ANALYST. JANUARY 1984. VOL. 100 31 Solid - Liquid Separation after Liquid - Liquid Extraction: Spectrophotometric Determination of Iron(l1) by Extraction of its Ternary Complex with 2,2’-Dipyridyl and Tetraphenylborate into Molten Naphthalene Masatada Satake and Toru Nagahiro Faculty of Engineering, Fukui University, Fukui 9 10, Japan Bal Krishan Puri* Department of Chemistry, Indian Institute of Technology, Hauz Khas, New Delhi-170016, India A method for the spectrophotometric determination of iron(l1) after extraction of its ternary complex with 2,2’-dipyridyl and tetraphenylborate into molten naphthalene has been developed. Iron(ll) reacts with 2,2’-dipyridyl to form a water-solu ble coloured complex. This complex cation forms a water-insoluble stable ternary complex in the presence of sodium tetraphenylborate, which is easily extracted into molten naphthalene in the pH range 2.8-7.6 by vigorous shaking for a few seconds. The solid naphthalene containing the iron 2,2’-dipyridyl tetraphenyl borate complex is separated by filtration and dissolved in acetonitrile.The absorbance is measured at 521 nm against a reagent blank. Beer‘s law is obeyed in the concentration range 2.2-65.5 pg of iron in 10 ml of acetonitrile solution. The molar absorptivity and Sandell’s sensitivity are 8.89 x lo3 I mol-1 cm-1 and 0.0063 pg cm-2 at 521 nm, respectively. The interference of various ions has been studied in detail. The method has been applied to the determination of iron in standard metallic samples and the results are compared with those obtained by the 1 ,lo-phenanthroline method.Keywords: Solid - liquid separation; spectrophotometry; iron determination; metal and alloy analysis 2,2‘-Dipyridyl has been widely employed in the detection and determination of certain metal ions. 1-3 It reacts with iron(I1) to form a water-soluble, deep red complex, Fe(di~y)~z+, which forms water-insoluble, stable ternary complexes with anions such as C104-, SCN-, Hg142-, HgBr42- and Cd142-. As a result, some metal ions and anions can be extracted into suitable organic solvents as ternary complexes.&6 This further increases the sensitivity and an improvement in selectivity can be achieved. Since 1969, we have developed a method7-11 for the spectrophotometric determination of trace metals using naph- thalene as an extractant. In this study, the technique was used for the extraction of iron(II), which forms a stable ternary complex with these complexing reagents at high temperatures. We have observed that the complex cation Fe(di~y)~2+ reacts with the tetraphenylborate anion to form a very stable, water-insoluble, coloured ternary complex.This ternary complex cannot be extracted into most organic solvents, such as benzene, toluene, xylene, dichloroethane, nitrobenzene, isoamyl alcohol, isobutyl methyl ketone and isoamyl acetate because of the low solubility of the complex in these solvents at room temperature. However, the complex can be quantit- atively extracted merely by contact with molten naphthalene or by vigorous shaking for a few seconds. The interference of various ions has been studied. The method has been applied to the determination of iron in alloys and metals with extraction of the ternary complex into naphthalene, and the results were compared with those obtained by the 1 ,lo-phenanthroline method.12 The main advantage of the present method is that equilib- rium distribution of the complex in the two phases is attained in a few seconds owing to the high temperature. As only a small amount of the organic solvent (1.2 g) is required for the complete extraction of the metal ion using about 50 ml of the aqueous phase, the sensitivity of the method is enhanced as the whole phase may be taken for the analysis.Ordinary * Author to whom correspondence should be addressed. non-aqueous organic solvents have more or less the same solubility in water, which causes errors in the determination of metals owing to the change in the volume of the organic to the aqueous phase, whereas naphthalene is a solid crystal, immiscible with water and completely separated from the aqueous phase at room temperature.Hence the proposed method is the most suitable for the extraction of the complex without encountering difficulties in phase separation. The iron( 11) dipyridyl complexes, which are produced with colour- less inorganic anions as described above, are unstable and dissociated in aqueous solution at high temperature, but in the presence of the organic tetraphenylborate anion and complex formed is very stable and hardly dissociated. Hence the procedure can be successfully applied to the extractive spectrophotometric determination of iron.Experimental Equipment A Hitachi, Model 200-20, double-beam spectrophotometer was used for the absorbance measurements. pH measure- ments were made with a Toa-Dempa HM-SA pH meter, equipped with a combined calomel and glass electrode assembly. Reagents Doubly distilled water and analytical-reagent grade acids and salts were used throughout unless stated otherwise. Standard iron(ZII) solution, 5 p.p.m. Dilute 5ml of 1000 p.p.m. standard iron(II1) solution to 1 000 ml with water. 2,2’-Dipyridyl solution, 0.4%. Dissolve 0.4 g of 2,2’- dipyridyl in a small volume of concentrated hydrochloric acid and dilute with distilled water to 100 ml in a calibrated flask. Acetic acid - ammonium acetate buffer. Mix 1 M acetic acid and 1 M ammonium acetate solution in suitable amounts to give buffer solutions in the pH range 3-6.32 ANALYST, JANUARY 1984, VOL.109 Ammonia - ammonium acetate buffer. Mix 1 M ammonia solution and 1~ ammonium acetate solution in suitable amounts to give buffer solutions in the pH range 8-11. Naphthalene. Check the purity spectrophotometrically before use in the range 200-700 nm. Acetonitrile. Check the purity spectrophotometrically before use in the range 200-700 nm. Sodium tetraphenylborate solution, 1 Yo. Hydroxylammonium chloride solution, 2%. General Procedure To an aliquot of iron solution in a beaker add 2.0 ml of 2% hydroxylammonium chloride, 2.5 ml of 0.4% 2,2'-dipyridyl solution and adjust the pH to 4.5 with 2.0 ml of acetate buffer solution and dilute aqueous ammonia.Transfer the solution into a 100-ml round-bottomed flask, add 2.0 ml of 1% sodium tetraphenylborate solution and allow the mixture to stand for 1-2 min. The flask was heated to about 60 "C on a water-bath. Add 1.2g of naphthalene, stopper the flask and continue to heat until the naphthalene melts. Remove the flask from the water-bath and shake it vigorously until naphthalene solidifies into very fine crystals. Separate the naphthalene from the aqueous phase by filtration through a small glass filter-funnel and dry it in an oven. Dissolve the naphthalene in acetonitrile and dilute it to lOml with acetonitrile in a calibrated flask. Place a portion of this solution in a 1-cm cell and measure the absorbance at 521 nm against a reagent blank. Results and Discussion Absorption Spectra The absorption spectra of the reagent and ternary iron(I1) complex in naphthalene - acetonitrile solution were measured against water (Fig.1). The iron(I1) - 2,2'-dipyridyl tetra- phenylborate complex shows maximum absorption in the range 518-523 nm, where the reagent does not absorb. All absorbance measurements were made at 521 nm in subsequent studies, Effect of pH The effect of pH on the extraction of the complex was studied at 521 nm with a sample containing 30 yg of iron, keeping the other conditions constant. A graph of absorbance against pH (Fig. 2) showed that the absorbance of the complex is dependent on pH, the maximum absorbance being obtained between pH 2.8 and 7.6. A pH of 4.5 was chosen for all subsequent measurements.Effect of Amount of Reducing Agent Various amounts of hydroxylammonium chloride were added to a solution containing 301.18 of iron(II1) and the general procedure was followed. Iron(II1) was reduced to iron(I1) quantitatively when 0.5-12 ml of 2% hydroxylammonium chloride solution were used. Therefore, 2.5 ml of 2% solution were used in all subsequent work. Effect of Reagent Concentration Extractions of the complex were carried out at fixed pH with 2.0 ml of 1% tetraphenylborate solution, and with the addition of various volumes of 0.4% 2,2'-dipyridyl solution (Fig. 3). The extractions were quantitative when more than 0.3 ml of 0.4% solution was used. Effect of Tetraphenylborate Concentration Various volumes of 1 % tetraphenylborate solution were added to a sample solution containing 30 pg of iron and 2.5 ml of 0.4% 2,2'-dipyridyl solution at pH 4.5, the general ' 0.: I 1 1 420 460 500 540 580 Wavelengt h/n rn Fig.1. Absorption spectra of 2,2'-dipyridyl - tetraphenylborate and Fe(I1) - 2,2'-dipyridyl- tetraphenylborate in naphthalene - acetonitrile solution. Fe(III), 30 pg; pH, 4.5; 2% NH20H.HCI, 2.0ml; 0.4% 2,2'-di yridyl, 2.5 ml; 1% tetraphenylborate, 2.0ml; naphthalene, 1.2 g; sgaking time, 30 s; reference, water o'6 ' I f I I I I 1 0 2 4 6 8 PH Fig. 2. conditions as in Fig. 1 Effect of pH on extraction. Wavelength. 521nm: other I I I 1 6 1 2 3 4 5 6 0.4% D i p y r i d y I /m I Fig. 3. in Fig. 2 Effect of reagent concentration on extraction. Conditions as 1 2 3 4 1 2 3 4 5 0.1% Kalibor/rnl 1% Kalibor/rnl Fig.4. Conditions as in Fig. 2 Effect of tetraphenylborate concentration on extraction.ANALYST, JANUARY 1984, VOL. 109 33 0.6 I 1 0.2 ' I I I I 20 40 60 80 100 Volume of aqueous phase/ml Fig. 5. Effect of aqueous phase volume on extraction. Conditions as in Fig. 2 procedure being followed. Extraction was complete when the volume of 1% tetraphenylborate solution was in the range 0.1-5.0ml (Fig. 4). Therefore, 2.0 ml of 1% tetraphenyl- borate solution were used in subsequent work. Effect of Buffer Solution It was found that the addition of 0.5-5.0ml of acetate buffer solution caused virtually no variation in the absorbance. Hence in all experiments 2.0 ml of buffer solution were used. Effect of Addition of Naphthalene The amounts of naphthalene were varied from 0.4 to 2.0 g and the extractions were carried out by the general procedure.The extraction was quantitative in this range, but above 2.0 g it was difficult to dissolve the naphthalene in the limited amount of acetonitrile (10ml). Hence 1.2g of naphthalene was used as the most suitable amount. Effect of Volume of Aqueous Phase As the volume of the organic phase is small compared with that of the aqueous phase, it was essential to study the effect of the volume of the aqueous phase on the extraction. When the latter was varied between 10 and 100m1, the absorbance remained constant up to a volume of 55ml. Above this volume, the extraction was not quantitative (Fig. 5). Effect of Shaking Time The extraction of the complex into molten naphthalene (81-85OC) was found to be very rapid and no change was observed when the shaking time was varied from 3 to 120s. Effect of Standing Time after Addition of Reagents To study the effect of standing time after addition of the reagents, the iron(I1) - 2,2'-dipyridyl tetraphenylborate com- plex in a solution containing 30 pg of iron was allowed to stand at room temperature for 1-30min and then extracted into molten naphthalene. The rate of formation of the complex was very fast even at room temperature and a standing time of 1-2 min before the extraction was sufficient.Effect of Standing Time A mixture of the ternary complex and naphthalene was dissolved in acetonitrile and the effect of standing time on the absorbance was investigated. The ternary complex in naphthalene - acetonitrile was stable for up to 10 d.Beer's Law and Sensitivity Using the optimum conditions described above, a calibration graph for iron was constructed at 521 nm and was found to be Table 1. Tolerance limits for diverse ions and salts. Iron, 30 pg; pH, 4.5; naphthalene, 1.2 g Tolerance Tolerance Ion Cr6+ . . . . . . . . co2+ . . . . . . Zn2+ . . . . . . Ni2+ . . . . . . . . cu+ . . . . . . . . Cd2+ . . . . . . Sn2+ . . . . . . . . Bi3+ . . . . . . . . Pb2+ . . . . . . . . Mg2+ . . . . . . Hg2+ . . . . . . limit/pg 5 000 2 000 8 000 100 300 800 400 100 8 000 150 3 000 Salt limit/mg NaI . . . . . . 7000 K C I O ~ . . . . . . 300 KN03 . . . . . . 3000 CH,COONa.3H20 4 000 KSCN . . . . . . 1000 Na2S04 . . . . 2000 NaCl . . . . . . 1000 Na2C204 . . . . 1 KH2P04 . . . . 400 KCN .. . . . . 1 Na2tartrate.2H20. . 100 Na3citrate.2H,0 . . 5 linear over the concentration range 2.2-65.5 pg of iron in 10 ml of acetonitrile. The molar absorptivity and Sandell's sensi- tivity were calculated to be 8.89 X 103 1 mol-1 cm-* and 0.0063 pg cm-2 of iron, respectively (for an absorbance of 0.001) at 521 nm. Ten replicate analyses of a sample solution containing 30 yg of iron gave a mean absorbance of 0.477 with a standard deviation of 0.0029 and a relative standard deviation of 0.6%. Choice of Solvent Various organic solvents for dissolving the mixture of the ternary complex and naphthalene were tried. The complex is soluble in dimethylformamide, dimethyl sulphoxide and acetonitrile, but insoluble in dioxane, chloroform, benzene, toluene, xylene , chlorobenzene, o-dichlorobenzene , isoamyl acetate, isobutyl methyl ketone, carbon tetrachloride, nitro- benzene, 1,2-dichloroethane, diethyl ether, hexane, amyl alcohol, isoamyl alcohol, butan-1-01 and ethyl acetate.It is unstable in dimethylformamide and the resulting solution is unsuitable for absorbance measurements. Hence acetonitrile was the most suitable solvent for dissolving the complex. Effect of Diverse Ions Sample solutions containing 30 yg of iron and various amounts of different alkali metal salts or metal ions were prepared and the determination of iron was studied. The pH of the solution was adjusted to 4.5 and the general procedure was applied. The tolerance limits for diverse ions are given in Table 1. Obviously the method is fairly selective. Determination of Iron in Standard Materials and Metal Alloy Samples The procedure has been applied successfully to the determina- tion of iron in standard materials and real samples.The results are in reasonably good agreement with those obtained by the conventional spectrophotometric procedure using 1 , 10- phenanthroline12 (Table 2). Silicon - bronze alloy (NBS, SRM-158a) A 100-mg amount of this alloy was dissolved in a mixture of 10 ml of hydrochloric acid (1 + 1), 1 ml of concentrated nitric acid, 3ml of concentrated sulphuric acid and l m l of 30% hydrogen peroxide. The mixture was then gently heated on a boiling water-bath. The excess of the acids was evaporated and the volume of the sample was adjusted to 200ml with water. A 3.0-ml aliquot of this sample was placed in a separating funnel and 40 ml of hydrochloric acid (1 + 1) were added.The iron in this sample was extracted by vigorous shaking for 5 min with 20 ml of isobutyl methyl ketone. It was34 ANALYST JANUARY 1984, VOL. 109 Conclusion Table 2. Determination of iron in standard materials and real samples Sample SRM-158a silicon - bronze alloy NBS NBS SRM-85b aluminium alloy . . . aluminium Metallic Iron can be determined spectrophotometrically using 2,2‘- dipyridyl but many metal ions interfere and the sensitivity is low. Liquid - liquid extraction can improve the selectivity but it cannot be applied in the present instance as the ternary complex has low solubility in almost all organic solvents or the interfacial phase separation is not clear. The proposed technique has not only solved this problem, but has also enhanced the sensitivity of the method because only 1.2 g of naphthalene is required for the complete extraction of the metal ion.Metal ions that form thermally unstable complexes can interfere in liquid - liquid extraction spectrophotometric methods, but have no effect in the proposed technique, Concentration ,lO-phenanth- Certified proposed method, rolirie method, Concentration obtained by Obtained by value, % % * /O o * 1.23 1.25 f. 0.02 1.24 k 0.02 0.24 0.233 f. 0.002 0.239 k0.002 resulting in better selectivity. (powder) . . - 0.133 kO.001 0.137 kO.001 CoS04.7H20 - 0.004 3 f 0.000 2 0.004 5 rt: 0.000 1 * Average for five individual samples. then back-extracted from the organic phase with 25ml of water and then the proposed method or the 1,lO- phenanthroline procedure was applied.Standard aluminium alloy (NBS, SRM-85b) or metallic aluminium powder A 1.0-g amount of the aluminium alloy or 2.0 g of aluminium powder were accurately weighed and then placed into a 100-ml beaker t o which were added 40-50ml of hydrochloric acid (1 + 1) and 3.0 ml of 30% hydrogen peroxide. On heating on a boiling water-bath, the mixture dissolved completely and the excess of hydrogen peroxide decomposed. After cooling, the sample was diluted to 200 ml with water in a calibrated flask. An aliquot of this sample solution was taken and the proposed procedure was followed. Analysis of Cobalt Salt A salt (CoS04.7H20) was dissolved in water and then analysed as described above for the standard material sample. References 1. Welcher, F. J . , “Organic Analytical Reagents,” Volume 111, Van Nostrand, London, 1964, p: 68. 2. Rao, A. L. J., and Puri, B. K., Zh. Anal. Khim., 1973,28,183. 3. Schilt, A. A., “Analytical Applications of 1 ,lO-Phenanthroline and Related Compounds,” Pergamon Press, Oxford, 1969. 4. Poluektov, N. S., and Nazarenko, V. A., J . Appl. Chem. U.S.S.R., 1937, 10, 2103. 5. Yamamoto, Y., and Kotsuji, K., Bull. Chem. Soc. Jpn., 1964, 37, 594. 6. Kotsuji, K., Yoshimura, Y., and Ueda, S., Anal. Chim. Acta, 1968, 42, 225. 7. Fujinaga, T., Satake, M., and Yonekubo, T., Bunseki Kagaku, 1970, 19, 216. 8. Fujinaga, T . , Satake, M., and Yonekubo, T., Bull. Chem. SOC. Jpn., 1973,46,2090. 9. Satake, M., and Yamauchi, T., Mem. Fac. Eng. Fukui Univ., 1977, 25, 107. 10. Satake, M., Anal. Chim. Acta, 1977, 92, 423. 11. Puri, B. K., and Gautam, M., Mikrochim. Acta, 1979, I, 515. 12. American Public Health Association, American Water Works Association and Water Pollution Control Federation, “Stan- dard Methods for the Examination of Water and Waste Water,” Thirteenth Edition, American Public Health Associa- tion, New York, 1971, p. 189. Paper A311 22 Received May 5th, 1983 Accepted August Ist, 1983