forms a series of anionic complex specie^.^^* As Zn(I1) also has some tendency to form anionic complexes,5 the decrease in the K values of Zn(I1) is greater than that observed for other elements. The remarkable affinity of silver for the thiosulphate compared with the resin is the reason for its very low K value^.^^^^^^.For the group I1 elements the K values decrease in the order Mg < Ca < Sr < Ba, which is in the order of their ability to form ion-pair complexes. The decrease in the K values of La(II1) and Ce(II1) is due to the formation of small charged cationic species, the concentra- tions of which increase with increasing concentrations of thiosulphate. Marcusll suggested the formation of anionic species of the type M(S203),3- with the lanthanides; this, however, could not be corroborated by the present studies. The species present with an excess of thiosulphate are not retained by an anion exchanger, whereas the K values are fairly high with a cation exchanger, which indicates that the charge on the species is positive.*May, 1981 AGENT IN CATION-EXCHANGE CHROMATOGRAPHY 531 Columnar Separation of Artificial Mixtures The K values predict certain possible separations, a number of which were carried out on Dowex 50W-X8 columns.Binary separations The feed was prepared by mixing an aliquot of a sdution containing one of the cations Cu(II), Cd(II), Ag(1) or Pb(I1) and one of the strongly sorbed elements Mn(II), Fe(II), Co(II), Ni(II), Zn(II), Mg(II), Ca(II), Sr(II), Ba(II), La(II1) or Ce(II1). The Cu(II), Cd(II), Ag(1) or Pb(I1) are eluted first and elution was therefore continued with 0.10 M sodium thiosulphate solution. A volume of 50 ml of the eluate (including 25 ml from the feed) was sufficient to recover Cu(II), Cd(II), Ag(1) or Pb(I1) quantitatively.Following t h s Mg(II), Mn(II), Fe(II), Co(II), Ni(I1) or Zn(I1) were eluted with about 80 ml and Ca(I1) with 125 ml of 1.0 M sodium thiosulphate solution. No elution of Ba(II), La(II1) or Ce(II1) was obtained even after the passage of about 125 ml of 1.0 M sodium thiosulphate solution. The eluting agent was therefore changed and La(II1) and Ce(II1) were eluted with about 80 ml of 1.0 M ammonium acetate solution and Ba(I1) with 100 ml of 4 M nitric acid. The elution characteristics of the metal ions are presented in Table 11. The results of the binary separations and standard deviations are reported in Table 111. The resolution of the artificial binary mixtures of Cd(I1) or Ag(1) and Mn(II), Fe(I1) or Zn(II), are shown in Fig. 1. The separations were repeated using various concentrations of each of the ions but only one representative value has been shown in each instance.(i) Czl(ll), C d ( I I ) , &(I) and Pb(l1) from other cations. TABLE I1 ELUTION CHARACTERISTICS OF METAL IONS ON DOWEX 50W-X8 Eluting agent I A 7 0.10 M sodium thiosulphate 1.0 M sodium thiosulphate \ I \ BTV*/ VEPt/ TEVI/ BTV/ VEP/ TEV/ ml ml ml ml ml ml A f A Cu(I1) Ag(I) Co(I1) Cd(I1) Pb(I1) Mn(I1) Fe(I1) Ni(I1) Zn(I1) Ba(I1) La(II1) Ce (111) 2(!$ .. . . . . .. . . .. . . .. .. . . .. .. .. .. .. .. .. . . . . .. .. .. .. .. .. .. 