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 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 HighWire
Right arrow Citing Articles via ISI Web of Science (11)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Neilson, J. W.
Right arrow Articles by Maier, R. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Neilson, J. W.
Right arrow Articles by Maier, R. M.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Neilson, J. W.
Right arrow Articles by Maier, R. M.
Related Collections
Right arrow Remediation
Right arrow Bioremediation and Biodegradation
Right arrow Heavy Metals
Right arrow Soil Pollution
Journal of Environmental Quality 32:899-908 (2003)
© 2003 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America

TECHNICAL REPORTS
Heavy Metals in the Environment

Characterization of Lead Removal from Contaminated Soils by Nontoxic Soil-Washing Agents

Julia W. Neilson, Janick F. Artiola and Raina M. Maier*

Department of Soil, Water, and Environmental Science, Shantz Building, Room 429, Univ. of Arizona, Tucson, AZ 85721

* Corresponding author (rmaier{at}ag.arizona.edu)

Received for publication May 3, 2002.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Few effective strategies exist for remediating and restoring metal-contaminated soils. We have evaluated the potential of two environmentally compatible, nondestructive, biological soil-washing agents for remediating aged, lead-contaminated soils. Two contaminated soils were washed with 10 mM rhamnolipid biosurfactant and 5.3% carboxymethyl-ß-cyclodextrin (CMCD). The metal removal efficiency of these agents was compared with 10 mM diethylenetriamine pentaacetic acid (DTPA) and 10 mM KNO3. Lead removal rates by both soil-washing agents exceeded the removal by KNO3, but were an order of magnitude less than removal by the synthetic chelator, DTPA. Analysis of soil extractions revealed that the Pb in the first soil (3780 mg kg-1) was primarily associated with the soluble, exchangeable, oxide, and residual fractions while the Pb in the second soil (23 900 mg kg-1) was found in the soluble, exchangeable, carbonate, and residual fractions. After 10 consecutive washes, rhamnolipid had removed 14.2 and 15.3% of the Pb from the first and second soils, respectively, and CMCD had removed 5 and 13.4% from the same two soils. The Pb removal rate by both agents either increased or was consistent throughout the 10 extractions, indicating a potential for continued removal with extended washing. Significant levels of Cu and Zn in both soils did not prevent Pb removal by either agent. Interestingly, the effectiveness of each agent varied as a function of Pb speciation in the soil. Rhamnolipid was more effective than CMCD in removing Pb bound to amorphous iron oxides, while both agents demonstrated similar potential for removing soluble, exchangeable, and carbonate-bound Pb. Neither agent demonstrated potential for the complete remediation of metal-contaminated soils.

Abbreviations: CMCD, carboxymethyl-ß-cyclodextrin • DTPA, diethylenetriamine pentaacetic acid • EDTA, ethylenediamine tetraacetic acid • TCLP, toxicity characteristic leaching procedure


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
HEAVY METAL CONTAMINATION of soils poses a serious health threat and has become one of the major constraining factors in remediating a wide range of contaminated sites (Peters, 1999). A review of the Records of Decision from the metal-contaminated Superfund sites on the USEPA National Priority List indicates that the predominant cleanup methods have been excavation and disposal in landfills or capping of the contaminated site, methods that do not actually remediate the soils. Alternatives must be investigated to comply with current Superfund regulations requiring treatment to include a reduction in the volume, toxicity, or mobility of the metal contamination (Steele and Pichtel, 1998; Peters, 1999).

One of the alternative strategies currently being explored is soil flushing with pump and treat technologies for in situ remediation (Reed et al., 1996; Pichtel and Pichtel, 1997; Davis and Hotha, 1998; Peters, 1999). Unfortunately, the few single reactants capable of mobilizing all metal contaminants are either toxic or destructive to the physical, chemical, or biological structure of the soil (Papassiopi et al., 1999; Peters, 1999). Thus, use of the most effective washing agents is not feasible in situations where it is requisite that the soil be returned to a healthy and productive state.

