ANALYST, MARCH 1986, VOL. 111 295 Interferences of Antimony(V) in the Differentiation of Antimony(ll1) from Antimony(V) by Extraction with Ammonium Tetramethylenedithiocarbamate Using Graphite Furnace Atomic Absorption Spectrometry Etsuro Iwamoto,* Yasu hiko Inoike, and Yuroku Yamamotot Department of Chemistry, Faculty of Science, Hiroshima University, Hiroshima 730, Japan and Yasuhisa Hayashi Department of Chemistry, Joetsu University of Education, Joetsu 943, Japan Procedures for the preparation of antimony sample solutions for the differentiation of antimony(ll1) from antimony( V) by extraction with ammonium tetra met hylenedithioca rba mate (am mon iu m pyr rol idi ned it h io- carbamate, APDC) were examined. It was found that, when APDC is added to the antimony(\/) solution of pH less than ca.3, the antimony(\/) - APDC complex is partially co-extracted with antimony(ll1) over the pH range 3 . 5 1 0. Further, the mixing of antimony(ll1) solution with acidic antimony(\/) solution, prepared by oxidising antimony(ll1) potassium tartrate solution, leads to the incomplete extraction of antimony(ll1). A standard procedure for removing the interferences was established. Keywords: Antimon y(lll) determination; antimon y(V) interference; ammonium tetrameth ylene dithiocarba- mate; graphite furnace atomic absorption spectrometry; extraction Although the combination of solvent extraction with atomic absorption spectrometry (AAS) is very effective for the selective determination of mg 1-1 levels of arsenic(II1) and arsenic(V)1-3 or antimony(II1) and antimony(V) ,3-5 much care is needed in the treatment of such small amounts of those elements.Compared with arsenic, antimony is much more subject to the influence of many parameters associated with the extraction of antimony(II1) and antimony(V) because of their hydrolysis reactions. A critical examination of the variables involved in the extraction - spectrophotometric determination of antimony as the ternary chloro complex of Brilliant Green has been reported6: a hydrolysis side-reaction gives products that do not form extractable ion association systems with the dye. Al-Sibaai and Fogg7 used an extraction - spectrophotometric procedure with Brilliant Green and found that dilute standard antimony solutions (4 mg 1-1) prepared by dissolving anti- mony potassium tartrate in water and diluting the solution with water are stable over a period of 50 d, but similar dilute standard antimony solutions containing hydrochloric acid deteriorated rapidly.It was suggested that the effective loss of antimony could be caused by the formation of one or more soluble hydrolysed species but not by adsorption of antimony on the walls of the containers. Ammonium te t rame th ylenedi thiocarbamate (ammonium pyrrolidinedithiocarbamate, APDC) forms complexes with antimony that can be extracted into organic solvents such as isobutyl methyl ketone (IBMK) and nitrobenzene, and these have been used effectively for the selective determination of antimony(II1) and antimony(V) .3-5 However, only a few developments of precise analytical procedures concerning, especially, the preparation of standard sample solutions have been made and some problems concerning the partial extrac- tion of antimony(V) in the pH range 2-10 and interference from antimony(V) for extraction of antimony(II1) have remained.In this work, the extraction behaviour of the APDC - antimony system was studied. Dichloromethane (DCM) and IBMK were selected as solvents in place of nitrobenzene because of their lower boiling-points. * To whom correspondence should be addressed. t Present address: Fukui Institute of Technology, Gakuen 3-618, Fukui 910, Japan. Experimental Reagents All solutions were prepared from analytical-reagent grade chemicals and de-mineralised water, and were stored in polyethylene bottles. Stock antimony(ll1) solution, 10 mg 1-I. Prepared by dissolving 2.742 g of antimony potassium tartrate in water, diluting to 1000 ml with water, taking 10 ml of this stock solution and diluting to 1000 ml with water; no acid being added.Stock antimony(V) solution, 10 mg 1-1. (A) Prepared by oxidising the stock antimony(II1) solution with potassium permanganate, by taking 5 ml of the 1000 mg 1-1 antimony (111) solution, adding about 2 ml of sulphuric acid and 4 ml of 1% potassium permanganate solution and heating at ca. 80 "C for 30 min, adding