JEQ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (14)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kaschl, A.
Right arrow Articles by Chen, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kaschl, A.
Right arrow Articles by Chen, Y.
Agricola
Right arrow Articles by Kaschl, A.
Right arrow Articles by Chen, Y.
Related Collections
Right arrow Humic Substances
Right arrow Bioremediation and Biodegradation
Right arrow Heavy Metals
Right arrow Organic Compounds
Right arrow Soil Pollution
Journal of Environmental Quality 31:1885-1892 (2002)
© 2002 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America

TECHNICAL REPORTS
Heavy Metals in the Environment

Cadmium Binding by Fractions of Dissolved Organic Matter and Humic Substances from Municipal Solid Waste Compost

Arno Kaschla, Volker Römhelda and Yona Chen*,b

a Institute of Plant Nutrition (330), University of Hohenheim, 70593 Stuttgart, Germany
b Dep. of Soil and Water Sciences, Faculty of Agricultural, Food and Environmental Quality Sciences, The Hebrew University of Jerusalem, P.O.B. 12, Rehovot 76100, Israel

* Corresponding author (yonachen{at}agri.huji.ac.il)

Received for publication July 24, 2001.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
The agricultural practice of amending soils with composted municipal solid waste (MSW) adds significant amounts of organic matter and trace metals, including Cd. Under these conditions, soluble organic complexes of Cd formed in the compost may be more significant than previously thought, due to Cd bioavailability and mobility in the soil environment. To study the relative importance of different types of organic ligands in MSW compost for the binding of Cd, six fractions of the dissolved organic matter (DOM) in addition to humic acid (HA) and fulvic acid (FA) were extracted and their complexation of Cd quantified at pH 7 using an ion-selective electrode (ISE). The highest complexing capacities (CC) for Cd were found for the most humified ligands: HA (2386 µmol Cd g-1 C of ligand), predialyzed FA (2468 µmol Cd g-1 C), and HoA, a fulvic-type, easily soluble fraction (1042 µmol Cd g-1 C). The differences in CC for Cd of the various organic ligands were not directly related to total acid-titratable or carboxylic groups, indicating the importance of sterical issues and other functional groups. The strength of association between Cd and the organic ligands was characterized by calculating stability constants for binding at the strongest sites (pKint) and modeling the distribution of binding site strengths. The pKint values of the DOM fractions ranged between 6.93 (HiN: polysaccharides) and 8.11 (HiB: proteins and aminosugars), compared with 10.05 for HA and 7.98 for FA. Hence, the highly complex and only partially soluble organic molecules from compost such as HA and FA demonstrated the highest capacity to sequester Cd. However, strong Cd binding of organic ligands containing N-functional groups (HiB) in addition to a high CC of soluble, humified ligands like HoA indicated the relevance of these fractions for the organic complexation of Cd in solution.

Abbreviations: ATG, acid-titratable groups • BS, binding site • CC, complexing capacity (for Cd) • CdB, bound cadmium • CdF, free (ionic) cadmium • CdT, total cadmium • DOC, dissolved organic carbon • DOM, dissolved organic matter • FA, fulvic acid • FA > 1000, fulvic acid after dialysis with a molecular weight cutoff of 1000 daltons • HA, humic acid • HoA, HoN, HoB, HiA, HiN, and HiB, dissolved organic matter fractions • HS, humic substances • ISE, ion-selective electrode (for Cd) • L, ligand • LC, ligand concentration expressed as dissolved organic carbon • MSW, municipal solid waste • OM, organic matter • pK, stability constant • pKi, incremental stability constant • pKint, intrinsic stability constant for binding at the strongest sites • {nu}, bound cadmium/ligand concentration (CdB/LC) • {nu}', bound cadmium/ligand concentration (CdB/LC) values lying equidistant between neighboring data points on the {nu} axis of the Scatchard plot


