JEQ Grow Your Career With ASA
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 Related articles in JEQ
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 ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Han, N.
Right arrow Articles by Thompson, M. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Han, N.
Right arrow Articles by Thompson, M. L.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Han, N.
Right arrow Articles by Thompson, M. L.
Related Collections
Right arrow Vadose Zone Processes and Chemical Transport
Right arrow Municipal Wastes
Right arrow Nonequilibrium Transport
Published in J. Environ. Qual. 32:1829-1836 (2003).
© 2003 ASA, CSSA, SSSA
677 S. Segoe Rd., Madison, WI 53711 USA

TECHNICAL REPORTS

Vadose Zone Processes and Chemical Transport

Impact of Dissolved Organic Matter on Copper Mobility in Aquifer Material

Nizhou Hana and Michael L. Thompson*,b

a Department of Geological Sciences, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061
b Agronomy Department, Iowa State University, Ames, IA 50011-1010

* Corresponding author (mlthomps{at}iastate.edu).

Received for publication July 12, 2002.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Naturally occurring dissolved organic matter (DOM) and biosolids-derived DOM have been implicated in the mobility of metals in soils and aquifer materials. To investigate the effect of DOM on copper mobility in aquifer material, DOM derived from sewage biosolids was separated into two apparent molecular-weight (MW) fractions, 500 to 3500 Da (LMW) and >14000 Da (HMW). In each MW fraction, the DOM was further fractionated into hydrophilic, hydrophobic acid, and hydrophobic neutral compounds by an XAD-8 chromatography technique. The mobility of these DOM components and their influences on copper transport in a sesquioxide-coated, sandy aquifer material were examined with column transport experiments. The LMW DOM was found to be highly mobile, whereas the HMW DOM had a greater tendency to be retained by the aquifer material. Within the same MW fraction, the mobility of DOM followed the order of hydrophilic DOM > hydrophobic acid DOM > hydrophobic neutral DOM. Copper breakthrough curves in the presence of various DOM components showed that, except for the HMW hydrophilic fraction, DOM components enhanced Cu transport through the aquifer columns at early stages of transport (the first 75 pore volumes). In the later stages, however, all the DOM components substantially inhibited Cu mobility. We hypothesize that several mechanisms could account for retardation of Cu movement in the presence of the DOM fractions, including the formation of ternary complexes between the aquifer material, Cu, and DOM; changes in the electrostatic potential at the solid-phase surface; and pH buffering by DOM.

Abbreviations: DOC, dissolved organic carbon • DOM, dissolved organic matter • HMW, molecular weight fraction of >14000 Da • LMW, molecular weight fraction of 500 to 3500 Da • MW, molecular weight


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
THE TRANSPORT of naturally occurring DOM and its components in soils and aquifer materials is a common phenomenon and has been previously studied at the scale of laboratory columns as well as at the field scale (e.g., Jardine et al., 1992; Dunnivant et al., 1992a, b; McCarthy et al., 1993). A general conclusion to be drawn from these studies is that low molecular weight, hydrophilic DOM components are highly mobile and could influence metal transport in aquifer material. It is difficult, however, to find a consensus among published studies that concern metal transport in the presence of DOM. For example, Dunnivant et al. (1992b) demonstrated an increase in Cd migration in the presence of stream-derived DOM, whereas Oden et al. (1993) observed a decrease in Cu mobility in the presence of DOM from a number of sources. Giusquiani et al. (1992) reported an increase in mobility for several heavy metals added to soils amended with urban waste compost; however, they demonstrated that the DOM in the compost had only a limited influence on that mobility. On the basis of field and laboratory studies, Temminghoff et al. (1997) reported that dissolved humic acids enhanced the mobility of Cu in a contaminated sandy soil. Thus, considering the variety of conclusions of research reports, the commonly expressed hypothesis that DOM will facilitate trace metal movement in subsurface zones needs to be further examined.

