| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Faculty of Agriculture, Kochi Univ., B 200 Monobe, Nankoku 783-8502, Kochi, Japan
b United Graduate School of Agricultural Science, Ehime Univ., Matsuyama 790-8566, Japan
Corresponding author (saha{at}cheerful.com)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Cd on the complexes resemble the reported metal selectivity sequences on amorphous Fe and Al hydroxides. At different pHs, partitioning the adsorbed metals into strongly and weakly held fractions indicated that specific adsorption rather than nonspecific adsorption might have largely controlled the metal selectivity, particularly on the complexes. This led us to assume a predominant involvement of interlayered HyA or HAS polymers in metal adsorption from such dilute solutions. On Mt, the metals were predominantly adsorbed on the permanent charge sites in an easily replaceable state. However, a substantial involvement of the edge OH- groups of Mt in specific adsorption of the metals was also evident, especially at higher pH. Obviously, on Mt and on the complexes, the relative abundance of each type of site and their affinity to heavy metals were substantially different.
Abbreviations: CEC, cation-exchange capacity • HAS, hydroxyaluminosilicate • HRTEM, high-resolution transmission electron microscopy • HSAB, hard–soft acid base • HyA, hydroxyaluminum • IAP, ion activity product • ICP-AES, inductively coupled plasma–atomic emission spectroscopy • M, metal • Mt, montmorillonite • SA, strongly adsorbed • TA, total adsorbed • Vt, vermiculite • WA, weakly adsorbed • XRD, x-ray diffraction
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
In spite of wide geographic distribution of hydroxy-interlayered phyllosilicates and their importance in many soils, there has been little work conducted on the adsorption of metals by the interlayer components. It has been known that, compared with smectites, the cation adsorption sites on allophane, imogolite, and amorphous oxy-hydroxides of Fe and Al exhibit higher selectivities for heavy metal cations (Wada, 1989; Alloway, 1990). But, the role of polymeric HyA and HAS ions on Mt in heavy metal adsorption is only poorly understood. Harsh and Doner (1984) reported that Cu adsorption was promoted by the presence of HyA polymeric components on Wyoming Mt. Their electron spin resonance data also confirmed the existence of chemisorbed Cu onto HyA polymeric components of the synthetically interlayered Wyoming Mt; however, there was also evidence for the existence of mobile aquoted Cu species, indicating the presence of some electrostatically bound Cu2+. Recently, adsorption phenomenon of Cd on the HyA–Mt complex has been reported (Keizer and Bruggenwert, 1991; Sakurai and Huang, 1995, 1996; Lothenbach et al., 1997, 1998; Taniguchi et al., 2000). Keizer and Bruggenwert (1991) mentioned that in presence of Ca, the Cd adsorption by HyA–Mt at pH 5.0 was equivalent to the decrease of adsorbed Ca, indicating ion-exchange process of Cd adsorption. But at pH 7.0, Cd adsorption was much higher than the decrease of Ca adsorption, suggesting a significant specific adsorption of Cd on HyA–Mt. Sakurai and Huang (1996) reported that desorption of Cd from Mt by 1 M KCl was 90% of that adsorbed, whereas that desorbed from HyA–Mt and HAS–Mt were 30 and 37%, respectively, much lower than that in the Mt system. This indicated that a substantial portion of the Cd was probably adsorbed by the interlayer material through specific adsorption.
