Studies on Mg/Fe Hydrotalcite-Like-Compound (HTlc)

Journal of Colloid and Interface Science 251, 26–32 (2002) doi:10.1006/jcis.2002.8319 Studies on Mg/Fe Hydrotalcite-Like-Compound (HTlc) I. Removal of Inorganic Selenite (SeO2− 3 ) from Aqueous Medium J. Das, D. Das, G. P. Dash, and K. M. Parida1 Regional Research Laboratory (CSIR), Bhubaneswar-751 013, India E-mail: [email protected] Received December 21, 2000; accepted February 21, 2002 valencies are not responsive to the same means. For example, effective removal of SeO2− 3 from waste water is possible by using ferric chloride or alum in a coagulation technique; however the same method does not significantly affect Se(VI). According to Trussell et al., for the same operating conditions and water compositions, Se(VI) is removed at a one-tenth the rate as Se(IV) (5). Selenite (Se(IV)) shows a strong affinity toward iron oxide surfaces (6–9) while selenate (Se(VI)) shows a weaker affinity for oxide surfaces (9, 10) and is easily transportable in ground waters and available for uptake by plants. The reactions of selenium in soil greatly affect the bioaccumulation of Se in plants and animals. Selenium interactions with soils and soil constituents primarily depend on selenite retention. Due to the ubiquitous occurrence of Fe and Mg in soil and sediments and their strong affinity in sequestering many elements, a great deal of literature on the removal/adsorption of selenium on oxide/oxyhydroxides of Fe and oxides of Mg is nowadays available (8, 11–16). Hydrotalcite-like compounds (HTlc) are the layered metal doux+ ble hydroxides having the general formula [MII1−x MIII x (OH)2 ] n− x− [Ax/n · mH2 O] . These materials consist of positively charged metal hydroxide sheets compensated by a large number of exchangeable charge-balancing anions and water molecules present in the interlayer spaces. Due to the presence of large interlayer spaces and the huge number of exchangeable anions, HTlcs act as a good ion exchanger and adsorbent. U.S. Patent No. 4,935,146 shows the removal of selenium (IV or VI) from wastewater with a substantial background of other anions such as SO2− 4 . Sato et al. (17) have studied the adsorptive behavior of thermally dehydrated synthetic Mg/Al HTlc relative to certain 2− other anions such as PO3− 4 and SO4 . So far no report on the adsorptive behavior of Mg/Fe HTlc toward selenium has been published. Since both Fe and Mg are common in soil and there is every possibility that a complex compound of Fe–Mg may be present in soil, the present study of adsorption of selenium on Mg/Fe HTlc has its significance. This study aims at investigating the sorption of selenite on Mg/Fe HTlc in aqueous medium and the factors affecting it such as pH, SeO2− 3 concentration, HTlc A Mg/Fe hydrotalcite-like-compound (HTlc) was prepared and its affinity toward the removal of SeO2− 3 from an aqueous medium was studied as a function of pH, time, temperature, particle dose, 2− and SeO2− 3 concentration. The fraction of SeO3 removal increases with decrease in both pH and temperature. The adsorption data are fitted to the Langmuir adsorption isotherm in the temperature range 303–333 K, and the thermodynamic parameters viz. standard Gibbs’ free energy change (G◦ ), enthalpy change (H◦ ), and entropy change (S◦ ) are calculated. The negative value of H◦ indicates that the adsorption process is exothermic. The apparent equilibrium constants (Ka ) are also calculated and found to decrease with increase in temperature. C 2002 Elsevier Science (USA) Key Words: Mg/Fe HTlc; adsorption; aqueous selenite. INTRODUCTION Selenium is an essential and at the same time toxic element from the human and animal health point of view (1, 2). In fact, there is a