In recent years, there has been concern that trace metals and other contaminants originating from sewage biosolids that are land-applied might be more mobile in soil than previously thought. Under certain conditions, such migration has been shown to occur in both field and laboratory settings (e.g., Darmody et al., 1983; Dowdy et al., 1984; Camobreco et al., 1996; Richards et al., 1998). These and other studies have led to interest in reevaluation of the biosolids application rules promulgated by the USEPA (USEPA, 1993; McBride, 1995; Harrison et al., 1997).

Land application of biosolids could introduce not only trace metals but also mobile organic compounds into soils (Han and Thompson, 1999a). Much of the potentially mobile organic material in biosolids-amended soils undergoes microbial transformation near the soil surface, but some fraction of it may move through the soil to shallow aquifers, particularly through continuous biopores and cracks. Biosolids-derived DOM has a strong metal-binding ability (Han and Thompson, 1999b), and DOM-bound metals might be more mobile than dissolved, uncomplexed metals ions. In this context, it is relevant to explore the influence of DOM on trace metal movement in subsurface aquifer materials that underlie sites of biosolids application.

The objective of this study was to determine the extent to which the molecular weight and the hydrophobicity of biosolids-derived DOM influenced the transport of Cu through an aquifer material. Although it is considered a low-mobility element in most soil environments, Cu was chosen because it is strongly complexed by organic ligands (McBride, 1994), allowing the role that DOM might play in its transport to be evaluated effectively. In addition, Cu is commonly found in municipal biosolids and is one of the first trace elements that would limit land application of biosolids (Alloway, 1990).


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Aquifer Material
Aquifer material was collected from the Skunk River aquifer near Ames, IA. The material was collected from a depth of 2.3 to 2.5 m, air-dried, and passed through a 600-µm plastic sieve. The particle size distribution of the aquifer material was 970 g sand kg-1 (53–600 µm), 4 g silt kg-1 (2–53 µm), and 26 g clay kg-1 (<2 µm), determined by the pipette method of Gee and Bauder (1986). Total organic C content was 0.3 g kg-1 [determined by the dry combustion technique described by Nelson and Sommers (1996)], the pH (1:1 solid to water) was 7.4, and the point of zero charge was 7.6 [determined by the methods of Van Raij and Peech (1972) and Hendershot and Lavkulich (1978)] (Fig. 1a) .



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 1. (a) Point of zero charge determination for untreated aquifer material. (b) Linearized Langmuir plot of the Cu sorption isotherm for the untreated and oxide-free aquifer material.

 
To investigate the effect of Fe and Mn oxides on Cu transport, a sample of the material was treated with Na citrate–bicarbonate–dithionite (CBD) to remove Fe and Mn oxides by a procedure similar to that of Jackson et al. (1986). Dithionite-extractable Fe and Mn contents of the aquifer material were 7.41 and 1.92 g kg-1, respectively. The treated material was washed extensively with 0.05 M KNO3 until free citrate was removed and then was air-dried before use.

The Cu sorption capacity of the aquifer material was estimated before and after the dithionite treatment by a batch equilibrium approach. Aquifer material samples of 1.000 g were weighed into six 50-mL Teflon centrifuge tubes. Aliquots of 30 mL of Cu2+ solution ranging from 3.14 x 10-5 to 3.14 x 10-4 mol L-1, made up in 0.025 M KNO3 at pH 6, were added to the tubes. The suspensions were shaken on a reciprocal shaker for 16 h, then centrifuged at 6458 x g for 10 min, and filtered through Whatman (Maidstone, UK) no. 40 ash-free filters. The Cu concentrations in the equilibrium solutions were determined by atomic absorption spectrophotometry. By fitting the data to a Langmuir model, the maximum Cu sorption capacity was calculated to be 8.44 mmol Cu kg-1 (536 mg Cu kg-1) of untreated aquifer material and 2.08 mmol Cu kg-1 (132 mg Cu kg-1) of oxide-free aquifer material (Fig. 1b).