Among the most effective soil-washing agents investigated are acids and chelating agents. Acid washes with HCl and HNO3 have produced satisfactory results, yet the high acid strength required is both lethal to soil microflora and destructive to soil physical and chemical structure due to mineral dissolution (Reed et al., 1996; Pichtel and Pichtel, 1997; Davis and Hotha, 1998). Extensive work has also been done with synthetic metal chelators with high metal stability constants. Effective chelators studied include ethylenediamine tetraacetic acid (EDTA), nitriloacetic acid (NTA), N-2 (acetamido) iminodiacetic acid (ADA), pyridine-2,6-dicarboxylic acid (PDA), diethylenetriamine pentaacetic acid (DTPA), and Metaset-Z (Reed et al., 1996; Neale et al., 1997; Pichtel and Pichtel, 1997; Davis and Hotha, 1998; Rampley and Ogden, 1998; Steele and Pichtel, 1998; Papassiopi et al., 1999). Unfortunately, as with the acids, environmental and health risks are associated with the majority of these chelators. The chelator NTA is a Class II carcinogen (Peters, 1999) and DTPA is identified by Sigma Chemical Co. (St. Louis, MO) as toxic and a potential carcinogen. Concern has also arisen with both the potential toxicity of EDTA and its recalcitrance in the environment (Borgmann and Norwood, 1995; Wasay et al., 1998; Peters, 1999; Bohuslavek et al., 2001). Residual chelators in the soil have the potential to mobilize metals in the future as demonstrated by a plume containing a radioactive cobalt–EDTA complex at a site in Oakridge, TN (McArdell et al., 1998). Both Tiedje (1977) and Bolton (1993) have demonstrated EDTA degradation by a variety of soils and sediments, but the residence time is extremely long. Tiedje (1977) documented 13 to 46% mineralization of 2 to 4 mg EDTA kg-1 soil in 13 surface soils after 15 weeks, but minimal degradation was observed in subsurface soil samples. Similarly, Bolton (1993) observed 1 to 18% degradation of 0.01 mM EDTA in subsoils after 16 weeks. Means et al. (1978) found EDTA associated with 12- to 15-yr-old radioactive wastes, thus documenting the extreme persistence of EDTA in the environment.

Anionic surfactants have also shown potential as soil-washing agents due to their ability to solubilize metals within micelles (Tan et al., 1994; Champion et al., 1995; Herman et al., 1995; Miller, 1995; Pichtel and Pichtel, 1997; Torrens et al., 1998; Mulligan et al., 1999). Biosurfactants have advantages over their chemical counterparts because they are not petroleum-based, are less toxic, and are more biodegradable (Banat et al., 2000). Extensive research is needed to evaluate the potential effectiveness of environmentally compatible, biological agents such as surfactants for soil flushing of aged, metal-contaminated soils exposed to diverse environmental conditions. Aged, contaminated soils, rather than artificially contaminated soils, must be used to realistically approximate the variation in metal lability found at contaminated sites (van Benschoten et al., 1997; Davis and Hotha, 1998).

The objective of this research was to compare the Pb removal efficiencies of two environmentally compatible, biological washing agents (rhamnolipid biosurfactant and carboxymethyl-ß-cyclodextrin [CMCD]) with that of a synthetic chelator (DTPA) in two different, aged, contaminated soils. Lead was selected as the contaminant, because it is a priority pollutant frequently found at hazardous waste sites (Reed et al., 1996; Brown et al., 1999). Rhamnolipid surfactant was selected because the conditional stability constants between rhamnolipid and Pb, Cu, and Cd are higher than those for other organic acids (Table 1) that have received favorable reviews for soil washing (Wasay et al., 1998; Peters, 1999; Ochoa-Loza et al., 2001). Conditional stability constants indicate the strength of complexation between a ligand and metal. Ligands with high metal complexation constants have a greater potential for removal of metals bound to soil constituents. In addition, the metal complexation behavior of rhamnolipid in artificially contaminated soils has been well characterized in previous research. We documented a 47 to 96% removal of Cd from four artificially contaminated soils using 12.6 mM rhamnolipid (Torrens et al., 1998). In addition, degradation of rhamnolipid in soil was documented (Maslin and Maier, 2000). Results following a 1 mM rhamnolipid application demonstrated that less than 10% was degraded in the first 4 d, but 60% had been mineralized within three weeks. Thus, the rhamnolipid degradation lag time is sufficient to allow for effective use in metal removal, but short enough to alleviate the concern that surfactants may be difficult to remove following soil washing (Peters, 1999). Cyclodextrins are cyclic oligosaccharides produced from the enzymatic degradation of starch by bacteria. Wang and Brusseau (1995) found that a 1% CMCD solution complexed 90% of the Cd from a 10 mg L-1 Cd(NO3)2 solution in a batch system. The complexation capacity of Cd with CMCD is less than rhamnolipid, but greater than or similar to other organic acids (Table 1). Another advantage of CMCD over other washing agents is that it does not sorb to soils.