hydrogen peroxide to remove the excess of permanganate and manganese dioxide produced, and diluting to 500 ml with 4 M hydrochloric acid. (B) Prepared by dissolving 2.778 g of potassium pyroantimonate {K[Sb(OH),].4H20} in water, diluting to 1000 ml with water to give a 1000 mg 1-1 solution and diluting 10 ml of this stock solution to 1000 ml with 4 M hydrochloric acid.APDC solution, 170 mIV. Buffer solution, p H 5.2. Prepared by mixing 1 M acetic acid Sodium tartrate solution, 1 % mIV. and 1 M sodium acetate in suitable proportions. Apparatus Atomic absorption measurements were made with a Nippon Jarrell-Ash Model AA-1 EW atomic absorption spectrometer equipped with a Model FLA-10 electrothermal atorniser and a Model HU-10 furnace. Peak heights were recorded with a Yanaco Model YR-110 chart recorder. A Hamamatsu TV antimony hollow-cathode lamp (L-223) was used as the light source. The background was checked by using a deuterium lamp. Samples were placed in the carbon tube with a Type 4700 Eppendorf pipette.An Iwaki Model KM shaking apparatus was used for solvent extraction.296 ANALYST, MARCH 1986, VOL. 111 General Procedure Preparation of sample solution Based on a critical examination of the experimental paramet- ers, the following procedure is recomended. Place ca. 80 ml of water and an aliquot of sample solution containing anti- mony(II1) and/or antimony(V) in a 100-ml calibrated flask. Add 2 ml of sodium tartrate solution, stir for 30-60 s, add 2-4 drops of methyl orange and then 2 M sodium hydroxide solution to adjust the pH of the sample solution to 4-5 and dilute to 100 ml. Antirnony(II4 determination Place an aliquot of sample solution containing not more than 1 pg of antimony(II1) in a separating funnel. Add 2 ml of APDC solution and 5 ml of acetate buffer solution.Dilute the mixture to 25 ml with water, the pH of the resulting solution being 5-6. Shake the funnel for 5 min with 10 ml of DCM or IBMK, allow it to stand for 30 min and separate the organic phase. Inject 20 yl of the organic phase with a micropipette into the carbon tube. Pass argon through the furnace at a flow-rate of 3 1 min-l, then atomise the sample with the following heating sequence: dry for 30 s at 30 A (ca. 300 "C), ash for 30 s at 70 A (ca. 700 "C) and atomise for 7 s at 230 A (ca. 2300 "C). Record the absorption signal at 217.6 nm. Run a reagent blank using the same instrumental settings and subtract the result from the analytical value. Total antimony determination Place an aliquot of a sample solution containing not more than 1 pg of antimony in a separating funnel and add 2 ml of APDC solution and 3 ml of 1 M hydrochloric acid.Dilute the mixture to 25 ml with water, the pH of the resulting solution being ca. 1. Carry out the extraction and measure the atomic absorption as for antimony( 111). The amount of antimony(V) is calculated from the differ- ence between the total antimony and antimony(II1). Results and Discussion Partial Extraction of Antimony(V) above pH 3.5 The differentiation of antimony(II1) from antimony(V) with APDC extraction is based on the principle that antimony(V) is not extracted above pH 3.5.3-5 However, it was found that when APDC is added to the antimony(V) solution at a pH below 3, followed by addition of the buffer solution, anti- mony(V) is partly extracted above pH 3.5.The effect of the acidity of the 0.1 mg 1-1 antimony(V) solution to which APDC is added on the extraction of antimony(V) for the APDC - DCM system is shown in Fig. 1. The 0.1 mg 1-1 standard solution was prepared by diluting the 10 mg 1-1 stock solution and adjusting the acidity with sodium hydroxide solution. The extraction, in which 10 ml of the standard solution (0.1 mg 1-1) were taken, was carried out at pH 5.2 and antimony was determined in the organic phase. The degree of extraction increased with increasing acidity, although antimony(V) was not extracted when APDC was added above pH 3. The behaviour was independent of the acid (hydrochloric, sulphuric and nitric acid) used to adjust the pH and organic solvents (DCM and IBMK). Stock solution B of antimony(V) gave the same result as stock solution A. Further, in tests lasting up to 12 d, the extractability of antimony(V) from each sample solution prepared from 10 mg 1-1 stock solutions A and B using 4 , l and 0.01 M acids was found to be unchanged, within experimental error.As antimony at low acidity tends to be adsorbed on the walls of glass containers, an attempt was made to check the amount of antimony adsorbed. The relative distribution of anti- mony(V) on extraction at pH 5.2 is given in Table 1 for pH 1.4 and 4 solutions of antimony(V). The percentage values for the O l d 0 PH Fig. 1. Effect of pH of the antimon (V) solution to which APDC is added on extraction at pH 5.2. Sb(V7, 1 pg Table 1. Distribution of antimony(V) in extraction at pH 5.2 Distribution, % pH1.4* pH4* Organic phase .. . . . . . . . . 