Biosolids-Derived Dissolved Organic Matter Components
Anaerobically digested sewage biosolids were sampled as a liquid suspension from the Water Pollution Control Facility at Ames, IA. The total solids content of the biosolids was 5%. The biosolids suspension was centrifuged at 478 x g for 10 min to remove large solid particles. The supernatant was further centrifuged at 6458 x g for 10 min to remove fine particles. The solution collected after the two centrifugation treatments was reduced in volume about 2.5-fold by vacuum-rotary evaporation at 40°C. The concentrated solution was adjusted to pH 2 and centrifuged at 6458 x g for 5 min to remove acid-insoluble organic compounds. In this sense, the DOM used in our investigation corresponds to the classical definition of fulvic acid (Stevenson, 1994).

The acid-soluble DOM solution was fractionated and isolated by using a combination of dialysis techniques and XAD resin chromatography (Supelite DAX8; Sigma-Aldrich, St. Louis, MO) (Han and Thompson, 1999b). The apparent molecular weight (MW) cutoffs of dissolved organic compounds were 500, 3500, and 14000 Da, with the <500-Da fraction being excluded from study because it contained large amounts of dissolved inorganic ions. Dissolved humic components were dialyzed against distilled and deionized water at pH 4 and approximately 4°C. The volume ratio of DOM solution to water outside the membrane was 1:20; the solution outside the membrane was changed daily until no dissolved organic carbon (DOC) could be detected. The collected solutions were combined, concentrated by rotary-vacuum evaporation, and dialyzed again with lower MW-cutoff membranes.

Each MW fraction was further fractionated by XAD-8 resin chromatography (Leenheer, 1981). First, the resin-cleanup procedure of Thurman and Malcolm (1981) was performed to remove organic and inorganic impurities from the XAD resin. The size of XAD-8 column was chosen according to Leenheer (1981). The DOM solution at pH 2 was pumped through a resin column at a flow rate of 15 bed volumes h-1. Effluent from the column was collected and designated as the hydrophilic fraction. The column was back-flushed with 0.1 M NaOH, the eluent representing the hydrophobic acids. Compounds that were adsorbed on the XAD-8 column and not desorbed by 0.1 M NaOH were defined as the hydrophobic neutral fraction. Hydrophobic neutrals were isolated by Soxhlet extraction of the resin with methanol and subsequent evaporation of excess methanol.

The two most abundant apparent molecular-weight fractions recovered were 500 to 3500 Da and >14000 Da, and we refer to them as the LMW and HMW fractions, respectively (Han and Thompson, 1999b). The total hydrophilic and the hydrophobic acid components of these MW fractions were used to study both DOM transport and transport of Cu in the presence of DOM. The hydrophobic neutral components were present in low concentrations and were difficult to recover from the biosolids, so they were used only in DOM transport experiments.

Column Experiments
Glass chromatography columns (1.5-cm i.d. x 10-cm length) were dry-packed with the aquifer material to a uniform bulk density of 1.45 Mg m-3, with porosity of 0.45 m3 m-3. The columns were wetted upward with 0.025 M KClO4 at 0.5 mL min-1 by using a peristaltic pump. The columns were saturated slowly over three days to remove air pockets resulting from the dry-packing technique.

Breakthrough curves of a conservative tracer (Cl-) were collected for evaluation of column transport characteristics. Peclet numbers (Pe), often used to evaluate flow characteristics of columns, were calculated from the parameters obtained from the Cl- breakthrough curves:

where V is the mean pore-water velocity (m s-1), L is the length of aquifer column (m), and D is the diffusion–dispersion coefficient (m2 s-1). The Peclet numbers for the columns in this study ranged from 42.9 to 102.4, with an average of 69.3. According to Dunnivant et al. (1992a), when Peclet numbers are greater than 35, convective processes are believed to be dominant over dispersive processes and preferential flow is insignificant during solute transport. Thus the chemical reactions that control transport can be evaluated. Before each transport experiment, the aquifer columns were eluted with dilute HCl to bring the initial effluent pH to 6.6 to 6.7.