View this table:
[in this window]
[in a new window]
  		Table 1. Conditional stability constants for Pb, Cu, Zn, Fe, Cd, and Ca with various organic ligands.

 

   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soils
Two similar soils were selected with differing contamination and environmental exposure histories. The Coeur d'Alene soil is a sandy loam from the floodplain of the lower Coeur d'Alene River in Idaho. The predominant soils in the area are Fluvaquentic Arents and Aquic Xerofluvents that are heavily contaminated by mine and ore wastes that were deposited directly into the river's tributaries before the 1920s from silver, lead, and zinc mines. This floodplain is saturated during annual periods of high water in the spring and summer, and significant areas are devoid of vegetation due to toxic metal concentrations (McGeehan and Williams, 2000). The Camp Navajo soil, from Bellemont, AZ, is also a sandy loam with similar physical properties (Table 2). Camp Navajo is an abandoned army depot in north-central Arizona. The contaminated soil was taken from a well-drained area exposed to lead from both lead-based paint leached from nearby buildings and lubricating oils from railroad cars.


View this table:
[in this window]
[in a new window]
  		Table 2. Soil analysis of Coeur d'Alene and Camp Navajo Soils.

 
Soil-Washing Agents
Extraction solutions were prepared in nanopure H2O and filter-sterilized (0.2-µm cellulose acetate filter) to prevent degradation before use.

The rhamnolipid biosurfactant was supplied by Jeneil Biosurfactant Co. (Saukville, WI) as a 0.45 M aqueous solution that contained approximately a 1:0.77 ratio of monorhamnolipid ({alpha}-L-rhamnopyranosyl-ß-hydroxydecanoyl-ß-hydroxydecanoate) to dirhamnolipid (2-O-{alpha}-L-rhamnopyranosyl-{alpha}-L-rhamnopyranosyl-ß-hydroxydecanoyl-ß-hydroxydecanoate). Jeneil rhamnolipid has a critical micelle concentration of 50 mg L-1 and a molecular weight of 567 g (molecular weight varies with the specific ratio of monorhamnolipid to dirhamnolipid in the product). Rhamnolipid was extracted from the aqueous solution and purified by silica gel column chromatography as described previously (Zhang and Miller, 1994). The purified rhamnolipid was used to prepare a 10 mM extraction solution with pH adjusted to 7.1 (±0.2) using KOH. The resulting ionic strength of the solution was 9 mM K+. The solution pH was optimized to minimize the size of the metal–ligand complex, thus avoiding filtration by small soil pores (Champion et al., 1995). The size of rhamnolipid aggregates is pH dependent and the structures are predominately small vesicles and micelles at pH > 6.0.

Analytical-grade carboxymethyl-ß-cyclodextrin (CMCD) was supplied by Cerestar USA (Hammond, IN). A 5.3% extraction solution was prepared with the pH adjusted to 7.0 based on conditions optimized by Wang and Brusseau (1995). The ionic strength of the solution was maintained between 9 and 10 mM. The optimal CMCD concentration is three orders of magnitude higher than the rhamnolipid solution concentration. This difference reflects a similar difference in their conditional stability constants with Cd (Table 1). Initial research optimizing extraction solution concentrations for both agents was conducted using Cd.

Analytical-grade DTPA (Sigma Chemical Co., St. Louis, MO) was selected as the basis of comparison for the biological agents because it is the synthetic chelator with the highest Pb stability constant of those typically used for soil-washing research (Table 1). As with rhamnolipid, a 10 mM DTPA extraction solution was prepared with pH adjusted to 7.3 to facilitate comparison of the extraction efficiencies of the two agents. A final ionic strength of 34 mM K+ was required for pH adjustment.

Soil Analysis
Soil samples were air-dried, sieved (2 mm), mixed well, and stored at room temperature in the dark. Soil physical and chemical properties were determined by the Soil, Water and Plant Analysis Laboratory of the University of Arizona using standard soil analysis methods (Table 2). Total soil metal concentrations were determined following a modification of USEPA Method 3051 (microwave-assisted acid digestion; USEPA, 1992) and analyzed with a PS 1000 inductively coupled plasma–atomic emission spectrometer (ICP–AES) (Leeman Lab, Lowell, MA). Soil pH was determined from the supernatant of a 1:2 soil to deionized H2O suspension following 1 h of mixing. Leachable metal concentrations were determined with the toxicity characteristic leaching procedure (TCLP) extraction following USEPA Method 1311 (USEPA, 1992). Briefly, the #1 TCLP extraction fluid (5.7 mL glacial acetic acid, and 64.3 mL 1 M NaOH brought to a final volume of 1 L with reagent-grade H2O, pH adjusted to 4.93 with HCl) was added to the soil in a 20:1 liquid to solid ratio and mixed on a rotary shaker for 18 h. Samples were then filtered through Whatman (Maidstone, UK) filter paper. Readily available Pb was analyzed using the DTPA soil test developed by Lindsay and Norvell (1978) to quantify plant-available metals. This extraction (0.005 M DTPA, 0.01 M CaCl2, and 0.1 M triethanolamine buffered at pH 7.3) will be referred to as the plant-available extraction to avoid confusion with the 10 mM DTPA soil-washing solution. The TCLP and plant-available extracts were acidified with HNO3 to a pH < 2, filtered with a 0.2-µm nylon filter, and analyzed by flame atomic absorption spectroscopy (FAA) with an Instrument Laboratory Video 12 aa/ae spectrophotometer (Allied Analytical Systems, Waltham, MA). The wavelengths used for FAA analysis were 217 nm for Pb, 213.9 nm for Zn, 324.7 nm for Cu, and 248.3 nm for Fe. Standard curves were generated from metal standard solutions obtained from Fisher Scientific (Pittsburgh, PA).

Batch Experiments
All batch soil-washing experiments were conducted in acid-rinsed 50-mL Nalgene Oakridge tubes (Nalge Nunc International, Rochester, NY). Triplicate tubes containing 2.5 g soil and 5 mL extracting solution were prepared for each treatment, incubated for 18 to 22 h on a rotary shaker at 200 rpm at 23°C, and then centrifuged at 48 000 x g for 20 min. The extract was then decanted, filtered through a 0.2-µm nylon filter, acidified to pH 2 with concentrated HNO3, and stored at 4°C until analyzed. Rhamnolipid has a pKa of 5.6 (Ochoa-Loza et al., 2001) and thus precipitates on acidification. Therefore, extracts from treatments containing rhamnolipid were incubated at 4°C for 48 h then centrifuged at 12 100 x g for 10 min to remove rhamnolipid from the matrix before analysis. Lead, copper, zinc, and iron concentrations were determined by FAA. For consecutive extractions, 5 mL of new extracting solution was added to the soil pellet and the tube returned to the shaker unless otherwise indicated.

Four series of batch experiments were conducted to evaluate the extraction potential of the soil-washing agents. In the first experiment, two sequential extractions were performed on each of the soils with 10 mM rhamnolipid and 10 mM DTPA. The results for the Camp Navajo soil demonstrated that more Pb was removed by rhamnolipid in the second extraction than the first while the Pb removal by DTPA decreased in the second extraction. Thus, a second batch experiment was designed with four sequential extractions using rhamnolipid, CMCD, and a 10 mM KNO3 ionic strength control solution. Soil metal toxicity levels, as measured by TCLP analysis, were evaluated following the four extractions. The pattern of constant or increasing metal removal by rhamnolipid was again observed over four consecutive washes for both soils. Thus, a third experiment was conducted with 10 consecutive washes, an increased soil to solution ratio of 2.5 g soil in 10 mL extracting solution, and an addition of a 50 mM Ca(NO3)2 control extraction solution. The Ca(NO3)2 control was added to evaluate the effect of washing with a divalent cation control solution in addition to the KNO3. The final experiment was conducted to evaluate the effect of the prolonged wetting of the soils during repeated extractions on metal removal efficiency. The 2.5-g samples were preincubated in 5 mL 10 mM KNO3 for 14 d at 4°C in the dark. The KNO3 supernatant was then removed by centrifugation. Single extractions were conducted simultaneously with rhamnolipid, CMCD, and KNO3 extracting solutions on both dry and prewetted soils as described previously. Metal concentrations were determined in all extracts.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Both the Coeur d'Alene and Camp Navajo soils exhibit Pb toxicity as measured by TCLP extract concentrations in excess of the 5 mg L-1 USEPA regulatory level. The TCLP values were 12.8 (±0.5) mg L-1 for the Coeur d'Alene soil and 660 (±61) mg L-1 for the Camp Navajo soil. Total metal concentrations for each soil are shown in Table 3. The TCLP values in Table 3 were expressed as milligram removed per kilogram soil for the purposes of comparison with other extract values.


View this table:
[in this window]
[in a new window]
  		Table 3. Metal analysis and comparison of metal removal by the toxicity characteristic leaching procedure (TCLP), diethylenetriamine pentaacetic acid (DTPA), and rhamnolipid in Coeur d'Alene and Camp Navajo soils.

 
First Experiment
Lead removal from the Coeur d'Alene and Camp Navajo soils using 10 mM DTPA was more than 10 and 40 times greater, respectively, than removal using 10 mM rhamnolipid (Fig. 1) . These results were not surprising considering the large difference in strengths of complexation between these chelators and Pb as shown in Table 1. Unfortunately, it is not cost effective to increase the concentration of the rhamnolipid extraction solution to improve extraction efficiency since previous research has documented dramatic increases in rhamnolipid sorption to soils exposed to solutions with ionic strengths greater than 10 mM. A 20 to 98% increase in rhamnolipid sorption to soils was observed when the solution ionic strength was raised from 10 to 20 mM (Torrens et al., 1998). Rhamnolipid is a weak acid that requires amendment with KOH to maintain a pH > 6.0, as explained in the methods. Thus, application of a rhamnolipid extraction solution with a concentration greater than 10 mM is less cost effective for mobilization of soil-bound metals due to the significant increase in the relative concentration of sorbed rhamnolipid.



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 1. Lead removal from the Coeur d'Alene River and Camp Navajo soils by two 10 mM diethylenetriamine pentaacetic acid (DTPA) soil washes, two 10 mM rhamnolipid soil washes, toxicity characteristic leaching procedure (TCLP) extraction, and the DTPA soil test for plant-available metals. The error bars represent the standard deviations of triplicate samples. Numbers above the bars indicate the Pb removal in mg kg-1.

 
Second Experiment
Results from the second experiment (data not shown) with four sequential extractions followed by a TCLP analysis of Pb content in the soil pellet showed a similar pattern. In this experiment, extractants included 10 mM KNO3, 10 mM rhamnolipid, and 5.3% CMCD. Lead removal rates by both rhamnolipid and CMCD were either constant or increasing over the four extractions. As in the first experiment, the relative extraction efficiencies of rhamnolipid and CMCD were different for each of the contaminated soils evaluated. Both agents removed similar amounts of Pb from the Camp Navajo soil, but rhamnolipid removed three times as much Pb from the Coeur d'Alene soil as CMCD. The final TCLP analysis of each of the soil pellets was also significant. While reductions in Pb concentration in the final TCLP extracts were observed for 25% of the samples, significant increases in TCLP-extracted Pb were observed for more than half of the soil replicates. The increases were observed for some of the soils washed with KNO3 in addition to those washed with rhamnolipid and CMCD, suggesting that the sequential washes solubilize not only the mobile fraction, but also a portion of the fractions previously characterized as immobile by the TCLP extraction.

The results from the second experiment emphasize that care must be taken in the analysis of data from sequential extractions. Metals are typically deposited in multiple layers and these layers may contain metals associated with different soil fractions having varying solubilities. Thus, dissolution of one layer by a preliminary extraction might expose a more soluble underlying layer. This nonuniformity in metal deposition, combined with the nonselectivity of extractants used in sequential extractions, is a common concern in their use for the identification of metal speciation in soil (Raksasataya et al., 1996). Van Benschoten et al. (1997) looked at seven contaminated soils and found that Pb associated with the carbonate fraction, according to the sequential extraction procedure, may include iron oxide–bound Pb as well. Due to these concerns, data from single extractions rather than sequential extractions were used in this study for the characterization of Pb species in these soils.

Third Experiment
The soil-washing solution volume was increased to 10 mL due to the low rhamnolipid extraction efficiency observed in the first two experiments. Following 10 consecutive washes, the molar ratio of total applied rhamnolipid to metal was 6.3:1 for the Coeur d'Alene soil and 2.8:1 for the Camp Navajo soil. The ratio was calculated using total Pb, Cu, and Zn and rhamnolipid-extractable Fe concentrations. Given that the molar ratio for rhamnolipid–metal complexes for Pb, Cu, and Zn is 2 (Ochoa-Loza et al., 2001), the effective chelant to metal ratios for this experiment were 3.2:1 and 1.4:1 for the Coeur d'Alene and Camp Navajo soils, respectively. The DTPA to metal ratios in the first experiment were 0.6:1 and 0.23:1 for the Coeur d'Alene and Camp Navajo soils, respectively. Cumulative results from the 10 consecutive extractions with 10 mM KNO3, 50 mM Ca(NO3)2, 10 mM rhamnolipid, and 5.3% CMCD are shown in Fig. 2 . Total Pb removal from the Coeur d'Alene soil was 14.2 and 5% for rhamnolipid and CMCD, respectively, and 15.3 and 13.4% for rhamnolipid and CMCD from the Camp Navajo soil. These extraction rates exceeded those of the KNO3 and Ca(NO3)2 ionic solutions by at least an order of magnitude, thus demonstrating the enhanced metal complexation potential of these agents as compared with electrolyte solutions. As in the previous experiments, Pb removal efficiencies were similar for rhamnolipid and CMCD in the Camp Navajo soil, but rhamnolipid efficiency exceeded CMCD in the Coeur d'Alene soil. In addition, Pb removal by rhamnolipid exceeded the combined removal of Cu, Zn, and Fe in the Coeur d'Alene soil despite the fact that their initial combined concentration far exceeded the Pb concentration. Likewise, in the Camp Navajo soil, where the total concentration of Cu, Zn, and Fe was equivalent to that of Pb, their combined removal was less than 10% that of Pb for both rhamnolipid and CMCD. Thus, these soil-washing agents are able to bind specifically with lead in the presence of significant concentrations of other metals.



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 2. Percent metal removed from Coeur d'Alene River and Camp Navajo soils following 10 sequential soil washes with 10 mM rhamnolipid, 5.3% carboxymethyl-ß-cyclodextrin (CMCD), 50 mM Ca(NO3)2, and 10 mM KNO3. The numbers in parentheses represent total metal in mg kg-1.

 
To further understand the removal patterns of these two washing agents, metal-removal rates were analyzed for each of the 10 sequential extractions in both of the soils. The removal rates were represented as percent removal per day since each of the extractions was conducted over a 24-h period. Removal of Pb and Zn by rhamnolipid from the Coeur d'Alene soil was sustained at a constant rate throughout the 10 extractions (Fig. 3) as demonstrated by linear regression (r2 = 0.99 for both metals). Copper and Fe removal rates were less consistent, but significant removal was observed throughout the 10 extractions. In the Camp Navajo soil, the Pb removal rate increased from 0.77 to 1.7% d-1 through the first three extractions and then remained constant at the highest rate for the following seven extractions (r2 = 0.99). Zinc removal was similar to Pb, while Cu removal rate decreased after the fourth extraction.



View larger version (30K):
[in this window]
[in a new window]
  		Fig. 3. Cumulative percent metal removed for ten 24-h sequential extractions using 10 mM rhamnolipid.

 
Metal removal by CMCD followed different patterns in the two soils (Fig. 4) . Removal rates from the Coeur d'Alene soil were highest during the first three extractions for Pb, Zn, and Fe and the first four extractions for Cu. Following the initial extractions, the rate decreased to a slower, but constant rate through Day 10. In contrast, a consistently increasing extraction rate was observed for Pb, Cu, and Zn removal from the Camp Navajo soil. The Fe removal rate also increased slightly with time, with the highest rates observed on Days 8 through 10. The constant or increasing extraction rates observed support the hypothesis that the metals are being solubilized with consecutive washes as suggested by the TCLP results. Thus, metal remediation using biological agents such as rhamnolipid and CMCD would require prolonged soil washing.



View larger version (30K):
[in this window]
[in a new window]
  		Fig. 4. Cumulative percent metal removed for ten 24-h sequential extractions using 5.3% carboxymethyl-ß-cyclodextrin (CMCD).

 
Fourth Experiment
In the last experiment, no increase was observed in Pb, Cu, Zn, or Fe removal from prewetted soils as compared with dry soils for either of the soil types using rhamnolipid, CMCD, or KNO3 (data not shown). In fact, rhamnolipid removed slightly more Pb and Fe from the dry Coeur d'Alene soil than from the prewetted soil. These results confirm that the steady rate of metal removal observed with repeated extractions was due to such factors as increased chelant availability and mineral dissolution rather than to soil wetting.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Neither of the biological soil-washing agents tested in the present study was found to be as effective as the synthetic chelator, DTPA, for the removal of total metal from aged, contaminated soils. Two washes with 10 mM DTPA (0.1 mmol) removed approximately three times the Pb from both soils as was removed by 10 washes with 10 mM rhamnolipid (1 mmol). These results can be explained by the higher strength of complexation between Pb and the synthetic chemical as compared with between Pb and the biological agents. Thus, low metal removal efficiencies must be anticipated from biological agents unless one is found with higher complexation constants. Such biological products do exist and include the microbially produced siderophores (Hernlem et al., 1996; Neubauer et al., 2000) and the mammalian metallothionein protein (Erk and Raspor, 1997). Products with binding constants similar to rhamnolipid could also be exploited more effectively if they were nonsorbing to soils. Any application of environmentally compatible alternatives such as rhamnolipid or CMCD would require large volumes of soil-washing solution applied using pump and treat or heap leaching strategies. Such strategies would obviously be much less cost effective and could only be justified in scenarios where the final soil health is a priority. Use of toxic chemicals such as acids to remove metals is extremely effective, but results in the production of large piles of inert material similar to mine tailings that would require extensive additional amendment before being considered viable soils.

The results of this study also demonstrate that the metal removal rates observed in this study for rhamnolipid were significantly lower than those previously documented from various soils artificially contaminated with Cd(NO3)2 and Pb(NO3)2 (Herman et al., 1995; Torrens et al., 1998; D. Gage and R. Maier, unpublished data, 1999). Thus, accurate assessment of the effectiveness of a soil-washing agent for remediation of metal-contaminated soils requires the use of aged, contaminated soils rather than artificially contaminated soils. In contaminated soils, bioavailability and mobility of metals are not only affected by soil physical properties and contamination history, but by exposure to different environmental conditions such as the fluctuating water table of the Coeur d'Alene river floodplain (Cook and Hendershot, 1996; Reed et al., 1996; Lee and Touray, 1998; Brown et al., 1999; Nikolaidis et al., 1999).

Analysis of Pb speciation in the two contaminated soils provides insights into the relative metal extraction potentials of rhamnolipid, CMCD, and DTPA in different contaminated soils. Rhamnolipid extracted 9% of the DTPA-extractable Pb from the Coeur d'Alene soil, but only 2% of the DTPA-extractable Pb from the Camp Navajo soil. This difference in relative extraction efficiency can be partially attributed to the fact that the overall metal concentration in the Camp Navajo soil is twice that of the Coeur d'Alene soil, but it is also a reflection of the differences in the properties of the soils themselves. The Pb speciation is different in the two soils and the clay content is higher in the Camp Navajo soil, which can lead to increased rhamnolipid sorption (Ochoa-Loza, 1998).

The Pb removal rates from the Coeur d'Alene soil were 14.2 and 5% for 10 washes with rhamnolipid and CMCD, respectively, and 43% for two washes with DTPA. The differences in these rates can be explained by analyzing the Pb speciation in this soil. As shown in Fig. 1, the plant-available and TCLP extractions removed similar amounts of Pb (6.7 and 6.8%, respectively). The plant-available soil test was designed to prevent excess dissolution of CaCO3 (Lindsay and Norvell, 1978) and thus extracts only the soluble and exchangeable fractions of contaminant metals. The TCLP extraction is an acetic acid digestion at pH 4.93 that dissolves the carbonates and some more soluble oxides and thus removes the metals associated with these fractions in addition to the soluble and exchangeable. The low Fe concentration in the TCLP extract of the Coeur d'Alene soil (Table 3) indicates that no significant amount of iron oxide was dissolved by the TCLP extraction. Thus, the similarity between the plant-available and TCLP-extractable Pb values indicates that they represent the soluble and exchangeable Pb fractions and that there is no significant amount of carbonate-bound Pb in this soil. In contrast, the similarity between the 10 mM DTPA Pb and Fe extraction results in Table 3 indicates that the majority of DTPA-extractable Pb in this soil is bound to Fe oxides. Because DTPA forms strong complexes with Fe (Table 1), it is likely that it is chelating the Fe and dissolving unstable Fe hydroxides. Dissolution of these Fe hydroxides causes release of bound metals, allowing chelation and removal by DTPA. The Coeur d'Alene soil was taken from the flood plain of the Coeur d'Alene River, which is saturated annually during periods of high water. Because of fluctuations in redox potential in these soils, Pb is probably bound to a range of amorphous Fe hydroxide species. Similar amounts of Pb and Fe were also extracted by both rhamnolipid (537 mg kg-1 Pb and 821 mg kg-1 Fe) and CMCD (189 mg kg-1 Pb and 284 mg kg-1 Fe) (Fig. 2), indicating that the variations in extraction efficiency observed for each of these chelators in this soil is controlled by their complexation strengths with Fe.

In the Camp Navajo soil, rhamnolipid and CMCD demonstrated a similar potential for Pb removal (15.3 and 13.4%, respectively) after 10 consecutive washes. Following two washes, DTPA removed 40% of the Pb. In the Camp Navajo soil, the plant-available (5.4%) and TCLP-extractable (55.3%) Pb values are quite different (Fig. 1). In addition, minimal Fe was removed by the TCLP, DTPA, or rhamnolipid extractions (Table 1 and Fig. 2). Thus, these data indicate that 5% of the Pb in the Camp Navajo soil is found in the soluble and exchangeable fractions and 50% is carbonate-bound (based on the difference between the plant-available and TCLP soil test results and the lack of readily extractable Fe species in this soil). This conclusion is further substantiated by the fact that one of the sources of contamination for this soil was lead-based paint. Lead carbonate and basic lead carbonate are two of the principle components of lead paint (Davis and Hotha, 1998). The higher Pb extraction efficiency observed for DTPA indicates that this synthetic chemical can scavenge carbonate-bound Pb much more effectively than the biological agents. The increasing removal rates observed for rhamnolipid and CMCD with subsequent washes can be attributed to the dissolution of carbonates by repeated washings. The Pb released into solution following the dissolution of the carbonates is then complexed by the rhamnolipid or CMCD, thus explaining the enhanced metal removal observed for the biological agents as compared with KNO3 and Ca(NO3)2.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
The different distribution of Pb in the carbonate and oxide fractions exhibited by the two soils analyzed reflects the variations in Pb speciation typically found in many contaminated soils (van Benschoten et al., 1997; Davis and Hotha, 1998). Thus, these soils proved to be good substrates for evaluating the Pb extraction potential of rhamnolipid and CMCD. Both of these agents were efficient at the removal of soluble and exchangeable Pb from the soils. Rhamnolipid, in contrast to CMCD, also had a limited potential for the removal of metals bound to amorphous iron oxides. The different efficacy of these two agents for metal removal (Fig. 2) indicates that no one ideal agent may be found for all situations and a suite of agents might be required to efficiently remediate the range of metals typically found at any one site.

Unfortunately, the low total Pb removal observed for these two environmentally compatible agents severely limits their potential application for the general remediation of metal-contaminated soils. Potential applications would be limited to situations where regulations prohibit the use of more effective but toxic agents and where extended pump and treat strategies are feasible. The only cost-effective strategy for the use of these agents would be a closed system where the metals were precipitated from the washing agents and the agents reused for continuous washing.

The research presented here clearly demonstrates the need to identify environmentally compatible agents with higher conditional stability constants than those exhibited by rhamnolipid and CMCD to remediate contaminated soils with carbonate and iron oxide–bound metals. Examples of biological materials with higher complexation constants are the mammalian metallothionein proteins (Emoto et al., 1996; Pazirandeh, 1996; and Erk and Raspor, 1997) and the bacterial siderophores (Hernlem et al., 1996; Neubauer et al., 2000) mentioned previously. Research in our lab also continues to seek novel surfactants with higher metal stability constants. Such benign alternatives to the currently available synthetic chelators will continue to be of interest for situations where the return of the soil to a healthy and productive state is of greater concern than the speed and efficiency of remediation. This research provides insights into critical parameters to be considered when evaluating such products.


   ACKNOWLEDGMENTS
 
This research was supported by Grant E5049490 from the National Institute of Environmental Health Sciences, NIH and Grant DE-FGD3-97ER62470 from the Department of Energy. We wish to thank Steve McGeehan of the Analytical Sciences Laboratory and Barbara Williams of the Idaho Water Resources Research Institute (both at the University of Idaho in Moscow, ID) for their assistance in obtaining the contaminated soil and for site characterization information from the Lower Coeur d'Alene River in Idaho. We also thank Jeneil Biosurfactant Co. (Saukville, WI) for supplying the rhamnolipid surfactant and Cerestar USA (Hammond, IN) for supplying the cyclodextrin used for this research.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 


Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2003 32: 745-750. [Full Text]  



This article has been cited by other articles:


Home page
 Environmental GeosciencesHome page
J. C. Seaman, V. M. Vulava, A. G. Sowder, B. P. Jackson, S. A. Aburime, and P. M. Bertsch
Metal extractability from contaminated SRS sediments: Comparison of column and batch results
Environmental Geosciences, December 1, 2005; 12(4): 235 - 242.
[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 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 HighWire
Right arrow Citing Articles via ISI Web of Science (11)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Neilson, J. W.
Right arrow Articles by Maier, R. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Neilson, J. W.
Right arrow Articles by Maier, R. M.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Neilson, J. W.
Right arrow Articles by Maier, R. M.
Related Collections
Right arrow Remediation
Right arrow Bioremediation and Biodegradation
Right arrow Heavy Metals
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