JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 7 June 2005
Published in J Environ Qual 34:1153-1164 (2005)
DOI: 10.2134/jeq2004.0403
© 2005 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
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 (5)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Karathanasis, A. D.
Right arrow Articles by Matocha, C. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Karathanasis, A. D.
Right arrow Articles by Matocha, C. J.
Agricola
Right arrow Articles by Karathanasis, A. D.
Right arrow Articles by Matocha, C. J.
Related Collections
Right arrow Ground Water Quality
Right arrow Mixed Wastes
Right arrow Colloid-Facilitated Transport
Right arrow Heavy Metals
Right arrow Preferential Flow

TECHNICAL REPORTS

Heavy Metals in the Environment

Biosolid Colloid-Mediated Transport of Copper, Zinc, and Lead in Waste-Amended Soils

A. D. Karathanasis*, D. M. C. Johnson and C. J. Matocha

Department of Agronomy, Univ. of Kentucky, N-122K Ag. Science-North, Lexington, KY 40546-0091

* Corresponding author (akaratha{at}uky.edu)

Received for publication October 28, 2004.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
Increasing land applications of biosolid wastes as soil amendments have raised concerns about potential toxic effects of associated metals on the environment. This study investigated the ability of biosolid colloids to transport metals associated with organic waste amendments through subsurface soil environments with leaching experiments involving undisturbed soil monoliths. Biosolid colloids were fractionated from a lime-stabilized, an aerobically digested, and a poultry manure organic waste and applied onto the monoliths at a rate of 0.7 cm/h. Eluents were monitored for Cu, Zn, Pb, and colloid concentrations over 16 to 24 pore volumes of leaching. Mass-balance calculations indicated significantly higher (up to 77 times) metal elutions in association with the biosolid colloids in both total and soluble fractions over the control treatments. Eluted metal loads varied with metal, colloid, and soil type, following the sequences Zn = Cu > Pb, and ADB > PMB > LSB colloids. Colloid and metal elution was enhanced by decreasing pH and colloid size, and increasing soil macroporosity and organic matter content. Breakthrough curves were mostly irregular, showing several maxima and minima as a result of preferential macropore flow and multiple clogging and flushing cycles. Soil- and colloid-metal sorption affinities were not reliable predictors of metal attenuation/elution loads, underscoring the dynamic nature of transport processes. The findings demonstrate the important role of biosolid colloids as contaminant carriers and the significant risk they pose, if unaccounted, for soil and ground water contamination in areas receiving heavy applications of biosolid waste amendments.

Abbreviations: ADB, aerobically digested biosolid • BTC, breakthrough curves • C/Co, effluent/influent concentration ratio • CEC, cation exchange capacity • DCB, dithionite–citrate–bicarbonate • DI, deionized water • EC, electrical conductivity • HIV, hydroxy-interlayered vermiculite • ICP, inductively coupled plasma • LSB, lime-stabilized biosolid • OM, organic matter • PMB, poultry manure biosolid • PVC, polyvinyl chloride • SAR, sodium adsorption ratio TG, thermogravimetric • XRD, X-ray diffraction


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
RECENT INCREASES in the application of organic wastes as soil amendments have raised concerns about their potential harmful effects on the environment. Although many biosolid materials are considered excellent sources of plant macronutrients, some contain high levels of heavy metals in labile forms that may pose a significant threat to soil and ground water quality (McBride et al., 1997; Scancar et al., 2001). The risk may be greater than anticipated considering that a significant metal load may be associated with dispersed biosolid colloid particles, which may be mobilized through soil macropores to lower soil depths and greater distances (McCarthy and Zachara, 1989; Kretzschmar et al., 1995; Karathanasis and Ming, 2002). Due to their high organic matter content, and increased surface area and reactivity, mobile biosolid colloids may exhibit greater affinity to sorb heavy metals than the soil matrix, thus facilitating metal transport (Karathanasis, 1999; Karathanasis and Ming, 2002). This colloid-mediated metal transport may explain the negative metal mass balances found by some researchers trying to account for the losses of sludge-applied heavy metals in soils (McGrath and Lane, 1989; Baveye et al., 1999).

Previous studies with mineral soil colloids have demonstrated that up to 3000 times higher Pb concentrations and up to 50 times more Cu and Zn could be transported through soils in the presence of colloid suspensions with varying mineralogical composition as compared with control treatments where no colloids were present (Karathanasis, 1999). These studies also demonstrated an increase in soluble metal elution in the presence of colloids, suggesting that colloids may enhance heavy metal mobility by forming soluble metal–organic complexes. Organically enriched and organic-coated colloids, such as biosolid colloids, have been shown to enhance colloid stability by acting as steric stabilizers, neutralizing positive edge sites on mineral colloids (Kaplan et al., 1993). Therefore, it is likely that organic biosolid colloids and their mixtures with soil colloids will maintain considerable stability in soil suspensions and facilitate heavy metal migration through the subsurface environment. Li and Shuman (1997) reported enhanced mobility of heavy metals associated with a poultry litter extract as a result of induced solubilization or the formation of soluble metal–organic complexes derived from the poultry litter. Lime-stabilized biosolid colloids have also been shown to significantly enhance heavy metal mobility in the soil, either by chemisorption or coprecipitation mechanisms onto the colloids or by the formation of soluble metal–organic complexes, particularly at high pH ranges (Karathanasis and Ming, 2002). Organic matter has also been shown to be an important factor of retention of many metals in soils (Adriano, 1986; Sims and Kline, 1991; Basta et al., 1993). Therefore, the high affinity that heavy metals exhibit for biosolid and biosolid–soil colloid mixtures coupled with the potential for these suspensions to remain stable and mobile over significant distances in subsurface environments provide significant risk for heavy metal contamination of soils and ground water resources.

The objectives of this study were: (i) to assess the mobility of Cu, Zn, and Pb associated with biosolid colloids through intact monoliths of three soils with diverse physical, chemical, and mineralogical properties; and (ii) to evaluate colloid and soil properties enhancing or inhibiting metal transport.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
Biosolid Samples
Bulk samples of a lime-stabilized biosolid (LSB) material in the final processing stage were obtained from a municipal wastewater treatment plant in Winchester (Clark County), Kentucky. A second aerobically digested biosolid (ADB) material (in final processing stage) was obtained from the West Hickman wastewater treatment plant in Fayette County, Kentucky. The third biosolid material consisted of poultry manure (PMB), collected from a Perdue chicken (Gallus gallus domesticus) production facility in McLean County, Kentucky. All samples were placed in sealable plastic bags to maintain their original moisture and stored under refrigeration.

Colloid Generation
Water-dispersible colloids were fractionated from bulk samples of each waste by placing 50 g of sample in a 1-L centrifuge bottle and filling with D.I. water. The slurry was mixed with a reciprocating shaker for 1 h and centrifuged at 130 x g for 3.5 min. The process was repeated twice for each 50 g of waste. The colloid particles remaining in suspension were decanted and saved as a stock solution in acid washed Nalgene carboys. The colloid mass was determined gravimetrically by placing 100-mL aliquots of each stock solution into 100-mL tared Teflon beakers and drying at 100°C for 24 h.

Soil Monolith Preparation
Undisturbed soil monoliths were collected from the upper Bt horizon of a Maury silt loam (fine, mixed, semiactive, mesic Typic Paleudalfs) and the AB horizon of a Woolper silt loam (fine, mixed, mesic Typic Argiudolls) at the Spindletop farm of the University of Kentucky. Another set of monoliths was collected from the Bw horizon of a Bruno fine sandy loam (sandy, mixed, thermic Typic Udifluvents) from Estill County, Kentucky. Representative soil subsamples from the three sites were also collected for physical, chemical, and mineralogical characterization. The soils used in the study were selected to represent diverse textural, macroporosity, and organic matter conditions. The soil monoliths were prepared by excavating and trimming soil pedestals before encasing them in polyvinyl chloride (PVC) tubes of 18 cm in diameter and 25 cm in height. The annulus between the soil and the PVC tube was sealed with expandable polyurethane foam (Poly-U-Foam) to enhance monolith stability and prevent preferential flow along the PVC walls.