The transport of DOM, Cu, or Cu + DOM solutions was investigated by applying a continuous input of solution to the aquifer columns at a Darcy flux of 0.0047 cm s-1. The initial solute concentrations were approximately 2.5 mmol L-1 (i.e., 30 mg C L-1, ranging from 29.5 to 31.8 mg C L-1) for DOM and about 0.074 mmol L-1 (i.e., 4.7 mg Cu L-1, ranging from 4.6 to 4.8 mg Cu L-1) for Cu. All the solutions were made up in 0.025 M KNO3 (to maintain a constant ionic strength) and 0.0005% NaN3 (w/w) (to prevent microbial activity) and were adjusted to pH 6.2. Total Cu concentrations in the column effluents were determined by atomic absorption spectrophotometry, and dissolved organic C was determined with a C analyzer (TOC-5050; Shimadzu, Kyoto, Japan).


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Dissolved Organic Matter Transport in the Aquifer Material
Transport of Dissolved Organic Matter with Different Molecular Weights
Breakthrough curves for the transport of unfractionated LMW and HMW DOM through the aquifer material feature an initial rapid increase in DOC concentration (within 5 pore volumes), followed by a leveling off of values of effluent concentration relative to influent concentration (C/C0) (Fig. 2a) . The effluent concentrations remained less than the influent concentrations and were changing slowly after approximately 16 or 20 pore volumes passed, when input of the DOM solution was terminated. The LMW fraction was retained less strongly than the HMW fraction. For example, at a pore volume of 2, the LMW fraction had already reached 70% of its input concentration, whereas the >14000 Da fraction had reached only 50% of its input concentration.



View larger version (20K):
[in this window]
[in a new window]
  		Fig. 2. (a) Breakthrough curves for 500 to 3500 Da (LMW) and >14000 Da (HMW) dissolved organic matter (DOM) fractions that had not been further separated according to hydrophobicity. The vertical axis represents effluent dissolved organic carbon (DOC) concentration relative to the influent concentration, C/C0, and the horizontal axis represents pore volumes of solution passed through each column. (b) Breakthrough curves of DOC derived from DOM of three hydrophobicity fractions of DOM with molecular weight of 500 to 3500 Da. (c) Breakthrough curves of DOC derived from DOM of three hydrophobicity fractions of DOM with molecular weight of >14000 Da. Arrows indicate when the influent was switched to KClO4.

 
The breakthrough curves indicated slow, asymmetric "tailing" of DOM, suggesting that the dominant mechanism of retention changed during the period of the experiment. One chemical mechanism that is consistent with solute tailing is the retention of dissolved organic compounds by interaction with DOM that had been previously sorbed to the aquifer material. We hypothesize that sorption of DOM components to the inorganic solid phase effectively created a new solid-phase surface partially coated with organic compounds to which subsequent DOM components were strongly attracted.

After DOM inputs were terminated at approximately 16 and 20 pore volumes, 0.025 M KClO4 solutions were pumped through the columns to remove readily desorbable DOM from the aquifer material. A rapid decrease in effluent DOC concentration was observed for both MW fractions (Fig. 2a), suggesting that both fractions in the column were strongly retained by the solid phase. Similar observations were also reported by Dunnivant et al. (1992a) and Gu et al. (1994) for the desorption of stream- or peat-derived DOM by an aquifer material and by an Fe oxide, respectively.

The difference in mobility between the MW fractions occurred throughout the experiment, indicating preferential retention of DOM with larger molecular weights. Dissolved organic compounds with lower molecular weights were less strongly retained and more mobile in the aquifer material, an observation similar to that reported by McCarthy et al. (1993).

Transport of Dissolved Organic Matter Components with Different Polarity
The mobilities of DOM components with different hydrophobicity properties followed the order of total hydrophilic DOM > hydrophobic acid compounds > hydrophobic neutral compounds (Fig. 2b,c). Differences in mobility were more pronounced for HMW DOM. For this MW fraction, at 2 pore volumes the effluent concentrations of the total hydrophilic, hydrophobic acid, and hydrophobic neutral components reached approximately 66, 20, and 10%, respectively, of their influent concentrations. The asymmetry of the breakthrough curves again suggested that retention of DOM compounds was gradually enhanced during the period of the experiment by interactions with previously sorbed organic matter.