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
HIGH AMOUNTS OF trace metals in composts from municipal solid waste (MSW) are a valid concern regarding waste application as a soil amendment. Among these elements, Cd is known for its high toxicity to almost all biota, including humans, and its high mobility in the terrestrial environment. In MSW compost, Cd is encountered in a variety of forms including a substantial fraction that is associated with organic matter (OM) (Bourque et al., 1994; He et al., 1992). In landfill leachates, Cd has been shown to associate with OM, including with high molecular weight ligands (Knox and Jones, 1979), while the strongest Cd complexes have appeared in leachates with stable OM, such as from mature composts (Lun and Christensen, 1989). In general, the interaction of Cd with organic ligands has not received the same attention as other trace metals such as Cu, since it is usually not as strongly complexed by OM. However, it has been shown that OM can also play an important role in Cd speciation in soils (Christensen, 1989; Naidu and Harter, 1998; Wong et al., 2000). In addition, since the OM in soils is tightly bound to clay minerals and its reactive sites are likely to be occupied by trivalent ions, "freshly" imported OM becomes very important for trace metal binding (Stevenson, 1976).

Compost OM is a heterogeneous mixture of substances that result from the breakdown and transformation of organic materials in the waste, which contain numerous functional groups for the complexation of trace metals. The dissolved fraction of OM (DOM) is of special interest, since it contributes to metal solubility and thus influences metal bioavailability and mobility in the soil environment after compost application. Due to lack of homogeneity, it is difficult to study the properties of compost DOM in bulk. To obtain fractions of DOM that are better-defined chemically, an operational extraction procedure originally developed for soil DOM by Leenheer (1981) has been successfully applied to compost DOM (Chefetz et al., 1998b). Using this procedure, the DOM is divided into six fractions depending on their adsorption to several exchange resins at different pH levels.

Many different methods have so far been employed to quantify the complexation of different trace metals by organic ligands (Stevenson, 1994). Among these, the ion-selective electrode (ISE) has been shown to be a reliable instrument for measuring Cd at concentrations ranging from 10-10 to 10-6.5 M in a complex organic matrix solution and it compares well with Cd activities calculated using the GEOCHEM program (Candelaria et al., 1995). It has also been applied successfully to studies of Cd complexation by fulvic acids (FA) (Saar and Weber, 1979), organic components of landfill leachates (Knox and Jones, 1979), and humic substances (HS) in water (Gardiner, 1974). Several theoretical models have been suggested to describe the metal complexing behavior of HS (Dzombak et al., 1986; Stevenson et al., 1993; Stevenson, 1994), since they are known to be important metal complexing agents in the soil environment and have therefore been widely investigated. A promising approach is to determine incremental stability constants from successive slope values in the Scatchard plot and to apply a continuous distribution model based on a normal distribution of binding site (BS) strengths (Stevenson and Chen, 1991; Stevenson et al., 1993; Logan et al., 1997).

The objective of this study was to quantify the complexation of Cd by various soluble organic ligands from MSW compost and to evaluate their relative importance. The DOM and HS were extracted from compost and the former subdivided into six fractions, each of which contained organic molecules of similar chemical properties. To model the binding of Cd, an approach developed originally for metal binding by HS molecules only was adopted.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
Humic Substance Extraction from Mature Municipal Solid Waste Compost
The MSW compost was obtained from a commercial plant in Afula, Israel. After 120 d of composting the temperature in the windrow pile had fallen to ambient levels. The OM content of the mature compost was 23% of the dry matter. The procedure for extracting the HS from the compost was described in detail by Chefetz et al. (1996). In short, HS are extracted from the composted OM with 0.1 M NaOH at 1:10 solid to solution (w/w) ratio under N2. By centrifugation the soluble HS are separated from the residue. Acidification to pH 1 with 6 M HCl causes humic acid (HA) to precipitate, while the fulvic fraction remains dissolved.

The HA precipitate was separated by centrifugation and shaken repeatedly for 24 h at room temperature with 1 L of dilute HCl + HF solution (5 mL concentrated HCl + 5 mL 52% HF + 990 mL deionized water) until the ash content was <1.0%. In addition, the HA was dialyzed against deionized water until the dialysate was free of chloride and exhibited a neutral pH. The fulvic fraction was separated into FA and the nonhumic fraction by adsorption of FA onto resin. After desorption from the column with 0.1 M NaOH, the FA was shaken repeatedly with a protonated cation exchange resin (Amberlyst 15; Merck, Darmstadt, Germany), until the Na concentration in the solution was <0.01 mmol L-1 (measured by a Sherwood 410 flame photometer). Both purified FA and HA solutions were freeze-dried and stored in a dessicator. Stock solutions were prepared by dissolving 100 mg of HA and FA in 100 mL of deionized water and stored under N2 in the refrigerator. To ensure complete dissolution of the HA, 5 µL of 1 M KOH were added to the solution. The FA > 1000 was obtained from FA stock solution after predialyzing against daily replaced deionized water for four days (molecular weight cutoff = 1000 Da). About a third of the dissolved organic carbon (DOC) was lost in this procedure.