39 2 Separating funnel . . . . . . . . . . 12 13 Aqueous phase . . . . . . . . . . 49 84 * pH values of the antimony(V) solution (0.1 mg 1-l). separating funnel refer to the amount of antimony adsorbed and present in small amounts of solvents adhering to the walls of the funnel after draining the organic and aqueous phases. Each concentration of antimony(V) was determined accord- ing to the procedure for the total antimony determination. As there is virtually no difference in the percentages for the separating funnel between the pH 1.4 and 4 solutions, it can be concluded that the non-extraction for the pH 4 solution is not due to adsorption. The above observations show clearly that a certain species of antimony(V) forms a complex which is extractable into the organic phase with APDC and the chloride anion is not responsible for its formation.Fig. 2 shows the distribution of species present using 10-5 M antimony(V) at 25 "C.8 It is interesting that the pH dependence of the extractability in Fig. 1 is very similar to that of the formation of Sb(OH)5 in Fig. 2. It seems likely that Sb(OH)5 forms a complex which is extractable in organic solvents with APDC whereas Sb(OH)6- does not, and once the complex has been formed below pH 3 it is stable at pH values higher than this. Interference from Antimony(V) for the Determination of Antimony( 111) In the absence of antimony(V), antimony(fi1) is completely extracted with APDC over a wide acidity range, from 4 M to pH lO.>5 However, it was found that the extraction of antimony(II1) is subject to interference from the presence of antimony(V), depending on the conditions of mixing of the antimony(II1) solution with the antimony(V) solution.Procedures for the preparation of sample solutions contain- ing antimony(II1) and antimony(V) are given in Table 2. The effect on extraction of the concentration of antimony(V) and the acidity of antimony(V) solutions in procedures I and I1 is shown in Fig. 3, where stock solution A was used for antimony(V). Evidently, the extraction of antimony(II1) is subject to interference from antimony(V) in stock solution A and the degree of interference increases with increasing acidity of the antimony(V) solution and the amount of antimony(V) present.In particular, when antimony(II1) was mixed with comparable amounts of antimony(V) in 4 M acid, antimony(II1) was not extracted at all. On the other hand, noANALYST, MARCH 1986, VOL. 111 297 Table 2. Procedures for the preparation of sample solutions Procedure Stock solution (10 mg 1-l) Standard solution (0.1 mg I - I ) d i l u t i o n y * loo m1 mixing- sample solution 1 I . . . . . . Sb(II1) - H20, 1 ml 'I Sb(V) - 4 M HCI, 1 ml-+ neutralisation? - dilution, 100 ml ' . ' ' ~ ~ ~ ~ ~ ~ mixing- neutralisation+ dilution, 100 ml- sample solution I I11 . . . . Sb(II1) - H 2 0 , 1 ml Sb(V)-4 M HCl, 1 ml + dilution (ca. 80 ml mixing + addition (tartrate) + neutralisation > dilution, 100 ml+ sample solution * Dilution with water. t Neutralisation with sodium hydroxide using phenolphthalein.$ Addition of auxiliary complex reagent (sodium tartrate or sodium citrate). I - 0 2 4 6 8 PH Fig. 2. Distribution of species present at 10-5 M Sb(V) in 0.5 M (CH&NCI at 25 "C. Taken from reference 8 80 60 E E .- UJ 40 . Y L a, r Y m t? 20 0 Fig. 3. Effect of amount of antimony(V) and acidity of antimony(V) solution A on the extraction of antimony(II1) with APDC - DCM at pH 5.2. Procedure I, 0; procedure 11, A, 0.01 M HC1; 0 , 1 M HCI; a, 4 M HCI. Sb(III), 1 pg interference was found for antimony(V) in stock solution B. This differs from the behaviour in the partial extraction of antimony(V) discussed in the preceding section. Sulphuric acid produced the same behaviour as hydrochloric acid and the use of methyl orange as an indicator also gave the same results.An increase in the amount of APDC did not increase the degree of extraction of antimony(III), showing that the amount present was sufficient. Although the antimony(II1) standard solution prepared from antimony potassium tartrate is very stable at the 1000 mg 1-1 level in aqueous solution, at low concentrations oxidation of antimony(II1) to antimony(V) may take place.9 Sun et ~ 1 . 1 0 reported that with the APDC - IBMK system no antimony(II1) was extracted at pH 6 in the absence of tartaric acid, which was added as a stabilising agent, but an equivalent Table 3. Recovery tests on antimony(II1) and antimony(\'). Dichloromethane was used as the solvent Antimony addedlpg I-' Antimony found/pg 1-1 Sb(II1) Sb(V) Sb(tota1) Sb(II1) Sb(V) 20 80 106 18 88 30 70 104 31 73 40 60 103 41 62 60 40 101 57 44 70 30 106 77 29 80 20 100 83 17 100 0 94 94 0 *O I 0 1 2 3 4 V/m Fig.4. extraction of Sb(II1) at pH 5.2. Sb(III), 1 pg; Sb(V), 1 pg Effect of volume of 1% sodium tartrate solution on the amount of antimony(V) was found by extraction at pH 1. In these tests lasting up to 6 d at the 1 mg 1-1 level, the dxrease in the degree of extraction at pH 5.2 was less than 10% in the absence of antimony(V). The rate of oxidation appears to depend on the quality of the water used to make the dilutions. Antimony(V) is reported to be hydrolysed rapidly even in 6 M hydrochloric acid.11 It is also speculated that more complex equilibria of hydrolysis would occur for solution A under acidic conditions, and a certain compound of antimony(V) forms a complex with antimony(II1) or adsorbs antimony(III), interfering with the complex formation of antimony(II1) with APDC.An attempt was made to remove this interference by adding auxiliary complexing agents before neutralisation according to procedure I11 in Table 2. It was found that sodium tartrate and sodium citrate are the most effective: antimony(II1) is protected from oxidation or interaction with antimony(V) species by complex formation with tartrate. Fig. 4 shows the effect of the concentration of sodium tartrate solution on the interference of antimony(V) on the extraction of anti- mony(II1) at pH 5.2. At least 2 ml of 1% sodium tartrate are required. The efficiency was independent of standing time for at least 2 h after the addition.EDTA showed no effect.298 ANALYST, MARCH 1986, VOL. 111 Addition of the auxiliary reagent to either antimony solution before mixing is equally effective. In our previous studies,3?4 the antimony(II1) solution was prepared according to proce- dure I. Subramanian and Meranger used ammonium citrate as a buffer solution.5 A recovery test on standard samples was carried out according to the standard procedure established here and satisfactory results were obtained, as shown in Table 3. Although the elucidation of the interference mechanism is beyond our present knowledge, it can be concluded that (i) in order not to extract antimony(V) over the pH range 3.5-10, the pH of the sample solutions must be higher than 3 when APDC is added, and (ii) an auxiliary complexing reagent, tartrate or citrate, is needed to prevent the interference of antimony(V) in the determination of antimony(II1) when using an acidic antimony(V) solution prepared from antimony potassium tartrate. Therefore, care must be taken, especially when using the method of standard additions, as unknown and standard solutions are often acidic. This research was supported in part by a Grant-in-Aid for Scientific Research (Nos. 57470031, 58540363 and 58030061) from the Ministry of Education, Science and Culture, Japan. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. References Karnada, T., Talanta, 1976, 23, 835. Yarnamoto, Y., and Kamada, T., Bunseki Kagaku, 1976, 25, 567. Chung, C. H., Iwamoto, E., Yarnamoto, M., and Yamamoto, Y., Spectrochim. Acta, Part B, 1984, 39, 459. Karnada, T., and Yamamoto, Y., Talanta, 1977, 24, 330. Subramanian, K. S . , and Meranger, J. C., Anal. Chim. Acta, 1981, 124, 131. Burke, R. W., and Menis, O., Anal. Chem., 1966, 38, 1719. Al-Sibaai, A. A., and Fogg, A. G., Analyst, 1973, 98, 732. Baes, C. F., Jr., and Mesmer, R., “The Hydrolysis of Cations,” Wiley, New York, 1976, p. 374. Andreae, M. O., Asmode, J.-F., Foster, P., and Van’t Dack, L., Anal. Chem., 1981, 53, 1776. Sun, H.-W., Shan, X.-Q., and Ni, Z.-M., Talanta, 1982, 29, 589. Newmann, H. M., J . Am. Chem. SOC., 1954, 76,2611. Paper A51314 Received September 9th, 1985 Accepted October 14th, 1985