Because of differences in experimental conditions and DOM samples among studies, there is no general agreement in the literature about the most likely mechanisms by which DOM is retained by silicates or by oxides of Fe and Mn. In a study of the sorption of peat-derived DOM on Fe oxides, Gu et al. (1994) concluded that specific ligand exchange played a dominant role in DOM retention. Conversely, Jardine et al. (1989) and Baham and Sposito (1994) demonstrated that peat-derived or biosolids-derived DOM was sorbed by various soil components as a result of favorable entropy changes when dissolved organic compounds approach the interface between a particle surface and water. In the present study, preferential retention of both the larger-MW DOM components (Fig. 2a) and of the hydrophobic components of the DOM (Fig. 2b,c) suggests that, under our experimental conditions (pH 6 and K saturation), entropic contributions may have been more important than ligand exchange in retarding the transport of DOM through the aquifer material.

Copper Transport in the Aquifer Material
Copper Transport without Dissolved Organic Matter Present
During transport experiments in the absence of DOM, the movement of Cu in the aquifer material was regulated by Cu interactions with the solid phase. The untreated aquifer material possessed a large capacity to sorb Cu ions, with an estimated maximum sorption capacity (Xm) of 8.44 mmol Cu kg-1 aquifer material (Fig. 1b). Consequently, when a Cu(NO3)2 solution (0.074 mmol Cu L-1) was passed through a column of untreated aquifer material, Cu was strongly retained by the solid phase, up to about 70 pore volumes (Fig. 3) . At 70 pore volumes, the column had adsorbed 1.7 mmol Cu kg-1 material. Thus, approximately 20% of the adsorbing sites, presumably the most strongly sorbing sites, were saturated, and Cu could then pass more readily through the column.



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 3. Breakthrough curves of Cu when passed through untreated aquifer material and aquifer material that was treated with Na citrate–bicarbonate–dithionite to remove sesquioxides. Arrows indicate when the influent was switched to KClO4.

 
After the first 70 pore volumes, the Cu breakthrough curve in the untreated aquifer material was characterized by slowly increasing effluent Cu concentrations. Up to a pore volume of 430, when the influent Cu was stopped, the effluent Cu concentration increased to 70% of its influent concentration. It is likely that Cu adsorption and transport processes occurred simultaneously, giving rise to the broad slope of the breakthrough curve. After the Cu input was terminated, the column was leached with 0.025 M KClO4. The Cu concentration in the effluent declined rapidly, indicating that the retained Cu had only limited tendency to be exchanged with K (i.e., there was a strong hysteresis in desorption of Cu compared with adsorption).

Copper breakthrough in the oxide-free aquifer material exhibited a different behavior from that in the untreated aquifer material (Fig. 3). In the treated, oxide-free aquifer column, Cu breakthrough occurred more rapidly at approximately 15 pore volumes, with a steep initial slope. We conclude that, in the absence of DOM, Fe and Mn oxides or hydroxides dramatically inhibited Cu mobility in the aquifer matrix.

Influence of Molecular Weight of Dissolved Organic Matter on Copper Transport
We have reported elsewhere that the Cu-binding capacity of the LMW DOM of this study was larger than that of the HMW DOM (Han and Thompson, 1999b). The difference was most pronounced in the hydrophilic fractions, in which LMW DOM had a Cu binding capacity of 14.3 mmol Cu mol-1 DOC and HMW DOM had a Cu binding capacity of 1.86 mmol Cu mol-1 DOC. Therefore, we initially hypothesized that the LMW DOM component would have the greater potential to accelerate Cu migration in the aquifer matrix. But further investigation demonstrated that this expectation was not completely correct.

In the presence of two different MW fractions of DOM (not fractionated by hydrophobicity), we passed Cu through untreated aquifer material (Fig. 4a) . Early enhancement of Cu transport by LMW DOM was expected because these compounds had high Cu-binding capacities (Han and Thompson, 1999b) and considerable intrinsic mobility in the aquifer material (i.e., with no Cu present) (Fig. 2). Up to approximately 85 pore volumes, Cu was transported more readily in the presence of LMW DOM than it was with no DOM present. At >85 pore volumes, however, Cu mobility was retarded in the presence of DOM.



View larger version (21K):
[in this window]
[in a new window]
  		Fig. 4. (a) Early breakthrough curves of Cu passed though untreated aquifer material in the presence of low and high molecular weight fractions of dissolved organic matter (DOM). (b) Breakthrough curves of Cu in the presence of hydrophilic acid compounds of two molecular-weight (MW) fractions. (c) Breakthrough curves of Cu in the presence of hydrophobic DOM compounds of two MW fractions. The Cu breakthrough curve in the absence of DOM is repeated in each graph for purposes of comparison. Arrows indicate when the influent was switched to KClO4.

 
One interpretation of the effect of the LMW DOM on Cu mobility is that a portion of the LMW DOM consisted of compounds that could bind Cu effectively in dissolved organic complexes with low or no net negative charge. These complexes could be transported rapidly through the column of aquifer material. This group of compounds would not be dominant in quantity because C/C0 values were only about 0.1 by the 75th pore volume. An alternative explanation of the early breakthrough of Cu is that sorption of DOM on sesquioxide surfaces of the aquifer material blocked access of free Cu to sorption sites. This mechanism seems less likely than the first, however, because LMW DOM was so readily transported at low pore volumes in the absence of Cu (Fig. 2a).

Up to approximately 75 pore volumes, Cu transport with HMW DOM was not much different than it was in the absence of DOM. But at higher pore volumes, HMW DOM retarded Cu mobility (Fig. 4a). Mass balance calculations for the first 150 pore volumes indicated that the quantity of Cu retained in the aquifer column in the presence of the HMW fraction (not separated by polarity) was 34% greater than in the absence of DOM. For the LMW fraction, mass retardation of Cu at 150 pore volumes was 24% greater than in the absence of DOM.

Part of the retardation in Cu mobility observed at higher pore volumes for both MW fractions may have been the result of sorption of the combined DOM–Cu complexes. Such enhanced Cu adsorption by the mineral phase in the presence of DOM has often been attributed to formation of ternary complexes among charged mineral surfaces, metal cations, and organic ligands (Oden et al., 1993; Tipping et al., 1983; Davis, 1984; Davies-Colley et al., 1984; Zachara et al., 1994; Murphy and Zachara, 1995). Hydroxyl groups of oxides or hydroxides and electron-donating functional groups of DOM are mainly responsible for the formation of ternary complexes. Ternary complexes of trace metals with DOM and oxide surfaces have been shown to have greater stability than trace metals chemisorbed alone on oxide surfaces (Davis, 1984). In our study, the presence of DOM in the influent enhanced overall Cu adsorption by the solid phase (Fig. 4a), and ternary complexes may account for part of the retention.

Influence of Dissolved Organic Matter Polarity on Copper Transport
We hypothesized that DOM components that varied with respect to hydrophobicity might affect Cu transport differently due to their variable Cu-binding abilities and their intrinsic mobilities in the aquifer material. One striking difference among these DOM components was their ability to bind Cu ions. Using complexation data that we have reported elsewhere (Han and Thompson, 1999b), we calculated that in the influents containing LMW hydrophilic DOM and hydrophobic-acid DOM, 48 and 24%, respectively, of the Cu was in a complexed form. In contrast, a similar calculation suggested that in the influents containing HMW hydrophilic DOM and hydrophobic-acid DOM, only 6 and 22%, respectively, of the Cu was complexed.

Copper transport through the aquifer material differed in the presence of hydrophilic compounds with different molecular weights. When compared with Cu transport in the absence of DOM, the LMW hydrophilic DOM facilitated early Cu transport in the aquifer column up to approximately 75 pore volumes (Fig. 4b), just as the "whole" LMW DOM did (Fig. 4a). Then a significant retardation of Cu transport was observed, probably enhanced by formation of ternary complexes. In contrast, the HMW hydrophilic DOM completely retarded Cu migration up to approximately 150 pore volumes, at which point Cu began to move more rapidly through the column. Since a small proportion (6%) of the total Cu was initially present in a soluble complex with the HMW hydrophilic component, formation of ternary complexes probably played a minor role in the retardation of Cu transport. Although the exact mechanisms are unknown, two other processes were likely to have increased Cu retention by the aquifer material in the presence of HMW hydrophilic DOM: pH buffering during chemisorption and an electrostatic effect at the water–solid interface.