Although macroscopic equilibrium studies and models give some important information about sorption–desorption phenomena, no molecular information is revealed. Scheidegger and Sparks (1996) suggested that molecular and/or atomic resolution surface techniques should be employed to corroborate the proposed mechanisms hypothesized from equilibrium and kinetic sorption experiments. Studies using surface spectroscopic and microscopic techniques have shown that the adsorption of heavy metals on clay and oxide surfaces results in the formation of multinuclear or polynuclear surface complexes much more frequently than previously thought (Fendorf et al., 1992; Charlet and Manaceau, 1993; Fendorf and Sparks, 1994; Junta and Hochella, 1994; Fendorf and Fendorf, 1996). Formation of multinuclear metal hydroxides of Pb, Ni, Cu, and Cr (III) on the surfaces of oxides and aluminosilicates has been observed (McBride et al., 1984; Bleam and McBride, 1986; Chisholm-Brause et al., 1990; Roe et al., 1991; Charlet and Manaceau, 1992; Fendorf et al., 1994; Bargar et al., 1995; Papelis and Hayes, 1996; Scheidegger et al., 1996a, 1996b). Such surface complexes or precipitates have been observed at metal surface loadings far below a theoretical monolayer coverage and in a pH-range well below the pH where the formation of metal hydroxide precipitates would be expected according to the thermodynamic solubility product (Fendorf et al., 1994; Scheidegger et al., 1996a, 1996b). Spadini et al. (1994) applied extended x-ray adsorption fine structure (EXAFS) spectroscopy to study the adsorption and coprecipitation mechanisms of Cd on hydrous ferric oxide (HFO) and 
-FeOOH (goethite) in aqueous solutions. Their findings revealed that Cd substituted for Fe within -FeOOH lattice. From the studies with increasing surface loading, they assumed the existence of at least two different surface sites. The high affinity/low affinity surface site density ratio was much higher on HFO than on -FeOOH. Siantar and Fripiat (1995) studied the Pb retention and complexation mechanisms on a Mg-hectorite and observed an increasing selectivity of the clay for Pb over Ca with increasing Ca concentrations in the clay. They assumed that the higher adsorption specificity for Pb could be partly due to the differences in ionic potentials and hydration energies between Pb and Ca. However, in the solid retrieved from exchange solution, an infrared spectroscopy study indicated the formation of a Pb–hdroxycarbonate complex from atmospheric CO2. Scheidegger et al. (1996a) applied high-resolution transmission electron microscopy (HRTEM) to elucidate Ni sorption mechanisms of pyrophyllite and reported that at low Ni sorption densities surface precipitation occurred preferentially along the edges of the particles. Based on HRTEM findings and on results from their previous EXAFS study, they hypothesized that the formation of a mixed Ni–Al hydroxide phase on the pyrophyllite was responsible for the sorption behavior above pH 7. Fendorf and Sparks (1994) used EXAFS, Fourier transform infrared, and HRTEM to study Cr (III) sorption on Si-oxide. At low Cr (III) surface coverage (<20%), adsorption was the dominant process and Cr (III) formed an inner-sphere monodentate surface complex on Si. As Cr (III) surface coverage increased (>20%), surface precipitation occurred and became the dominant sorption mechanism at higher surface coverages. Surface clusters were observed at the high surface loadings. This EXAFS study, as well as others (Fendorf et al., 1994; Scheidegger et al., 1996a), demonstrates that the total coverage of surface sites is not necessary for the formation of multinuclear surface complexes and implies that the natural mineral surfaces promote hydrolysis and multinuclear complex formation.
Because of the importance of hydroxy-interlayered clays in acid soils and sediments and the paucity of research results on the role of hydroxy-interlayered components in trace metal adsorption, much is left to be done (Barnhisel and Bertsch, 1989). Therefore, the present study was carried out to compare the Cd, Zn, and Pb adsorption behavior of Mt, HyA–Mt, and HAS–Mt complexes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
The HyA and HAS ionic solutions were prepared through interaction of orthosilicic acid, AlCl3, and NaOH solutions as follows. Orthosilicic acid prepared from tetraethyl orthosilicate (Farmer et al., 1979) was mixed with 0.1 M AlCl3 solution to obtain an Si/Al molar ratio of 0.00 and 
0.50. The solutions were then titrated with 0.1 M NaOH at the rate 0.2 to 0.5 mL min-1 with continuous stirring to give NaOH/Al molar ratios of 2.5. The solutions were diluted to 2 L (final Al concentration 4 mM) and aged for 7 d at 20°C. The pHs of the solutions were recorded and clear filtrates were obtained by passing through a Toyo Roshi cellulose NO3 membrane filter (Toyo-Roshi Co., Tokyo, Japan) of 0.2-µm pore size to remove the solid particles of Al(OH)3 or any aluminosilicates that might have formed. The Al and Si concentrations in the filtrates were determined according to the methods reported by Davenport (1949) and Weaver et al. (1968), respectively. The OH/Al and Si/Al molar ratios, pH, and Al concentrations of the HyA and HAS parent solutions reacted with Mt are presented in Table 1.
|
The negative charge characteristics of Mt and HyA– and HAS–Mt complexes were determined through measurement of Ca2+ retained in the pH range of 4 and 7.5 following the procedure described by Wada and Okamura (1980), except that CaCl2 was used instead of NH4Cl for saturating the samples.
To prepare samples for x-ray diffraction (XRD) analysis, untreated Mt and HyA– and HAS–Mt complexes were treated repeatedly with 1 M KCl or with 0.5 M MgCl2 to make them K- and Mg-saturated, respectively. The K- and Mg-saturated specimens were washed successively with a small amount of water and 80% CH3OH; then parallel-oriented clay specimens were prepared by spreading the clay slurries on glass slides. The XRD analysis of the clay specimens was carried out with a Rigaku x-ray diffractometer RAD-1A (Rigaku Co., Tokyo, Japan) by using Fe-filtered CoK radiation generated at 30 kV and 10 mA. In order to investigate more precisely the mineralogical characteristics of the HyA– and HAS–Mt complexes, the K-saturated specimens were heated at 110, 300, and 550°C; the Mg-saturated specimens were solvated with glycerol; and their x-ray diffractograms were recorded.
Metal Adsorption Experiments
Fifty milligrams of clay specimen were mixed well with 10 mL of 0.02 M NaClO4 in 50-mL polypropylene copolymer (PPCO) Nalgene centrifuge tubes (Nalge Nunc International, Rochester, NY). Later, 10 mL of 2 x 10-6 M Cd(ClO4)2, Zn(ClO4)2, or Pb(ClO4)2 solution were added and mixed well, so the initial concentration of background electrolyte in the equilibrium solution was 0.01 M and that of metal was 10-6 M or 1 µM. The pHs of the suspensions were adjusted to around 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, 7.0, 7.5, and 8.0 with 0.1 M HClO4 or 0.1 M NaOH. The tubes with samples were shaken for 24 h at 25°C for equilibration with another interim adjustment of their pHs. A 24-h reaction period was chosen based on the result of a preliminary kinetic experiment conducted under the experimental conditions of pH 5.0 and metal concentration of 10-6 M. After equilibration, separation of solid and liquid phases was done by centrifugation at 3100 g for 30 min. A 10-mL aliquot of supernatant solution from each tube was drawn for determination of equilibrium pH and metal concentration in the equilibrium solution by inductively coupled plasma–atomic emission spectroscopy (ICP-AES) equipped with UAG-1 (Shimadzu, Kyoto, Japan) carrying ultrasonic nebulizer. Total adsorption (TA) of each metal was calculated from the decrease in metal concentration from initial to equilibrium solution.
Partitioning of adsorbed metal was done by a carefully controlled washing procedure similar to that reported by Tiller et al. (1984) using 0.01 M NaClO4 solution. First, replacement of a 10-mL aliquot of the removed supernatant solution with 10 mL of metal-free 0.01 M NaClO4 solution was made. The clays were then quickly resuspended and were immediately centrifuged. The supernatant wash solution was taken for determination of metal concentration by ICP-AES equipped with UAG-1 carrying ultrasonic nebulizer. The amount of adsorbed metal remaining on the clay following such washing with metal-free 0.01 M NaClO4 solution was defined as strongly adsorbed (SA). Weakly adsorbed (WA) metal was obtained from the difference of TA minus SA.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
The CEC of untreated Mt showed only a slight pH dependence. Fixation of both HyA and HAS ions resulted in a significant reduction in permanent negative charge and a substantial increase in pH-dependent negative charge (Table 1). However, the CEC of HAS–Mt complex exhibited a stronger pH dependence than that of HyA–Mt. The PZSE value was to some extent higher for the HyA–Mt than for the HAS–Mt. Because hydroxides or oxides of Si have much lower PZSE values (lower than 2.0) than those of Al (as high as >8.5) (Parks, 1965), a lower PZSE value of the HAS–Mt complex than that of HyA–Mt complex could be considered logical.
As calculated using the method of Hsu (1968), both HyA and HAS materials fixed on Mt had almost equal OH/Al ratios around 2.7 (Table 1), exactly the same as the ratio calculated for a variety of preparations by Barnhisel (1977). The HyA/HAS–Mt complexes used by Harsh and Doner (1984), Sakurai and Huang (1998), and Taniguchi et al. (2000), however, had this average OH/Al ratio ranging from 2.8 to 2.93.
The K-saturated HyA–Mt complex had expanded d(001) spacing of 1.47 nm, in comparison with 1.22 nm as shown by Mt (Table 2), suggesting that at least a portion of HyA was fixed in the interlayer. Heating at 110°C resulted in layer collapse to 1.39 nm. When heated to 300°C, a broad band from 1.19 to 1.32 nm appeared. This implies that the difference between d(100) spacings of the HyA–Mt and that of Mt ranged from 0.18 to 0.31 nm, suggesting that the adsorption of HyA ions was not uniform throughout the interlayer space. The layers collapsed to 1.07 nm after heating at 550°C. A broad band from 1.39 to 1.65 nm was observed in the K-saturated air-dried HAS–Mt complex. This broad band remained unchanged upon heating at 110°C. Heating at 300°C caused partial layer collapse, shifting this broad band between 1.10 to 1.32 nm. Appearance of such broad band also indicates a nonuniform adsorption of HAS throughout the interlayer space. A layer collapse similar to Mt was observed after heating at 550°C.
|
Metal Adsorption Behavior
pH Dependence and Adsorption Selectivity
Compared with Mt, the HyA–Mt and HAS–Mt complexes showed an abruptly altered behavior with respect to the adsorption of Cd, Zn, and Pb (Fig. 1)
. Among the three metals, Cd and Zn adsorption patterns were abruptly different from Pb adsorption on the clays, particularly on the complexes.
|
5), Mt adsorbed larger amounts of Cd and Zn than the complexes. However, substantial fractions of these two metals were adsorbed also on the complexes at pH below 5. As the pH rose above 5, Cd and Zn adsorption on the complexes steeply increased and reached plateau levels at pH between 6 and 7 when adsorption was almost 100%. Obviously, the adsorption on the complexes greatly superseded that on Mt at pH values >5. The maximum fractional adsorption of Cd and Zn on Mt at the highest pH values were no more than 49%. In Fig. 1, a clear adsorption edge (the pH range where adsorption increased abruptly) can be detected for the adsorption of Cd and Zn on both HyA– and HAS–Mt complexes. The Cd adsorption edges for the two complexes were, indeed, not very different from each other. The Zn adsorption edge for HAS–Mt complex, however, was comparatively lower than that for HyA–Mt complexes.
In comparison to Cd and Zn, Pb adsorption on both HyA–Mt and HAS–Mt were substantially higher at the lowest pH (4). Unlike Cd and Zn, Pb adsorption on HyA–Mt and HAS–Mt complexes was also significantly higher than that on Mt even at the lowest pH (around 4.0). However, both the complexes showed an almost identical adsorption patterns across the entire pH range. Clear adsorption edges for Pb adsorption on the complexes were not possible to detect. At low pH values, manifold higher fractional adsorption of Pb relative to that of Cd and Zn on the clays, particularly on the complexes, indicates that adsorption edges of Pb on the clays, especially on the complexes, might exist in a lower pH range than that of other two metals. For a given type of surface sites, the pH-dependent adsorption edge of various metals is a function of their adsorption strength (affinity), if all other conditions (e.g., affinity for anions) remain constant. It is often argued that the affinity of a surface for cations is high if adsorption occurs at low pH.
Linear regression was used to fit data lines of -log [metal adsorbed (mol g-1)] (y) vs. pH (x1) with the model y = ß0 + ß1x1. The values for ß1 range from 0.027 to 0.338 (Table 3) for different adsorption systems and are significantly different from 0 at = 0.01, indicating that the proton (pH) had a significant influence on all the adsorption systems. The ß1 values can be used as an indicator of the degree of pH dependence of the adsorption systems. Therefore, on Mt, influence of pH on adsorption increased in the order: Cd < Zn < Pb. Adsorption of Zn and Cd on the complexes were more pH-dependent than that on Mt. In the case of Cd adsorption, influence of pH was not very different between HyA–Mt and HAS–Mt, but HAS–Mt showed a greater influence of pH than HyA–Mt in Zn adsorption. In case of Pb adsorption, there was an apparent decrease in ß1 values for the complexes in comparison with Mt. Considering very high fractional adsorption (85–100%) in the entire pH range, we assume that Pb adsorption on the complexes would show greater pH dependence, if a higher initial Pb concentration were used.
|
|
|
|
In general the most important factors that influence the relative selectivity of cations in solution are valence and ionic radius. While selectivity behavior for Group IA and IIA elements of the periodic table can generally be explained using these factors, no such generalization can be made for the heavy metals (Puls and Bohn, 1988). The ionic potentials (Z2/r) of the three metals estimated based upon the charge (Z) and radius of the ion (r) follow the order: Zn > Cd > Pb (Table 4). If the metal adsorption on the clays were entirely electrostatic, ions of higher ionic potentials should have adsorbed more strongly. The selectivity sequences observed in the present study, therefore, suggest that the Mt–metal and HyA– and HAS–Mt–metal bonds are not entirely electrostatic. The presence of 104 times higher concentration of Na+ relative to M2+ in the equilibrating solutions also does not support the predominance of a simple electrostatic metal bonding in the adsorption process. However, slightly higher amounts of Cd and Zn adsorption on Mt relative to the complexes only at low pH values might be a reflection predominantly of the electrostatic bonding of these two metals involving permanent charges on the surface of the clays.
Many authors explained selectivity by examining the metals and clays hard–soft acid base (HSAB) behavior (Puls and Bohn, 1988). The HSAB principle states that hard Lewis acids prefer to complex or react with hard Lewis bases and soft acids prefer to complex or react with soft bases. Hard indicates high electronegativity, low polarizability, a small ionic radius, and a lack of an unshared pair of electrons in their valence shell, while the opposite is true for soft ions (Pearson, 1963). In general, the above sequences demonstrate a greater affinity of oxides for hard metal cations. This implies that the reactive adsorption sites of oxides are hard in character. Soil clay minerals seem to behave as soft bases relative to water, which is very hard base (Sullivan, 1977). On the basis of the Misono softness parameter (Misono et al., 1967), softness decreases in the order: Pb > Cd > Zn, and metal selectivity on Mt (soft Lewis base) should follow the same sequence according to the HSAB principle. But the selectivity sequences of Pb > Zn > Cd on the Mt and Pb >> Zn = Cd on the complexes as observed in the present study do not match well with the HSAB principle. The literature also shows that there are discrepancies in the application of the HSAB principle, particularly with respect to the trace metal cations (Abd-Elfattah and Wada, 1981). McBride (1991), analyzing the reported metal selectivity sequences on an amorphous oxy-hydroxides of Fe, Al, and Si, concluded that on the oxide surfaces, harder transition metals tend to be preferred over softer transition metals, but softer nontransition metals (e.g., Pb2+) are preferred over harder nontransition metals (e.g., Cd2+). A high ionic potential of Zn2+ causes it to be adsorbed more strongly than Cd2+, but Zn2+ is insufficiently soft to be adsorbed as strongly as Pb2+ (McBride, 1991).
The observed selectivity sequences of Pb > Zn > Cd on the Mt and Pb >> Zn = Cd on the complexes, however, resemble the metal selectivity sequences reported for amorphous Fe and Al hydroxides (Kinniburgh et al., 1976) and for the silanol groups of silica gel (Dugger et al., 1964; Schindler et al., 1976). As mentioned above, in the low pH range (up to pH 5), the metals especially Cd and Zn were predominantly adsorbed on the permanent charge sites of Mt and the complexes as well. At pH >5, we assume a predominant involvement of oxide-like surfaces on the complexes such as edge surfaces of Mt and broken edges of the fixed HyA or HAS polymers in the adsorption of the metals. Also on Mt, a substantial portion of the metals were probably adsorbed on the edge surfaces especially in higher pH range. Significant metal adsorption on the complexes below their PZSE indicates specific adsorption predominantly on the HyA and HAS components fixed on Mt. Many authors also concluded similar oxide-like metal adsorption behavior of layer silicate clays based on the observations that adsorption was strongly pH dependent while independent of ionic strength (Peigneur et al., 1975; Maes et al., 1976; Inskeep and Baham, 1983; Maes and Cremers, 1986; Puls and Bohn, 1988; Schulthess and Huang, 1990). They ascribed such observations to metal complexation, principally with AlOH and SiOH sites on the edges of the layer silicate crystallites. Ziper et al. (1988) suggested that Cd adsorption on Mt occurs on edge, planer, and interlayer sites; however, they stated that specific adsorption by edge and planar sites is much greater than by interlayer sites.
Adsorption of metal by oxides is believed to be controlled by competitive ion-exchange reactions between the metal and the proton (Schulthess and Sparks, 1989). The variable pH-dependent hydrolysis of metals changes the Lewis acid strength of the aqueous metal species in solution and thus stands as an important determinant of competitive ion exchange. Therefore, consideration of adsorption of hydrolyzable metals (Cd, Co, Cu, Pb, and Zn) via adsorption of MOH+ and M(OH)0 species through hydroxyl bridges (James and Healy, 1972) and through competitive ion exchange (Schulthess and Sparks, 1989) are complementary to each other. Although there is considerable variation in the reported hydrolysis constants of the metals, the first hydrolysis constants (pK1) follow the increasing order: 7.93 for Pb (Forbes et al., 1976) < 9.0 for Zn (Perrin, 1969) < 10.2 for Cd (Perrin, 1969). Thus, it is apparent in our results that the cation that is most readily hydrolyzed in solution also has the greatest affinity for the clays. Schulthess and Huang (1990) observed that aqueous Pb, which speciates to PbOH at much lower pH values than Cd and Zn, was significantly adsorbed on Mt at a much lower pH value than the others. Kinniburgh et al. (1976) also observed a good correlation between cation adsorption and cation hydrolysis (usually with 1% of the metal hydrolysis occurred) of several divalent cations on Fe and Al oxides.
Ion activity products [IAPM(OH)2] relative to respective metal (M) hydroxides [M(OH)2] were calculated for all the solid–solution systems utilizing their equilibrium pHs and metal concentrations. The IAPM(OH)2 values were lower than the Ksp of respective M(OH)2 for all the solid–solution systems (Fig. 3) of the present study, suggesting undersaturation of the systems with respect to concerned M(OH)2 solid phase. However, such a situation does not necessarily indicate that the systems were free from precipitation reactions. Using molecular and/or atomic resolution surface techniques, formation of multinuclear metal hydroxides of Pb, Ni, Co, Cu, and Cr (III) on the surfaces of oxides and aluminosilicates have been observed at metal surface loadings far below a theoretical monolayer coverage and in a pH-range well below the pH where the formation of metal hydroxide precipitates would be expected according to thermodynamic solubility product (Fendorf et al., 1994; Scheidegger et al., 1996a, 1996b). Therefore, the possibility of formation of multinuclear metal hydroxides of Cd, Zn, and Pb on the clay surfaces merits due attention in describing the metal retention mechanisms under the present experimental conditions. Although it may not be a significant process, the possibility of formation of hydroxycarbonate complexes and precipitates from atmospheric CO2, especially at higher pH, should not be totally ignored, since CO2 was not eliminated from the adsorption systems. Siantar and Fripiat (1995), employing infrared spectroscopy, studied the Pb retention and complexation mechanisms on a Mg-hectorite and suggested Pb-hdroxycarbonate complex formation in the clay interlayers.
|
0.5) and around 10 times higher on allophane (Si/Al 1.0), which correlated well with their surface areas (McBride, 1991). Zinc could have higher affinity for Si-OH group than for Al-OH group; it has been reported to form Zn–Al–Si precipitates on the clay mineral surface (Ford and Sparks, 2000).
Partitioning of Strong and Weak Adsorption
Following the adsorption of Cd, Zn, and Pb by Mt and the complexes, the adsorbed metals were separated into strongly and weakly bound forms as described in Materials and Methods. The results are presented in Fig. 4. Total adsorbed and SA Cd, Zn, and Pb increased with pH for the range studied. The proportion of the SA to TA metals (i.e., SA/TA) increases with pH and also depends on the kind of adsorbent and metal as well. In comparison with Cd and Zn adsorption, SA/TA figures for Pb adsorption at any given pH are remarkably higher, and this was true for Mt as well as for the complexes. In the case of Cd and Zn adsorption on the complexes, TA metals may become an approximation for SA metals when pH = 7. In contrast, for Pb, SA could represent TA on the complexes from much lower pH values (5). The HyA–Mt and HAS–Mt complexes were not much different from each other with respect to strong and weak adsorption of the three metals. Thus, the results presented above indicate that the complexes are more reactive than Mt in strong adsorption of the metals. Among the metals, strong adsorption of Pb is more pronounced than that of Cd and Zn on the adsorbents included. Therefore, as pH values decrease, less reactive adsorbents are involved, the proportion of metals adsorbed strongly may decrease substantially. The Zn and Cd weakly adsorbed on the complexes generally increase somewhat with pH up to a maximum at pH around 5 to 6, and then decrease probably due to preferred reactions with adsorption sites of higher affinities at higher pH values (Fig. 4).
With increasing pH, the increasing difficulties in desorption of all the three metals on Mt, HyA–Mt, and HAS–Mt (see WA in Fig. 4) could result in part due to the probable formation of complexes and precipitates as discussed above. Nevertheless, such results indirectly support our speculated predominance of specific binding of metals in our adsorption systems. Such difficulties in desorption were, however, not similar for all the clay–metal combinations. The difficulties in desorption increased among the clays in the order: Mt << HyA–Mt 
HAS–Mt and among the metals in the order: Zn Cd << Pb, which was similar to the sequence of metal selectivity. Therefore, we believe that specific adsorption rather than nonspecific adsorption principally controlled the observed metal selectivity, especially on HyA–Mt and HAS–Mt in our experiments. On Mt, metal selectivity was probably controlled by a combination of both nonspecific and specific adsorption.
|
ACKNOWLEDGMENTS |
|---|
Received for publication June 20, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
-Al2O3–water interface. 209th American Chemical Society National Meeting, extended abstracts, Vol. 35(1).This article has been cited by other articles:
|
|
 |
|
  U. K. Saha, C. Liu, L. M. Kozak, and P. M. Huang Kinetics of Selenite Adsorption on Hydroxyaluminum- and Hydroxyaluminosilicate-Montmorillonite Complexes Soil Sci. Soc. Am. J., July 1, 2004; 68(4): 1197 - 1209. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. K. Saha, U. K. Saha, K. Iwasaki, and K. Sakurai DESORPTION BEHAVIOR OF Cd, Zn AND Pb SORBED ON HYDROXYALUMINUM-AND HYDROXYALUMINOSILICATE-MONTMORILLONITE COMPLEXES Clays and Clay Minerals, October 1, 2003; 51(5): 481 - 492. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. K. Saha, S. Taniguchi, and K. Sakurai Simultaneous Adsorption of Cadmium, Zinc, and Lead on Hydroxyaluminum- and Hydroxyaluminosilicate-Montmorillonite Complexes Soil Sci. Soc. Am. J., January 1, 2002; 66(1): 117 - 128. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