very narrow range between the deficient and toxic levels of selenium in animals, which necessitates a clear knowledge of the processes and factors responsible for the distribution of selenium in environment (3). Selenium as a contaminant may be found from the combustion of fossil fuels, roasting and refining of sulfide ores, and the wastewater from power plants, particularly those using lignite as a fuel source. While selenium is usually present in small amounts averaging from 0.1 to 20 ppm, such levels are too high to permit safe environmental discharge as the current drinking water regulation requires selenium levels to be less than 0.01 ppm. Selenium contaminants are also available in some types of soils, on which exposure to water substantially enhances the toxic level in environment and can be deleterious to animal life. For most aqueous medium, selenium exists in two primary oxidation states: Se(IV) usually as HSeO− 3 (biselenite) or SeO2− (selenite) and Se(VI) usually present as SeO2− 4 (sele3 nate) (4). The removal of selenium is highly selective and both 1 To whom correspondence should be addressed. Fax: (0674) 581637. 0021-9797/02 $35.00  C 2002 Elsevier Science (USA) All rights reserved. 26 STUDIES ON Mg/Fe HYDROTALCITE-LIKE-COMPOUNDS (HTlcs) dose, and temperature with an objective to provide means for the substantial removal of SeO2− 3 from wastewater streams. Furthermore, the thermodynamic parameters for the adsorption process were calculated to determine the viability and effectiveness of the process. EXPERIMENTAL Materials and Method Mg/Fe HTlc with MgII : FeIII molar ratio of 2 : 1 was prepared by co-precipitation (at constant pH) method (18). In this method two solutions, solution 1 containing mixed metal nitrates of MgII and FeIII of desired concentrations and solution 2 prepared by taking NaOH (0.35 mol) and Na2 CO3 (0.15 mol), of 200 ml each were added simultaneously to a 1-L beaker containing 100 ml of deionized water at 50 ml/h at 35◦ C under constant stirring conditions. The pH of the solution was around 9.6. The precipitate was aged in a thermostatic bath at 65◦ C for 18 h. It was then filtered, washed thoroughly with deionized water till neutral pH, and dried at 100◦ C overnight. 27 Yorco thermostatic water bath cum shaker. During pH measurement, the solutions were constantly bubbled with nitrogen and the flasks were kept closed for the whole period of adsorption in order to avoid the possible interference of atmospheric carbon dioxide. The pH of the solutions was adjusted by using 0.01 M HNO3 and/or NaOH (Merck, GR), and the pH measurements were carried out in an Elico digital pH meter (Model LI-120) using a combined glass electrode (Model CL 51). Buffer was not used to avoid any possible interference of foreign anions in the adsorption process. Preliminary investigations revealed that 2 h is sufficient for the system to reach equilibrium. The mixture was then filtered through a G-4 crucible (Borosil make). Residual SeO2− 3 content in the filtrate was determined spectrophotometrically (21) with the help of a Varian Cary 1E UV-VIS spectrophotometer (Model EL96043181) fitted with Cary 100 software using 10-mm matched quartz cells at 380 nm. For spectrophotometric analysis, 2-Mercapto ethanol (Aldrich) was used as both the complexing agent and reductant. All the experiments were carried out in duplicate, and the results were reproducible within ±5%. So the average was taken for calculation and data analysis. Chemical Analysis The chemical analysis of Mg and Fe in Mg/Fe HTlc was carried out by dissolving 0.5 g of material in 1 : 1 HCl followed by their estimation with standard EDTA and dichromate methods, respectively (19). Before the analysis of Mg, Fe(III) was removed from the solution by using NH4 Cl and HN4 OH as hydroxide, and the filtrate thus obtained was taken for Mg analysis. From the analysis the Mg/Fe ratio was found to be 1.97 as compared to 2.0 (taken). Textural Characterization The pHpzc (point of zero charge) of the sample was determined by measurement of the surface charge of the particles using a particle charge detector (PCD-03-pH) from Mutek, Germany, by the polyelectrolyte titration method (20). The powder XRD pattern was recorded in a Philips (Model 1710) semiautomatic X-ray diffractometer with an autodivergent slit fitted with a graphite monochromator using CuK α radiation at a scanning speed of 2◦ /min, operated at 40 kV and 20 mA. The XRD data were matched with standard JCPDS data files. The FTIR spectrum was recorded in the range 4000–400 cm−1 in KBr phase with a Specord 75 IR (Carl Zeiss, Germany) spectrophotometer. The specific surface area of the sample was determined by the N2 adsorption/desorption method at liquid N2 temperature (77 K) using Quantasorb (Quantachrome, USA). Prior to surface area measurement, the sample was degassed at 90◦ C in vacuum (1 × 10−4 Torr). RESULTS AND DISCUSSION Surface Properties, XRD, and FTIR Analysis The pHpzc of HTlc is determined to be 8.94. The XRD pattern of a 100◦ C dried sample is shown in Fig. 1. From this it is observed that at lower 2θ values, the peaks are sharp and symmetric as compared to those at higher 2θ values, which is characteristic of clay minerals (hydrotalcites) having a layered structure (18). In addition, the 100◦ C dried sample is well crystalized and consists of one phase only, i.e., Mg/Fe HTlc phase. The FTIR pattern (Fig. 2) of the sample shows that the spectrum is broadly divided into three noticiable absorption regions: (1) 4000–2300, (2) 1800–1200, and (3) 1100–400 cm−1 . A strong absorption band at 3435.95 cm−1 is due to the vibration of structural OH groups hydrogen bonded with interlamellar water Adsorption of Selenite (SeO2− 3 ) on HTlc Adsorption of SeO2− 3 on Mg/Fe HTlc was carried out with 50 ml of solution (prepared from Na2 SeO3 , Merck, AR) in 100-ml stoppered conical flasks under constant shaking in a FIG. 1. XRD pattern of Mg/Fe HTlc with Mg : Fe molar ratio 2 : 1. 28 DAS ET AL. FIG. 2. FTIR pattern of Mg/Fe HTlc with Mg : Fe molar ratio 2 : 1. molecules or OH groups in adjacent layers (22). The shoulder at around 3100–3200 cm−1 may be due to the hydrogen bonding between water and the anions in the interlayer region. In the second region, the peaks at 1628.31 and 1359.37 cm−1 may be due to the bending mode of water molecules and interlamellar carbonate ions, respectively (carbonate ions are expected to be present as hydrolysis of the metal salts was done in NaOH and Na2 CO3 ). In the third region, a band at 1016 cm−1 could be due to monodentate nitrate (23) and a strong absorption at 584.87 cm−1 may be related to Mg–O–Mg or Mg–O–Fe vibration. The specific surface area of the 100◦ C dried Mg/Fe HTlc was determined by the single-point BET method and found to be 59 m2 /g. pH adjustments, 0.01 M HNO3 and 0.01 M NaOH are used (use of NaOH is minimized and much care has been taken to restrict its use). The buffer is not used in order to avoid the addition of a foreign anion because it is well established that the presence of a secondary anion extensively affects SeO2− 3 adsorption (5, 25). The pH of the 50-ml water suspension of 0.05 g material is ∼9.2, which indicates the presence of a large number of exchangeable − ions such as CO2− 3 and/or OH , and these ions play a vital role in the adsorption process. At pH 6.0 the fraction of adsorption is marginally above 50% (Fig. 3) at 30◦ C, with SeO2− 3 concentration and HTlc dose at 50 mg/L and 1 g/L, respectively. From Fig. 3 it is clear that there exists a considerable percentage of adsorption above the value of pHpzc , i.e., the pH at which the adsorbent surface is negatively charged. This indicates that the adsorption is not only due to the electrostatic force of attraction between the adsorbent surface and SeO2− 3 ion but also due to a combined effect of both chemical, i.e., ligand exchange type (26), and electrostatic interaction. Further, in order to prove an ion exchange mechanism, the XRD of one selenite exchanged sample was taken, and a shift in the peak position (003 line) toward lower 2θ values compared to the parent hydrotalcite was observed. This could be due to the exchange of CO2− 3 in the interlayer space by selenite. Similar observations have been made by Legrouri et al. (27). Again it shows a steady and continuous increase in the fraction of adsorption when pH is decreased below the pHpzc value. So it can be predicted that the adsorption below pHpzc is a combined effect of both chemical and electrostatic interaction between the HTlc surface and SeO2− 3 ion. However, that above the pHpzc value is due to some sort of chemical interaction that comes into force even when both the surface and pH-Dependent Adsorption One of the most important processes for regulating the concentration and mobility of selenium in environment is adsorption on solid surfaces. Much effort has been devoted to determine the factors, that influence the interaction between the adsorption site and the anion, such as the rate of reaction, the equilibrium, and the nature of the anion. However, in general adsorption depends upon pH, solid composition, surface site concentration, nature of the adsorbate and its concentration and formation of solution complexes along with ionic strength, competing adsorbate ions, etc. (7). The studies investigating the pH dependence of selenite adsorption on Mg/Fe HTlc (Fig. 3) show that selenite adsorption is a function of pH (6, 9, 24). The fraction of SeO2− 3 removal increases with decrease in pH in the pH range 3 to 10. However, at lower pH, i.e., pH below 6, there is a possibility of disordering the HTlc structure and this has been confirmed by the analysis of Mg present in the resulting solution after adsorption with the help of AAS. Keeping this in mind, the ideal pH for experiments is considered to be 6.0. This is also important from an environmental point of view, as the pH of most water streams remains neutral to slight alkaline range unless it is an acid drainage. For on Mg/Fe HTlc as a function of pH. FIG. 3. Adsorption of SeO2− 3 [SeO2− 3 ] = 50 mg/L and particle dose = 1 g/L. STUDIES ON Mg/Fe HYDROTALCITE-LIKE-COMPOUNDS (HTlcs) 29 TABLE 1 Desorption of SeO23− (%) from Adsorbed HTlc Concentration of adsorbed solid (g/L) 2 2 2 2 2 pH of water Temperature (K) Time of exposure (h) % desorbed 5.0 6.5 6.5 6.5 8.5 303 303 318 333 303 48 48 48 48 48 1.72 2.06 14.62 15.15 37.32 adsorbing ion are negatively charged. This type of observations has also been reported earlier (8, 28) over ferric oxyhydroxide surfaces. When desorption experiments were carried out taking the ◦ SeO2− 3 ion-saturated HTlc (Table 1) at pH ≤ pHpzc (at 30 C), 2− there is practically no release of SeO3 ions. However, at temperatures 45 and 60◦ C, the release is quite appreciable. When pH ≥ 8.2 the release of SeO2− 3 ion is more. The irreversibility of the release process of SeO2− 3 ions at pH < pHpzc can be explained as follows: The HTlc surface strongly holds SeO2− 3 ions either through electrostatic or chemical interaction. The chemical interaction may proceed through a monodentate or multidentate ligand formation mechanism with the hydroxyl surface as explained by Zhang and Spark in the case of selenite adsorption on goethite (29). However, for pH ≥ pHpzc , the release of SeO2− 3 especially at higher temperatures may be due to the ionexchange process of loosely held SeO2− 3 ions which comes out either from the surface or from interlayer spaces. There exists a reversible mechanism in between the weakly held exchangeable anions present in the interlayer spaces and the anions present in the bulk of the solution. However, a detailed study on this aspect is required and will be communicated separately. FIG. 4. Adsorption of SeO2− 3 on Mg/Fe HTlc as a function of adsorbent dose. [SeO2− 3 ] = 50 mg/L and pH 6.0. SeO2− 3 concentrations for a fixed particle dose (Fig. 5), all types of surface sites are fully exposed. With increase in concentration there is an increase in adsorption density upto a certain value, which may be due to a high intramolecular competitiveness to occupy the lower energetic surface sites left behind. After the saturation point, this process becomes purely a reversible one and there is no further adsorption, which is evident from Fig. 5. Concentration-Dependent Adsorption The fraction of SeO2− 3 removal tends to increase with increase in HTlc dose at a fixed SeO2− 3 concentration (50 mg/L), but the increase is not proportional (Fig. 4). This is consistent with the argument that the surface sites of HTlc are heterogeneous (30, 31). According to the surface site heterogeneity model, the surface is composed of sites with a spectrum of binding energies. At low HTlc doses, all types of sites are entirely exposed for adsorption and the surface gets saturated faster. However, at higher particle concentration, the availability of higher energy sites decreases and a larger fraction of lower energy sites becomes occupied. This results in an overall decrease in binding energy of the surface, and there may exist a reversible type of process between the SeO2− 3 ions attached to low energetic sites and those present in bulk solution. This may be the reason for the higher uptake of selenite with increase in the HTLc dose upto 4 g/L (Fig. 4) and thereafter remains nearly constant. However, it was found that the loading capacity (mg/g) decreases with increase in HTLc dose (data not shown). However, at different FIG. 5. Selenite adsorption density as a function of initial SeO2− 3 concentration. Particle dose = 1 g/L and pH 6.0. 30 DAS ET AL. Temperature Dependence and Calculation of Thermodynamic Parameters Figure 6 depicts the fraction of SeO2− 3 removal as a function of temperature (from 303 to 333 K) and the fraction of removal decreases with rise in temperature. This further supports the surface heterogeneity model of the HTlc surface as discussed earlier. This observation leads to the argument that the process is exothermic, which can be concluded from the calculation of H ◦ (standard enthalpy changes) and will be discussed later on. In addition to this, with a rise in temperature the thermal energy of the adsorption sites may increase with the equilibrium, shifting more toward the left (desorption favored). This type of observation was also made by Pradhan et al. for adsorption of Cr(IV) on red mud (32). When the fraction of SeO2− 3 removal is investigated as a function of contact time (Fig. 7), the equilibrium is reached in about 2 h. Even the change in temperature does not influence the equilibrium time, and the removal curve is smooth and continuous. So there is a possibility of the formation of monolayer coverage. The temperature dependence of the adsorption process is associated with several thermodynamic parameters. The data obtained from the adsorption process is fitted into the linearly transformed Langmuir adsorption isotherm Ce /(X/m) = 1/bQ + Ce /Q, [1] where Ce is the equilibrium adsorbate concentration in solution (mol/L), Q denotes the amount adsorbed per unit mass of adsorbent (mol/g) for formation of the monolayer, (X/m) denotes the amount adsorbed per unit mass of the adsorbent, and b is FIG. 6. Adsorption of SeO2− 3 on Mg/Fe HTlc as a function of temperature. [SeO2− 3 ] = 50 mg/L, particle dose = 1 g/L, and pH 6.0. FIG. 7. Adsorption of SeO2− 3 on Mg/Fe HTlc as a function of time at ] = 50 mg/L, particle dose = 1 g/L, and pH 6.0. different temperatures. [SeO2− 3 the Langmuir constant or binding constant. Plots of Ce /(X/m) vs Ce at various temperatures (Fig. 8) result in straight line, which favors the applicability of the Langmuir Eq. [1] at pH 6.0. The values of b and Q are determined from the slopes and intercepts of the isotherms by regression method and have been tabulated in Table 2. Both values decrease with increase in temperature, indicating the adsorption process to be exothermic. The FIG. 8. Langmuir isotherm for the adsorption of SeO2− 3 on Mg/Fe HTlc. 31 STUDIES ON Mg/Fe HYDROTALCITE-LIKE-COMPOUNDS (HTlcs) TABLE 2 Langmuir Constants and Apparent Equilibrium Constant Calculated from Langmuir Equation TABLE 3 Gibbs Free Energy, Enthalpy, and Entropy Changes of SeO23− Adsorbed HTlc Sample Temperature (K) b Q (×105 ) Ka Sample Temperature (K) G ◦ (KJ/mol) H ◦ (KJ/mol) S ◦ (J/mol/K) HTlc HTlc HTlc 303 323 333 21,891 16,627 14,734 3.704 2.812 2.145 0.811 0.468 0.316 HTlc HTlc HTlc 303 323 333 −25.173 −26.099 −26.572 −11.023 46.7 apparent equilibrium constant K a , which indicates the affinity of the adsorbents toward the ions, were calculated (33, 34) from the product of Langmuir parameters Q and b and tabulated in Table 2. The continuous decrease in K a values with increase in temperature expresses the lowering in affinity with temperature. Thermodynamic parameters such as standard Gibbs’ free energy change (G ◦ ), enthalpy change (H ◦ ), and entropy change (S ◦ ) for the process were calculated from G ◦ = −RT ln b ◦ [2] ◦ ln b = S /R − H /RT, [3] where R = universal gas constant, T = temperature (K), and b = Langmuir constant. G ◦ values were directly calculated from Eq. [2] and tabulated in Table 3. The negative values of G ◦ reflect the feasibility and spontaneity of the process. The standard enthalpy and entropy changes were calculated from the slope and intercept of van’t Hoff plot ln b vs T −1 (Fig. 9), which was found to be linear. The negative value of H ◦ confirms that the process is exothermic as predicted earlier. The positive value of S ◦ reflects the affinity of HTlc toward the SeO2− 3 ion in aqueous medium and suggests some structural changes in the adsorbate and adsorbent. CONCLUSIONS Mg/Fe HTlc can be used as a good adsorbent for the removal of SeO2− 3 from an aqueous medium. The above studies show that the adsorption increases with the decrease in pH, but at lower pH dissolution of HTlc takes place. A significant fraction of the adsorption occurs even at pH ≥ pHpzc , i.e., where the surface is negatively charged, which indicates that the interaction between the surface sites and SeO2− 3 ions is not necessarily of electrostatic type; rather it is a combined effect of both chemical and electrostatic in nature. The chemical interaction may proceed through the formation of monodentate or multidentate ligands with the hydroxyl surface. With the increase in temperature the surface sites become less active and the fraction of removal decreases. The negative value of H ◦ indicates that the process is exothermic. The decrease in the value of apparent equilibrium constants also confirms that the activity of the surface sites decreases with increase in temperature. ACKNOWLEDGMENTS The authors are thankful to Dr. R. S. Thakur, Head, E.M. & I.C. Department, for his constant encouragement throughout the work and to Dr. V. N. Misra, Director, R. R. L., Bhubaneswar for his permission to publish this paper. REFERENCES FIG. 9. Plot of ln b vs T −1 for SeO2− 3 adsorption. 1. Lakin, H. W., Adv. Chem. Ser. 123, 96 (1973). 2. Mayland, H. F., James, L. F., Panter, K. E., and Sonderegger, J. L., in “Selenium in Agriculture and Environment” (L. W. Jacobs, Ed.). SSSA Spec. Publ. No. 23, p. 15, 1989. 3. Mc Neal, I. M., and Balistrieri, L. S., in “Selenium in Agriculture and Environment” (L. W. Jacobs, Ed.). SSSA Spec. Publ. No. 23, p. 1, 1989. 4. Neal, R. H., Sposito, G., Holtzeland, K. M., and Traina, S. J., Soil Sci. Soc. Am. J. 51, 1161 (1987a). 5. Trussell, R. R., Trussel, A., and Kreft, P., U.S. Environmental Protection Agency Official Research and Development [Rep.] EPA-600/2-80-153, p. 159 (1980). 6. Hingston, F. J., Posner, A. M., and Quirk, J. P., Adv. Chem. Ser. 79, 82 (1968). 7. Benjamin, M. M., Hayes, K. F., and Leckie, J. O., J. Water Pollut. Control Fed. 54, 1472 (1982). 32 DAS ET AL. 8. Parida, K. M., Gorai, B., Das, N. N., and Rao, S. B., J. Colloid Interface Sci. 185, 355 (1997). 9. Balistrieri, L. S., and Chao, T. T., Soil Sci. Soc. Am. J. 51, 1145 (1987). 10. Fujii, R., and Deveral, S. J., in “Selenium in Agriculture and Environment” (L. W. Jacobs, Ed.). SSSA Spec. Publ. No. 23, p. 195, 1989. 11. Ghosh, M. M., Cox, C. D., and Yuan-pan, J. R., Environ. Prog. 13, 79 (1994). 12. Wang, S., Mishra, M., Reddy, R. G., and Milbourne, J. C., in “Residues and Effluents-Processing and Environmental Considerations” (R. G. Reddy, W. P. Imrie, and P. B. Queneau, Eds.). Minerals, Metals and Material Soc. (1991). 13. Balistrieri, L., and Chao, T. T., Geochim. Cosmochim. Acta 54, 739 (1990). 14. Zhang, P., and Sparks, D. L., Environ. Sci. Technol. 24, 1848 (1990). 15. Measures, C. I., and Burton, J. D., Earth Planet Sci. Lett. 46, 385 (1980). 16. Geering, H. R., Cary, E. E., Jones, L. H. P., and Allaway, W. H., Soil Sci. Soc. Am. Proc. 32, 35 (1968). 17. Sato, T., Wakabayashi, T., and Shimada, M., Ind. Eng. Chem. Prod. Res. Dev. 25, 89 (1986). 18. Cavani, F., Trifiro, F., and Vaccari, A., Catal. Today 11, 173 (1991). 19. Vogel, A. I., “A Text Book of Quantitative Inorganic Analysis,” pp. 309– 434. Longman, London, 1969. 20. Richter, H. J., Jobst, K., Friedrich, H., Heinze, W., Friedrich, E., and Hermel, W., Aufbereit. Techn. 34, 257 (1993). 21. Afsar, H. A., Apak, R., and Tor, I., Analyst 114, 1319 (1989). 22. Lopez-Salinas, E., Garcia-Sanchez, M., Ramon-Garcia, M. A. L., and Schifter, I., J. Porous Mater. 3, 169 (1996); Miyata, S., Clays Clay Miner. 23, 369 (1975). 23. Curtis, N. F., and Curtis, Y. M., Inorg. Chem. 5, 383 (1966). 24. Hingston, F. J., Atkinson, R. J., and Posner, A. M., Nature 215, 1459 (1971). 25. Brown, M. J., and Carter, D. L., Soil Sci. Soc. Am. Proc. 33, 563 (1969). 26. Neal, R. H., Sposito, G., Holtzclaw, K. M., and Traina, S. J., Soil Sci. Soc. Am. J. 51, 1165 (1987b). 27. Legrouri, A., Badreddine, M., Barroug, A., De Roy, A., and Besse, J. P., J. Mater. Sci. Lett. 18, 1077 (1999). 28. Stumm, W., Huang, C. P., and Jenkins, S. R., Croat. Chem. Acta 42, 223 (1970). 29. Zhang, P., and Spark, D. L., Environ. Sci. Technol. 24, 1848 (1990). 30. Honeyman, B. D., and Santschi, P. H., Environ. Sci. Technol. 22, 862 (1988). 31. Catts, J. G., and Langmuir, D., Appl. Geochem. 1, 255 (1986). 32. Pradhan, J., Das, S. N., and Thakur, R. S., J. Colloid Interface Sci. 217, 137 (1999). 33. Lopez-Gonzalez, J. D., Velenzuela-Aluhorro, C., Jimenez-Lopez, T., and Ramierez-Saenz, A., An. Quim. 74, 225 (1978). 34. Mishra, T., Parida, K. M., and Rao, S. B., Sep. Sci. Technol. 33, 1057 (1998).