When much of the soluble Cu was present as free Cu2+ ions in the Cu + DOM input solution, Cu retention by the solid phase probably occurred, to a large extent, via inner-sphere reactions with surface hydroxyl groups of sesquioxides that occurred in the aquifer material (McBride, 1994; Baham and Sposito, 1994). For iron oxide surfaces (the dominant sesquioxides in these aquifer materials) the reaction can be written:

The consequence of chemisorption of Cu2+ ions onto an Fe hydroxide or oxide surface is the release of protons. During the sorption reaction, if the protons released were buffered by the dissolved organic compounds, the sorption equilibrium would be shifted toward the right side of the equation and thus more Cu sorption would be favored. Table 1 shows evidence of a pH buffering effect of the HMW hydrophilic acid components. The pH of the effluent in the DOM-free column decreased from 6.6 to 6.1 in the first 100 pore volumes, whereas in the presence of the hydrophilic DOM, effluent pH values decreased only slightly within the same number of pore volumes, coinciding with greater retention of Cu by the solid phase, as shown in Fig. 4b.


View this table:
[in this window]
[in a new window]
  		Table 1. The pH of effluent at selected pore volumes coming from aquifer-material columns with three different types of influent solutions.

 
Electrostatic effects may have also played a role in Cu sorption by the aquifer material in the presence of DOM. Because of its relatively low Cu-binding capacity, the HMW hydrophilic component is believed to be rich in polysaccharides. According to Stumm and Morgan (1981), adsorption of such polymers can significantly reduce the potential of the diffuse double layer near solid-phase surfaces. In our study, because the pH of the influent column (pH 6) was less than the point of zero charge of the aquifer material (pH 7.6), the solid phase probably included many sites with positive charge from which Cu ions would be repelled. Thus, a decrease in the potential across the diffuse double layer caused by DOM adsorption would reduce repulsion between the surface and Cu cations, creating an environment for more Cu cations to accumulate. Tipping et al. (1983) calculated that a change in the diffuse layer potential by approximately 30 mV by adsorption of humic substances would result in a change in the electrostatic free energy of attraction of about -6 kJ mol-1 for divalent ions like Cu2+. Indirect evidence supporting this hypothesis is shown by comparison of the slopes of the desorption curves in Fig. 4b (>400 pore volumes). The Cu retained by the solid phase in the presence of HMW hydrophilic DOM exhibited a greater tendency to be desorbed compared with Cu in the DOM-free column, consistent with the presence of weakly adsorbed (e.g., exchangeable) Cu near solid-phase surfaces.

Similar to the hydrophilic DOM, both MW components of the hydrophobic acids tended to retard overall Cu transport (Fig. 4c). An initial period of enhanced transport (compared with DOM-free transport) was observed up to approximately 75 pore volumes. Then, between pore volumes 100 and 400, the LMW hydrophobic acids showed the greater tendency to inhibit Cu transport. The Cu-binding capacity of the LMW hydrophobic acid fraction was only slightly larger than that of the HMW hydrophobic acid fraction (7.38 vs. 6.71 mmol Cu mol-1 DOC; Han and Thompson, 1999b), and the retardation of first Cu breakthrough was roughly similar for the two fractions. The desorption curves for both MW fractions (>400 pore volumes) indicate that some Cu was released from the solid phase at about 450 total pore volumes. We do not speculate here about why Cu was released at this point in the experiment, but clearly the rates and mechanisms of trace metal desorption from aquifer materials deserve the attention of future research.

Mass Balance
For a better comparison of the influences of hydrophilic or hydrophobic DOM with different molecular weights on Cu mobility, the mass of Cu retained in the columns at selected pore volumes was calculated (Table 2). Small proportions of the Cu added to the columns were found in the early effluent samples (0–75 pore volumes) where the LMW hydrophilic DOM fraction and both of the hydrophobic DOM fractions were present, but Cu mobility was largely inhibited. When compared with the DOM-free system, all four DOM components inhibited Cu mobility at pore volumes between 75 and 400 (Table 2). Thus, over all 400 pore volumes, considerably more Cu was retained by the columns when DOM fractions were present.


