| 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 (sahajp2001{at}yahoo.com)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cd in most ternary systems were consistent with predictions based on the first hydrolysis constants. We propose that HyA-Mt and HAS-Mt are composed of nonuniform metal adsorption sites. For a given type of site, metal selectivity was predominantly determined by the metal properties, softness and ease of hydrolysis, while ionic potential had a limited predictive index for metal adsorption. The overall adsorption behavior of the metals along with their extraction patterns showed that the strength of adsorption followed the order of Pb >> Zn > Cd among the metals and of HyA-Mt HAS-Mt >> Mt among the adsorbents.
Abbreviations: CEC, cation-exchange capacity • CM, competing metal • ESR, electron spin resonance • HAS, hydroxyaluminosilicate • HSAB, hard-soft acid base • HyA, hydroxyaluminum • ICP–AES, inductively coupled plasma-atomic emission spectrometry • IAP, ion activity product • Me, metal • MeI, initial concentration of each metal • Mt, montmorillonite • PM, primary metal • PZSE, point zero salt effect • SA, strongly adsorbed • TA, total adsorbed • WA, weakly adsorbed • XRD, x-ray diffraction
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
or double 
gibbsite-like rings at OH/Al ratios <2.1. A further increase in OH/Al ratio (2.2 to 2.7) transforms these cyclic derivatives into larger polymers through 
and 
to 
with a reduced net positive charge per Al atom, but with ionic charge remaining positive until OH/Al equals 3.0. Another proposed model of Al polynucleation (Bertsch et al., 1986) includes Al13 polynuclear species 
and larger condensed units of this basic structure, usually in combination with a minimum of other species, such as the dimer 
or trimer 
. The HAS ions are formed through the interaction of HyA ions with orthosilicic acid. They resemble allophane-like constituents (Wada and Greenland, 1970) or illdefined fraction of allophane-imogolite complex (Farmer et al., 1983) or illdefined aluminosilicate complexes (Inoue and Huang, 1986). As reactive Al species, the HyA and HAS ions interact with both organic and inorganic soil components and play a significant role in regulating the mobility of plant nutrients and pollutants in the acidic soil environments. In 2:1 type silicate clays, HyA and HAS interlayers are common in acid to slightly acid soils. As a consequence, several attempts have been made to synthesize such interlayers in smectites and vermiculites (Rich, 1960, 1968; Coulter, 1969; Barnhishel and Bertsch, 1989). Laboratory investigations concluded that irreversible adsorption of HyA and HAS on the silicate surface results in a drastic modification of electrochemical and mineralogical properties of the host clays (Inoue and Satoh, 1992, 1993; Sakurai and Huang, 1998). Effects of hydroxy-interlayering on phosphate retention (Saha and Inoue, 1997a; Saha et al., 1998), K+ and NH+4 fixation and exchange (Saha and Inoue, 1997b, 1998), and K+/Ca2+ and NH+4/Ca2+ selectivities (Kozak and Huang, 1971; Saha et al., 1999) have been reported. However, there has been little work conducted on the adsorption of metals by the interlayer components (Keizer and Bruggenwert, 1991).
Wada (1989) and Alloway (1990) observed that in comparison to smectites, the cation adsorption sites on allophane, imogolite, and amorphous oxy-hydroxides of Fe and Al had higher selectivities for heavy metals, but the role of polymeric HyA and HAS ions on Mt in heavy metal adsorption is poorly understood. Harsh and Doner (1984) observed enhanced Cu adsorption because of the presence of HyA polymeric components on Wyoming Mt. The electron spin resonance (ESR) data also confirmed the existence of chemisorbed Cu onto HyA polymeric components of HyA-Mt, however, there was also evidence for the existence of some electrostatically bound Cu2+. Recently, adsorption behavior 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) explained Cd adsorption on HyA-Mt as an electrostatic process at pH 5.0, but they observed significant specific adsorption at pH 7.0. Based on different desorption patterns of adsorbed Cd on Mt, HyA-Mt, and HAS-Mt, Sakurai and Huang (1996) also indicated that a substantial portion of the Cd was probably adsorbed on the interlayer material through specific adsorption. For single metal systems, Saha et al. (2001) observed that the adsorption selectivity on both HyA-Mt and HAS-Mt complexes followed the order of Pb >> Zn Cd. Based on desorption patterns and other data, they concluded that specific adsorption involving chemical bonding might have dominated over electrostatic bonding in the metal adsorption processes.
Although experimental evidence demonstrates that anion adsorption in soils is significantly affected by the presence of other anions, relatively few studies have investigated simultaneous adsorption of trace metal cations by model sorbents and soils (Murali and Alymore, 1983). Many studies have investigated the selectivity sequences for trace metal adsorption on various model sorbents and soils. However, it is not clear whether strong preferential adsorption of a particular heavy metal ion occurs because of its relative affinity for a specific site or its adsorption to the sites that are unavailable to other metals (Benjamin and Leckie, 1981). Information generated through simultaneous adsorption-desorption experiments can be used to distinguish these two possibilities. Thus, the present study was carried out to compare the adsorption behavior of Cd, Zn, and Pb from mixed solutions on Mt, HyA-Mt, and HAS-Mt complexes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The HyA and HAS ionic solutions were prepared through interaction of orthosilicic acid, AlCl3 and NaOH solutions as follows. For HyA, 0.1 M AlCl3 solution was titrated with 0.1 M NaOH at the rate 0.2 to 0.5 mL min-1 with continuous stirring to give a NaOH/Al molar ratio of 2.5. For HAS, orthosilicic acid prepared from tetraethyl orthosilicate (Farmer et al., 1979) was mixed with a 0.1 M AlCl3 solution to obtain a Si/Al molar ratio of 
0.50. The solution was then titrated with 0.1 M NaOH in the manner described above. The HyA and HAS 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 0.2-µm pore size cellulose-NO3 membrane filter (Toyo-Roshi Co., Tokyo, Japan) to remove the solid particles of Al(OH)3 or any aluminosilicates. The Al and Si concentrations in the filtrates were determined according to 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.
|
Characterization of the Complexes
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 to 7.5 following the procedure described by Wada and Okamura (1980), with CaCl2 substituted for NH4Cl (see also Inoue and Satoh, 1993). The point zero salt effect (PZSE) of the clay complexes was determined by a modified salt titration method (Sakurai et al., 1988). The total and external surface areas of the clay samples were determined by ethylene glycol monoethyl ether (EGME) method (Eltantawy and Arnold, 1973) and by adsorption of N2 gas at -195°C using a BET surface area analyzer (Shibata P-850, Shibata-Kagaku Co., Tokyo, Japan), respectively. The internal surface area was calculated as the difference between total and external surface areas.
X-ray diffraction (XRD) analysis of the parallel oriented clay specimens was performed on K- and Mg-saturated clay specimens with a Rigaku X-ray diffractometer RAD-1A (Rigaku Co., Tokyo, Japan) using Fe-filtered CoK
radiation generated at 30 kV and 10 mA. 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 and the Mg-saturated specimens were solvated with glycerol, and their X-ray diffractograms were recorded.
Metal Adsorption Experiments
On Mt, HyA-Mt, and HAS-Mt, simultaneous adsorption of Cd, Zn, and Pb from their binary and ternary mixed solutions was studied with the following initial concentrations of each metal (MeI):(i) Cd-Zn, Cd-Pb, and Pb-Zn, Binary Systems with 1 x 10-6 M MeI; and (ii) Cd-Pb-Zn Ternary System with 1 x 10-6, 2 x 10-5, and 5 x 10-5 M MeI.
Thus, the ratio of two metals in the binary systems was 1:1 and that of three metals in the ternary systems was 1:1:1. The metal salts, Cd(ClO4)2, Zn(ClO4)2, and Pb(ClO4)2 were used to prepare the mixed solutions. The equilibration method and conditions are briefly described below. Fifty milligrams of clay specimen were mixed well with 10 mL of 0.02 M NaClO4 in 50-mL polypropylene copolymer (PPCO) centrifuge tubes (Nalgene brand product of Nalge Co., Rochester, NY). Immediately after mixing, a 10-mL aliquot of appropriate mixed metal solution with double MeI was added and mixed well. The clay/solution ratio was 2.5 g L-1, the concentration of background electrolyte was 0.01 M and the MeI was as targeted. The pHs of the suspensions were adjusted to 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 sample tubes 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. After equilibration, solid and liquid phases were separated by centrifugation at 3100 x g for 30 min. Duplicate 5-mL (10 mL in total) aliquots of supernatant solution from each tube were drawn for determination of pH and concentration of metals by ICP–AES (Inductively Coupled Plasma Atomic Emission Spectrometry) (ICPS-1000IV, Shimadzu, Kyoto, Japan) equipped with UAG-1 (Ultrasonic Aerosol Generator) ultrasonic nebulizer. The ICP-AES instrument was programmed for triplicate measurements on each sample. Total adsorption (TA) of each metal was calculated from the decrease in concentration of the respective metal from initial to equilibrium solution. A treatment blank carrying 20 mL of 0.01 M NaClO4 with 50 mg of clay specimen was included in samples in all adsorption experiments. The suspension pH of the blank was adjusted to 4.0 and equilibrated together with the samples, for which the pH was adjusted a second time during equilibrium. After equilibration, Cd, Zn, and Pb were not detected in the supernatant solution of the blank in any case.
Sorption in ternary metal system was evaluated at an initial concentration of 5 x 10-5 M (MeI) for each metal. At first, the 10-mL aliquot of the removed supernatant solution was replaced with 10 mL of metal-free 0.01 M NaClO4 solution. The clays were then quickly resuspended and were immediately centrifuged. The supernatant wash solution was taken for determination of metal concentrations. The amount of adsorbed metals 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 |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2.7 (Table 1), the same as the ratio calculated for a variety of preparations by Barnhisel and Bertsch (1989). The HyA/HAS-Mt complexes used by Harsh and Doner (1984), Sakurai and Huang (1998) and Taniguchi et al. (2000), however, reported an average OH/Al ratio ranging from 2.8 to 2.93. The K-saturated HyA-Mt complex had a d(001) spacing of 1.47 nm in comparison to 1.22 nm for Mt (Table 1), suggesting that at least a portion of HyA was fixed in the interlayer. Heating at 110°C resulted in the layer collapsing to 1.39 nm. When heated to 300°C, a broad band from 1.19 to 1.32 nm appeared. This implies that the thickness of HyA-interlayer in HyA-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 a broad band also indicates nonuniform adsorption of HAS throughout the interlayer space. Layer collapse similar to Mt was observed after heating at 550°C. The XRD results of air-dried and glycerated Mg-clays show that on solvation, both HyA-Mt and HAS-Mt complexes expanded from 1.55 to 1.79 nm, indicating that the presence of HyA and HAS ions in the interlayer did not lead to irreversible bonding between the silicate layers (Table 1). This is consistent with the behavior observed in most preparations of hydroxy interlayered smectites (Harsh and Doner, 1984).
The cation-exchange capacity (CEC) of untreated Mt showed only a slight pH-dependence (Table 1). 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, once again confirming the nonexchangeable adsorption of HyA or HAS ions on Mt. 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 in comparison to that for the HAS-Mt. Hydroxides or oxides of Si have much lower PZSE values (<2.0) than those of Al (8.5) (Parks, 1965), thus a lower PZSE value was observed for the HAS-Mt compared with HyA-Mt.
Fixation of HyA or HAS on Mt resulted in lower total and internal surface areas with a slight increase in external surface area (Table 1), as observed in previous studies (Inoue and Satoh, 1993; Sakurai and Huang, 1998).
Metal Adsorption Behavior
Binary Systems
Simultaneous adsorption behavior of Cd, Zn, and Pb on Mt, HyA-Mt, and HAS-Mt from Cd-Zn, Cd-Pb, and Pb-Zn mixed solutions (1 x 10-6 M MeI with 1:1 ratio) was evaluated. In the Cd-Zn system, the percentage of adsorption of Cd on Mt was greater than Zn throughout the entire pH range (Fig. 1)
. Adsorption of Cd and Zn on the complexes showed a stronger pH dependence in comparison with that on Mt. Below pH 5, Mt adsorbed larger amounts of Cd and Zn than the complexes. While a considerable fraction of Cd was adsorbed on HAS-Mt at pH <5, little Zn adsorption was observed below pH 5. However, substantial fractions of both metals were adsorbed on HyA-Mt at low pH (<5.0). As the pH rose >5, Cd and Zn adsorption on the complexes increased steeply and reached plateau levels at pH values between 6 and 8, where adsorption was almost 100%. Obviously, the adsorption of both metals on the complexes greatly exceeded that on Mt at pH values >5. Cadmium adsorption on both the complexes was higher than Zn adsorption at low pH values, but Zn adsorption exceeded Cd adsorption at higher pH (4.46 for HyA-Mt and 5.34 for HAS-Mt).
|
In the Pb-Zn system, adsorption of Pb on Mt was remarkably higher than that of Zn throughout the whole pH range. While Zn adsorption on the complexes was strongly pH-dependent, Pb adsorption showed only a slight pH-dependence. The percentage of adsorption of Zn on Mt was higher than that on HyA-Mt at pH <4.42 and on HAS-Mt at pH <5.45. Obviously, the adsorption of Zn on both complexes greatly exceeded that on Mt as the pH increased above those points. Adsorption of Pb on the complexes was remarkably higher than that on Mt at any pH value. Up to certain pH values, both HyA-Mt and HAS-Mt adsorbed more Pb than Zn in spite of their same concentration in the equilibrating solution.
Adsorption of metals from the mixed solutions of Cd-Zn, Cd-Pb, and Pb-Zn on Mt also showed some pH dependent behavior (Fig. 1), even though Mt is known to possess predominantly permanent negative charges. However, the pH-dependent adsorption edges (the pH range where adsorption increased abruptly) on Mt were generally weak. In Fig. 1, clear adsorption edges can be detected for the adsorption of Cd and Zn on both HyA- and HAS-Mt complexes, but an adsorption edge for Pb was not observed over the pH range of this study. At low pH values, higher fractional adsorption of Pb relative to that of Cd and Zn was observed, particularly on the complexes. This indicates that adsorption edges for Pb, especially on the complexes, might exist in a lower pH range in comparison to that of other two metals. For a given type of surface sites, the pH-dependent adsorption edge of various metals is a function of the adsorption strength (affinity). It is often argued that the affinity of a surface for a specific cation is high if the adsorption edge occurs at low pH. The adsorption edges for Cd and Zn on the complexes in different binary systems are plotted in Fig. 2 . On both complexes, almost identical Zn-adsorption edges were observed regardless of the type of accompanying metal (either Cd or Pb). Adsorption edges for Cd on HAS-Mt were also similar in both Cd-Zn and Cd-Pb system. On HyA-Mt, the Cd adsorption edge in the Cd-Zn system was higher than that in the Cd-Pb system. In comparison with HAS-Mt, HyA-Mt showed adsorption edges for both Cd and Zn in the lower pH ranges regardless of the type of accompanying metal (Fig. 2).
|
4 to 6.5 in case of the complexes. On Mt, adsorption of Cd was always higher than that of Zn. As observed in Cd-Zn system, Cd adsorption on both the complexes was higher than Zn adsorption at low pH values, while Zn adsorption exceeded Cd adsorption at higher pH. This trend in the ternary systems was very clear for the highest (5 x 10-5 M) MeI. At low pH values, adsorption of Cd and Zn on Mt from 1 x 10-6, 2 x 10-5, and 5 x 10-5 M MeI solutions was higher than that on HyA-Mt (at pH < 4.5) and HAS-Mt (at pH < 5). Adsorption of Cd and Zn on both complexes, however, greatly exceeded that on Mt at higher pH values. Adsorption of Pb on Mt and HAS-Mt also showed a similar trend as Cd and Zn for the highest (5 x 10-5 M) MeI only. For 1 x 10-6 and 2 x 10-5 M solutions, Pb adsorption on the complexes was substantially higher than that on Mt throughout the studied pH range. In general, a higher MeI tended to reduce the peragecent of the adsorption of each metal on any particular clay, especially at low to medium pH values. This trend became evident when adsorption data for 1 x 10-6 and 5 x 10-5 M (MeI) solutions were compared.
|
When the metal adsorption patterns for all binary systems and the ternary system with the same 1 x 10-6 M MeI were plotted with that of the respective metals in monometal systems under identical conditions (the plots are not shown), there was no evidence of significant competition between the metals for adsorption on any adsorbents. Likewise, in the ternary systems with 1 x 10-6 and 2 x 10-5 M MeI, Cd, Zn, and Pb adsorption patterns did not reflect any remarkable competitive effect (the plots are not shown). With 5 x 10-5 M MeI, competing metals (CM) in Cd-Zn-Pb system, presumably, suppressed the adsorption of Cd and Zn on Mt as well as on the complexes up to certain pH value (Fig. 4) . However, Pb adsorption on any adsorbent, was not affected by the competing metals even with this high MeI in Cd-Zn-Pb system.
|
|
Extractability of Adsorbed Metals
The TA metals were separated into SA and WA forms based on their relative extractability as described above in the Materials and Methods. The results are presented in Fig. 5
. The TA and SA forms of Cd, Zn, and Pb increased with pH over the range studied. The proportion of the SA to TA forms of metals, i.e., SA/TA, increased with pH and was dependent on the adsorbent and metal. Strong adsorbed/TA ratios for Pb adsorption at any given pH were much higher than that of Cd and Zn, for Mt as well as for the complexes. For Mt, most of the Cd and Zn were of WA form throughout the studied pH range. On the complexes, a great deal of the Cd and Zn was WA at pH values <5. The SA form of Cd and Zn greatly exceeded WA as the pH increased. Relative to Cd, adsorbed Zn on the complexes was less extractable. For the complexes, except in the low pH range (pH < 5), almost all of adsorbed Pb was strongly bound. In general, metal adsorption on the complexes was strong in nature when compared with that on Mt. The HyA-Mt and HAS-Mt complexes were, however, not much different from each other with respect to the SA and WA forms of Cd, Zn, and Pb.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The relative adsorption affinity or selectivity of the metals on the clays has been described by pH50 values (Table 2). Several factors such as differences in MeI, adsorbent concentration, type and concentration of background electrolyte preclude a rigorous comparison of the reported pH50 values (Sposito, 1984; Puls and Bohn, 1988). However, the pH50 values obtained for the adsorption of all three metals on the complexes in the present study are lower than the values reported for the respective metal adsorption on either Mt (Texas bentonite) clays (Puls and Bohn, 1988) or on amorphous Al-oxide adsorbents (Benjamin and Leckie, 1980). This is an indication that metal adsorption affinities for amorphous hydroxy-materials are enhanced when they exist in combination with clays (Harsh and Doner, 1984; Kaizer and Bruggenwert, 1991). Harsh and Doner (1984) attributed such promoting effects to a greater specific surface areas for the interlayered hydroxy-materials than their discrete counterparts, perhaps resulting from the atoll-like distribution, where the HyA or HAS polymers tend to populate near the edges of the particles (Frink, 1965).
Except in the Cd-Zn-Pb system with 5 x 10-5 M MeI, the addition of a second and even a third metal had relatively little effect on the adsorption of a given metal, which suggests that each metal preferentially binds to different sites, i.e., their preferred binding sites do not overlap each other. This observation is further supported by the fact that the total concentration of the metals in all these MeI-Clay systems was sufficient to occupy at most a small percentage of the surface sites and the ratio of the CM/the primary metal (PM) was always 1. Significant suppressive effects of a CM have been reported for soils and model sorbents (Zaosaki, 1974; Benjamin and Leckie, 1980; Basta and Tabatabai, 1992). In these studies, the total concentration of the metals or the CM/PM ratio in the equilibrating solution was 1 x 102 to 1 x 104 times greater than that of the present study. Benjamin and Leckie (1981), however, observed no competitive effects for Cd, Cu, Zn, and Pb adsorption on amorphous Fe oxyhydroxide, even when the CM/PM ratio was as high as 100.
In the absence of competition, the observed greater adsorption affinity of Pb than that of Cd and Zn would be related to some relevant metal properties. The predicted affinity sequences of metals based on of their ionic potential, hydrolysis constant, and softness are shown in Table 3. For electrostatic adsorption of metals with equal charge (Z) on ion-exchange materials, metal affinity should be inversely related to unhydrated radii (r). If the metal adsorption on the clays were entirely electrostatic, ions of higher ionic potentials (Z2/r) should be adsorbed more strongly. The poor agreement of the observed metal affinity sequences in the present study (Table 2) with those predicted by ionic potential (Table 3) suggests that the Mt-Metal (Me) and HyA/HAS-Mt-Me bonds are not predominantly electrostatic in nature. Based on CEC of the clay specimens, the total concentration of metal was sufficient to occupy at most a few percent of all surface sites. The presence of 1 x 104 times higher concentration of Na+ relative to Me2+ in the equilibrating solutions does not support the predominance of a regular ion-exchange phenomenon controlling metal adsorption. Puls and Bohn (1988) observed that while selectivity behavior for Group IA and IIA elements of the periodic table can generally be explained using their ionic potentials, no such generalization can be made for the heavy metals. Nevertheless, higher amounts of Cd and Zn adsorption on Mt relative to the complexes at low pH values may reflect electrostatic bonding of these two metals involving permanent charge sites on the surface of Mt.
|
The degree to which metals are hydrolyzed is likely to be a major factor determining the amount retained at any given pH. Reactions between metal ions and water to form hydrolysis products (i.e., MeOH+) are common to most metals and can be expressed in general form (Baes and Mesmer, 1976):
 |
The affinity sequence based on the first hydrolysis constants (formation of MeOH+ complexes) is reported in Table 3. Metal affinity sequences consistent with the adsorption of MeOH+ species have been used to describe metal adsorption by soils and Fe and Al oxides (Kinniburgh and Jackson, 1981; McBride, 1989). Although the high affinity of Pb can be predicted based on metal hydrolysis regardless of the adsorbent type, or CM, the observed affinity sequences of Zn and Cd show such consistence only for the complexes. Even for the complexes in Cd-Pb-Zn systems with 1 x 10-6 M MeI and in few other instances, the predicted order of Zn > Cd was not strictly followed.
The above discussion relating metal affinity sequences with those predicted by HSAB and other models indicates that metal adsorption behavior cannot be explained considering any single mechanism. Consequently, one must assume that the surfaces of HyA-Mt and HAS-Mt are composed of nonuniform metal adsorption sites. Therefore, the relative abundance and affinity for different metals varied for Mt, HyA-Mt, and HAS-Mt. For a given type of site, metal selectivity was perhaps determined by the metal properties, principally softness and ease of hydrolysis, while ionic potential was of limited predictive value. The relative extractability of adsorbed metals indicated that almost all of the adsorbed Pb on the complexes was strongly bound, whereas a substantial fraction of adsorbed Cd and Zn was weakly bound (Fig. 5). Compared with the complexes, metal adsorbed on Mt was easily extractable. Thus, the overall strength of adsorption followed the order of Pb >> Zn > Cd among the metals and of HyA-Mt HAS-Mt >> Mt among the adsorbents.
As the pH increased, aqueous metal cations hydrolyze, resulting in a suit of soluble metal-hydroxy complexes according to the generalized reaction given earlier. This hydrolysis may be accompanied by precipitation of metal hydroxides [Me(OH)2], which is experimentally indistinguishable from metal adsorption, which precludes comparison of relative adsorption affinities at pH 7. Ion activity products (IAP) 
relative to respective metal hydroxides [Me(OH)2] were calculated for all the solid-solution systems based on equilibrium pHs and metal concentration. Using the solubility product constants (Ksp) of respective Me(OH)2 (Baes and Mesmer, 1976), the ratios of IAPMe(OH)2/KspMe(OH)2 for all the solid-solution systems were calculated and plotted against the respective equilibrium pHs (Fig. 6)
. The IAP/Ksp ratios for Pb and Zn were nearly 1 at pH 7, suggesting supersaturation of the systems with respect to the Me(OH)2 solid phase. The 100% adsorption of Pb and Zn in this pH range irrespective of the MeI level (Fig. 3) also indicates the possibility of metal retention through precipitation, either in bulk phase or on the surface. The IAP/Ksp ratios of Cd suggested undersaturation of all the Clay-Cd systems with respect to Cd(OH)2 solid phase. Considering uncertainity in the solubility product and the possibility that surface precipitation may occur before saturation in the bulk fluid is attained, an IAP/Ksp < 1 does not necessarily indicate that the system is free from precipitation reactions. Utilizing molecular or atomic resolution surface techniques, the formation of multinuclear metal hydroxides of Pb, Ni, Co, Cu, and Cr (III) on the surfaces of oxides and aluminosilicates has been observed at metal surface loadings far below a theoretical monolayer coverage and in a pH range well below where the formation of metal hydroxide precipitates would be expected according to thermodynamic solubility product (Fendorf et al., 1994; Scheidegger et al., 1996a, b). Therefore, the formation of multinuclear metal hydroxides of Cd, Zn, and Pb on the clay surfaces merits due attention in describing the metal retention mechanisms under present experimental conditions. With increasing pH, the increasing difficulty in extracting all three metals on Mt, HyA-Mt, and HAS-Mt (see WA form in Fig. 5) could also result in part because of the formation of complexes and precipitates as discussed above.
|
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  U. K. Saha, L. M. Kozak, and P. M. Huang KINETICS OF CITRATE-INDUCED SELENITE DESORPTION FROM MONTMORILLONITE AS AFFECTED BY COMPLEXATION WITH HYDROXYALUMINUM AND HYDROXYALUMINOSILICATE IONS Clays and Clay Minerals, February 1, 2007; 55(1): 71 - 88. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Antoniadis, J. D. McKinley, and W. Y. W. Zuhairi Single-Element and Competitive Metal Mobility Measured with Column Infiltration and Batch Tests J. Environ. Qual., January 9, 2007; 36(1): 53 - 60. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 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]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||
