Published online 17 July 2007
Published in J Environ Qual 36:1273-1280 (2007)
DOI: 10.2134/jeq2006.0373
© 2007 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
TECHNICAL REPORTS
Heavy Metals in the Environment
Colloid Mobilization and Arsenite Transport in Soil Columns: Effect of Ionic Strength
Hua Zhang and
H.M. Selim*
Sturgis Hall, School of Plant, Environmental and Soil Sciences, LSU, Baton Rouge, LA 70803. Contribution from Louisiana State Univ. Agric Center as manuscript no. 07-14-0027
* Corresponding author (mselim{at}agctr.lsu.edu).
Received for publication September 16, 2006.
 |
ABSTRACT
|
|---|
Colloid generation and transport in soils is of significance because of suspected colloid-facilitated transport of contaminants to the groundwater. In this study, colloid mobilization and its effect on the transport of arsenite [As(III)] were investigated in Olivier (fine-silty, mixed, active, thermic Aquic Fraglossudalfs) and Windsor (mixed, mesic typic Udipsamments) soil columns. Input solution of 10 mg L–1 As(III) in 0.01 M NaCl was applied to water-saturated columns, and followed by leaching with deionized water (DIW). Flow interruptions were performed during the As(III) input and DIW leaching phases. Turbidity, electrical conductivity (EC), and pH of column effluents were monitored with time. Total and dissolved concentrations of As, Fe, and Al were analyzed. Effluent results demonstrated that colloid-facilitated transport contributed little to arsenic movement when the solution ionic strength was maintained constant. Mobilization of colloidal amorphous material and enhanced transport of As(III) were observed as a result of changes in ionic strength of the input solution. The peak of colloid generation coincided with peak concentrations of Fe, suggesting mobilization of Fe oxides and facilitated transport of As(III) adsorbed on oxide surfaces. Colloid mobilization was enhanced due to flow interruption in the Olivier column, which suggests slow dissociation of aggregated colloidal particles. Moreover, effluent results indicate significant effect of organic matter in stabilizing aggregates of colloidal particles.
Abbreviations: BTC, breakthrough curve • DIW, deionized water • EC, electrical conductivity • NTU, nephelometric turbidity units, SDC, sodium dispersible clay • WDC, water dispersible clay • XRD, X-ray diffraction
|
INTRODUCTION
|
|---|
HIGH CONCENTRATIONS of arsenic (As) in soils and aquifers have been observed worldwide due to both geologic and anthropogenic sources. In the soil environment, the dominant forms of arsenic are arsenate [As(V)] under aerobic conditions and arsenite [As(III)] under anaerobic conditions. At natural pHs, arsenite (pKa1 = 9.2, pKa2 = 12.7) predominantly exists as H3AsO3 and H2AsO3–, while arsenate (pKa1 = 2.3, pKa2 = 6.8, pKa3 = 11.6) is mainly in the forms of H2AsO4– and HAsO42– (Goldberg and Johnston, 2001). Because of the slow kinetics of the redox reactions, both oxidation forms are often found in soils regardless of pH and Eh (Masscheleyn et al., 1991). Chemical speciation of arsenic in groundwater often reveals that a significant fraction is in the arsenite form, which is believed to be more toxic than As(V) (Le et al., 2000).
Adsorption to soil and sediment is the major pathway of attenuating arsenic bioavailability and toxicity in the environment. Several spectroscopic and kinetic studies have demonstrated that iron oxides/hydroxides have a particularly high affinity for arsenic by forming inner-sphere complexes (Sun and Doner, 1998; Manning et al., 1998; Raven et al., 1998). Dixit and Hering (2003) reported that As(V) have a higher affinity to iron oxides than As(III) below pH 5–6, whereas more As(III) was sorbed above pH 7–8. Adsorption studies on soils demonstrated that the amounts of arsenite adsorption were less than arsenate (Manning and Goldberg, 1997; Smith et al., 1999). In addition, it was found that the adsorption of arsenite on soils is time-dependent and essentially irreversible (Elkhatib et al., 1984a, 1984b). Most studies on the mobility of arsenite or arsenate were conducted in well-mixed batch experiments, whereas few studies investigated the mobility of arsenite under dynamic flow conditions. Recently, the column studies of Radu et al. (2005) demonstrated that As(III) was more mobile at pH 4.5 than pH 9, and that increasing pore water velocity increased the mobility of As(III) in goethite-coated sand.