View this table:
[in this window]
[in a new window]
  		Table 2. Percentage of added Cu that was retained by the aquifer material in the absence and presence of dissolved organic matter (DOM) fractions at selected pore volumes. Between 0 and 75 pore volumes, 44 µmol of Cu were added to the columns. Between 75 and 400 pore volumes, 191 µmol of Cu were added.

 
Influence of Sesquioxides on Copper Transport in the Presence of Dissolved Organic Matter
As shown in Fig. 3, sesquioxides influenced Cu transport without DOM present, and removal of sesquioxides was expected to favor Cu transport in the aquifer material, regardless of DOM properties. This was confirmed by the Cu + DOM breakthrough curves (BTCs) in dithionite-treated, oxide-free aquifer columns in the presence of selected DOM fractions (Fig. 5) . The Cu + DOM BTCs in the oxide-free material showed faster initial Cu breakthrough (<30 pore volumes) than in the untreated material (approximately 75 pore volumes) (compare with Fig. 4). Early Cu transport was promoted by both hydrophilic and hydrophobic LMW DOM fractions, yet after approximately 30 pore volumes Cu movement was slowed in comparison with the no-DOM BTCs. Still, the degree of retardation by the oxide-free material was less than that found in untreated aquifer material. The results indicated that after removal of sesquioxides, the interactions among the DOM, Cu, and the solid surface were reduced, and more Cu could be transported through the aquifer material faster. In the presence of HMW hydrophilic DOM, early Cu transport was retarded in both oxide-free and untreated columns. But the parallel slopes of the breakthrough curves suggest that Cu moved through the columns at a rate comparable with that when no DOM was present (Fig. 4 and 5). This observation suggests that the capacity of the aquifer material to retain Cu + DOM complexes was exceeded more rapidly for the HMW DOM than for the LMW DOM.



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 5. Breakthrough curves of Cu passed through sesquioxide-free aquifer material (AM) in the presence of selected dissolved organic matter (DOM) fractions. The Cu breakthrough curve in the absence of DOM is also presented for purposes of comparison.

 

   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Our study does not necessarily reconcile all earlier interpretations of metal mobility in the presence of DOM. It does demonstrate that early Cu breakthrough in sandy, sesquioxide-rich aquifer material may be promoted by DOM and that longer-term Cu movement may be retarded by DOM. Copper mobility depended on several factors, including the Cu-complexing ability of the solid phase, the Cu-complexing ability of the DOM, and the molecular weight of the DOM fractions. In our experiments, except for the HMW hydrophilic fraction, DOM components facilitated some Cu transport through the aquifer columns at early stages (about the first 75 pore volumes). From approximately 75 to 400 pore volumes, however, all DOM components inhibited Cu mobility. We hypothesize that several mechanisms could account for retardation of Cu movement in the presence of the DOM fractions, including the formation of ternary complexes between the aquifer material, Cu, and DOM; changes in the electrostatic potential at the solid-phase surface; and pH buffering by DOM. Further work to determine the relative importance of these mechanisms for the various DOM fractions is required.


   ACKNOWLEDGMENTS
 
We thank Hua Ren, Linda Schultz, and Chao Shang for technical assistance in various aspects of this project. We appreciate many editorial and scientific suggestions offered by several anonymous reviewers. We also acknowledge the Iowa Agriculture and Home Economics Experiment Station, the City of Ames, and the USEPA (Grant no. R81-9996-01-0) for funding this research. Journal Paper no. J-18032 of the Iowa Agriculture and Home Economics Experiment Station, Ames. Project no. 3359 supported by Hatch Act and State of Iowa funds.


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


Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2003 32: 1577-1582. [Full Text]  




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 Related articles in JEQ
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 ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Han, N.
Right arrow Articles by Thompson, M. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Han, N.
Right arrow Articles by Thompson, M. L.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Han, N.
Right arrow Articles by Thompson, M. L.
Related Collections
Right arrow Vadose Zone Processes and Chemical Transport
Right arrow Municipal Wastes
Right arrow Nonequilibrium Transport

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
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome