402 KAKABADSE AND WILSON THE COMPLEX OF VANADYLV [Vol. 86 The Complex of VanadylV with Ethylenediaminetetra-acetic Acid BY G. KAKABADSE AND H. J. WILSON (Department of Chemistry, College of Science and Technology, Manchesttv 1) From conductimetric measurements, the existence of a 1 to 1 complex of vanadylv ions and ethylenediaminetetra-acetic acid is established, and its application to volumetric analysis is suggested. The ratio of hydrogen ions to vanadium in the initial solution is critical and must exceed 5 ; the pH must be above 1-8. Conductimetric end-point detection is compared with a method in which xylenol orange is used as indicator. COMPLEXOMETRIC investigations of the vanadylV ion, V02+, with phosphate, citrate, tartrate, oxalate and ethylenediaminetetra-acetic acid (EDTA) ions have revealed marked comples- ation around pH 2 with oxalate and EDTA, and an attempt was made to define the quant- itative character of the V02+ - EDTA complex1 by the titration technique.It is obvious that the existence of this complex will, to a large extent, depend on the hydrogen-ion concentration, i.e., the minimum acidity required for its quantitative formation from the vanadate ion2- whereas the maximum acidity will be determined by the general stability requirements of a metal - EDTA complex- With increasing acidity, this equilibrium will be displaced to the left and the stability of the complex decreased? In this investigation the ratio of hydrogen ions to vanadium was varied between 1 to 1 and 60 to 1 and the vanadium concentration between 0.015 and 0.009 gram-atom of vanadium per litre.Owing to a rather limited choice of low-pH complexometric indicator^,*^^^^ end- point determinations were in the first instance carried out conductimetrically. When the existence of the complex had been established the indicator method was tried, with xylenol orange as indicator. EXPERIMENTAL REAGENTS- . . - - (1) .. - - (2) V0,- + 2H+ + VO,+ + H20 . . . . Men+ + H,Y2- + (MeY)n-4 + 2H+ . . Ammonium metavanadate-Analytical-reagent grade. Perchloric acid, 60 per cent. w,h-Analytical-reagent grade. EDTA solution-A standard solution of the disodium salt of ethylenediaminetetra- acetic acid. XyZenoZ orange solution-A 0.1 per cent. solution in dilute ethanol. PREPARATION OF STOCK VANADYL SOLUTION- Dissolve 1.1699 g of ammonium metavanadate in approximately 400 ml of distilled water, heat to assist solution, add 60 ml of M perchloric acid, heat to boiling to ensure complete conversion of the polyvanadate to the pale-yellow vanadyl ion, and make up to 1 litre with distilled water.This stock solution is 0.01 M with respect to vanadium and 0.06 M in perchloric acid. Alternatively, the vanadyl stock solution has been prepared by passing ammonium metavanadate solution through a column containing Zeo-Karb 225 resin' and receiving the solution of polyvanadic acid in perchloric acid of the requisite concentration. For conducti- metric work the latter method has the advantage of eliminating ammonium ions. Its dis- advantage is the need for subsequently determining vanadium, some of which is lost by adsorption on the resin.TITRATION- Transfer 5 ml of the stock vanadyl solution to a conical flask, dilute to the volume shown in Table 111, p. 405, add 5 drops of xylenol orange solution,s and titrate in the cold withJune, 19611 WITH ETHYLENEDIAMINETETRA-ACETIC ACID 403 0.01 M EDTA solution until the colour changes from red to pure yellow. Pyrocatechol violetg did not show a colour change. For conductimetric titration (see Figs. 1 and 2), transfer 25ml of the stock solution to a conductivity titration flask, dilute to about 200ml with distilled water, and titrate at room temperature with 0-05 M EDTA solution to minimise dilution effects. 2-o c Volume of 0.01037 M EDTA, ml Fig. 1. Conductimetric titrations of solutions (200 ml) containing polyvanadic and perchloric acids (concentrations as shown in Table I) with 0 .0 5 0 ~ EDTA a t about 20" c 2.4c L I Volume of 0.05 M EDTA, mi Fig. 2. Conductimetric titrations of solutions (200 ml) containing ammonium metavanadate and perchloric acid (con- centrations as shown in Table 11) with 0.01037 M EDTA at about 20" C Accurate conductivity measurements (see Fig. 3) were made at 25" i 0.001" C on solutions prepared individually with conductivity water. In each test 10 ml of vanadyl stock solution (polyvanadic acid plus perchloric acid), being 0.01845 M in vanadium and having a ratio of hydrogen ions to vanadium of 6.57 to 1, were transferred to a 50-ml cali- brated flask. The required amount of 0-0104 M EDTA solution was then added, and the volume made up to 50ml.This solution was transferred to a conductance cell, and, after it had been set aside for about 2 hours in a thermostatically controlled bath, its resistance was measured until constant readings were attained. These solutions were then retained for 7 days to attain equilibriurn,l0 when their resistances were measured again.404 KAKABADSE AND WILSON: THE COMPLEX OF VANADYL~ [Vol. 6 EQUIPMENT- Conductimetric titrations were carried out with a Cambridge conductivity bridge and dip-type electrodes. Accurate conductance measurements were made with a Jones and Bollinger type cellll kept at constant temperature in a Townson & Mercer S200 thermostatically controlled bath containing Puremore 210 light oil. The bridge consisted of Muirhead decade resistance boxes, the null-point was detected on a Cossor oscilloscope and the alternating-current supply was obtained from an Airmac oscillator at 1500 cycles per second.RE su LTS Results based on conductimetric end-point determinations are shown in Tables I and 11. In Table I, for which polyvanadic acid was used, the vanadium concentration was varied, whereas in Table 11, for which ammonium metavanadate was used, it was kept constant (except for curve 2 E). The titration flask was fitted with, a stirrer. TABLE I CONDUCTIMETRIC TECHNIQUE WITH POLYVANADIC ACID AS THE SOURCE OF VANADIUM Concentration of Concentration of Ratio of Theoretical Experimental vanadium, hydrogen ions, * hydrogen ions end-point, end-point, Curve gram-atoms per litre gram-ions per litre to vanadium ml ml 1A 0.00382 0.00382 1 15-28 - 1B 0.001 84 0.01595 7-59 7.36 7.4 1c 0-00368 0.02808 7.59 14-72 14-0 1D 0.00736 0.05234 7.59 29.44 28.8 * Including protons obtained from ion-exchange column during formation of polyvanadic acid.10 13 16 19 22 25 Volume of 0.01037 M EDTA, rnl Fig. 3. Accurate conductance measurements:at 25' 4 0.001" C of solutions containing polyvanadic and perchloric acids (0.00369 gram-atom of vanadium per litre; ratio of hydrogen ions to vanadium = 6.57) and different volumes of 0.01037 M EDTA (total volume made up to 50 ml): curve A, after 2 hours; curve B, after 7 days. Theoretical end-point at117.79 ml Conductimetric titration curves are shown in Figs. 1 and 2. As the breaks in these curves are in general somewhat ill-defined, only approximate experimental end-points are quoted in Tables I and 11.Fig. 3 shows accurate conductivity measurements on solutions 0.00369 M in vanadium and having a ratio of hydrogen ions to vanadium of 6.57 to 1. After 2 hours, the experi- mental end-point was reached on addition of 17.60 ml of standard EDTA solution, but, when set aside for 7 days, this break occurred at 18.20 ml (theoretical end-point at 17.79 ml).June, 19611 WITH ETHYLENEDIAMINETETRA-ACETIC ACID TABLE I1 405 CONDUCTIMETRIC TECHNIQUE WITH AMMONIUM METAVANADATE AS THE SOURCE OF VANADIUM Concentration of vanadium, Curve gram-atoms per litre 2A 0.00088 2B 0.00088 2C 0.00088 2D 0.00088 2E 0.001 78 Concentration of hydrogen ions, gram-ions per litre 0.00188 0.00555 0-01 11 0.0175 0.0111 Ratio of hydrogen ions t o vanadium 2.14 6.3 12.6 20 6.3 Theoretical end-point, ml 17.0 17.0 17.0 17-0 34.0 Experimental end-point, ml 6-5 16.8 16.8 17.5 34.5 Table I11 shows titration results with xylenol orange as indicator and a stock vanadyl solution prepared from ammonium metavanadate.TABLE 111 INDICATOR TECHNIQUE Total volume, ml (a) 100 (b) 100 (d) 100 (e) 100 (6) 100 (f) 10 ["h', 1:: (2) 12.5 ( 3 ) 30 Concentration of vanadium, gram-atoms per litre 0-0005 0.0005 0*0005 0.0005 0-0005 0-005 0.00 1 0.0005 0-004 0.0025 Concentration of hydrogen ions, gram-ions per litre 0.003 0.0055 0.013 0.018 0-023 0.03 0.03 0.03 0-044 0-065 Ratio of hydrogen ions to vanadium 6 11 26 36 46 6 30 60 11 26 End-point ITheoretical,Experimentai, ml of 0.01 M ml of 0.01 M EDTA EDTA 5.07 5.12 6-07 5-09 5.07 5.10 5-07 4.8 5.07 - 5.07 6-05 5-07 5-10 5-07 - 5.07 5.14 5.07 - (f = 0.986) (f -- 0.986) Error, / O +1 + 0.4 + 0.5 -5 - 0.4 0' - $- 0.6 + 1-3 - - PH - - 1-94 1.90 1-80 1-80 1-70 1-33 - - DISCUSSION OF RESULTS By using the absorptiometric technique Ringbom, Siitonen and Skrifvarsl established the formation of two complexes between quinquivalent vanadium and EDTA-V02Y3- (pH 6 to 9) and V02HY2- (below pH 3-5).These findings are difficult to understand, since cations are com- plexed by EDTA and the vanadyl cation is formed in appreciable quantities only beyond the isoelectric point of quinquivalent vanadi~m,l~9~3 i.e., below a pH of about 2. In our investi- gation, solutions with a low ratio of hydrogen ions to vanadium gave very low vanadyl titres (see curves A in Figs.1 and 2), which are attributed to incomplete formation of vanadyl ions. To establish minimum acidity conditions, or rather the minimum ratio of hydrogen ions to vanadium required for the quantitative conversion of vanadate to vanadyl ion in the reaction expressed by equation (1) we studied the hydrolysis of a 0.0017 M aqueous solution of vanadium pento xide with perchloric acid by means of accurate conductivity measurements (see Fig. 4)- V20, aq. + 2H+ + 2V02+ + H20 . . .. . . - (3) In Fig. 4, curve A, denoting the change in conductivity of the aqueous vanadium pentoxide - perchloric acid system, becomes parallel to curve B-conductivity of perchloric acid-when the ratio of hydrogen ions to vanadium is about 4 to 1, indicating that on further addition of acid no changes occur in the reaction denoted by equation (3).Moreover, at point P, the fall in conductivity from theoretical corresponds approximately to the removal of one proton per vanadium atom. Although the lower hydrogen ions to vanadium limit is fixed by the minimum acidity requirements for the quantitative conversion of vanadate to vanadyl ions,* the upper limit is governed by the general instability of the metal - EDTA complex in the presence of excess of acid (equation 2), which can be particularly serious with a univalent cation.l* In addition, the oxidising power of the vanadylv cation increases with increase in hydrogen-ion concentration,l5 thus rendering the risk of a redox reaction between VO,+ and EDTA more pronounced. * This ratio would correspond to 5 to 1 for ammonium metavanadate.406 KAKABADSE AND WILSON: THE COMPLEX OF VANADYL~ [Vol.86 The quantitative character of complex formation between EDTA and the univalent vanadylv cation near the lower hydrogen ions to vanadium limit was established by a series of accurate conductivity measurements (see Fig. 3) and by volumetric experiments (see Table 111) by the indicator method. At the'higher acidity limit, complexation is still quantita- tive when the ratio of hydrogen ions to vanadium is 26 (experiment (c), Table 111), whereas for a ratio of 36 the experimental end-point is 5 per cent. low, and, for even higher ratios, the colour change is so gradual as to make the end-point undetectable. This is also true when the total acidity is high (experiment ( j ) , Table 111).If it is large enough to cause the minimum acidity to drop the pH below 1.8, a satisfactory end- point is not obtained. At the other extreme, for very small amounts of vanadium in solution the metal-indicator colour becomes too faint, even above pH 1.8. In the indicator method, the concentration of vanadium has an indirect effect. I P 7- - IE 4 - % 5 - - I E m 0 - x 4 - Q) c 2 3 - U 8 u 2 - E a e, tn I - 0 0 3 6 9 I Ratio of hydrogen ions to vanadium Fig. 4. Conductivity curves a t 25' f 0.001" C for: curve A, solutions of vanadium pentoxide in perchloric acid (0.001 7 gram-atom of vanadium per litre) ; curve B, solutions of perchloric acid The rather ill-defined breaks in the conductivity curves (Figs. 1 and 2) may be attributed to the low concentrations of the solutions, rendering the conductivity measurements less accurate. As regards the redox character of the vanadylv - EDTA system, we found that partial reduction to the blue vanadylIv ion occurs in solutions of high hydrogen-ion concentration, Around pH 2, this reduction appears to be slow and does not interfere with the titrimetric determination of the vanadylv ion.Even when the solution is set aside for 7 days, the conductivity break was only 2.2 per cent. high (see Fig. 3). Although the vanadylv - EDTA complex seems to be only moderately stable and involves a delicate balance between minimum and maximum acidity requirements, we suggest the method described below for the volumetric determination of vanadium. Place the vanadium compound, containing between 7.5 and 25 mg of vanadium, in a 250-ml conical flask together with 30 ml of 0.1 M perchloric acid, gently boil the solution, cool, and dilute to about 100ml. Add 3 drops of indicator solution, and titrate with 0.01 M EDTA until the last trace of orange disappears and the solution is pure yellow.50.95 g of vanadium 5 1000 ml of M EDTA :. 100 ml of 0.01 M EDTA = 51 mg of vanadium. We thank Dr. T. S. West for recommending the use of xylenol orange and for supplying We also thank the Manchester This is overcome to a large extent by the improved instrumentation (Fig. 3). us with samples of xylenol orange and pyrocatechol violet. Oil Refinery for a gift of Puremore 210 light oil.June, 19611 WITH ETHYLENEDIAMINETETRA-ACETIC ACID 407 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. REFERENCES Ringborn, A., Siitonen, S., and Skrifvars, B., Acta Chem. Scand., 1957, 11, 551. Ducret, L.-P., Ann. Chim., 1951, 6, 705. Barnard, A. J., Broad, W. C., and Flaschka, H., Chemist Analyst, 1956, 45, 86. Belcher, R., Lab. Practice, 1960, 9, 568. Belcher, R., Leonard, M. A., and West, T. S., J . Chem. Soc., 1958, 2390. Headridge, J. B., Analyst, 1960, 85, 379. Russel, R. U., and Salmon, J. E., J . Chem. Soc., 1958, 4708. Korbl, J., and Pfibil, R., Chemist Analyst, 1956, 45, 102. Flaschka, H., and Sadek, F., Mikrochim. Acta, 1957, 1. Martell, A. E., and Calvin, M., “Chemistry of Metal Chelate Compounds,” Prentice-Hall, New York, Jones, G., and Bollinger, G. M., J . Amer. Chem. Soc., 1931, 53, 411. Sidgwick, N. V., “The Chemical Elements and Their Compounds,” The Clareiidon Press, Oxford, Jander, G., and Jahr, K. F., 2. anorg. Chem., 1933, 212, 1. Genge, J. A. R., Lab. Practice, 1960, 9, 592. Kakabadse, G., and Wilson, H. J., J . Chem. SOC., 1960, 2475. 1952, p. 87. 1950, p. 812. Received January 12th, 1961