A value of 19.5% of the total compost OM was extractable as HS. Of these, the most abundant were HA (67.7% of HS), while FA constituted 15.3% of HS. Humic acid and FA have been characterized extensively as reported previously (Kaschl, 2001).

Dissolved Organic Matter Fractionation
The DOM fractions were extracted from an aqueous extract of mature compost (1:10 solid to solution ratio, w/w). The procedure has been described in detail by Chefetz et al. (1998b). Adsorption to XAD-8 separates the DOM into so-called hydrophobic (Ho, adsorb onto the resin) and hydrophilic (Hi, do not adsorb) substances. Further fractionation is achieved according to resin adsorption at different pH values: HoB and HoA are desorbed from XAD-8 with 0.1 M HCl and 0.1 M NaOH, respectively; HoN is desorbed with methanol. HiB is adsorbed to a protonated cation exchange resin (Amberlyst 15) and desorbed with 0.1 M NH4OH; HiA is sorbed to an anion exchange resin (Amberlyst A-21) and desorbed with NaOH; and HiN remains in solution after all column separation steps. The fraction HoB constituted less than 1% of the DOM in mature compost and did not provide sufficient material for metal-binding studies. The fraction HiA contained high amounts of inorganic salts. The DOM fractions were freeze-dried and stored in a dessicator. Stock solutions were prepared by dissolving 50 mg in 50 mL of deionized water and stored under N2 in the refrigerator. The DOM fractions were characterized extensively with cross polarization magic angle spinning C13 nuclear magnetic resonance (CPMAS NMR) and Fourier-transform infrared spectroscopy (FTIR) as described by Chefetz et al. (1998a)(b).

Chemical Analysis
All glass and plastic containers were acid-washed and all reagents used were reagent grade. The OM content of compost was calculated after ignition of dry samples at 550° for 8 h. Total C was calculated as 58% of OM (Inbar et al., 1990; Chefetz et al., 1996). The DOC was measured by a total C monitor (TCM 480, Carlo Erba Instruments, Milan, Italy). Acid-titratable groups (ATG) (carboxyl and phenolic functional groups) were determined by an acid–base titration of stock solutions as described by Inbar et al. (1990). Total metal content was obtained by measuring stock solutions with an ICP–AES (Spectroflame Modula E; Spectro Analytical Instruments, Kleve, Germany) with a detection limit of 1.4 µg L-1 for Cd.

Cadmium Ion-Selective Electrode
The ISE measures the activity of the free (hydrated) Cd ions in the solution, while total Cd is equal to the amount of metal salt added in a titration. The Cd titrations were performed in enclosed Metrohm (Herisau, Switzerland) titration vessels with openings for the electrodes and an N2 tube. They were placed in a temperature-controlled water bath at 25 ± 1°C. The ligand was added from its stock solution to 0.01 M KClO4 to give 50 mL of 10 mg DOC L-1 in the titration vessel. Nitrogen gas was initially bubbled through the solution and then a constant N2 atmosphere was maintained above the solution. The pH of the solution was adjusted to 7.00 by addition of minute amounts of diluted KOH or HClO4 and it was left to equilibrate overnight.

The Cd ISE, the reference electrode and the pH electrode were all obtained from Metrohm. Calibration of the Cd electrode was performed by adding Cd(ClO4)2 at a concentration of 10-7 to 10-4.8 M Cd to 50 mL of 0.01 M KClO4 in a second vessel at pH = 7.00 and temperature = 25°C, and in a N2 atmosphere. For concentrations of 10-6 to 10-4.5 M Cd, the measured electrode voltage was linear to Cd concentrations with a Nernstian slope (29 mV). For Cd concentrations between 10-7 and 10-6 M, the slope decreased to 20 mV. This was accounted for by adjusting a second- or third-degree polynomial to four data points measured in this range in accordance with Saar and Weber (1979), who used polynomial fitting successfully for measurements at low Cd activities. Cadmium concentrations below 10-7 M were not measurable. The calibration was always done immediately before the ligand titration. The Cd electrode was cleaned before each experiment according to manufacturer's instructions. The reference electrode was a double-bridge electrode filled with 3 M KCl inside and 1% KClO4 in the outer compartment.

During the titration, aliquots of 5 to 100 µL from Cd(ClO4)2 stock solutions were pipetted into the vessel. Volume changes were accounted for in the subsequent calculations for complexing capacity (CC) and stability constants (pK). The pH was measured continuously and kept constant (±0.02) by addition of diluted HClO4 and KOH; a constant pH is critical to the electrode response at pH > 4 (Truitt and Weber, 1981). Four minutes after each Cd addition, the electrode potential stabilized and was recorded. One titration consisted of about 40 Cd additions covering the CdF (free [ionic] cadmium) ranges from 10-7 to 1.6 x 10-5 M and was completed within 3 to 4 h. Titrations for each ligand were performed in triplicate.

Mathematical Data Evaluation
The complexing capacity (CC) was determined by plotting CdF/LC vs. CdT/LC (where CdT is total cadmium in mol L-1, equivalent to the amount of Cd added to the vessel, and LC is total ligand concentration expressed in mg L-1 DOC). If there was no association between the ligand and the Cd ions, a straight line through the origin with a slope of 1.0 would be expected. In the case of metal complexation by the ligand, a line with slope = 1 and an x axis intercept > 0 (CC) is expected, since as long as CC > CdT/LC, CdF will be near zero and will accumulate in the solution substantially only after the CC of the ligand has been reached. Hence, a linear regression of all data points equal to and greater than the initial CdF/LC concentration at which a slope of 1 is reached gives the CC as the x axis intercept and the corresponding standard error.

Incremental stability constants for Cd binding were derived from the titration data as described in detail by Stevenson and Chen (1991) and Stevenson et al. (1993) for HS. In short, it is assumed that the organic molecule (L) is the central group to which each Cd ion (Cd) is bound to a single reactive site (formation of LCd, LCd2,..LCdn complexes) (Stevenson, 1994). The Scatchard plot ({nu}/CdF vs. {nu}, with {nu} = CdB/LC, where CdB is cadmium bound to ligand, which equals the difference between CdT and CdF) yields the stability constant K0 as the slope. For HS the Scatchard plot representation gives a curvilinear graph, because HS have a range of binding sites (BS) with differing binding energies rather than a discrete number of sites of defined binding strength; electrostatic and sterical effects influencing the binding strength of functional groups and the complexation of one metal ion with two ligand molecules might also contribute to the curvilinear nature (Stevenson, 1994; Logan et al., 1997). The Cd binding data obtained for the different DOM fractions showed the same curvilinearity in the Scatchard plot as generally found for HS. This is not surprising, since the DOM fractions are, just as HS, operationally defined mixtures of organic molecules including macromolecules with a myriad of functional groups. Hence, we believe it is justified to use the same mathematical approach as was used for HS to obtain incremental stability constants, that is, successive slope values were calculated from neighboring data points in the Scatchard plot (slope = {Delta}[{nu}/CdF]/{Delta}{nu}). These slope values are conditional stability constants Ki at a certain value of {nu}', which in turn lies equidistant between the original neighboring data points on the {nu} axis of the Scatchard plot. A plot of pKi (log Ki) against {nu}' displays the range of BS strengths observed. Extrapolation toward {nu}' = 0 gives the intrinsic constant Kint. In the natural environment this constant associated with the binding of a metal at lowest concentrations is important, since it describes the strongest binding sites for this metal (Stevenson, 1994). The value of Kint was determined by fitting a third-degree polynomial to the data of the (Ki vs. {nu}') plot and extrapolating to the y intercept with the linear regression feature using the graphic computer program SigmaPlot 3.0 (SPSS, 2002). The strictly mathematical approach of fitting a polynomial to the Scatchard plot data and then calculating the derivative was not followed, since the inherent error of polynomial fitting is multiplied when the derivative is formed.

A continuous distribution model successfully applied to metal titration data of HS is based upon the Gaussian bell curve (Stevenson and Chen, 1991; Stevenson et al., 1993). It is assumed that the BS strengths have a normal distribution for the macromolecules comprising the HS solution. From the polynomial curve fit to the data in the (pKi vs. {nu}') plot, values for pKi were calculated at equal intervals of {nu} (Stevenson and Chen, 1991). With the Gaussian distribution function Ci/CL = (a/{sigma})[(µ - log Ki)/{sigma}]2{delta}logKi, the mole fraction of binding sites Ci/CL in the interval {delta} log Ki can be calculated (a = e-1/2/2{pi}; {sigma} is the standard deviation for the distribution of pKi around the mean µ).


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
SUMMARY AND CONCLUSIONS
 REFERENCES
 
In the ISE titrations with Cd, good correlation was found between replicates for a single ligand in CdF/LC vs. CdT/LC diagrams, since data values overlap (Fig. 1) . When the specific CC of a ligand is exceeded, the slope becomes 1, so that extrapolation toward the x axis yields the CC. The CC values thus obtained for HA and FA > 1000 were compared with the results obtained for the identical ligands with the method of equilibrium dialysis as reported previously (Kaschl, 2001). The two different experimental approaches were in excellent correlation regarding the numerical value for CC of Cd (Fig. 2) . In contrast to the method of equilibrium titration, the ISE has the advantage of producing many data points, which improves the quality of the graphical determination of the binding constants. The capacity of soil FA for Cd binding, as reported by Truitt and Weber (1981) and Rainville and Weber (1982), was in the same range as our results for the compost HS. In addition to its ability to complex Cd, compost HS was measured to have a similar CC for the binding of Zn and a significantly higher CC for Cu (Kaschl et al., 2000; Kaschl, 2001). Hence, HS from compost will significantly increase the ability of soils with a low OM content to retain Cd and other trace metals, as compost HS are expected to adsorb on mineral surfaces or precipitate in the soil environment.



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 1. Data obtained with the ion-selective electrode for titrations of humic acid (HA), fulvic acid (FA), fulvic acid after dialysis with a molecular weight cutoff of 1000 Da (FA > 1000), FA, and dissolved organic matter (DOM) fractions (HoA, HoN, HiA, HiN, and HiB) from municipal solid waste (MSW) compost with Cd(ClO4)2 at pH 7. The x axis intercept of the regression line with a unity slope is defined as the complexing capacity (CC) for Cd of the respective ligand. Data shown were obtained from triplicated separate titrations for each ligand. CdF, free (ionic) cadmium.

 


View larger version (19K):
[in this window]
[in a new window]
  		Fig. 2. Complexing capacities (CC) at pH 7 for Cd of humic acid (HA) and fulvic acid after dialysis with a molecular weight cutoff of 1000 Da (FA > 1000) from municipal solid waste (MSW) compost as determined by two different methods: Cd ion-selective electrode (ISE) and equilibrium dialysis (EQ; data from Kaschl, 2001).

 
In the DOM derived from MSW compost, HoA (22% of DOC) was the dominant hydrophobic fraction by quantity, while HoN represented 17% and HoB less than 1% of the DOC (Chefetz et al., 1998a,b). The hydrophilic fractions were present in the order HiA (26%), HiN (21%), and HiB (13%) relative to total DOC (Chefetz et al., 1998a, b). For Cd binding, HoA clearly stands out among the DOM fractions, binding the largest quantity (based on ATG or LC) (Fig. 3) . In addition, HoA was the only DOM fraction that retained residual Cd (but only 0.19% of the total CC for this element, Table 1) after the extraction from compost and the subsequent purification procedure that involved drastic pH changes and the use of ion-exchange resins. Since HoA-type molecules also constitute a quantitatively important fraction of the DOM in the compost, they can be regarded as the most prominent (compost-derived) organic ligands for Cd in the solution phase. The chemical structure of HoA molecules has been described as "young" FA, of a polyphenolic, humic structure with associated carbohydrates (Chefetz et al., 1998b; Guggenberger et al., 1994). Rather than being identical to FA, HoA represents the most soluble fraction of FA (Chefetz et al., 1998b). This is in accordance with our observation that the CC of total FA is closer to that of HoA, while FA > 1000, which comprises the FA fraction after dialysis at a cutoff of 1000 Da and thus contains the larger, more complex (humified) molecules, has a distinctly higher CC (1652 [FA] vs. 2553 [FA > 1000] µmol Cd g-1 DOC; Fig. 2 and Fig. 3). After compost application, HoA ligands from the compost could be important in keeping trace metals such as Cd in solution and in this way more readily available to microorganisms and plants just as increasing Cd mobility in the soil.



View larger version (26K):
[in this window]
[in a new window]
  		Fig. 3. Complexing capacities for Cd of dissolved organic matter (DOM) fractions (HoA, HoN, HiA, HiN, and HiB) and fulvic acid (FA) from municipal solid waste (MSW) compost as determined with the Cd ion-selective electrode at pH 7.

 

View this table:
[in this window]
[in a new window]
  		Table 1. Acid-titratable groups (ATG) and residual Cd for humic acid (HA), fulvic acid (FA), fulvic acid after dialysis with a molecular weight cutoff of 1000 Da (FA > 1000), FA, and dissolved organic matter (DOM) fractions (HoA, HoN, HiB, HiN, and HiA) from municipal solid waste (MSW) compost.

 
The HoA from other sources has similarly been identified as an important complexing agent of Cd and other trace metals. Hydrophobic acids (HoA) from a spodosol soil had a higher affinity for Cd than hydrophilic acids (HiA) (Guggenberger et al., 1994). Likewise, in DOM obtained from lakewater and soils, the HoA fraction complexed various trace metals more strongly than other fractions (Vogt et al., 2000). However, for the binding of Cu by compost DOM, significantly more Cu was bound by the HoN fraction as opposed to the HoA, due to the high content of ATG on HoN molecules (Kaschl et al., 2000). HoN and HoA are chemically similar, comprising the most humified molecules in the DOM (Chefetz et al., 1998a). Hence, it appears that in the liquid phase of a variety of environments humic-type molecules play an important role as metal complexers due to their high loading capacity.

When bound cadmium (CdB) is related to C content in solution (LC), the measured CC of FA > 1000 and HA were very similar: 2553 and 2321 µmol Cd g-1 DOC, respectively (Fig. 2; measured with the ISE). However, in relation to acid-titratable groups (ATG) on the molecules, the CC for HA (227 mmol Cd mol-1 ATG) was higher, since the FA > 1000 fraction (CC = 179 mmol Cd mol-1 ATG) has a higher acidity than HA (Table 1). This is in contrast to studies examining the binding of Cu and Zn by compost HS, where differences of CC between HA and FA correlated with ATG content of the ligand (Kaschl et al., 2000; Kaschl, 2001). For Cd, the HS molecules of higher molecular weight and complexity exhibit greater Cd-binding capacity than the smaller, easily soluble DOM fractions without an apparent relation to density of ATG on the molecules (Table 1). In addition, the differences between the individual DOM fractions were not related to the total number of ATG or the total COOH group content. Other authors have pointed out the importance of phenolic and especially the carboxyl groups on HS for the binding of Cd (Stevenson, 1994; Manunza et al., 1995; Leenheer et al., 1998). However, our findings suggest that Cd binding is also strongly affected by other factors (such as sterical issues) as a result of the higher complexity of HS vs. the simpler DOM molecules and the presence of other functional groups (see below), more so than the more commonly studied binding of Cu by organic ligands in the literature.

Incremental stability constants (pKi) describing the strength of Cd binding by the various ligands are displayed in Fig. 4 . A fairly good third-degree polynomial fit could be achieved for data of HA, FA > 1000, HoA, HiN, and HiA. Data points for FA, HoN, and HiB from different titrations were somewhat divergent, possibly because of the more heterogeneous nature of these fractions. The strongest binding sites (pKint) for Cd at low ratios of Cd to ligand were found on HA (pKint = 10.05), followed by HiB (pKint = 8.11) (Table 2). The pKint for all other fractions was similar: 7.98 for FA > 1000, 6.80 for FA (here the polynomial fit was not satisfactory, hence pKi values for the continuous model were not calculated), 7.74 for HoA, 7.69 for HoN, 7.02 for HiA, and 6.93 for HiN. These stability constants are high in comparison with literature data obtained for HS (Stevenson, 1994). This may be due to the relatively high pH (7) and the low ionic strength (0.01 M) of the tested solution. Both factors have indeed been reported to result in higher formation constants (Saar and Weber, 1979; Stevenson et al., 1993; Logan et al., 1997). It has also been recognized that association constants for metal–organic complexes in the literature are often unrealistically low, since they are based on measurements at high metal loadings on the complexing sites of OM (McBride et al., 1999). The metal binding at the strongest BS at low metal to ligand ratios is not monitored well by the discrete (noncontinuous) models due to a limitation of the method sensitivity. Trace metals, in fact, occupy different and more selective bonding sites at low metal levels (McBride et al., 1999). The magnitude of the stability constants determined in this study underlines the fact that organic complexation is indeed important for the speciation of Cd at high pH and should be appropriately addressed in modeling approaches.



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 4. Incremental stability constants (pKi) for Cd binding by humic acid (HA), fulvic acid after dialysis with a molecular weight cutoff of 1000 Da (FA > 1000), FA, and dissolved organic matter (DOM) fractions (HoA, HoN, HiA, HiN, and HiB) from municipal solid waste (MSW) compost at pH 7. The pKi data were obtained by calculating the slope of adjacent points in the Scatchard plot representation ({nu}/CdF vs. {nu}, where CdF is free [ionic] cadmium) of the titration data. The scale of {nu}' varies in accordance with the total complexing capacity (CC) of each ligand (the maximum value of {nu}' corresponds to 100% CC). Three titrations were performed for each ligand except for HiA (1). The data shown were obtained from triplicated separate titrations for each ligand.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Stability constants for binding of Cd at the strongest sites (pKint) for humic acid (HA), fulvic acid after dialysis with a molecular weight cutoff of 1000 Da (FA > 1000), FA, and dissolved organic matter (DOM) fractions (HoA, HoN, HiA, HiN, and HiB). The pKint data were obtained by linear fitting of the listed polynomials to data in Fig. 3.

 
Among the easily soluble DOM fractions, HiB had the highest stability constant for Cd binding at low Cd to ligand ratios (pKint = 8.11). HiB consists mainly of proteins, peptides, and amino–sugar polymers (Chefetz et al., 1998b). Nitrogen-containing groups in the HiB molecules could therefore explain its high strength of binding, since there is much evidence that N-containing groups in OM interact strongly with trace metals and since Cd preferably binds to soft donor groups like N and S (Stevenson, 1994). Similarly, the high pKint of HA and FA > 1000 may be related to their high N content (7.4 and 5.1%, respectively; Kaschl, 2001). High N contents and very low S contents (not true for HS from composted sewage sludge) are characteristic for HS from MSW composts compared with HS from other sources (Chen et al., 1996; Kaschl, 2001). Hence, organic ligands from compost may have a special significance for the speciation in the soil of Cd and other trace metals binding to soft donor groups.

Using the normal distribution approach, the range and relative abundance of BS strengths can be illustrated (Fig. 5) . All curves are asymmetrical (skewed), with a higher percentage of weaker binding sites, as was noted for HS by Stevenson and Chen (1991). This phenomenon was most prominent in HA, FA, HoA, and HiB. The pKint appears in the continuous model as the extreme value on the right hand side of each distribution curve where Ci/CL approaches zero; consequently it is obvious from the curves that functional groups with a BS strength close to pKint represent only a minute percentage of total metal-binding groups on the ligands. Nonetheless, Stevenson (1994) stressed the importance of the strongest binding groups on a ligand molecule in a natural setting, since these are the first to be occupied by the surrounding cations.



View larger version (28K):
[in this window]
[in a new window]
  		Fig. 5. Distribution of binding site strengths for Cd at pH 7 determined using the normal distribution approach. The upper diagram displays data for humic acid (HA) and fulvic acid after dialysis with a molecular weight cutoff of 1000 Da (FA > 1000) and the lower, for dissolved organic matter (DOM) fractions (HoA, HoN, HiA, HiN, and HiB) from municipal solid waste (MSW) compost. The term µ is the average, {sigma} is the standard deviation, and s is skewness.

 
The pK values for the most abundant BSs according to the normal distribution were highest for HA (6.01) and HoN (6.21) molecules (Fig. 5). All other fractions had distinctly lower average pK values: 5.54 for FA > 1000, 5.51 for HoA, 5.43 for HiA, 5.35 for HiN, and 5.73 for HiB (Fig. 5). The flattest distribution curve was observed for HA with pKi values extending from 4.7 to 10.05, indicating the heterogeneity of the strength of BS on the HA molecule. The FA > 1000 (pKi = 4.7–8.0), HoA (pKi = 4.6–7.7), and HiB (pKi = 4.9–8.1) also had a relative wide range of BS strengths, while HiN (pKi = 4.8–6.9), HiA (pKi = 4.5–7.0), and HoN (pKi = 5.2–7.7) had a narrow range of pKi and their distribution curve was rather peaked. The latter shape is an indication of a more homogeneous nature of the BS, a finding backed up by the composition of HiN and HiA as seen by C13 nuclear magnetic resonance. The HiN fraction consists mostly of poly- and oligosaccarides, while HiA are small, highly oxidized organic compounds including a high amount of inorganic salts (Chefetz et al., 1998b). The overall BS distribution curve for HoN is dissimilar from that of the other DOM fractions, being shifted toward higher pKi. This fraction has been characterized as a mixture of highly apolar, aromatic compounds containing 44 to 48% aliphatic structures and is thought to be closely related to HA, thus comprising the least soluble organic molecules of the DOM (Chefetz et al., 1998a, b). Indeed, the range of the most common binding sites (peak values of the distribution curve) on the molecules of HA and HoN is similar. However, the HoN molecules lack in structural and functional complexity compared with HA and thus show less heterogeneity in overall BS distribution (peaked shape of curve). The relevance of HoN ligands for Cd binding in solution is expected to be low, because of its low CC compared with HoA and HiB.


   SUMMARY AND CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
REFERENCES
 
Cadmium demonstrated a tendency to preferably associate with larger, humified, and less soluble organic materials such as HA and FA > 1000. These substances both demonstrated the highest capacity for binding and possessed the strongest Cd-binding groups. For water-soluble ligands, the HoA, which resembles FA, had the highest capacity for Cd complexation, while the strongest binding groups were found on HiB molecules, which include proteinaceous ligands. The values for CC obtained were similar to those established for soil HS. Hence, we believe that OM entering the soil with the compost will have an important effect on Cd speciation in the soil. From the results we obtained for the HA it also appears that Cd may be tightly bound by insoluble OM such as large HA molecules and humin, thereby increasing the capacity of the soil to adsorb Cd and prevent leaching.


   ACKNOWLEDGMENTS
 
The authors wish to thank the Deutsche Forschungsgemeinschaft (DFG), the Israeli Ministry of Science, Culture and Sport (MOS), and the Bundesministerium fuer Bildung, Wissenschaft, Forschung und Technologie (BMBF) for their support of this research.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 




This article has been cited by other articles:


Home page
 J. Environ. Qual.Home page
A. J. Franzluebbers, S. R. Wilkinson, and J. A. Stuedemann
Bermudagrass Management in the Southern Piedmont, USA: IX. Trace Elements in Soil with Broiler Litter Application
J. Environ. Qual., March 1, 2004; 33(2): 778 - 784.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (14)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kaschl, A.
Right arrow Articles by Chen, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kaschl, A.
Right arrow Articles by Chen, Y.
Agricola
Right arrow Articles by Kaschl, A.
Right arrow Articles by Chen, Y.
Related Collections
Right arrow Humic Substances
Right arrow Bioremediation and Biodegradation
Right arrow Heavy Metals
Right arrow Organic Compounds
Right arrow Soil Pollution

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Vadose Zone Journal Journal of Plant Registrations
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal