| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Departamento de Ciência do Solo, Universidade Federal de Lavras, 37200-000 Lavras, MG, Brasil
b Departmento de Solos, Universidade Federal de Viçosa, 36571-000, Viçosa, MG, Brasil
c Instituto de Geociências, Rua Barão de Geremoabo s/n, Campus Universitário de Ondina, Universidade Federal da Bahia, 40170-290, Salvador, BA, Brazil
* Corresponding author (mpfontes{at}mail.ufv.br)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: ALF, Alfisol • CECef, effective cation-exchange capacity • CECtot, total cation-exchange capacity • Gibb, gibbsite • Goet, goethite • Hem, hematite • Kao, kaolinite • Kd, distribution coefficients • OC, organic carbon • OX1, Oxisol number 1 • OX2, Oxisol number 2 • OX3, Oxisol number 3 • OX4, Oxisol number 4 • UL1, Ultisol number 1 • UL2, Ultisol number 2 • 
Oxi, sum of oxide content
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The adsorption of heavy metals has been studied and reported in the literature for silicate minerals (Swift and McLaren, 1991); for Fe-, Al-, and Mn-oxides (Schwertmann and Taylor, 1989), and for humic substances (Stevenson and Fitch, 1986). This adsorption has been described in terms of two molecular mechanisms: (i) nonspecific adsorption where the metallic cations behave as counter-ions in the diffuse layer; and (ii) specific adsorption resulting from surface complexation (Msaky and Calvet, 1990). The retention of different metals in a B horizon of a Brazilian Oxisol, based on relative amounts recovered in different soil fractions by sequential extraction, has demonstrated that Cd, Ni, Pb, and Cu were adsorbed both specifically and as exchangeable cations while Cr was more adsorbed specifically than as exchangeable cations (Gomes et al., 1997).
Several models have been proposed to explain the specific adsorption of metal cations, such as the exchange of H+ for Mn+, the preferential adsorption of hydrolyzed products, and the induced hydrolysis of cations on the surface of hydroxides (James et al., 1975). Hsu (1989) suggested that the exchange adsorption should be viewed as the competition between M2+ and H+ for the surface O on the basis of the cation relative affinity for this surface O. Accordingly, the affinity of cations for O or their relative tendency to form covalent bonds with O should be related to their electronegativity. Schwertmann and Taylor (1989) postulated that pH is the main force governing the adsorption of metal cations, and the fact that pH of maximum increase in adsorption is found to be linearly related to the first hydrolysis constant of the metal K1 = (MOH+)/(M2+)·(OH-) indicates that the hydrolyzed species (MOH+) is preferentially adsorbed over the unhydrolyzed one (M2+). Sposito (1989) defined the tendency of the metals to form covalent bonds on the basis of the ionic radius and the ionization potential quantified by the Misono softness parameter. This parameter measures the ability of the metal cations to form strong complexes according to their ability to form covalent bonds, in the following order: Pb > Cd > Cu > Co > Ni > Zn. According to McBride (1994), electronegativity is an important factor in determining which of the trace metals chemisorb with the highest preference and, on this basis, the predicted order of bonding preference would be Cu > Ni > Co > Pb > Cd > Zn > Mg > Sr. On the other hand, still according to this author, if the ability of the metals to chemisorb were based on only electrostatics, the strongest bond should be formed by the metal with the greatest charge-to-radius ratio, which would produce a different order for the same metals, that is, Ni > Mg > Cu > Co > Zn > Cd > Sr > Pb.
Selectivity sequence of heavy-metal cation adsorption has been found for goethite in the order Cu > Pb > Zn > Cd > Co > Ni > Mn, while hematite gave the same sequence except for an exchange in positions of Cu and Pb (Schwertmann and Taylor, 1989). Aluminum hydroxides have the order Cu > Pb > Zn > Ni > Co > Cd according to Hsu (1989), and for humic substances Schnitzer (1969) reports the order Cu > Pb > Ni > Co > Zn. Abd-Elfattah and Wada (1981) report that most of the observed sequences are correlated neither with the sequence of ionic radii, which is Pb (1.20) > Cd (0.97) > Zn (0.74) > Cu (0.72) > Ni (0.69) Å, nor with the sequence of electronegativity given by Cu (1.9) > Pb (1.8) = Ni (1.8) > Cd (1.7) > Zn (1.6). There is, however, a parallel between the adsorption sequence and the hydrolysis properties of the heavy-metal cations, as pointed out by several investigators (Forbes et al., 1976; Elliott et al., 1986; Schwertmann and Taylor, 1989).
Distribution coefficients (Kd) indicate the capability of a soil to retain a solute and also the extent of its movement in a solution phase (Reddy and Dunn, 1986). The mobility and fate of metals in the soil environment are directly related to their partitioning between soil and soil solution (Evans, 1989) and, therefore, are directly related to their distribution coefficients. According to Alloway (1995a), Kd is a useful parameter for comparing the sorptive capacities of different soils or materials for any particular ion, when measured under the same experimental conditions. Distribution coefficients have been used in studies of mobility and retention of heavy metals as related to their competition (Hendrickson and Corey, 1981; Reddy and Dunn, 1986; Anderson and Christensen, 1988; Gao et al., 1997). They also have been used in soil adsorption and mobility experiments for Cd (Sánchez-Martín and Sánchez-Camazano, 1993; Lee et al., 1996) and, more recently, they have been used to study mobility and solubility of trace metals (McBride et al., 1997; Römkens and Salomons, 1998) and long-term leaching of trace elements in a sludge-amended soil (McBride et al., 1999).
The main goal of this study was to evaluate the relative retention and mobility of several heavy metals applied together to soils. The specific objectives were to evaluate the selectivity sequence of these heavy-metal cations in several Brazilian soils by means of distribution coefficients, and to investigate the relationship between soil properties and adsorption of heavy metals by Brazilian soils with different chemical and mineralogical characteristics.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
25°C), then the soil was separated from the solution by centrifugation. The supernatant liquids obtained after centrifugation were analyzed and the heavy-metal cation concentrations remaining in solution were ascertained by atomic absorption spectrophotometry. The difference between the initial amount of metal in solution and the amount remaining in solution after the reaction period was assumed to be adsorbed by the soil. The distribution coefficients (Kd) were calculated according to Reddy and Dunn (1986), Anderson and Christensen (1988), and Alloway (1995a):
 |
A correlation analysis was performed by using the statistical package SAEG 5.0 (1993).
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Due to competition among the metal cations, plots of the concentration of metal adsorbed versus equilibrium concentration of metal in solution, for the majority of the cations, did not yield a straight line. Therefore, distribution coefficients as the slope of the isotherm line, as utilized by Gao et al. (1997) for simultaneous sorption of metal cations, could not be determined. Apparently, the smaller concentrations used by those authors did not induce a strong competition among the metal cations studied. Therefore, to be able to compare the different metal cations in each different soil, based on the reasoning of Sánchez-Martin and Sánchez-Camazano (1993), a Kd at a given concentration (Kd25), which represents the ratio of metal sorbed to equilibrium concentration at 25 mg L-1 metal in the added solution, was utilized to give one comparable coefficient for each metal and soil (Table 2, italicized print). This concentration was chosen because for Cd, Ni, and Zn, the amount of metal adsorbed consistently started to decrease above this concentration. The behavior of these elements was strictly in line with the results of Fontes et al. (2000) who, working with competitive adsorption of Cd, Cu, Pb, and Zn, showed that for elements such as Pb and Cu, competition had a very small effect on their adsorption. Linear, Langmuir, and Temkin isotherm models gave the best fit for their adsorption data. On the other hand, competition strongly influenced the adsorptive capacity of Cd and Zn, inducing a decrease in their adsorption in the more concentrated solutions such that their adsorption data were best modeled by quadratic and square root polynomial equations.
From the Kd25 results presented in Table 2, an adsorption order for the heavy metals was derived and a selectivity sequence is shown in Table 3. It should be noted here that due to the nonlinearity of the isotherms, the fact that the metals were added in equal mass, rather than equimolar amounts, introduces a possible bias in the comparison of the metals. There would be a bias in favor of the higher atomic mass elements, Cd and Pb, because of the decrease in Kd values at higher concentrations, while at the same time there would be a bias against these elements because they are at a competitive disadvantage for a finite number of sites. The direction of the overall bias is not predictable but would show up as differences in apparent selectivity at different concentrations of the metals. Inspection of the data in Tables 2 and 4 reveals that there is very little difference in the order of Kd values between the 5 and 25 mg L-1 concentrations.
|
|
The positions of Pb and Cu in the sequence, with Pb the more retained of the two for all but one soil, are reversed with respect to that expected on the basis of electronegativity values. But the preference for Pb over Cu in these soils agrees with predictions on the basis of the Misono softness parameter as postulated by Sposito (1989), as well as predictions based on the first hydrolysis constant, both of which are greater for Pb. Copper and Pb exchanging places is not unusual as seen by the results of Kinniburgh et al. (1976), Bruemmer et al. (1988), Hsu (1989), and Schwertmann and Taylor (1989) and for the synthetic minerals goethite, hematite, and Al oxides.
According to the sequences presented by Schwertmann and Taylor (1989), Zn is always adsorbed to a larger extent than Cd for the synthetic samples, which is not true for any of the soils in our study. In all cases, Cd was adsorbed to a larger extent than Zn, which is in line with results of de Matos et al. (1996) for the retention of these two heavy metals in a Brazilian Oxisol profile. Nickel and Zn were the least adsorbed metals in all but two soils and they also exchanged places, Ni being preferred in the neutral pH soils, and Zn being preferred in all four Oxisols. Nickel, despite its higher electronegativity value compared with Cd, was more retained than Cd in only two soils, and was the least retained in most of the other soils.
Taking into consideration that different soil materials such as silicate clays, Fe and Al oxides, and humic substances are present in different kinds and amounts in soils, these are not unexpected results. None of the selectivity sequences found followed the charge-to-radius ratio mentioned by McBride (1994), emphasizing that electrostatics alone do not explain bonding of divalent metals to soil particles and organic matter.
Chemical inputs to soils can take several pathways, including rapid leaching into ground waters, uptake by plants, volatilization to the atmosphere, and storage and retention by soil (Stigliani, 1988). Therefore, an important attribute of many soils is their ability to adsorb and store heavy metals. Table 4 presents the data for adsorbed metals and, for comparative purposes, the sum of the sorbed metal concentrations for each soil in order to assess the relative capacity of each soil to hold and, maybe, store metal cations. Table 4 shows that the order of relative metal-cation adsorption by the different soils was ALF > OX2 > UL2 > UL1 > OX1 > OX4 > OX3.
The soils from the Alfisol and Ultisol orders (ALF, UL1, and UL2), with higher values of pH and effective CEC compared with the other soils in this study (Table 1), were among the ones with highest relative capacity to adsorb metal cations. From the Oxisol order, only Oxisol number 2 (OX2), which has the highest organic matter and CEC of all of the Oxisols examined, was among the ones with highest adsorption values, whereas the other three adsorbed the least amounts of heavy metals. It seems that the younger soils with higher pH values adsorbed more metals compared with most of the older soils, suggesting that the increase in the hydrolyzed forms of the cations and effective CEC of the soils with the increase in pH was as important as the nature of the adsorbent material.
The overall results for the selectivity sequence of divalent heavy-metal adsorption seem to be related to a balance between the predominance of the metals in their unhydrolyzed (M2+) form in soils with lower pH (acid conditions) and the hydrolyzed (MOH+) form in the soils with higher pH. In the first case, the larger the metal electronegativity the easier the dissociation of the H from the functional groups of the soil mineral and organic particles, forming covalent bonding. In the second case, the influence of the metal hydrolysis on metal adsorption becomes more important for the more alkaline range of pH, as suggested by Forbes et al. (1976), Elliott et al. (1986), Bruemmer et al. (1988), and Schwertmann and Taylor (1989). Therefore, electronegativity and cation affinity for O should play more important roles in the order of adsorption of heavy metals in acidic older soils whereas pH and hydrolyzed forms of metal cations should be more important in the less acidic younger soils.
Correlation Analysis
The evaluation of the influence of soil characteristics on the metal-adsorption capacity of the soil was examined by correlation analysis. Table 5 shows the simple linear correlation coefficients between Kd values for the metal concentration of 25 mg L-1 (Kd25) and soil chemical and mineralogical characteristics.
|
The correlation coefficients significant at the 5% probability level showed that Cu adsorption, as inferred by its Kd values (Table 5), was related to the amounts of organic C, clay, and gibbsite. The correlation between goethite content and Cu adsorption with a correlation coefficient r = 0.646 was almost significant (P = 0.0587). These results indicate that Cu complexation by organic-matter compounds and specific adsorption by Fe and/or Al oxides may play significant roles on Cu adsorption by these soils.
The adsorption of Ni was related to pH as shown by the significance of the correlation coefficient between this variable and the Kd values for this element. Cation-exchange capacity, both effective and total, also is also correlated to Ni adsorption. These results suggest that the hydrolyzed form, which is affected by the pH, may be important in Ni adsorption, but also the capacity for sorption on exchange sites cannot be neglected.
Cation-exchange capacity was the variable that most strongly correlated to Pb adsorption. Also, it can be seen in Table 5 that pH may have influenced Pb adsorption as shown by the correlation coefficient r = 0.576 of P = 0.0879, which may be related to the Pb hydrolysis. Lead was most strongly adsorbed in the younger and higher pH soils.
Zinc gave no significant correlation between its Kd values and soil chemical and mineralogical characteristics. Goethite content with the correlation coefficient r=0.597 of P=0.0785 was the soil characteristic that gave the highest correlation, suggesting that the iron oxides may exert some influence on Zn adsorption by these soils.
The clear lack of significant effect of organic C on heavy-metal cation adsorption, with the exception of Cu, is probably related to the fact that most of the samples have a low organic-matter content, due to the fact that they were collected in the upper part of the B horizon. In this situation, the humic substances are likely to be chemically and physically stabilized by the interaction with the mineral fraction. A stronger influence of the humic substances on metal adsorption in samples from the surface horizons would be expected.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The most frequent heavy-metal cation selectivity sequences were Cr > Pb > Cu > Cd > Zn > Ni and Pb > Cr > Cu > Cd > Ni > Zn.
Younger soils from the Alfisol and Ultisol orders adsorbed more than most of the older soils from the Oxisol order. The order of decreasing total amount of heavy-metal cations adsorbed was, by soil order, Alfisol > 1 Oxisol > 2 Ultisol > 3 Oxisols.
Correlation analysis showed that for the competitive adsorption, the soil properties that most strongly related to heavy-metal cation adsorption were pH and CEC for Cd and Cr; organic C, clay, and gibbsite contents for Cu; pH and CEC for Ni, and CEC and possibly pH for Pb.
|
ACKNOWLEDGMENTS |
|---|
Received for publication May 28, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y.-J. Wang, C.-B. Li, W. Wang, D.-M. Zhou, R.-K. Xu, and S. P. Friedman Wien Effect Determination of Adsorption Energies between Heavy Metal Ions and Soil Particles Soil Sci. Soc. Am. J., January 11, 2008; 72(1): 56 - 62. [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]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