Traditionally, the transport of arsenic is assumed to occur entirely in the aqueous or soluble phase. However, a significant portion of mobile arsenic in the groundwater is present in colloidal form, i.e., arsenic minerals or adsorbed on mineral surfaces. For example, Le et al. (2000) observed that more than 50% of the arsenic was present in the particulate fraction (>0.45 µm) for the well water samples collected from northern Alberta. Ishak et al. (2002) investigated arsenic release from intact columns of Appling loamy sand through leaching with deionized water (DIW). Their results showed that arsenic levels present in the leachate roughly correlated with effluent turbidity, which support the supposition that arsenic movement was generally associated with mobilized colloids.
The potential of colloids mobilized from soils or sediments in facilitating the transport of strongly sorbed contaminants have been studied by several researchers. Based on the classical Derjaguin-Landau-Verwey-Overbeek (DLVO) theory, a low concentration of background electrolytes increases repulsive electrostatic force between colloids and therefore leads to their dispersion (Sposito, 1989). Using columns of a highly weathered aquifer material, Seaman et al. (1995) observed the mobilization of colloidal particles with positive eletrophoretic mobility when the input solution (CaCl2, NaCl, and MgCl2) was replaced with DIW. Grolimund et al. (1996) observed substantial colloid release and Pb2+ mobilization in columns of an aquic dystric Eutrochrept when input solution was switched from 50 mM NaCl to 0.15 mM CaCl2. Recently, they developed a mathematical transport model coupling the transport of colloids and solutes, where a large array of reactions was considered (Grolimund and Borkovec, 2005). Similarly, Roy and Dzombak (1997) observed substantial transport of strongly sorbed Ni2+ in Lincoln sand in the presence of colloids mobilized by flushing with low ionic strength solution. Puls and Powell (1992) conducted column experiments with aquifer material to investigate possible effects of colloidal iron oxide on facilitating As(V) transport. They observed the mobilization of colloid-associated arsenate when flushing the column with DIW. Moreover, they evaluated several factors (particle size, pH, velocity, ionic strength, and anions) on the transport of colloidal iron oxides. Their results demonstrated that the presence of HAsO42– and HPO42– anions substantially increased the mobility of iron oxides due to increased particle-particle repulsive forces.
The reductive dissolution of iron oxide and subsequent release of sorbed or precipitated arsenic were proposed as the mechanisms of arsenic contamination of aquifers in Bangladesh (Smedley and Kinniburgh, 2002; Herbel and Fendorf, 2006; Pedersen et al., 2006). In addition, the reduction of iron(III) (hydr)oxides may mobilize colloidal arsenic by dissolving the ferric oxyhydroxide coatings cementing the aggregates (Ryan and Gschwend, 1990; Thompson et al., 2006). Ryan and Gschwend (1990) indicated that mobilization of colloidal clay particles in anoxic groundwater was a result of the depletion of oxidized iron coatings which mobilized colloids. Recently, Thompson et al. (2006) suggested that colloid mobilization during iron redox oscillations with a Hawaiian soil were dependent on pH shifts due to the Fe redox reaction.
The objective of this study was to determine the relationships between the mobilization of colloidal Fe oxides and transport of As(III) in Olivier (fine-silty, mixed, thermic Aquic Fragiudalf) and Windsor (mixed, mesic Typic Udipsamments) soils. Specifically, we conducted sorption/mobilization column experiments to assess the potential for rapid mobilization of arsenic from soils by the release of colloids on reducing ionic strength of the influent.
|
Materials and Methods
|
|---|
Surface soils from the Ap horizon (0–10 cm) of Olivier loam and Windsor sand were collected from Louisiana and New Hampshire, respectively (Table 1). The soils were air-dried and passed through a 2-mm sieve, and were analyzed for pH using 1:1 soil/water paste, for organic matter using the acid dichromate oxidation method (Nelson and Sommers, 1982), for free iron oxides by the dithionite-citrate-bicarbonate method (Mehra and Jackson, 1960), and for cation exchange capacity of the acid soils by exchange with 0.1 M BaCl2–0.1 M NH4Cl (Gillman, 1979).
X-Ray Diffraction
Random powder samples were prepared by grinding in isopropyl alcohol using an agate mill micronizer. Sodium dispersed clay (SDC) samples were prepared by immersing overnight 40 g soil in 320 mL 0.01 wt % Na3PO4. Water dispersible clay (WDC) samples were treated by mixing 40 g soil in 320 mL of DIW and shaking for 16 h on a reciprocal shaker. The coarse (0.2–2 µm) and fine (<0.2 µm) fractions of SDC and WDC were extracted by gravity settling and centrifugation, respectively. The oriented slides for clay mineral analysis were treated after K saturation with heat treatment at 300 and 550°C and Mg saturation with salvation using glycerol and ethylene glycol. Composite effluent solution from the Olivier column was filtered through a 0.1-µm membrane filter and particles retained on the filter were used for XRD analysis. Oriented and random powder samples were analyzed by X-ray diffraction (XRD) using a Siemens/Bruker model D5000 (Bruker AXS Inc., Madison, WI) with Cu-K
radiation at 40 kV and 30 mA in the 2–36°2
and 2–70°2 range, respectively, with a step size of 0.02°2 and a count time of 1.0 s. Compensation variable-slit, sample spinning, and position reflection correction with quartz (100) internal standard were used. MacDiff 4.2.5 software (Petschick, 2000) was used for peak decomposition and qualitative identification of minerals. Semi-quantitative mineralogical compositions were estimated from relative peak areas of characteristic intensities for each mineral species.
The XRD analysis of bulk soil samples indicated that both Olivier and Windsor soils consisted of quartz, with much less plagioclase feldspar, and clay minerals (Table 1). The mineralogy of the fine (<0.2 µm) and coarse (0.2–2 µm) fractions of SDC and WDC are given in Table 2. Based on XRD results, smectite, illite, kaolinite, and quartz were found in the fine clay fraction (<0.2 µm) of both soils, whereas chlorite was only found in the Windsor soil.
View this table:
[in this window]
[in a new window]
|
Table 2. Semi-quantitative mineral composition of coarse (0.2–2 µm) and fine (<0.2 µm) fractions of sodium-dispersible clay (SDC) and water-dispersible clay (WDC) for Olivier and Windsor soils.
|
|
Miscible Displacement Experiments
Saturated column transport experiments were conducted to study arsenite transport in the soils described above. Two acrylic columns (6.4-cm i.d. and 10 cm in length) were uniformly packed with <2 mm air-dried soil samples resulting in bulk densities ranging from 1.05 to 1.35 g cm–3. Influent was supplied by a piston pump to each column to sustain steady-state flow and the effluent was collected using a fraction collector. The columns were saturated with 0.01 M NaCl background solution where upward flow was maintained to ensure water saturation. Input solution of 10 mg L–1 arsenite (in the form of NaAsO2) in 0.01 M NaCl was applied subsequent to full saturation. After about 10 pore volumes of arsenite pulse application, the flow was completely interrupted for 5 d to evaluate non-equilibrium or kinetic-controlled arsenic retention and colloid mobilization. Some 8 to 10 pore volumes of arsenite solution were applied subsequent to flow interruption and were then leached with DIW. Another 5-d flow interruption was performed during leaching with DIW. Electrical conductivity (EC), turbidity, and pH of column effluent were monitored over time (Fig. 1–3
). The samples collected from the column experiments were analyzed by inductively coupled plasma atomic emission spectrometry (ICP–AES; Spectro Citros CCD, model CCD; Spectro Analytical Instruments, Kleve, Germany). Total (<20 µm) concentrations of As, Fe, and Al in the <20 µm fraction in effluent samples were measured by digestion using 16 M HNO3, whereas the dissolved (<0.20 µm) concentrations were determined after filtration with 0.20-µm membrane filter. The particulate (0.2–20 µm) As, Fe, and Al concentrations were calculated from the difference between total and dissolved concentrations. The volume of the arsenite pulse along with soil parameters associated with each column is given in Table 3.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 1. Effluent turbidity during injection of 10 mg L–1 As(III) in 0.01 M NaCl followed by leaching with deionized water (DIW) for the Olivier and Windsor soil columns. Arrows indicate pore volumes when flow interruptions occurred.
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 2. Effluent pH during injection of 10 mg L–1 As(III) in 0.01 M NaCl followed by leaching with deionized water (DIW) for the Olivier and Windsor soil columns. Arrows indicate pore volumes when flow interruptions occurred.
|
|
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 3. Effluent electrical conductivity (EC) during injection of 10 mg L–1 As(III) in 0.01 M NaCl followed by leaching with deionized water (DIW) for the Olivier and Windsor soil columns. Arrows indicate pore volumes when flow interruptions occurred.
|
|
View this table:
[in this window]
[in a new window]
|
Table 3. Column soil physical parameters for miscible displacement experiments. Values of the dispersion coefficient (D) were estimated from tritium breakthrough results.
|
|
Sequential Extraction
At the end of each miscible displacement experiment, the 10-cm-long soil column was dismantled and sectioned into five segments (2 cm each). The amounts of arsenic bound at different strengths by the soil matrix were determined through sequential extraction. A simplified version of the extraction procedures proposed by Keon et al. (2001) was used. Specifically, four fractions were quantified, referred to here as exchangeable (extracted with 1 M MgCl2, pH = 8), strongly bound (extracted with 1 M NaH2PO4, pH = 5), coprecipitated with Fe/Al oxides (extracted with 0.2 M ammonium oxalate, pH = 3, reacted in dark), and recalcitrant fraction (extracted with 16 M HNO3). The first three phases were measured by mixing 3 g of soil with 20 mL of the extractant solution, shaking for 24 h, and centrifuging, whereas recalcitrant arsenic was determined by mixing with 20 mL of 16 M HNO3 solutions and shaking for 2 h in a water bath maintained at 80°C.
|
Results
|
|---|
Arsenite Transport
The results from saturated miscible displacement experiments are presented in Fig. 4 as breakthrough curves (BTCs) for Olivier and Windsor soils. The transport of arsenite in Olivier soil showed early arrival with a diffuse BTC front, characterized by gradual increase of total arsenic concentration to 3.5 mg L–1 after 17 pore volumes of arsenite input. Flow interruption during arsenic input pulse resulted in a decrease of arsenic concentration by 1.0 mg L–1. Onefold increase in the total arsenic concentration in the effluent was observed as a result of and after the arsenic input solution was replaced with DIW. Stopping the flow during leaching with DIW resulted in an increase of arsenic concentration by 0.6 mg L–1. After 26 pore volumes of leaching with DIW, the total arsenic concentration in the effluent decreased to 0.4 mg L–1. Moreover, the desorption or leaching side of the BTC indicates extensive tailing or slow release of As from the soil (see Fig. 4). The particulate arsenic (difference between <20 µm and <0.2 µm arsenic) was observed only when the input solution was replaced by DIW and had a peak concentration of 2.0 mg L–1.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 4. Breakthrough curves (BTC) of total (<20 µm) and dissolved (<0.2 µm) arsenic for the Olivier and Windsor soil columns. Arrows indicate pore volumes when flow interruptions or leaching with deionized water (DIW) occurred.
|
|
The breakthrough of arsenite from the Windsor column showed extensive retardation (>15 pore volumes) with a sharp front between 25 and 27 pore volumes (see Fig. 4). The peak concentration of dissolved arsenic (<0.2 µm) occurred after the column was leached with more than 10 pore volumes of DIW. Following continued leaching with DIW for 25 pore volumes, a relatively high concentration (
1.4 mg L–1) of arsenic was present in the effluent. Flow interruption during the leaching phase resulted in the release of particular arsenic with a peak concentration of 1.5 mg L–1.
Arsenic Retention in Soils
Arsenic fractions retained by each soil vs. depth as determined using sequential extraction are presented in Fig. 5. In addition, the percentages of arsenic recovered in the effluent and sequential extraction are given in Table 4. Effluent results demonstrate extensive retention of arsenite. In fact, after leaching with DIW for 20 to 30 pore volumes, the percentage of applied arsenic retained in the soils were 59.2 and 70.9% in Column 101 (Olivier) and Column 102 (Windsor), respectively. Based on sequential extraction, large fractions of arsenic were strongly bound (extracted with NaH2PO4) or coprecipitated with Fe/Al oxides (extracted with ammonium oxalate). The results in Fig. 5 illustrate that 27 and 32% of arsenic were retained in the 0- to 2-cm layer of Olivier and Windsor soils, respectively, indicating strong retention of arsenic by both soils.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 5. The percentage recoveries of arsenic from different soil column depths. Agents used for extractions were: exchangeable (1 M MaCl2), strongly sorbed (1 M NaH2PO4), precipitated (0.2 M ammonium oxalate), and recalcitrant (16 M HNO3).
|
|
Mobilization of Colloidal Particles
Turbidity represents the concentration of colloidal particles in the effluent solution and is shown by the BTCs in Fig. 1. For Olivier soil, the initial effluent turbidity was 50 NTU and decreased rapidly to less than 10 NTU after 3 pore volumes of arsenic application. Flow interruption implemented during arsenic input phase resulted in a small increase of turbidity to 22 NTU. Rapid release of large amounts of colloidal particles was observed when the input solution was replaced by DIW, as illustrated by the sharp increase in effluent turbidity (>100 NTU). In addition, flow interruption during leaching with DIW resulted in an enhanced release of colloidal particles with peak turbidity >100 NTU. For Windsor soil, the release of colloidal particles was observed when the DIW was introduced, as reflected by the peak turbidity around 50 NTU. However, flow interruptions did not result in an increase of turbidity. The XRD pattern shown in Fig. 6 demonstrates that the major portion of the mobile particles in the effluent solution from Olivier soil was amorphous with only trace amounts of quartz particles.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 6. X-ray diffractograms (XRD) of colloids in the composite effluent solution of the Olivier soil column.
|
|
Release of Iron and Aluminum
Elution results of Fe and Al from soil columns are presented by the BTCs in Fig. 7 and 8 for Olivier and Windsor Soils, respectively. Large amounts of Fe were released during the first 10 pore volumes of the arsenic pulse with peaks of 10 and 52 mg L–1 for Olivier and Windsor soils, respectively. Flow interruption during arsenic input phase resulted in the release of Fe from both soils. However, the amount of release from Windsor soil (peak concentration of 41 mg L–1) was much higher than that from Olivier soil (of 3.6 mg L–1). Replacing the input solution with DIW resulted in a rapid increase of total Fe concentration in the effluent from the Olivier column. In contrast, the total Fe concentration in the effluent from the Windsor column continued to decrease during leaching with DIW. Furthermore, flow interruption during the DIW leaching phase resulted in a rapid release of Fe (8.5 mg L–1) from the Olivier soil but not from the Windsor soil. For the Olivier column, the particulate Fe fraction was only observed after the arsenic input solution was replaced with DIW. In contrast, a small fraction of particulate Fe was observed for the Windsor soil throughout the arsenic input phase.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 7. Mobilization of total (<20 µm) and dissolved (<0.2 µm) iron fractions from the Olivier and Windsor soil columns. Arrows indicate pore volumes when flow interruptions or leaching with deionized water (DIW) occurred.
|
|
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 8. Mobilization of total (<20 µm) and dissolved (<0.2 µm) aluminum fractions from the Olivier and Windsor soil columns. Arrows indicate pore volumes when flow interruptions or leaching with deionized water (DIW) occurred.
|
|
During the arsenic input pulse, Al concentration in the effluent was consistently low for both Olivier and Windsor soil (see Fig. 8). Significant mobilization of Al was observed only after the DIW was introduced. Flow interruption during DIW leaching resulted in a significant release of Al from Olivier soil. Furthermore, Al was more mobile in Oliver soil as illustrated by the higher Al concentration of the effluent than that for Windsor soil. A significant amount of Al was present in the particulate fraction, especially for Windsor soil.
|
Discussion
|
|---|
For Olivier soil, there was no apparent difference between the mineralogical composition of SDC and WDC. This is in agreement with Seta and Karathanasis (1997) and is believed to be a result of the low organic content in this soil. However, the impact of dispersing agent (sodium vs. DIW) was observed in the XRD patterns of Windsor soil. Specifically, smectite constitutes a significant fraction of SDC but was completely missing in WDC, indicating that the aggregates between smectite particles were relatively stable. Similarly, much less kaolinite was found in WDC than in SDC. In addition, for Windsor soil, kaolinite was only found in the coarse clay fraction (0.2–2 µm), indicating large size aggregates of kaolinite. The relative stability of smectite and kaolinite aggregates was explained based on the high organic matter content of Windsor soil. The important role of organic matter as a major binding agent for water-stable aggregates has been well documented in the literature (see Tisdall and Oades, 1982).
Even though XRD analysis demonstrated that the clay minerals in Olivier soil were readily dispersed by DIW, we did not observe significant mobilization of clay minerals in our column study based on the XRD pattern of colloidal particles in the effluent (Fig. 6). The peaks of colloid BTCs coincided with peaks of Fe BTCs (Fig. 4 and 7) for the Olivier column, suggesting that a fraction of the amorphous material was colloidal Fe oxides. Relatively few studies have been performed to investigate the mineral composition of colloid particles mobilized from soils and sediments. Kretzschmar et al. (1999) suggested that mobilized colloids were representative of the mineralogy composition of the clay fraction. However, Seaman and Bertsch (1997) found that colloids mobilized from southeastern coastal plain sediments consisted mainly of Al-rich goethite, with lesser amounts of kaolinite, and some crandallite, which is different from the mineral composition of bulk sediment. Several studies reported that a wide range of chemical and physical factors could impact the attachment-detachment, flocculation-dispersion, and transport of colloids in a natural environment (Bertsch and Seaman, 1999; Kretzschmar et al., 1999). Our results also indicated that the characteristics of colloids mobilized from soil may be significantly different from that of the clay fraction in the bulk soils.
In our column experiments, the initial release of colloidal particles might be partly due to the dispersion of colloidal particles induced by increasing exchangeable sodium percentage (ESP). Another possible mechanism for colloid mobilization during the initial stages was the formation of innersphere surface complexes between arsenite and Fe/Al oxides resulting in reduced aggregate stability. Results from our experiments indicate that significant amounts of colloidal particles were mobilized when the background solution (0.01 M NaCl) was displaced with DIW. Comparisons between column effluent results indicate that significantly more colloids were mobilized from Olivier than Windsor soil, which is in agreement with the relative stability of aggregates based on the mineralogical analysis. Moreover, flow interruption in the Olivier column resulted in a significant increase of effluent turbidity, which is indicative of slow release of colloids. Colloid mobilization following flow interruption was perhaps a physical process impacted by flow rate or a chemical process such as slow dispersion. However, the colloid concentration did not increase as a result of stop flow for the Windsor column, suggesting other mechanisms for colloid mobilization.
The shape of Fe BTCs (Fig. 7) from the Olivier column closely followed the BTC of colloid mobilization (Fig. 4). Based on these similarities, the amorphous particles appearing in the effluent from the Olivier column may be ferric hydroxide minerals. For Windsor soil, the high effluent concentration of Fe (see Fig. 7) might be due to reductive Fe(III) dissolution (Ryan and Gschwend, 1990). Furthermore, we suggest that the differences between the observed Fe behavior in the Olivier and Windsor columns can be explained based on the higher organic matter content of Windsor soil. For Olivier soil, iron oxides were associated with the clay particles or quartz grains. Changing ionic strength resulted in the dispersion of particles and exposed iron oxides in the aggregates to a reductive solution, which stimulated the reduction of iron oxides. In contrast, complexes were likely formed between iron hydroxides and the organic matter in Windsor soil (Davis, 1982). Organic coating on those iron oxides modified its reaction and protected it from dispersion by low ionic solution. The role of organic matter in controlling the release of Fe oxides under anaerobic condition warrants further investigation. In our experiments, substantial release of Al was only observed after the input solution was displaced by DIW. While free Al content of Windsor soil was approximately three times that of Olivier soil, substantially less Al was released from Windsor than from Olivier soil. We suggest that Al oxides in Windsor soil were immobilized by complexation with organic matter.
The transport of arsenite was largely controlled by adsorption–desorption on the surfaces of Fe/Al oxides (Radu et al., 2005). Our sequential extraction results demonstrate that a large fraction of the input arsenic was strongly retained on Fe/Al oxides, which was similar to the sequential extraction results of Keon et al. (2001). In addition, the sorption of As(III) in these soils were rate-limited as illustrated by the long tailing and concentration change after flow interruption. Under steady flow and constant ionic strength, the contribution of colloid mobilization to arsenic movement was negligible in both soils. However, the transport of As(III) was greatly enhanced by the introduction of DIW and the subsequent decrease in ionic strength. Goldberg and Johnston (2001) observed that the arsenite adsorption on amorphous Fe and Al oxides actually increased with decreasing ionic strength. Other studies have shown that ionic strength has relatively little effect on the As(III) adsorption (Manning and Goldberg, 1997; Smith et al., 1999). Therefore, desorption is unlikely the cause of the release of arsenic after the high ionic strength input solution was replaced with DIW water. In fact, a high concentration of particulate arsenic was observed after introducing DIW and accounted for a significant portion of the enhanced arsenic concentration in the effluent. In conclusion, the change in ionic strength dispersed colloidal particles and released colloidal arsenic into solution.
