286 Analyst, May, 1968, Vol. 93, +$. 286-288 Determination of Antimony in Rocks by Neutron-activation Analysis BY A. 0. BRUNFELT AND E. STEINNES (Mineralogical-Geological Museum, University of Oslo, Sarsgt. 1, Oslo 6, Norway) (Inslitutt for Atomenergi, Isotofie Laboratories, Kjeller, Nwway) Antimony has been determined in some geochemical standards (andesite AGV-1, diabase W-1, dunite DTS-1, granite GB and tonalite T-1) by neutron-activation analysis. A radiochemical procedure has been developed, in which antimony(V) is extracted into isopropyl ether and subsequently back-extracted after reduction to the tervalent state with tin(I1). The antimony activity is measured by y-spectrometry. Chemical-yield determina- tion is carried out by re-activation. The precision of the method is better than 5 per cent.for samples with antimony content exceeding 1 p.p.m. NEUTRON-ACTIVATION analysis can be used to advantage for the determination of antimony at low concentrations in geochemical samples. Antimony has been determined by this technique in rnete~rites,l*~*~ s4 tektites: silicate rock.~,~s~,6 and rnineral~.~ s5 To facilitate an efficient determination of micro amounts of this element, a specific radiochemical procedure is required before the radioactivity measurements are made. Tanner and Ehmann4 have used a radiochemical procedure based on multiple sulphide precipitations, followed by a reduction step to metallic antimony, while KiesP has used a procedure involving distillation, anion-exchange separation and sulphide precipitation. The procedure of Kiesl was part of a radiochemical separation scheme that also allowed for the simultaneous separation of other elements.The present paper presents a radiochemical procedure based on the extraction of antimony(V) into isopropyl ether. The antimony is back-extracted from the organic phase by reduction to antimony(II1) with tin(II), and subsequently precipitated as sulphide. The chemical-yield determination is carried out by re-activation. The silicate rocks analysed were some primary geochemical reference samples listed in Table I. EXPERIMENTAL APPARATUS- crystal was used. REAGENTS- A 400-channel y-spectrometer with a well-type, 3 x 3-inch, sodium iodide (thallium) in Reagents of analytical-reagent grade quality were used. Antimony standard solution-Prepare a stock solution by dissolving 25 mg of the metal a few millilitres of aqua regia and diluting with a solution of 0.1 M citric acid and M hydro- chloric acid to give a concentration of 50 pg of antimony per ml.Antimony carrier solution-Prepare a stock solution by dissolving antimony in con- centrated hydrochloric acid to give a solution containing 5 mg of antimony per ml. IRRADIATION- Finely crushed rock samples of about 100mg were accurately weighed into small polythene bags that were then heat-sealed. About 0.7 ml of the antimony standard solution was sealed in a quartz ampoule. The irradiation of the samples and standards was carried out in the reactor JEEP-1 (Kjeller, Norway) at a thermal-neutron flux of about 2*1012 neutrons per cm2 per second for 3 days.The irradiated samples were stored for 3 days to allow the short-lived activities to decay. 0 SAC and the authors.BRUNFELT AND STEINNES 287 RADIOCHEMICAL PROCEDURE- Open the polythene bag containing the sample and transfer it into a 250-ml polypropylene beaker containing 1.00ml of the antimony carrier solution. Add a mixture of 10ml of hydrofluoric acid - nitric acid (1 + 1). Heat the beaker on a water-bath, and stir with a polythene rod to ensure the complete dissolution of the rock powder. After evaporation to dryness, dissolve the residue in 10 ml of 9 M hydrochloric acid. Transfer the solution into a 100-ml glass beaker and heat with about 0-25 ml of bromine. Remove the excess of bromine by boiling. After cooling, transfer the solution into a 250-ml separating funnel and extract twice with 20 ml of di-isopropyl ether.Discard the aqueous phase, and wash the combined ether phases twice with 2 ml of 9 M hydrochloric acid. Back-extract antimony with 40 ml of a 3 M hydrochloric acid solution containing 50 mg of tin(I1) as tin(I1) chloride. Transfer the aqueous phase into a 100-ml glass beaker and precipitate antimony as antimony tn- sulphide with about 200mg of thioacetamide. Filter the precipitate on to a membrane filter, transfer it into a 50-ml glass beaker and dissolve in about 5.5 ml of concentrated hydro- chloric acid. With a pipette, withdraw 5.00 ml of the solution and transfer it into a polythene vial for y-activity measurements. Withdraw 0.500 ml of the standard solution from the quartz ampoule and dilute to 250 ml with concentrated hydrochloric acid.Transfer 5-00 ml of this solution into a counting vial. ACTIVITY MEASUREMENTS- The samples and standards were counted inside the well of the scintillation crystal for 5 to 20 minutes, depending on the activity level. The y-measurements were based on the 565-KeV photopeak of antimony-122 (half-life 2-75 days), which also contained a minute contribution of the 600-KeV y-ray of antimony-124. The area of the peak was evaluated according to the method of Covell.7 The radiochemical purity was checked by repeating the measurements after 3 to 4 days. DETERMINATION OF CHEMICAL YIELD- After measuring the y-activity, the samples were diluted to 100 ml with 0.4 M citric acid. About 1.2 ml of each solution were sealed in a polythene vial and activated for 3 hours, together with aliquots of the carrier solution diluted in the same way, at a thermal flux of about 2*1012 neutrons per cm2 per second.After the re-activation, the solutions were allowed to “cool” for 1 day. Aliquots of 1.00 ml were transferred into polythene vials for y-activity measurements. The y-counting was again based on the 565-KeV photopeak of antimony-122. The chemical yield was rather low, typically 30 to 50 per cent., but the re-activation technique renders precise chemical-yield determination possible, even if the yield is as low as a few per cent. RESULTS AND DISCUSSION The experimental results obtained by the present method for five different reference rock samples are presented in Table I. The relative standard deviation appears to be below 5 per cent. for silicate rocks with an antimony content higher than 1 p.p.m.At concentrations in the range 0.1 to 1 p.p.m., the precision seems to exceed 5 per cent. TABLE I RESULTS IN P.P.M. OBTAINED FOR ANTIMONY BY NEUTRON ACTIVATION OF STANDARD ROCKS Relative standard deviation Rock samples Experimental results Mean of single value, per cent. Andesite AGV-I* . . . . 4.33, 4.52, 4.27, 4.18, 4-44 4.35 3.1 Diabase W-l* . . . . 0.91, 0-87, 0.93, 0-92, 0.89 0.90 2.7 Dunite DTS-I* . . . . 0.53, 0.41, 0.51, 0.57, 0.49 0.50 11.6 GraniteGBt .. . . 0.21, 0.21, 0-25, 0.24, 0.24 0.23 8.1 Tonalite T-l$ . . . . 0.73, 0.68, 0.77, 0.76, 0.75 0.74 4.7 Rock suppliers : * U.S. Geological Society. t Carpatho-Balkan Geological Science. $ Geological Survey of Tanganyika.288 BRUNFELT AND STEINNES TABLE I1 ANTIMONY CONTENT IN P.P.M.OF U.S. GEOLOGICAL SURVEY STANDARD ROCKS Neutron-activation method Spectrophotometric Mass Pre'sent Rock sample method spectrometer Previous results results AGV-1 .. .. - - 5-42, 2.214, 4.76* 4.30 w-1 .. .. 1-28 0*809, 0*8'O, 0.3'1 0.9512, 0 ~ 1 4 ~ , 0.96, l*15t 0.90 DTS-1 .. .. - - 0.53, 0-404 0.60 * Relative determination to W-1 with assigned value of 0.90 p.p.m. t Results obtained with different bottles of W-1. In Table 11, the results are compared with those given in the available literature, which were obtained partly by neutron activation and partly by different analytical techniques. The present value of 0.90 p.p.m. for sample W-1 is in good agreement with that of other workers who used neutron-activation analysis, the only exception being the value of 0.14 p.p.m.reported by Tanner and Ehmann.4 The small differences may well be caused by significant differences in the antimony content of different bottles of W-1, as indicated by the results of Esson, Stevens and Vin~ent.~ The agreement with results obtained by the use of other techniques must also be considered as satisfactory.* y9 ,lo Serious disagreement with the results of Tanner and Ehmann4 has also occurred for sample AGV-1, and to a lesser extent for sample DTS-1. The present value for AGV-1, however, is supported by the work of Gorden, Randle, Goles, Corliss, Beeson and Oxley,6 who have determined the ratio of the amounts of antimony in AGV-1 and W-1 by using lithium-drifted germanium detectors in connection with instrumental activation analysis.The good agreement between other workers indicates that systematic errors yielding low results must be inherent in the results of Tanner and Ehmann. The presence of large systematic errors in the present method seems improbable. No y-emitting impurities were detected in the activity measurements. Possible interference from competing nuclear reactions, i.e. , uranium fission or (n,p) reaction in tellurium-122, is negligible, as the independent fission yield of the shielded nuclide antimony-122 is very low, and tellurium rarely occurs in silicate rocks. Neutron-shielding effects are also insignificant, because none of the major elements in the rocks is a strong neutron absorber, and the dilution of the standard was sufficient to avoid self-shielding caused by the large resonance integral of antimony-121.The accuracy of the present method is, therefore, in all probability about & 10 per cent. For samples with antimony content above 1 p.p.m., the accuracy is perhaps as good as + 5 per cent. The sensitivity of this method is adequate for the determination of antimony in silicate rocks, with sufficient precision and accuracy for general studies in geochemistry. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. REFERENCES Smales, A. A., Mapper, D., Morgan, I. W., Webster, R. K., and Wood, A. J., Int. Conf. Peaceful Hamaguchi, H., Nakai, T., and Endo, T., Nippon Kugakzt Zusshi, 1961, 82, 1485. Kiesl, W., 2. analyt. Chem., 1967, 227, 13. Tanner, J. T., and Ehmann, W. D., Geochim. Cosmochim. Acta, 1967, 31, 2007. Esson, I., Stevens, R. H., and Vincent, E. A., Min. Mug., Lond., 1965, 35, 88. Gorden, G. E., Randle, K., Goles, G. G., Corliss, J. B., Beeson, M. H., and Oxley, S. S., Geochim. Covell, D. F., Anulyt. Chem., 1959, 31, 1785. Ward, F. N., and Lakin, H. W., Ibid., 1954, 26, 1168. Taylor, S. R., Nature, 1965, 205, 34. Nicholls, G. D., Graham, A. L., Williams, E., and Wood, M., Analyt. Chem., 1967, 39, 584. Brown, R., and Wolstenholme, W. A., Nature, 1964, 201, 598. Hamaguchi, H., Kuroda, R., Tomura, K., Osava, M., Watanabe, K., Onuma, N., Yasunaga, T., Received December 18th, 1967 Uses Atom. Energy, Geneva, 1958.2, 242, paper 282. Cosmochim. Actu, 1967, in the press. Hosohara, K., and Endo, T., Geochim. Cosmochim. Acta, 1961, 23, 296.