3.5 20 3.5 30, 55 3.5 30 5.0 25 > 200 L > 200 - > 200 - > 200 - > 200 - > 200 - > 200 - > 200 - > 200 - > 200 - - - - 55 75 55 35 - - - - - - - - - - 3.5 20 50 - 3.5 20 60 - 3.5 20 60 - 3.5 20 50 - 3.5 10 25 - 3.5 15 80 - 3.5 20 125 - > 100 - > 125 - > 125 - - - - - - * BTV = Breakthrough volume.t VEP = Volume of elution peak. $ TEV = Terminal elution volume. (ii) B a ( l l ) , L a ( l l 1 ) or C e ( l l 1 ) from M g ( l l ) , C a ( l l ) , M n ( l l ) , F e ( l l ) , Co(ll), N i ( l 1 ) or Zn(l1). Both of the metal ions to be separated are retained on the resin bed. After sorption of the metal ions the column was washed with 0.10 M sodium thiosulphate solution and the weakly sorbed metal ions were then eluted with 0.28 M sodium thiosulphate solution. The elution characteristics of these cations are reported in Table IV. No Ba(I1) could be detected in the first 225 ml of eluate nor had La(II1) or Ce(II1) been detected after 325 ml of 0.28 M sodium thiosulphate solution had been passed.The eluent was therefore changed and the strongly sorbed metal ions Ba(II), La(II1) or Ce(II1) were desorbed as described under (i) . Some of the separations could also be achieved when the weakly sorbed cations were eluted with 1.0 M sodium thiosulphate. Small amounts of Ca(I1) could be separated from Ba(I1)532 EUSEBIUS et al. : THIOSULPHATE AS A COMPLEXING Analyst, VoZ. 106 0 40 80 0 20 40 60 Effluent vol u me/m I Fig. 1. Elution curves for Cd(I1) or Ag(1) - Zn(I1). Fe(I1) or Mn(I1). The Cd(I1) or Ag(1) was eluted with 0.1 M sodium thiosulphate solution and the other ion with 1.0 M sodium thiosulphate solution. whereas this could not be achieved in 0.28 M sodium thiosulphate solution. Separations of Mg(I1) from Ba(I1) and Ca(I1) from Ba(I1) are shown in Fig.2. Here the volume of the eluate from the feed has not been taken into account and separations were achieved in alkaline thiosulphate medium. Results for the resolutions of binary mixtures are reported in Table V. Ternary separations The following separations were studied: Cu(II), Cd(II), Ag(1) or Pb(I1) from Mnf from Ba(II), La(II1) r i Ce(II1). For resolving these mixtcLes the height of the resin bed was increased to 10-11 cm. The feed was pre- pared by mixing solutions of (i) Cu(II), Cd(II), Ag(1) or Pb(II), (ii) Ba(II), La(II1) or Ce(II1) and (iiz) Mg(II), Mn(II), Fe(II), Co(II), Ni(II), Zn(I1) or Ca(I1). After sorption Cu(II), Cd(II), Ag(1) or Pb(I1) were quantitatively recovered with about 65 ml of 0.10 M sodium thiosulphate solution.Mn(II), Fe(II), Co(II), Ni(II), Mg(I1) or Ca(I1) were eluted with 0.28 M sodium thiosulphate solution. Finally the most strongly sorbed metal ions were desorbed as described previously. Mn+ is one of the ions listed under (iii) below. I 22.10 1 0 20 40 60 80 0 40 80 Effluent volume/ml Fig. 2. Elution curves for Mg(I1) or Ca(I1) - Ba(I1). The Mg(I1) or Ca(I1) was eluted with 1.0 M sodium thio- sulphate solution and the Ba(I1) with 4.0 M nitric acid.May, 1981 AGENT IN CATION-EXCHANGE CHROMATOGRAPHY TABLE I11 533 RESULTS FOR THE SEPARATION OF VARIOUS BINARY MIXTURES Mean amount Mean amount Standard Metal ion taken/mg n found/mg deviation/mg Error, yo First component ion- Cu(I1) . . . . 25.42 20 25.44 0.07 0.08 . . 20.39 20 22.41 0.12 0.10 .. 20.72 4 20.70 0.13 0.10 Cd(I1) . . . . 81.36 20 81.38 0.12 0.02 Second component ion- Mg(I1) . . . . Ca(I1) . . . . Mn(I1) .. .. Fe(I1) . . . . Co(I1) . . . . Ni(I1) . . . . Zn(I1) . . . . Ba(I1) . . . . La(II1) . . . . Ce(II1) . . .. 12.08 19.92 25.16 27.91 34.62 27.30 30.40 64.55 19.00 42.74 8 6 6 6 6 8 4 10 4 4 12.06 20.10 25.30 27.99 34.68 27.43 30.44 64.59 18.98 42.87 0.17 0.02 0.02 0.14 0.14 0.03 0.14 0.16 0.02 0.01 0.17 0.90 0.56 0.29 0.17 0.48 0.13 0.06 0.58 0.30 EZution cwves. The results of the analyses of synthetic mixtures are presented in Table VI. It is evident from the elution characteristics that the metal ions investigated can be grouped Group 1 : Cu(II), Cd(II), Ag(1) and Pb(II), which were not sorbed by the resin from 0.10 M sodium thiosulphate solution. Group 3: Mg(II), Ca(II), Mn(II), Fe(II), Co(II), Ni(I1) and Zn(II), which were retained by the resin from 0.10 M sodium thiosulphate solution but could be con- veniently desorbed with a reasonable volume of more concentrated sodium thiosulphate solution (0.28 or 1.0 M).Ba(II), La(II1) and Ce(III), which were very strongly sorbed by the resin from 0.10 or 0.28 M sodium thiosulphate solution and could not be eluted with a reasonable volume of 1 .O M sodium thiosulphate solution. Figs. 3 and 4 show the separation of Cu(I1) from Fe(I1) or Mn(I1) from Ce(II1) and Pb(I1) from Fe(I1) from Ba(II), respectively. according to their column elution behaviour : Group 3 : 'I R 20 40 600 40 120 200 0 40 80 Effluent volume/ml Fig. 3. Elution curves for Cu(I1) - Fe(I1) or Mn(I1) - Ce(II1).The Cu(I1) was eluted with 0.1 M sodium thiosulphate solution, the Fe(I1) or Mn(I1) with 0.28 M sodium thiosulphate solution and the Ce(II1) with 1.0 M ammonium acetate solution.534 EUSEBIUS et al. : THIOSULPHATE AS A COMPLEXING Analyst, VoZ. I06 TABLE IV ELUTION CHARACTERISTICS OF METAL IONS ON DOWEX 50W-X8 USING 0.28 M SODIUM THIOSULPHATE SOLUTION AS ELUTING AGENT Metal ion Mg(I1) .. .. Ca(I1) . . .. Mn(I1) .. .. Fe(I1) . . .. Co(I1) . . .. Ni(I1) . . .. Zn(I1) . . .. Ba(I1) . . .. La(II1) . . .. ce(II1) . . .. BTV/ml 25 100 28 15 30 37 5 > 225 > 325 > 325 VEP/ml 70 180 80 60 70 80 20 - - - TEV/ml 170 310 190 180 180 180 75 - - - TABLE V RESULTS FOR THE SEPARATION OF VARIOUS BINARY MIXTURES Mean amount Mean amount Standard Metal ion takenlmg n found/mg deviationlmg Error, % First component ion- Ba(I1) .. . . 64.55 12 64.52 0.27 0.05 La(II1) . . . . 47.58 14 47.61 0.15 0.02 Ce(II1) . . . . 42.74 14 42.70 0.18 0.09 Second component ion- Mg(I1) .. . . 12.08 4 12.04 0.05 0.33 Mn(I1) .. . . 13.09 6 13.04 0.14 0.38 Fe(I1) . . . . 27.91 6 28.05 0.03 0.50 Co(I1) . . . . 24.24 4 24.36 0.04 0.50 Ni(I1) . . . . 13.16 4 13.10 0.02 0.46 Zn(I1) . . . . 15.20 4 15.24 0.02 0.26 Ca(I1) . . .. 9.96 6 9.94 0.15 0.10 TABLE VI RESULTS FOR THE SEPARATION OF VARIOUS TERNARY MIXTURES Mean amount Mean amount Standard Metal ion First component ion- CU(I1) . . .. Cu(I1) . . .. Third component ion- Ba(I1) . . .. La(II1) . . .. Ce(II1) . . .. Second component ion- Mg(I1) .. .. Ca(I1) .. .. Mn(I1) .. .. Fe(I1) . . .. Co(I1) . . .. Ni(I1) . . .. Zn(I1) . . .. takenlmg 25.42 33.04 11.35 56.76 81.36 20.40 20.72 103.30 64.55 19.03 23.80 47.58 17.09 42.74 9.67 12.08 9.96 13.05 27.91 17.31 13.65 9.93 15.00 n 14 26 14 12 14 40 14 26 48 28 12 14 28 28 10 14 16 24 24 20 20 8 16 found/mg deviationlmg Error, yo 25.43 33.06 11.37 56.71 81.33 20.39 20.76 103.50 0.16 0.13 0.10 0.13 0.17 0.13 0.14 0.23 0.04 0.06 0.18 0.09 0.04 0.04 0.19 0.19 64.60 0.19 0.08 19.02 0.22 0.05 23.89 0.22 0.38 47.56 0.17 0.04 17.13 0.23 0.23 42.83 0.17 0.21 9.70 12.05 10.05 13.01 28.01 17.39 13.60 9.91 15.02 0.17 0.04 0.15 0.17 0.18 0.14 0.14 0.03 0.06 0.31 0.25 0.90 0.31 0.36 0.46 0.37 0.20 0.13May, 1981 AGENT I N CATION-EXCHANGE CHROMATOGRAPHY 535 v) c .- c 8 t 2 p 6 r' 2 4 +- c-0 .- 4- +- 0) c 8 2 0 - 0 Eff I uent vol ume/m I 50 1 00 Fig.4. Elution curves for Pb(I1) - Fe(I1) - Ba(I1). The ions were eluted with 0.1 and 0.28 M sodium thiosulphate solutions and with 4.0 M nitric acid, respectively. The elution behaviour of metal ions in 0.28 M sodium thiosulphate solution (Table 111) indicates that Zn(I1) can be separated from Ca(I1) because the breakthrough volume of Ca(I1) is 100 ml whereas the terminal elution volume of Zn(I1) is 75 ml. The separation of these two ions from La(II1) and Ce(II1) with 0.28 M sodium thiosulphate solution and from Cu(II), Cd(II), Ag(1) and Pb(I1) suggested the possibility of separating the following quater- nary mixtures: Cu(II), Cd(II), Ag(1) or Pb(I1) from Zn(I1) from Ca(I1) from La(II1) or Ce(II1).Cu(II), Cd(II), Ag(1) or Pb(I1) may be eluted with 0.10 M sodium thiosulphate solution as described previously. The elution of Zn(I1) with 0.28 M sodium thiosulphate solution was followed by the elution of Ca(I1) with the same eluent. Finally Ce(II1) or La(II1) were desorbed as described previously. Fig. 5 represents the resolution of a mixture containing Cu(II), Zn(II), Ca(I1) and La(II1). All separations were achieved in neutral aqueous sodium thiosulphate medium except for mixtures containing Ba( 11), when alkaline sodium thiosulphate was used to avoid precipita- tion of barium thiosulphate. The elution behaviour of Pd(I1) has not been studied but its distribution coefficient indicates that it should not be retained on the column with 0.06- 0.10 M sodium thiosulphate solution.The easy removal of sodium thiosulphate from the c v ) 8 - .- C S 2 h 6 - s c .- .- 5 4 - p 2 - c. h c s S O - #l 0 30 60 0 40 80 120 160 200 240 280 0 30 60 Effluent volume/ml Fig. 5. Elution curves for Cu(I1) - Zn(I1) - Ca(I1) - La(II1). The Cu(I1) was eluted with 0.1 M sodium thiosulphate solution, the Zn(I1) and Ca(I1) with 0.28 M sodium thiosulphate solution and the La(II1) with 1.0 M ammonium acetate solution.536 EUSEBIUS, GHOSE AND DEY eluate is an advantage of this method. Perhaps the most serious disadvantage of the use of sodium thiosulphate is its instability in acidic media but as all of the separations can be achieved in alkaline medium (pH 8-10) this does not present any serious problems. 1 . 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. References Eusebius, L. C. T., Ghose, A. K., Mahan, A., and Dey, A. K., Analyst, 1980, 105, 52. Novakovskii, M. S., and Ryazantsena, A. P., Tr. Khim. Fak. Nauk Issled. Inst. Khim. Kharkov Gos. Sil’nichenko, V. G., Uch. Zap. Mosk., 1959, 84, 119; Ref. Zh. Khim., 1961, 6V36. Sillen, L. G., and Martell, A. E., “Stability Constants of Metal Ion Complexes,” Spec. Publ. No. 17, Yatsimirskii, K. B., and Gus’kora, L. V., Zh. Neorg. Khim., 1957, 2, 2039. Ziegler, M., Naturwissenschaften, 1959, 46, 353. Ryabchikov, D. I., Omagiu Raluca Ripan., 1966, 485; Chem. Abstr., 1967, 67, 39696q. Toropova, V. F., Sirotina, I. A., and Lisora, T. I., Uch. Zap. Kaz. Gos. Univ. V. I . Ul’yanova - Lenina, 1955, 45, 43; Chem. Abstr., 1958, 52, 2633b. Dey, A. K., Khim., J . Inorg. Nucl. Chem., 1958, 6, 71. Marcus, Y., Acta Chem. Scand., 1957, 11, 619. Marcus, Y., U.S.A.E.C., Pub./UP/R-20, 1959, p. 9; Chern. Abstr., 1962, 57, 6859b. Univ., 1954, 12, 227; Ref. Zh. Khim., 1955. 1964; Spec. Publ. No. 25, 1971, Chemical Society, London. Received July 15th, 1980 Accepted September 17th, 1980