Environmental Implications
A major implication of this study is that changes in chemical composition of solutions in aquifer and vadose zones might result in colloid-facilitated transport of contaminants such as As (Grolimund and Borkovec, 2005). For example, landfill leachate often contains high concentrations of heavy metals with high ionic strength. Displacement by rainfall or irrigation water, which typically has low ionic strength, could result in mobilization of arsenic associated with colloidal particles and potential contamination of surface or groundwater. Another possible case is freshwater intrusion into a contaminated coastal aquifer, which is often saturated with saltwater. Moreover, the extent of colloid generation and its effect on contaminant transport relies heavily on the chemical and mineralogical composition of the geological materials.
|
ACKNOWLEDGMENTS
|
|---|
The authors would like to acknowledge the technical assistance of Dr. Ray Ferrell, Dep. of Geography and Geophysics at Louisiana State Univ., Baton Rouge, LA for XRD analyses and Tony Palazzo, U.S. Army Corps of Engineers (Cold Regions Research and Engineering Lab), Hanover, NH for providing the Windsor soil used in this study.
|
NOTES
|
|---|
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.
|
REFERENCES
|
|---|
- Bertsch, P.M., and J.C. Seaman. 1999. Characterization of complex mineral assemblages: Implications for contaminant transport and environmental remediation. Proc. Natl. Acad. Sci. USA 96:3350–3357.[Abstract/Free Full Text]
- Davis, J.A. 1982. Adsorption of natural dissolved organic-matter at the oxide water interface. Geochim. Cosmochim. Acta 46:2381–2393.[CrossRef][Web of Science]
- Dixit, S., and J.G. Hering. 2003. Comparison of arsenic(V) and arsenic(III) sorption onto iron oxide minerals: Implication for arsenic mobility. Environ. Sci. Technol. 37:4182–4189.[Medline]
- Elkhatib, E.A., O.L. Bennett, and R.J. Wright. 1984a. Kinetics of arsenite adsorption in soils. Soil Sci. Soc. Am. J. 48:758–762.[Web of Science]
- Elkhatib, E.A., O.L. Bennett, and R.J. Wright. 1984b. Arsenite sorption and desorption in soils. Soil Sci. Soc. Am. J. 48:1025–1030.[Web of Science]
- Gillman, G.P. 1979. A proposed method for the measurement of exchange properties of highly weathered soils. Aust. J. Soil Res. 17:129–139.[CrossRef]
- Goldberg, S., and C.T. Johnston. 2001. Mechanisms of arsenic adsorption on amorphous oxides evaluated using macroscopic measurements, vibrational spectoscopy, and surface complexation modeling. J. Colloid Interface Sci. 234:204–216.[CrossRef][Web of Science][Medline]
- Grolimund, D., and M. Borkovec. 2005. Colloid-facilitated transport of strongly sorbing contaminants in natural porous media: Mathematical modeling and laboratory column experiments. Environ. Sci. Technol. 39:6378–6386.[Medline]
- Grolimund, D., M. Borkovec, K. Barmettler, and H. Sticher. 1996. Colloid-facilitated transport of strongly sorbing contaminants in natural porous media: A laboratory column study. Environ. Sci. Technol. 30:3118–3123.
- Herbel, M., and S. Fendorf. 2006. Biogeochemical processes controlling the speciation and transport of arsenic within iron coated sands. Chem. Geol. 228:16–32.
- Ishak, C.F., J.C. Seaman, W.P. Miller, and M. Summer. 2002. Contaminant mobility in soils amended with fly ash and flue-gas gypsum: Intact soil cores and repacked columns. Water Air Soil Pollut. 134:287–305.
- Keon, N.E., C.H. Swartz, D.J. Brabander, C. Harvey, and H.F. Hemond. 2001. Validation of an arsenic sequential extraction method for evaluating mobility in sediments. Environ. Sci. Technol. 35:2778–2784.[Medline]
- Kretzschmar, R., M. Borkovec, D. Grolimund, and M. Elimelech. 1999. Mobile subsurface colloids and their role in contaminant transport. Adv. Agron. 66:121–193.[CrossRef]
- Le, X.C., S. Yalcin, and M. Ma. 2000. Speciation of submicrogram per liter levels of arsenic in water: On-site species separation integrated with sample collection. Environ. Sci. Technol. 34:2342–2347.
- Manning, B.A., S.E. Fendorf, and S. Goldberg. 1998. Surface structure and stability of arsenic(III) on goethite: Spectroscopic evidence for inner-sphere complexes. Environ. Sci. Technol. 32:2383–2388.
- Manning, B.A., and S. Goldberg. 1997. Arsenic(III) and arsenic(V) adsorption on three California soils. Soil Sci. 162:886–895.[CrossRef]
- Masscheleyn, P.H., R.D. Delaune, and W.H. Patrick. 1991. Effect of redox potential and pH on arsenic speciation and solubility in a contaminated soil. Environ. Sci. Technol. 25:1414–1419.
- Mehra, O.P., and M.L. Jackson. 1960. Iron oxide removal from soils and clays by a dithionite-citrate system with sodium bicarbonate. Clays Clay Miner. 7:317–327.[CrossRef]
- Nelson, D.W., and L.E. Sommers. 1982. Total carbon, organic carbon, and organic matter. p. 539–577. In A.L. Page (ed.) Methods of soil analysis. Part 2: Chemical and microbiological properties 2nd ed. ASA, Madison, WI.
- Pedersen, H.D., D. Postma, and R. Jacobsen. 2006. Release of arsenic associated with the reduction and transformation of iron oxides. Geochim. Cosmochim. Acta 70:4116–4129.[CrossRef][Web of Science]
- Petschick, R. 2000. MacDiff 4.2.5 manual. Available at http://www.geologie.uni-frankfurt.de/Staff/Homepages/Petschick/PDFs/MacDiff_Manual_E.pdf (verified 16 Apr. 2007)
- Puls, R.W., and R.M. Powell. 1992. Transport of inorganic colloids through natural aquifer material: Implication for contaminant transport. Environ. Sci. Technol. 26:614–621.
- Radu, T., J.L. Subacz, J.M. Phillippi, and M.O. Barnett. 2005. Effects of dissolved carbonate on arsenic adsorption and mobility. Environ. Sci. Technol. 39:7875–7882.[Medline]
- Raven, K.P., A. Jain, and R.H. Loeppert. 1998. Arsenite and arsenate adsorption on ferrihydrite: Kinetics, equilibrium, and adsorption envelopes. Environ. Sci. Technol. 32:344–349.
- Roy, S.B., and D.A. Dzombak. 1997. Chemical factors influencing colloid-facilitated transport of contaminants in porous media. Environ. Sci. Technol. 31:656–664.
- Ryan, J.N., and P.M. Gschwend. 1990. Colloid mobilization in two Atlantic Coastal Plain aquifers: Field studies. Water Resour. Res. 26:307–322.[CrossRef]
- Seaman, J.C., and P.M. Bertsch. 1997. Characterization of colloids mobilized from southeastern coastal plain sediments. Environ. Sci. Technol. 31:2782–2790.
- Seaman, J.C., P.M. Bertsch, and W.P. Miller. 1995. Chemical control on colloid generation and transport in a sandy aquifer. Environ. Sci. Technol. 29:1808–1815.
- Seta, A.K., and A.D. Karathanasis. 1997. Stability and transportability of water-dispersible soil colloids. Soil Sci. Soc. Am. J. 61:604–611.[Web of Science]
- Smedley, P.L., and D.G. Kinniburgh. 2002. A review of the source, behavior, and distribution of arsenic in natural waters. Appl. Geochem. 17:517–568.[CrossRef]
- Smith, E., R. Naidu, and A.M. Alston. 1999. Chemistry of arsenic in soils: I. Adsorption of arsenate and arsenite by selected soils. J. Environ. Qual. 28:1719–1726.
- Sposito, G. 1989. The chemistry of soils. Oxford Univ. Press, New York.
- Sun, X., and H.E. Doner. 1998. Adsorption and oxidation of arsenite on goethite. Soil Sci. 163:278–287.[CrossRef]
- Thompson, A., O.A. Chadwick, S. Boman, and J. Chorover. 2006. Colloid mobilization during soil iron redox oscillations. Environ. Sci. Technol. 40:5743–5749.[Medline]
- Tisdall, J.M., and J.M. Oades. 1982. Organic-matter and water-stable aggregates in soils. J. Soil Sci. 33:141–163.[CrossRef]
TOP
ABSTRACT
INTRODUCTION
