JUNE, 1967 THE ANALYST Vol. 92, No. 1095 The Reversed-phase Thin-layer Chromatography of Metal Ions with Tributyl Phosphate BY L. S. BARK, G. DUNCAN AND R. J. T. GRAHAM (The Department of Chemistry and Applied Chemistry, University of Salford, Salford 5, Lancashire) About sixty-five metal-containing ionic species have been chromato- graphed on thin layers of cellulose impregnated with tributyl phosphate (TBP) at various concentrations of aqueous hydrochloric acid. A mechanism for the chromatography may be explained in terms of the ability of the metal ions to form chloro complexes, viz., that those metal ions which readily complex with chloride ion are readily extracted by the TBP and consequently have low RF values. Conversely, those metals which do not form chloro complexes are not retained by the TBP and hence have high RF values.Attention is drawn to the similarities between the known liquid - liquid extraction behaviour of the metal ions in the TBP - hydrochloric acid systems and the behaviour of these ions in the TBP - hydrochloric acid chromato- graphic systems. A strong resemblance has also been found between the RF spectra (Rp V ~ Y S U S hydrochloric acid) of the metal ions and the behaviour of these ions in resinous anion-exchange - hydrochloric acid systems. The latter similarity has been used as evidence for the suggestion that the TBP on the layers functions as a liquid anion exchanger, Le., that TBP-solvated protons, ion-associated with chloride ions, can undergo ion exchange with the metal chloro complex- (TBP), + (HCl),, =+ (TBPH+Cl-), (nTBPH+Cl-), + (MCl,-*) e MC1,-n(TBPH+),o + (nCl-)aa.The suffix “0” refers to the organic phase and the suffix “aq” refers to the aqueous phase. THE easy detection of metals as both major and trace components is of obvious industrial importance. While solvent extraction and ion exchange have been used for the separation of some of the metal ions, the use of a system that would simultaneously both separate and maintain these ions in a concentrated form is of potential importance. A technique combining solvent extraction, ion exchange and chromatography, obtained by using a suitable liquid ion exchanger or solvent extractant as the stationary phase in a chromatographic system, makes this feasible. We have previously reported1y2y3 the separation of some metals, including some of toxico- logical imp~rtance,~ by using a reversed phase of cellulose impregnated with tributyl phos- phate (TBP), with hydrochloric acid as the eluent.This system has been extended to a study of a great many metal ions, including the alkali metals, the alkaline earths, some first, second and third row transition metals, the lanthanides, scandium, yttrium, thorium and uranium. The separation of some of these metals by reversed-phase paper chromatography, with TBP as the stationary phase, has been previously r e p ~ r t e d . ~ , ~ ~ ~ Hu and Liu7 have studied the separation of niobium, tantalum and some noble metals by using TBP-impregnated silica gel, with hydrochloric acid (M and 2 M) as the mobile phases; the R, values quoted by these authors indicate that the spots obtained were not discrete but were badly streaked. We have previously shown that the TBP loadings used by these workers cannot give chromato- grams suitable for comparative purposes.I t is necessary to control the loading of the layers to the same rigorous standards that are necessary for the other extramolecular factors to obtain highly reproducible chromatographic values. 347348 BARK, DUNCAN AND GRAHAM: REVERSED-PHASE THIN-LAYER [Analyst, Vol. 92 METAL IONS- Solutions were prepared from analytical-grade chloride or nitrate salts of the alkali metals (lithium to caesium), the alkaline earths (beryllium to barium), Al(III), Ti02+, VO,+, Cr(III), Mn(II), Fe(II), Fe(III), Co(II), Ni(II), Cu(II), Zn(II), Sb(III), Zr02+, Ru(IV), Rh(II), Pd(II), Ag(I), Cd(II), In(III), Sn(II), Ir(III), Pt(II), Au(III), Hg(II), Tl(I), Pb(II), Bi(III), La(III), Ce(IV), Lu(III), Y(III), Th(1V) and UO,,+.Se(IV), Te(IV), Re(VI1) and Os(VII1) were used in the form of the sodium or potassium salt of the oxy-anion (Se0,2-, Te032-, Re0,- and 0sOs2-). Sc(III), Pr(III), Nd(III), Sm(III), Eu(III), Gd (111), Tb(III), Dy(III), Ho(III), Er(III), Tm(II1) and Yb(II1) ions were obtained from the appropriate metal oxide. CHROMOGENIC REAGENTS- 3,5,7,2’,4’-Pentahydrox~$avone (Morin), 0.1 per cent. in ethanol. 8-Hydroxyquinoline, 2 per cent. in chloroform. 1-(2-Pyridylazo)-2-naj5hthol (PAN), 0.1 per cent. in ethanol. p-Dimethylamino benzylidenerhodanine, 0.1 per cent. i n ethanol. 2,7-Bis- (o-arsono~henylaxo)-1,8-dihydroxynaphthalene-3,8-disul~honic acid, sodium salt.Arsenazo I I I , 0.1 per cent. in water. Diethyldithiocarbamate ( D D T C ) , 0-1 per cent. in ethanol. EXPERIMENTAL PREPARATION OF THE METAL-ION SOLUTIONS- Water was used as the solvent, except in those instances when the metal salt was insoluble in water, viz., antimony trichloride was dissolved in sodium hydroxide solution, scandium and several of the lanthanide oxides (Sm,O,, Eu203, Tb203, Er203 and Yb,O,) were dissolved in dilute hydrochloric acid, while others (Pr,O,, Gd2O3 and Tm2O3) were dissolved in about 8 M nitric acid and diluted with water. The pentoxides of niobium and tantalum were converted to the pentachlorides by boiling in sulphur dichloride for several hours. The yellow crystals of the pentachlorides were dissolved in sufficient “A”-grade carbon tetrachloride to give a concentration of 5 mg per ml. PREPARATION OF THE LAYER- Cellulose powder (15 g of Machery Nagel MN 300 HR) was mixed with a solution of purified8 TBP in carbon tetrachloride (70 ml of a 5 per cent.solution) to give a homogeneous slurry that was spread as an even layer, 0-3 mm thick, on five 20 x 20-cm glass plates. The layers were allowed to dry in an air-oven at room temperature for 1 hour to remove the carbon tetrachloride. APPLICATION OF THE METAL-ION SOLUTIONS- When the plates were dry, aliquots (1 pl) of the metal-ion solutions to be chromato- graphed were applied to the layers by using a multiple spotting devices at a fixed distance from the base of the plate; the applied spots were allowed to air-dry for a further standard time (15 minutes).ELUTION OF THE CHROMATOPLATE- The chromatoplates were developed in a small-volume double-saturation chamber, as previously described,l with various concentrations of hydrochloric acid (0.1 to 9 M). By using a vertical development, the eluent was allowed to move a fixed distance (12.5 k 0.25 cm) from the point of application of the metal ions. The development times ranged from 1 to 3 hours, depending on the acid concentration of the mobile phase (there was an increase in time with increase in acid concentration). When the eluent front had moved to the pre-determined distance from the point of application of the spots, the plates were removed from the saturation chamber and heated for a standard time (20 minutes) in an air-oven (temperature 110” 5 2°C) to remove the acid and most of the TBP.VISUAL IDENTIFICATION OF THE METAL IONS- to form complexes with some of the “free” metal ions. sprayed with one of the chromogenic reagents mentioned above. Solutions of the metal ions (5 mg per ml) in suitable solvents were prepared. It is necessary to remove most of the TBP as indicated, to allow the chromogenic indicator Each plate was examined in visible light and under ultraviolet radiation. It was thenJune, 19671 CHROMATOGRAPHY OF METAL IONS WITH TRIBUTYL PHOSPHATE 349 Treatment Visible light Ultraviolet light Sprayed with morin, exposed to NH, and held under ultraviolet light Sprayed with 8-hydroxyquinoline, exposed to NH, and held under ultraviolet light Sprayed with arsenazo I11 Sprayed with a 1 + 1 mixture of PAN and F-dimethylamino- benzylidenerhodanine and viewed (i) in visible light (ii) in ultraviolet light Sprayed with DDTC and visible light viewed in Appearance Dark spots on white layer Dark spots quenching background Yellow fluorescent spots fluorescence Dark spots quenching background Yellow fluorescent spots Green spots on blue background fluorescence Red spots Blue spots Grey spots Yellow spots Dark spots quenching background Bright yellow fluorescent spot Blue fluorescent spot Green spots fluorescence Metal Ti02+, U022+, Ru Rh, Ir, Nb, Ta, Sb Alkali metals Ti02+, UOz2+ Alkaline earths and A1 Sc, Y , the lanthanides, Cr, Mn, Ni, Cu, Zn, Th, U022+ Zr02+, Cd, In, Sn, Hg, T1 Co, Pd, Pt Fe(II), Fe(II1) Ag Ru, Rh, TeOS2- Au Se0,2- Pb, Bi Re0,-, 0~0,~- RESULTS The mean R, values of the metal ions chromatographed in the various eluent systems are shown in Tables I to V.The corresponding RF spectra are shown in Fig. 1. Each R, value is the mean of at least four determinations, each within 50.02 of the mean value. Such highly reproducible R, values were achieved by strictly controlling the experimental conditions. 0 2 4 6 810. Fig. 1. RF spectra of metal ions chromatographed in various eluent systems350 BARK, DUNCAN AND GRAHAM REVERSED-PHASE THIN-LAYER [ArtaZySt, VOl. 92 DISCUSSION OF RESULTS It is not possible to obtain a single system that will separate each of the ions from a mixture of the whole. The tables of results have been grouped according to the periodic classification, as often, mixtures requiring separations are of metals that are closely related in the periodic table, vix., transitional or alkali metals.MEAN RF VALUES OF ON CELLULOSE - TBF Metal ion Lithium . . .. Sodium . . . . Potassium . . .. Rubidium . . .. Caesium . . .. Beryllium . . .. Magnesium .. Calcium . . .. Strontium . . . . Barium . . .. Aluminium .. I TABLE I THE ALKALI AND ALKALINE EARTH METALS AND ALUMINIUM (5 PER CENT.) LAYERS AT VARIOUS ACID CONCENTRATIONS 4M HCl 6~ HCI 7M HCI 8 M HCI .. 0-94 0.87 - 0.88 . . 0-98 0.94 0.87 0-89 .. 0.98 0.93 0.88 0.87 .. 0.98 0.92 0.88 0-86 , . 0-98 0.91 0.88 0.85 . . 0.94 0.95 0.82 0.78 .. 0.94 0.87 0.85 0.8 1 .. 0.94 0.87 0.77 0.73 . . 0-94 0.82 0-78 0-70 . . 0.94 0.82 - .. 1.00 1.00 1.00 1.00 - The spots obtained in these systems were generally elongated.THE ALKALI AND ALKALINE EARTH METALS- Table I shows the average R, values for the alkali and alkaline earth metals on layers of 5 per cent. TBP - cellulose. It was possible by using either the 4 M or 6 M eluent, partially to separate lithium from the other alkali metals (sodium, potassium, rubidium and caesium), but none of the other metal ions listed in this table could be separated. The spots, except for lithium and sodium, were generally elongated (2 to 2.5 cm), and the length of the spot increased with atomic weight. Identification of the alkali metal ions after development was difficult, and the spray reagent used (morin) gave the most definite reaction of those tried TABLE I1 MEAN R, VALUES OF THE LANTHANIDES, SCANDIUM, YTTRIUM, THORIUM AND URANIUM ON CELLULOSE - TBP (5 PER CENT.) LAYERS AT VARIOUS ACID CONCENTRATIONS Metal ion Scandium(II1) .. Yttrium(II1) . . Lanthanum(II1) . . Praseodymium(1 I I) Neodymium(II1) . . Samarium(II1) . . Europium(II1) . . Gadolinium(II1) . . Terbium(II1) . . Dysprosium(II1) . . Holmium(II1) . . Erbium(II1) . . Thulium(II1) . . Ytterbium(II1) . . Lutetium(II1) . . Thorium(1V) . . (Uranyl) UO,z+ . . Cerium(1V). . .. M 2M 0.99 0.98 0.99 0.98 0.99 0.98 0.99 0.98 0.99 0.98 0.95 0.97 0.96 0.97 0-96 0.96 0.95 0.93 0.97 0.98 0.97 0.98 0.95 0.98 0.95 0.97 0.98 0.97 0.98 0.99 0.98 0.99 :T} 0.97T 0.65 0.55 4M 0.98 0.98 0.96 0-94 0-95 0.96 0.96 0.96 0.96 0-97 0.97 0.97 0.97 0.97 0-97 0.97 0-84T 0.20 5 M 0.97 0.97 0.96 0.94 0.94 0.92 0.94 0-94 0.95 0.96 0.97 0.97 0.96 0.97 0.96 0.97 0.83T 0.13 6M 0.82 0.90 0.86 0.85 0.85 0.85 0.85 0.83 0.86 0.86 0-86 0.86 0.86 0.86 0-87 0.89 0-76T 0.11 7M 0.57 0.89 0.83 0.84 0.84 0.83 0.84 0.82 0.82T 0.86 0.86 0.86 0-88 0.89 0.86 0.90 0.64T 0.41T) 0-10 8 M 0-25 0.89 0-84 0.83 0.84T 0.83T 0-84T 0-82T 0.81T 0.85T 0.83T 0.88T 0.84T 0-87T 0.87T T EZ} 0.12 9M 0.10 0-78T 0.76T 0.75T 0-74T 0.74T 0*76T 0.78T 0.78T 0.80T 0.81T 0.80T 0.76T 0.84T 0-83T T 0.59T 0.11 The spots obtained were rather large (3 cm), especially for the higher acid concentrations.T = Tailed or streaked spot. hlultiple values indicate multiple spots.June, 19671 CHROMATOGRAPHY OF METAL IONS WITH TRIBUTYL PHOSPHATE 351 (zinc uranyl acetate, violuric acid and 8-hydroxyquinoline) . Other workerslO have also found difficulty in identifying the alkali metals on thin layers.Similar difficulties were encountered in finding a sufficiently sensitive reagent for barium; 8-hydroxyquinoline gives only a faint indication of this metal. However, it was decided to retain 8-hydroxyquinoline as aspray reagent for the alkaline earth metals because of its very sensitive reaction with beryllium, and also because of the experimental difficulties in spraying any single metal ion with its individual spray reagent (the metal ions were applied to the layers at l-cm intervals, and so attempts at spraying individual spots generally led to overlapping of the sprayed bands). The R, values quoted in Table I1 are in agreement with the extraction behaviour of these metal ions in the corresponding TBP - hydrochloric acid Of the alkali metals, only lithium measurably extracts into TBP. SCANDIUM, YTTRIUM, THE LANTHANIDES, THORIUM AND URANIUM- From the results in Table I1 it can be seen that only scandium, thorium and uranium are appreciably retained by the stationary phase.Except for scandium and uranium, the spots were generally very large (3 cm) and diffused laterally. The thorium spot was badly streaked although well defined and reproducible, and at acid concentrations above 6 M hydro- chloric acid separated into two spots, the upper spot being the more intense. The phenomenon of multiple-spot formation will be discussed later. The streaking is probably caused by a slow rate of attainment of equilibrium between the thorium and the hydrochloric acid or between the thorium chloro - aquo complex and the TBP, or both.Similar arguments can be used to explain the tailing of the lanthanides at high eluent acid concentration. The results show that the separation of scandium from yttrium, the lanthanides and uranium is possible by using 6, 7 and 8 M hydrochloric acid as eluents. Scandium can be separated from uranium at all of the eluent concentrations used, except 9 M hydrochloric acid. The R F behaviour of these metals is in agreement with the available extraction results in the corresponding TBP - hydrochloric acid system.ll MEAN R, VALUES OF Acid con- centration Ti(IV) 0.1 M 1.00 0.75 M 1-00 0.53 1.0 M 0'99 - 2.0 M 3.0 M 0.89 4.0 M 0-97 4.5 1\1 0.94 5-0 M 0.91 6.0 M 0.89 7.0 M 0.79 7.5 M - 8.0 M 0.75 9.0 M - TABLE I11 THE FIRST ROW TRANSITION METALS ON CELLULOSE - TBP (5 PER CENT.) LAYERS AT VARIOUS ACID CONCENTRATIONS Cr(I1I) 0.99 0.98 0.98 0.97 0.96 0.94 0.93T 0.93 0.91 0-89 - 0.86 0.84 Mn(I1) 1.00 0.99 0.99 0.97 0.96 0.95 0.95 0.93 0.89 0.87 - 0.76 0.66 Fe(I1) 1.00 0.99 0.98 0.98 0.97 0.97 0.97 0.94 0.87 T T T - Fe(II1) 0.92 0.8 1 0.55 - 0.00 0.00 - 0.00 0.00 0.00 0.00 0.00 0.00 Co(I1) 1.00 0.99 0.99 0.97 0.96 0.95 - 0.95 0.85 0.83 - 0.72 0.65 Ni(II) 1.00 0.99 0.98 0.97 0.96 0.95 - 0.94 0.88 0.87 0.86 0.80 0.75 Cu(I1) 1-00 0.99 0-97 0.97 0.96 0.93 - 0.91 0.82 0.79 0-76 0.74 0.67 Zn(I1) 0.94 0.80 0.65 0.20 0.18 0.3 1 - 0-40 0.47 0.52 0.55 0.59 0-62 T = Tailed spot.Multiple values indicate multiple spots. THE FIRST ROW TRANSITION METALS- The average RF values for the first row transition metals are given in Table 111.The results for manganese, cobalt, copper and zinc have been reported previously.1,2 The spots obtained for these materials were, in nearly all instances, small and well defined. Vanadium352 BARK, DUNCAN AND GRAHAM: REVERSED-PHASE THIN-LAYER [AnaZySt, VOI. 92 and iron(I1) showed severe tailing in some eluent systems, although in the case of vanadium, the tail consisted of a series of partly separated spots. These spots are probably caused by the presence of several oxidation states of vanadium in the hydrochloric acid mobile phase ; chromatographic separation of these oxidation states will occur, resulting in the tail described above. Titanium showed multiple-spot formation in the M hydrochloric acid eluent system.For acid concentrations below 6 M, none of the first row- transition metals, except zinc, is appreciably retained by the stationary phase. Complete separation of titanium, vanadium, manganese, cobalt and copper is not possible in any of the eluent systems used. However, separation of iron(I1) from iron(II1) is possible in all eluent systems, and separation of chromium, iron(I1)and 111), nickel and either cobalt or manganese from each other is possible by using 8 M hydrochloric acid as eluent. Zinc can be separated from the other first row transition metals in nearly all of the eluent systems. TABLE IV MEAN RF VALUES OF THE SECOND ROW TRANSITION METALS ON CELLULOSE - TBP (5 PER CENT.) LAYERS AT VARIOUS ACID CONCENTRATIONS Acid con- centration 0.1 M 0.75 M 1.0 M 2.0 M 3.0 M 4.0 M 4.5 M 5.0 M 6.0 M 7.0 M 7.5 M 8.0 M 9.0 M Zr(1V) 0.92T 0.84T T T 0.79T - 0-83T 0.80T T - 0.13 0.00 0.00 N W ) T T T T 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Mo(V1) Ru(IV) T T 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0-00 0.00 0-00 0.00 Rh(I1) Pd(I1) Ag(1) 0-89 0.87 0.88 0-89 0.91 0.90 0.89 0.90 0.84 0.83 0.82 0.79 0.74 0.75 0-79 - - 0-7 1 0-72 0-73 0.70 0.67 - 0.63 0.63 0.63 0.00 0.00 - - 0.29T - 0.66T 0-72 0.80 - 0.82 0.79 0.55 Cd(I1) In(II1) Sn(I1) 0.92 0.86 - - 0-18 0.22 0.26 0.24 0.32 - 0-41 0.52 0.59 0-92 0.86 - - 0.10 0.10 0.11 0.07 0.06 - 0.06 0.01 0.07 T 0.14 - - 0.00 0.00 0.00 0.00 0.00 - 0.00 0.00 0.00 T = Tailed spot.Multiple values indicate multiple spots. THE SECOND ROW TRANSITION METALS- Well defined spots were generally obtained for the second row transition metals.Zir- conium was the only metal ion that showed severe tailing (Table IV) over a wide range of acid concentration; niobium, molybdenum and tin showed moderate tailing in the low ‘acid concentration systems. Several separations of the second row transition metals are possible. A t 3 M hydrochloric acid, niobium, molybdenum, ruthenium, rhodium, palladium, cadmium and indium can all be separated from each other. At 6 M hydrochloric acid, rhodium, palladium, silver, cadmium, indium and tin can be separated, and at 7.5 M hydrochloric acid, zirconium can be separated from the other metals. A comparison of the R, values in Table IV with those in Table I11 shows that the second row transition metals are generally much more strongly held by the TBP than the first row transition metals.Thus, separation of the metals within groups of the periodic table is possible in nearly all of the eluent systems. Such separations will be discussed.June, 19671 CHROMATOGRAPHY OF METAL IONS WITH TRIBUTYL PHOSPHATE 353 TABLE V MEAN IZ, VALUES OF THE THIRD ROW TRANSITION METALS, ANTIMONY, SELENIUM AND TELLURIUM OF CELLULOSE-TBP (5 PER CENT.) LAYERS AT VARIOUS ACID CONCENTRATIONS Acid concentration 0.5 M 1.0 M 2-0 M 3-0 M 4.0 M 5.0 M 6.0 M 7.0 M 8.0 M 9.0 M Acid concentration 0.5 M 1.0 M 2.0 M 3.0 M 4.0 M 5.0 M 6.0 M 7.0 M 8.0 M 9.0 M Se(1V) 1.00 1.00 1.00 0.98 0.98 0.98 0.85 0.88 0.73 0-46T Ir(II1) ;:;iT} ::;F} T T T T T T T T Te (IV) 0-87 0.83 0.24T 0.09 0.02 0.02 0.00 0.01 0.00 0.00 Pt(I1) E T } K:} ::;“F} E} ::E} 0.20} K} 0.36 0.13) 0.41 0.50 0.51 Sb (I I I) o:o} oao} o:o} 0:0> o:o} 0:o } A} o:o} o:o} T 1 0.00 j- Au(II1) 0.00 0.00 0-00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Ta(1V) 0.02 O:o} o:o> 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Hg(II) 0.11 0.09 0.06 0.05 0.07 0.10 0-13 0.22 0.29 0.38 W V I ) K} 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.09 1 0.02 J T1P) 0.90 T T T 0:2} 0;9} oio} &3} T 1 0.61 J Re(VI1) 0-78 0.12} Z } E} ;:E} ;:g} ::g} K3} E} 0.40 1 0.02 j- Pb(I1) 0.8 1 0.80 0.81 0.83 0.89 0.92 0.93 0.95 0.95 0.90 Os(VII1) 0.93 0.84 0.40) 0.95 0.95 0.95 0.88 0-84 0.84 0.85 0.80 Bi(II1) 0.19 0.32 0.40 0.57 0.79 0.92 0-94 0.95 0.88 0.90 T = Tailed or streaked spots.Multiple values indicate multiple spots. THE THIRD ROW TRANSITION METALS, ANTIMONY, SELENIUM AND TELLURIUM- Table V gives the RF values of the third row transition metals, antimony, selenium and tellurium.Many of the metal ions listed in this table showed multiple-spot formation. Iridium was particularly troublesome in this respect in that several spots of equal intensity appeared on spraying with indicator, and considerable streaking occurred between these spots. Thus no accurate RF values could be measured for iridium in systems with acid con- centrations above 1 M. Antimony appeared as a series of tailed spots and the positions of these spots on the chromatograms were not reproducible, except for the very small, intense spot that remained at the point of application. This small spot is probably caused by the354 BARK, DUNCAN AND GRAHAM : REVERSED-PHASE THIN-LAYER [Analyst, Vol.92 reaction of the cellulose layer with the sodium hydroxide in which the antimony trichloride was dissolved; shrinkage of the layer occurred at the point of application of the spot, resulting in the applied spot breaking away from the rest of the layer. The severe tailing may have been caused by a slow rate of hydrolysis of the antimony compound and partial separation of some of the intermediate species of the hydrolysis reaction. Despite some tailing at intermediate acid concentrations, selenium can be separated from tellurium in all of the eluent systems used. Osmium, platinum, gold and mercury can be separated by using 0.5 or M hydrochloric acid as eluents. Mercury, thallium, lead and bismuth can also be separated at low acid concentrations, although thallium shows streaking at acid concentrations between 1 M and 9 M hydrochloric acid.Rhenium and osmium can be separated in all of the acid systems above 1 M, while rhenium, osmium, platinum, gold, mercury and lead can be separated by using 8 M hydrochloric acid as eluent. Antimony, tantalum, tungsten and gold have R, values of zero in all of the eluent systems used. Tables 111, IV and V show that the RF values of the metals chromatographed generally decrease in the order: first row > second row > third row, and thus separations of the metal ions within a group of the periodic table are, in many instances, possible. Some of these separations have already been discussed. SEPARATION OF TITANIUM (IV) AND ZIRCONIUM (IV)- chloric acid.Titanium and zirconium are readily separated in any eluent system above 7 M hydro- SEPARATION OF CHROMIUM(III) FROM MOLYBDENUM(VI) AND TUNGSTEN(VI)- systems used. SEPARATION OF THE PLATINUM METALS (Ru, Rh, Pd, Os, Ir, Pt)- It is not possible to separate all of the platinum metals by using a single eluent system. However, it is possible to separate the first three metals (ruthenium, rhodium and palladium) in either the 3 or 5 M hydrochloric acid systems and to separate osmium from platinum in all of the eluent systems. Iridium invariably tailed in all of the eluent systems. Ruthenium, palladium, osmium and platinum can be separated in the 3 M hydrochloric acid system. SEPARATION OF COPPER, SILVER AND GOLD- Gold invariably has an RF value of zero, and so this metal can be separated from silver and copper in all systems above 3 M hydrochloric acid.The three metals can be separated from each other in the 3, 4.5 and 5~ hydrochloric acid systems. SEPARATION OF ZINC, CADMIUM AND MERCURY- Although the R, values of zinc and cadmium are very similar in several systems, separation of both of these metals from each other and from mercury can be achieved in any eluent system between 4.5 and 8 M hydrochloric acid. While a large number of separations within periods or groups, or both, of the periodic table have been discussed, industrial mixtures are, in fact, not so systematic in their com- position, e.g., the coinage and jewellery metals, metals of toxicological interest, and ferrous and non-ferrous alloys, etc.SEPARATION OF COINAGE AND JEWELLERY METALS (Ni, Cu, Ru, Rh, Pd, Ag, Ir, Pt, Au, Sn)- A complete separation of all of these metals is not possible in a single eluent system. However, if iridium is absent, a separation of the other metals can be achieved by a suitable choice of two or more solvent systems, viz., at 3 M hydrochloric acid, all, except nickel and copper, and gold and tin, may be separated, one from the other. Nickel and copper may be separated from one another, and from the others, by the use of 7-5 M hydrochloric acid as eluent. SEPARATION OF THE TOXIC METAL IONS (Bi, Cd, Co, Cu, Hg, Mn, Ni, Pb, Sb, Se, Te, U, 2n)- The application of the technique to the separation of metals of toxicological interest has been reported previ~usly.~ Chromium can be separated from molybdenum and tungsten in any of the eluent Gold and tin can be separated by using 0.75 M hydrochloric acid as eluent.June, 19671 CHROMATOGRAPHY OF METAL IONS WITH TRIBUTYL PHOSPHATE 355 Possible separations of many other industrial mixtures could be discussed.However, the feasibility of separating any given mixture of metal ions can be readily ascertained by reference to the R, spectra shown in Fig. 1 . MULTIPLE-SPOT FORMATION- A theoretical treatment of multiple-spot formation has been given by Keller and Gid- dings,12 who have shown that slow chemical equilibrium in either one or both phases can be the cause. However, in most instances, the formation of multiple spots from a single applied substance actually involves separation of different forms of the substance, these forms being relatively stable in the chromatographic conditions used.13 The latter is probably true in TBP - hydrochloric acid systems where several extractable species can co-exist .14 THE INFLUENCE OF CHLORO-COMPLEX FORMATION ON THE R, VALUES AND THE NATURE OF The R, values of the metal ions show that, in general, those metal ions that readily form chloro complexes have low RF values, e.g., zinc, cadmium, mercury, platinum and gold, while those metals which do not readily form chloro complexes have high R, values, e.g., chromium, cobalt and nickel.Much of the literature on the nature of chloro complexes formed by most of the metal ions in different concentrations of hydrochloric acid has been reviewed.15 From the chloro complexes formulated in the above review, and from the R, values quoted in the results, it is evident that TBP favours the extraction of both neutral and anionic chloro complexes.In particular, TBP favours the extraction of acido -halo complexes, and those metals which form such complexes generally have R, values of zero. Marcus16 has reviewed the use of TBP in the solvent extraction of metal ions and mineral acids, and has formulated some of the extracted species existing in the organic phase. He has drawn attention to the possibility that TBP can extract in two different ways- (;) by direct co-ordination of the TBP to the metal, e.g., in the TBP - hydrochloric acid system, uranium is reported to be extracted as U0,Cl2.2TBPl7 at low acid concentration ; (ii) by solvating the proton in the extraction of the transition metal halo-acids, e.g., H (TBP),FeCl, exists in the organic phase at high hydrochloric acid concentrations.18 Only the first possibility has been considered by most authors, e.g., the cobalt and iron species have been stated to be CoC12.2TBP and FeC1,.2TBP.l9 However, it has been shown in many instances that the second possibility is much more likely, and that the metal ions are extracted as “ion-association” complexes, e.g., cobalt may be extracted as (TBPH+),CoC1,2- and iron(II1) as (TBP)2H+FeCl,.1g It can be seen that the TBP-to-metal ratio is the same for both solvated and ion-associated formulae, and thus it is not possible to formulate the species unambiguously from a knowledge of this ratio alone.Morris and Short14 have shown that the complexes HZnC1,.3TBP and H2ZnC1,.2TBP, both of which are present in the TBP phase, exist as ion pairs [TBPHS] [ZnC13.2TBP-] and [TBPH’], [ZnC1J2-, respectively.It is well known that at high acid concentration, both cobalt and iron(II1) exist as their acid chloro complexes, and hence it is reasonable to assume that at high acid concentration the ion-association species predominates. From the similarities in the R, spectra of zinc, cadmium and mercury, it is also reasonable to assume that the species extracted will be ion-associated, rather than solvated. At high hydrochloric acid concentration, the acid itself is extracted into the TBP, the amount extracted becoming appreciable at concentrations above 7 M hydrochloric acid.20 The extraction of hydrochloric acid by TBP has been reviewed and several possible extraction mechanisms have been given.16 The extraction process is not a simple one and it appears that partly dissociated ion pairs of the type [H(H,O),]+[R(TBP solvated)]- co-exist in the TBP phase with TBP.H,O and un-ionised TBP.HA.More TBP is associated with the ion pair at low acid concentration and more hydrochloric acid is associated with the ion pair at very high acid concentration. From the above discussion, it is evident that the R, value of a metal ion in any TBP- hydrochloric acid system is dependent on the ability of the metal ion to form chloro complexes in the aqueous phase in order that the TBP can either solvate or ion-associate as TBPH+ with those complexes. The dividing line between solvation and ion-association is not clear, THE SPECIES EXTRACTED INTO TBP356 BARK, DUNCAN AND GRAHAM : REVERSED-PHASE THIN-LAYER [Analyst, Vol.92 and it may well be that many of the complexes showing TBP as a solvating molecule are actually ion-associations involving TBP-solvated protons (TBPH+) . The similarities between TBP-extraction behaviour and ion-exchange behaviour (discussed below) certainly support this view. COMPARISON BETWEEN THE RF SPECTRA AND CORRESPONDING LIQUID - LIQUID EXTRACTION BEHAVIOU R Ishimori, Watanabe and Nakamurall have plotted the partition coefficient of metal ions in pure TBP - hydrochloric acid systems against the concentration of the hydrochloric acid in the aqueous phase. The curves obtained by these workers compare favourably with the R, spectra of the same metal ions, i.e., an increase in partition coefficient with acid concentration can be compared with a decrease in R, value over the same range of acid concentration.In the acid concentration range 4 to 9 M hydrochloric acid, the alkali and alkaline earth metals have very low partition coefficients that increase slightly with increas- ing acid concentration. This compares favourably with very high R, values that decrease slightly with increasing acid concentration. The transition metals also compare f avourably , and the maxima in the extraction curves of zinc, cadmium, mercury, platinum and lead occur at the same acid concentration as the minima in the corresponding Rp spectra. Those metal ions which have partition coefficients above 100 generally have R, values of zero, e.g., niobium, tantalum, molybdenum, tungsten, iron( 111), tellurium and gold.From the extraction curves for niobium and tantalum, it should be possible to separate these metals at low acid concentration, although no separation was detected by us because of the difficulty in identifying these metals on the layers. Niobium is not readily extracted into TBP at low acid concentrations (M hydrochloric acid), whereas tantalum is. O’Laughlin and Banks4 obtained an RF value of 0.1 for niobium on TBP-treated papers eluted with 0.5 M hydrochloric acid, but they did not chromatograph tantalum. It is of interest to compare the R, values quoted by O’Laughlin and Banks4 with those obtained in the present study, and with the extraction results of Ishimori, Watanabe and Nakamura.ll It has already been shown that there is good agreement between the last two of these.The first of these gives the R, values of several metal ions on TBP-treated papers eluted with various concentrations of hydrochloric acid. Several of the RF values show a maximum at 9 M hydrochloric acid (yttrium, titanium, manganese, cobalt, nickel, copper, vanadium and chromium), while others show a minimum at 6 M hydrochloric acid (tin and mercury). Such behaviour is not in agreement with extraction results or the R, behaviour of the same metal ions in the present work. COMPARISON BETWEEN R, SPECTRA AND ION-EXCHANGE BEHAVIOUR Few authors have drawn attention to the similarities between the extraction behaviour of metal ions in the TBP - hydrochloric acid system and the anion-exchange behaviour of the same metal ions in a resinous ion exchanger - hydrochloric acid system.11s21 Ishimori, Watanabe and Nakamurall have shown that the shapes of the extraction curves compare favourably with the shapes of the ion-exchange curves and these authors are of the opinion that the mechanism of solvent extraction by TBP is similar to the ion-exchange mechanism.The maxima in the ion-exchange and extraction curves for zinc, cadmium and lead occur at about the same acid concentration as the minima in the RF spectra of these metal ions; all of the noble metals are strongly retained by the resin, corresponding to high TBP extraction and low Rp values. The alkali and alkaline earth metals are not appreciably retained by the resin, corresponding to poor extraction by the TBP and high RF values.Kraus and Nelson22 explained their results in terms of chloro-complex formation and they classified the metal ions into three groups according to their ion-exchange behaviour- (i) Those not adsorbed at any concentration (the alkali and alkaline earth metals, aluminium, nickel and thorium). (ii) Those which show increasing adsorption with increasing acid concentration and only good adsorption at high acid concentration. (iii) Those which show only decreasing adsorption with increasing acid concentration (metals of the central region of the periodic table), i e . , those metals which readily complex with chloride ion.June, 19671 CHROMATOGRAPHY OF METAL IONS WITH TRIBUTYL PHOSPHATE 357 is not 1.2. 3. 4. 5 . 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. The same classifications could equally well be applied to the R, spectra of the metal ions. Elution of zirconium from the resin resulted in tailing at low acid concentration. This tailing is in agreement with the tailing that zirconium exhibits in the TBP - hydrochloric acid chromatographic system. From the above considerations it appears likely that in the chromatographic systems used, the TBP extracts the metal ions by an anion-exchange process in which the metal - chloro complex exchanges with chloride associated with the TBP-solvated proton. We have previously suggested3 that the TBP in such a system should be considered to act more as an ion exchanger than as an extractant.[TBPH+]Cl-, viz., TBP-solvated protons associated in an “ion-association” system with chloride ions can exchange the chloride ion for negatively charged chloro complexes of the metal ions. Thus ion exchange results as the consequence of the ion-association of the TBPH+ and the metal - chloro complexes. Such a mechanism has been suggested by Morris and Short14 for the extraction of zinc into TBP. We suggest that the systems studied exhibit behaviour typical of ion-exchange chromato- graphy and are better regarded as such than as purely liquid - liquid extraction systems. Any difference may be one of technique rather than one involving the mechanism of the process. Ion-exchange chromatography may be regarded as a steady-state process and liquid - liquid extraction as an equilibrium phenomenon ; whenever TBP and metallochloro complexes are used the [TBPH],+[M,Cly]n- complex is probably formed and retained by the TBP.The stability of this complex and hence the degree of retention by the TBP determines the R, value of the metallochloro anion (and hence of the metal) and the partition coefficient. The positioning of the TBP and the method by which it is brought into contact with the other phase are all that determine whether the process is ion-exchange chromatography or solvent extraction. It is thus probable that other materials capable of giving ion-association systems and complex-ion systems of different stabilities may be of use in separating metals from one another. After the initial submission of this paper for publication, work was published by Pierce and Flint,23 and by Brinkman and V e l t k a m ~ , ~ ~ dealing with some of these metal ions chromato- graphed in reversed-phase systems with TBP.In general, the resultsz4 are in agreement with those recorded here; the results for copper(I1) and manganese(I1) reported by Pierce and Flint23 do not agree with the general findings. An interpretation of Pierce and Flint’s results possible in terms of the extraction - ion association described earlier. REFERENCES Duncan, G., Bark, L. S., and Graham, R. J. T., Proc. SOC. Analyt. Chem., 1966, 3, 145. Bark, L. S., Duncan, G., and Graham, R. J. T., Analyst, 1967, 92, 31. - _ _ - , , “Proceedings of the 4th International Symposium, Chromatography and Elec- trophoresis,” Belgian Pharmaceutical Society, Bruxelles, 1966, in the press. O’Laughlin, J. W., and Banks, C. V., Analyt. Chem., 1964, 36, 1222. Hu, 2.-T., Shi, S.-C., Acta Chim. Sin., 1964, 30, 352. Hu, 2.-T., Kexue Tongbao, 1966, 17, 166. Hu, 2.-T., Liu, C.-L., Acta Chim. Sin., 1964, 31, 267. Healy, T. V., and Brown, P. E., U.K. Atomic Energy Authority Report, AERE/C/R 1970, H.M. Bark, L. S., Graham, R. J. T., and McCormick, D., Talanta, 1965, 12, 122. Hashmi, M. H., Maqbool, A. S., and Ayaz, A. A., Ibid., 1955, 12, 713. Ishimori, T., Watanabe, K., and Nakamura, E., Bull. Chem. SOC. Japan, 1960, 33, 636. Keller, R. A., and Giddings, J. C., J . Chromat., 1963, 3, 205. MaCek, K., and Hais, I. M., “Paper Chromatography,” Academic Press, London, 1963, p. 158. Morris, D. F. C., and Short, E. L., J . Chem. Soc., 1962, 2662. Jerrgensen, Ch. K., “Complexes in Inorganic Chemistry,” Academic Press, London, 1963, p. 42. Marcus, Y., Chem. Rev., 1963, 63, 139. Gal, I. J., and Ruvarac, A., Bull. Inst. Nucl. Sci. Boris Kidrich, 1958, 8, 67. Majumdar, S. K., and De, A. K., Talanta, 1960, 7, 1. Chatelet, M., and Nicaud, C., C. 22. Hebd. Se‘anc. Acad. Sci., Paris, 1956, 242, 1471 and 1891. Kertes, A. S., J . Inorg. Nucl. Chem., 1960, 14, 104. Brinkman, U. A. Th., and De Vries, G., J . Chromat., 1965, 18, 142. Kraus, K. A., Nelson, F., Proc. I n t . Conf. Peaceful Uses of Atom. Energy, Geneva, 1958, 7, 837. Pierce, T. B., and Flint, R. F., J . Chromat., 1966, 24, 141. Brinkman, U. A. Th., and Veltkamp, H., Ibid., 1966, 24, 489. Stationery Office, London, 1956. Received Sefitember 29th, 1966