Phenomena Affecting the Equilibrium of Al(III) and Zn(II) Extraction with Winsor II Microemulsions

Journal of Colloid and Interface Science 227, 244–246 (2000) doi:10.1006/2000.6854, available online at http://www.idealibrary.com on

NOTE Phenomena Affecting the Equilibrium of Al(III) and Zn(II) Extraction with Winsor II Microemulsions MATERIALS AND METHODS The extraction of Al(III) and Zn(II) from an aqueous solution with two water-in-oil microemulsions, one containing di(2ethylhexyl)phosphoric acid (DEHPA), was investigated to aid the understanding of the role of the extractant and the metal specific characteristics in the mechanism of microemulsion extraction. The extraction of Al with the DEHPA microemulsion increased by a factor of about 10 with respect to that in the conventional DEHPA system, whereas the extraction of Zn was lower than that in the single DEHPA system. Extraction with the DEHPA-free microemulsion was very low, showing that metal ion solubilization was not important in the mechanism of microemulsion extraction. It is proposed that the effect of the mixed microemulsion on the metal distribution coefficient is the result of the balance between a decrease in the complexation reaction yield due to the interaction between butanol and DEHPA, and the adsorption of the metal complex at the macro- and microinterfaces. The former leads to a decrease in Zn(II) extraction and the latter to Al(III) extraction synergism. °C 2000 Academic Press Key Words: microemulsions; reversed micelles; metal extractions; synergism.

INTRODUCTION Di(2-ethylhexyl)phosphoric acid (DEHPA) is a selective reagent which in the absence of additives does not form reversed micelles in nonpolar solvents. Bauer et al. (1–3) reported substantial improvements in the extraction of trivalent and quadravalent metals with DEHPA when a water-in-oil (W/O) microemulsion was formed by adding a surfactant and a cosurfactant to an organic phase containing the extractant. However, Brejza (4) studied the extraction of Zn(II) with one of their microemulsion formulations and observed a decrease in the distribution coefficient with respect to the conventional DEHPA system. The same contrast in the behavior of a microemulsion in the extraction of two metals of different groups was reported by Pepe and Otu (5) for the extraction of Bi(III) and Zn(II). In general, increases in the metal ion distribution coefficients have been explained in terms of the ability of W/O microemulsions to solubilize both hydrophilic aqueous species and hydrophobic extractants (6). However, the effect of the metal ion characteristics and its mechanism of extraction has not yet been related to microemulsion extraction. The purpose of this work was to investigate the role of the specific characteristics of Zn(II) and Al(III) on the mechanism of extraction with a microemulsion containing DEHPA. Extraction with DEHPA, and with a DEHPA-free microemulsion, was also investigated in order to compare the behavior of the three systems. 0021-9797/00 $35.00

C 2000 by Academic Press Copyright ° All rights of reproduction in any form reserved.

Selected Systems The reference systems consisted of an aqueous sulfate or nitrate medium of ion strength I = 1 M and pH 2.5, containing either 0.001 M Zn(II) or 0.002 M Al(III), and a 0.2 M solution of DEHPA in kerosene. The microemulsions contained 2.5% wt/wt sodium dodecylbenzene sulfonate (NaDBS) and 15% wt/wt nbutanol in kerosene. The mixed microemulsion contained in addition DEHPA at a concentration of 0.2 M.

Experimental Procedure Distribution coefficient experiments were conducted by shaking equal volumes of the aqueous and the organic phases in a water bath at 25◦ C. Samples were taken at regular intervals to measure the metal concentration and follow the approach to equilibrium. Equilibrium in the Zn systems was reached within 30 min, whereas Al systems required 5 h contact. In each case the pH was measured again after equilibration, adjusted when required, and followed by phase reshaking. After being shaken the solutions were left to rest for at least 2 h until two clear phases were obtained. Metal concentrations were measured using an atomic absorption spectrometer Perkin Elmer 1100B. Samples taken from aqueous phases containing Zn were diluted to fall within the calibration linear range. Aqueous samples containing Al were within the linear range and were analyzed directly. In all systems the standard solutions had the same matrix as the phase to be analyzed. Samples from the organic phase were stripped with a 2 M solution of sulfuric or nitric acid. Water content was measured by Karl Fischer coulometric titration using a Mettler DL 37 Karl Fischer coulometer. Between three and six tests were run per sample.

Chemicals DEHPA was obtained from B.D.H and purified according to the method recommended by Union Carbide (7). Analytical grade Na2 SO4 , NaNO3 , ZnSO4 · 7H2 O, Al2 (SO4 )3 · 16H2 O, NaDBS, n-butanol, and n-dodecane AnalaR 95% were purchased from B.D.H. and used without further purification. Kerosene, all from the same batch, was kindly provided by B.P. Chemicals.

RESULTS AND DISCUSSION The distribution coefficients of Zn and Al in the reference and the micellar systems are given in Table 1. There was a remarkable difference in the extraction of the two metals with the DEHPA/NaDBS/n-butanol microemulsion (subsequently called the mixed microemulsion). Whereas the distribution coefficient of Zn decreased from 23 in the reference system to 0.5 in the mixed microemulsion, that of Al increased from 0.3 to 2.2. In the DEHPA-free microemulsion the distribution coefficient of both metals was very low (0.2 for Zn and 0.1 for Al), thus indicating that metal ion solubilization in the microemulsion core was

244

245

NOTE

TABLE 1 Water Content and Metal Distribution Coefficients in the Reversed Micellar and Nonmicellar Systemsa Sulfate medium

Nitrate medium

Metal

Organic phase

[H2 O]org , M

D

[H2 O]org , M

D

—

DEHPA NaDBS + n-butanol DEHPA + NaDBS + n-butanol

0.0039 2.9 5.2

— — —

— 3.2

— —

Zn

DEHPA NaDBS + n-butanol DEHPA + NaDBS + n-butanol

— 2.8 3.5

23 0.2 0.5

— — 2.9

169 1.8 2.5

Al

DEHPA NaDBS + n-butanol DEHPA + NaDBS + n-butanol

— 3.4 4.1

0.3 0.1 2.2

— — 2.9

0.3 0.1 2.2

a Bulk aqueous phase: initial [Zn2+ ] = 1.0 × 10−3 M; initial [Al3+ ] = 1.9 × 10−3 M; pH 2.5; I = 1.0 M. Organic phase: [DEHPA] = 0.2 M in kerosene.

not important in the mechanism of extraction, making metal complexation with DEHPA an essential step. The complexation reactions of the two metals in the reference systems have been reported to take place at the interface (8–10), but their mechanisms of extraction are not the same due to their different characteristics. For Al, Sato et al. (10) concluded that dehydration of the hexaaqua Al ion must take place before complexation, followed by complete dehydration of the complex and desorption into the bulk organic phase, [Al(H2 O)6 ]3+ (aq) ⇔ [Al(H2 O)5 ]3+ (aq) + H2 O 3+

[Al(H2 O)5 ]

−

(aq) + 3(A2 H) (i) ⇔ Al[(H2 O)5 (A2 H)]

2+

[1] −

(i) + 2A2 H

m Al(A2 H)3 (org) + 5H2 O(aq),

[2]

where A2 H− is the ionized form of the DEHPA dimer, and (aq), (i), and (org) indicate the aqueous phase, the interface, and the organic phase, respectively. However, the possibility of the hydrated ion being adsorbed at the interface loosely associated with the ligand has been proposed by Roddy et al. (11) for the extraction with DEHPA of Fe3+ , an ion which is also hydrated. A similar behavior of the Al ion would partly explain the low Al distribution coefficient in the reference system. The Zn ion, on the other hand, is not hydrated to any considerable extent; therefore there is no dehydration step in its proposed mechanism of extraction (8, 9): Zn2+ (aq) + A2 H− (i) ⇔ ZnA2 (org) + H+ (aq) ZnA2 (org) + HA(i) ⇔ ZnA2 · HA(org).

negligible. However, the presence of DEHPA in the mixed microemulsion more than doubled the Zn distribution coefficient with respect to that of the DEHPAfree one, showing that the amount of free DEHPA, although reduced, was not negligible. The metal complex adsorption–desorption equilibrium depends on the characteristics of the complex. Using light and neutron scattering techniques, Neuman et al. (13) investigated the structure of several DEHPA salts in n-alkanes in the presence of an excess water phase, and observed that at high concentrations these salts aggregate to form reversed micelles. In the case of Zn, they report that, although it forms cylindrical reversed micelles at high concentrations, n-octanol had to be added to the organic phase in order to get water uptake, suggesting that the salt is essentially hydrophobic. Thus, the Zn complex may have a higher affinity for the organic phase than for the interface, which would explain its higher distribution coefficient in the reference system as compared to that of Al. Therefore, the presence of the extensive micellar oil–water interface of the mixed micellar system would provide a substantial extra site for the more hydrophilic Al complex to adsorb, leading to the observed synergism, whereas the hydrophobic Zn complex would still be preferentially dissolved in the continuous organic phase. The water content in the different systems, also given in Table 1, provided support to this hypothesis. The organic-phase water content in the reference systems was negligible, which was taken as a confirmation that the reagent did not form a microemulsion. In the metal-free microemulsions, the water content in the organic phase of the mixed microemulsion was 5.2 M whereas that in the DEHPA-free one was only 2.9 M. The difference corresponds to about 10 molecules of water per molecule of DEHPA, which suggested that in the mixed microemulsion DEHPA molecules were included into the reversed micelles (6). The water content in the mixed microemulsion decreased with metal extraction, from 5.2 to 3.5 M for Zn and to 4.1 M for Al. The fact that the decrease was higher for Zn than for Al, despite the fact that the extraction of Zn was considerably lower, could be explained by a reduction in the number of reversed micelles containing DEHPA following its complexation with Zn and the subsequent desorption of the complex. The Al– DEHPA complex, being more hydrophilic, would remain attached to the reversed micelles. Table 1 also shows much larger distribution coefficients for Zn in the systems with a nitrate medium than in those with a sulfate medium. This is a result of the lower concentration of Zn ion in the sulfate medium due to the sulfate–bisulfate equilibrium.

[3] [4]

The distribution coefficient of both metals depends on the concentration of DEHPA, and on the adsorption–desorption equilibrium between the interface and the organic phase. In the mixed micellar system the addition of the surfactant and the alcohol may block the acid group in the DEHPA molecule, decreasing the concentration of the reactive group. Hala (12) reviewed the interaction between alcohols and dialkylphosphoric acids in organic solution and concluded that they form a hydrogen bond, leading to inactivation of the extractant. Brejza (4) observed a similar effect, as the extraction of Zn with the same mixture of DEHPA and n-butanol as in the reversed micellar system was

CONCLUSIONS Extraction with the DEHPA-free microemulsion was low for both metals, showing that solubilization in the microemulsion core was not an important step in the mechanism of microemulsion extraction. The presence of the alcohol in the single DEHPA system reduced the extraction of Zn substantially, indicating that the concentration of free DEHPA in the mixed microemulsion may be similarly reduced. The synergism in the Al extraction, and the reduction in the extraction of Zn, produced by the addition of butanol and NaDBS to the conventional DEHPA system in order to form a microemulsion, were due to the combination of a reduced yield of the metal extraction reaction in the presence of the alcohol and the interfacial behavior of the metal complex. The Al complex being more hydrophilic than the Zn one, it is likely to adsorb at the extensive microemulsion interface, thus overcoming the effect of the lower concentration of free DEHPA, whereas the hydrophobic Zn complex would preferentially dissolve into the continuous organic phase.

ACKNOWLEDGMENTS Financial support by SERC (Grant No. GR/F28502) is gratefully acknowledged.

246

NOTE

REFERENCES 1. Bauer, D., and Komornicki, J., Proc. Int. Solv. Extn. Conf. (ISEC ’83) 315 (1983). 2. Bauer, D., Cote, G., Komornicki, J., and Mallet-Faux, S., Can. Soc. Chem. Eng. 2, 425 (1989). 3. Fourre, P., Bauer, D., and Lemarie, J., J. Anal. Chem. 55, 662 (1983). 4. Brejza, E. V., Doctoral Thesis, University of London, 1994. 5. Pepe, E. M., and Otu, E. O., Solv. Extr. Ion Exch. 14 (2), 247 (1996). 6. Osseo-Asare, K., Sep. Sci. Technol. 23 (12 & 13), 1269 (1988). 7. Union Carbide Mining Chemicals (monograph), October, 1972. 8. Ajawin, L. A., Perez de Ortiz, E. S., and Sawistowski, H., Chem. Eng. Res. Dev. 61 (1), 62 (1983). 9. Huang, T. C., and Juang, R. S., Ind. Eng. Chem. Fundam. 25 (4), 752 (1986). 10. Sato, T., Yoshino, T., Nakamura, T., and Kudo, T., J. Inorg. Nucl. Chem. 40, 1571 (1978). 11. Roddy, J. W., Coleman, C. F., and Arai, S., J. Inorg. Nucl. Chem. 33, 1099 (1971). 12. Hala, J., in “Solvent Extraction from Aqueous Organic Media in Ion Ex-

change and Solvent Extraction” (J. A. Marinsky and Y. Marcus, Eds.), Vol. 8 p. 369, Dekker, New York, 1981. 13. Neuman, R. D., Zhou, N. F., Wu, J., Jones, M. A., Gaonkar, A. G., Park, S. J., and Agrawal, M. L., Sep. Sci. Technol. 25, 13 (1990). Edwina V. Brejza1 E. Susana Perez de Ortiz2 Department of Chemical Engineering and Chemical Technology Imperial College of Science, Technology, and Medicine London SW7 2BY, U.K. Received March 22, 1999; accepted March 23, 2000 1 Present address: Department of Pharmacy, University of Malta, Msida MSD

06, Malta. 2 To whom correspondence should be addressed.