Physical, Chemical, and Mineralogical Characterizations
Particle size distribution of biosolid colloid samples was determined by fractional centrifugation in the >2, 1 to 2, 0.5 to 1, 0.2 to 0.1, and <0.1 µm size range and gravimetric determinations at the following settings, respectively: 170 x g (1.5 min), 280 x g (3.0 min), 430 x g (6.0 min), 860 x g (8.5 min), and 860 x g (30 min). Electrical conductivity on a µS/cm basis and pH (in standard units) were measured on each colloid suspension with a Denver Instruments Model 250 pH*ISE*electrical conductivity meter (Arvada, CO). Organic C determination was performed with a Shimadzu TOC 5000A (Kyoto, Japan) C analyzer on approximately 100 mg of air-dried and finely ground colloid and soil samples passed through a 0.23-mm sieve. Crystalline Fe and Al oxides in colloid and soil samples were extracted with the dithionite–citrate–bicarbonate (DCB) method, and amorphous Fe–Al oxides by the ammonium oxalate method (NRCS, 1996). Extractable Fe and Al were analyzed with a Shimadzu GFA-EX7 graphite furnace atomic absorption spectrophotometer (GFAAS) (Kyoto, Japan).

Soil particle size analysis was performed on air-dried samples using the sodium hexametaphosphate [Na(PO3)6] pipette method (NRCS, 1996). Soil porosity estimates from bulk density measurements were used to calculate pore volumes based on the volume of each soil monolith (NRCS, 1996). Saturated soil hydraulic conductivity was determined via a modified constant head permeameter method (NRCS, 1996). Colloid- and soil-cation exchange properties were determined by the ammonium acetate method (NRCS, 1996). Mineralogical composition was performed by X-ray diffraction (XRD) and thermogravimetric (TG) analysis (Karathanasis and Hajek, 1982a). A set of Phillips PW 1840 diffractometer/PW 1729 X-ray generator (Mahwah, NJ) equipped with a cobalt X-ray tube (40 kV, 30 mA), and a Bragg-Bretano design goniometer were used for X-ray analysis. The scanning rate was set at 0.05° 2{theta} per minute from 2° to 40° and the scattering slit at 0.1°. A TA 2000 thermogravimetric analyzer interfaced with a 951 DuPont TG module at a heating rate of 20°C/min under N2 atmosphere was used for TG analysis. The colloid surface area was estimated from thermogravimetric water sorption experiments (Karathanasis and Hajek, 1982b).

Adsorption isotherms were also generated to evaluate the affinity of the soils and biosolid colloids for Cu, Zn, and Pb. A 5-mL colloid suspension of 200 mg/L concentration was added to 10-mL Teflon test tubes containing 0 to 5 mg/L metal concentrations. Stock metal solutions were prepared from CuCl2, ZnCl2, and PbCl2 reagents (>99% purity, Aldrich Chemicals, Milwaukee, WI). Samples were shaken on a reciprocating shaker for 24 h at room temperature and centrifuged for 1 h at (2750 x g) 3500 rpm. Supernatants were collected and analyzed for Cu, Zn, and Pb via inductively coupled plasma analysis (ICP). Freundlich isotherms fitted on log-scale by linear regression were used to describe the experimental adsorption data. All analyses, except for the mineralogical characterizations, were performed on duplicate samples.

Colloid Leaching Experiments
Eight soil monoliths were used for each soil type, to provide replicated sets for the three biosolid colloids (200 mg/L), and one for the deionized water control. Before application of the biosolid colloid suspensions and deionized water solutions onto the soil monoliths, the biosolid colloids were spiked with 5 mg/L of Cu, Zn, and Pb and equilibrated overnight to increase metal loads above inherent levels and enhance metal detectability.

The soil monoliths were immersed (saturated) in deionized water for approximately 24 h to remove air pockets and loose material and then leached with 5 L of a 1 mM CaCl2 solution as a conservative tracer before the application of the biosolid colloid suspensions. Leaching was conducted via unsaturated gravity flow at a rate of ~0.7 cm/h, using a peristaltic pump. Water and suspension loads were applied on the top of each soil monolith via a sprinkler head. Natural lower boundary conditions in the bottom of the soil monoliths were simulated by suction in the elution funnel applied with a vacuum pump. The lower boundary was controlled with a Marriott device similar to that of Ritter et al. (2004), utilizing a constant level reservoir at –10 cm suction, which was monitored with a tensiometer. Eluents were collected at 1-L intervals for a total of 21 L, corresponding to 16 to 24 pore volumes. Each sample was analyzed for EC and pH using a Denver Instruments Model 250 pH–ISE–conductivity meter. Chloride concentrations were measured using an adaptation of the automated ferricyanide method (APHA, 1998) on 60-µL samples. Colloid concentrations were determined turbidimetrically on each 1-L sample, using a Bio-Tek Instruments microplate reader (Bio-Tek Instruments, Winooski, VT) scanning at 540 nm. Metals associated with the eluted colloids were fractionated into soluble and sorbed forms. The soluble metal fraction was determined by filtering 25 mL of a colloid subsample through a 0.2-µm filter and saving an aliquot for metal analysis via Inductively Coupled Plasma (ICP) analysis. An acid extraction with 25 mL of 1 M HCl–HNO3 was used to extract the sorbed metal associated with the colloids retained on the filter. Breakthrough curves of colloid and metal (total and soluble) loads eluted from each monolith were plotted as reduced concentrations (C/Co) by pore volume for each soil.

Statistical Analysis
Duncan's multiple range test was used to compare differences in total metal elution among colloids and controls for different soil types. Single correlation and multiple stepwise regression analyses were conducted to establish relationships between soil and colloid properties and colloid-mediated metal transport. All statistical analyses were performed with a Statgraphics Plus version 5.0 software program (Manugistics, Rockville, MD) at P < 0.05 (*) and P < 0.01 (**) probability levels.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
SUMMARY AND CONCLUSIONS
 REFERENCES
 
Biosolid Colloid Properties
Results from physical, chemical, and mineralogical characterization indicate considerable variability in the biosolid colloids properties (Table 1). The ADB and PMB colloids, were dominantly organic in nature (48–60% OM), while the LSB colloids contained only ~20% organic matter. The LSB colloids were also characterized by a very high pH and CaCO3 content, as a result of the lime stabilization process, compared with the other biosolid colloids. Mineralogical characterizations suggested the presence of minor quantities of quartz, mica, kaolinite, or interstratified minerals in all biosolid colloids. Surprisingly, the mean colloid diameter of the biosolid colloids was similar, ranging from 0.38 to 0.46 µm, with 70 to 85% (on a mass basis) being >0.2 µm. Their cation exchange capacity (CEC) spanned the range of 80 to 101 cmolc/kg, being highest in the ADB and lowest in the LSB colloids. Their surface area was much greater than that of the soils used in the study, following the sequence PMB > ADB > LSB colloids (Table 1). Sodium adsorption ratios (SAR) of soils and biosolid colloids were too low to play a significant role in the stability. Nevertheless, the SAR of the ADB and PMB colloids was 5 to 17 times higher than the LSB colloids and soils, while the EC of the biosolid colloid suspensions was 30 to 340 times higher than that of the soil monolith solutions.


View this table:
[in this window]
[in a new window]
  		Table 1. Physical, chemical, and mineralogical properties of colloids and soils used in the study.{dagger}

 
Soil Properties
The Woolper and Maury soils exhibited higher porosity and hydraulic conductivity than the Bruno soil (Table 1). Better structural development and increased bioturbation in these soils have been documented to contribute to greater macroporosity in spite of their heavier texture. The Maury and Woolper soils also had similar pH values (5.6 vs. 6.1), compared with the more acidic Bruno soil (4.0). The higher pH of the Maury and Woolper soils is expected to have an effect on colloid mobility, since colloid stability is enhanced at high pH values (Degueldre et al., 1996). Of the three soils, the Woolper soil exhibited the greatest CEC, probably due to its high clay content and OM content. Woolper and Maury soils also contained considerable amounts of Fe and Al oxides. All soils had similar mineralogy, dominated by quartz, and minor quantities of hydroxy-interlayered vermiculite (HIV), kaolinite, mica, or smectite, and interstratified mica–vermiculite–smectite.

Metal Sorption Affinity by Soils and Biosolid Colloids
The Freundlich isotherms indicated varying degrees of sorption affinity for different metals by the biosolid colloids and the soils used in the study (Table 1). Contrary to what was anticipated, Kf values of soil matrices for Cu and Zn were higher than those of the biosolid colloids, except for the affinity of LSB colloids for Zn. Lead proved to be the only metal with a consistently higher affinity for the biosolid colloids than the soils. Most soil isotherms were parabolic, with 1/n values <1.0, suggesting a decreasing energy of sorption with increasing surface coverage, particularly with respect to Cu and Zn. In contrast, the majority of the colloid isotherms were linear with 1/n values near 1.0, indicating uniform surface coverage. Only the isotherms of the LSB colloids for Cu and Pb, and the ADB colloids for Pb exhibited 1/n values <1.0, suggesting variable sorption energy sites. This indicates that the fate of the heavy metals used in the study is intrinsically associated with the fate of the biosolid colloids and their dynamic interactions with the soil matrix during the transport process. With the exception of copper, LSB colloids exhibited the greatest metal sorption affinity, retaining 3 to 15 times more Pb and Zn than the ADB and PMB colloids (Table 1). This may be attributed to the high carbonate content of the LSB colloids, which has been found to significantly affect metal retention with increasing pH (King, 1988; Davis and Hotha, 1998). Statistical analysis confirms these findings, since carbonate content and pH were positively correlated (r = 0.72** and 0.81**, respectively) with the retention of these metals. Although the LSB colloids contained the least amount of organic matter among the biosolid colloids, the pH apparently was high enough to effectively overcome organic C dissociation effects. This is supported by the negative correlation (r = –0.59*) between organic matter and the retention of Zn and Pb by the biosolid colloids. Increased sorption of Pb and Zn by the LSB colloids may be due to extensive hydrolysis of heavy metals at alkaline pH values (Leeper, 1970), and potential coprecipitation with carbonate colloids (de Matos et al., 2001).

The ADB and PMB colloids exhibited significantly lower metal sorption capacity than the LSB colloids for Zn and Pb. Aerobically digested biosolid colloids had a greater affinity for Pb than the PMB colloids, but PMB colloids had a greater affinity for Zn (Table 1). The greater affinity of the ADB colloids for Pb is mainly associated with their high organic matter content and the significant covalency of the Pb–carboxylic or Pb–phenolic group bond (McBride, 1989). The slightly higher Zn sorption affinity exhibited by the PMB colloids was probably due to their higher surface area, although it was not as high as that reported by Li and Shuman (1997) for organic materials derived from poultry litter.

Surprisingly, the biosolid colloids showed a significantly lower Cu sorption capacity than the soils, in spite of their higher OM content (Table 1). It is likely, that Cu sorption on biosolid colloid surfaces was depressed by the formation of complexes with dissolved organic ligands (Zhou and Wong, 2001). Christl and Kretzschmar (2001) reported that at pH > 6 Cu sorption was reduced significantly due to increasing concentrations of dissolved metal–organic complexes that competed for surface sites. Dissolved amines and amino acids from alkaline stabilized biosolids are particularly effective in complexing Cu and keeping it in solution, thus explaining the high solubility of Cu relative to Pb and Zn in the pH range 8 to 12 (McBride, 1998). This may explain the slightly greater affinity of the PMB colloids for Cu (pH 5.5) compared with the LSB and ADB colloids (pH 11.3 and 7.2, respectively).

Among the three soils, Woolper exhibited the highest overall metal sorption affinity, while Maury and Bruno showed lower affinities. These trends may be the result of the higher organic matter, pH, and Fe oxide content of the Woolper soil, as indicated by the positive respective correlations (r = 0.86*, 0.77*, and 0.64*). Surprisingly, with the exception of Pb, the Woolper soil exhibited higher metal affinity than the ADB and PMB colloids in spite of their increased levels of organic matter and surface charge properties. Apparently, the type of organic sites or other components present in the Woolper soil out-competed the organic colloid sites for metal sorption (Jin et al., 1996). Consequently, this may lead to decreased metal mobility associated with these colloids through the Woolper soil. There were no significant differences in metal retention between the Maury and Bruno soils even though the Bruno soil contained five times more organic matter than the Maury soil. Apparently, the effect of OM was compensated by organic site quality and higher pH and Fe–Al oxides in the Maury soil. The metal sorption capacity in the Bruno soil may have also been influenced by its low pH and surface area (r = –0.56*, and 0.55*, respectively). These findings suggest that the universal assumption of colloids having greater affinity for metals than the soil matrix due to increased surface area and charge properties may not always be true. Therefore, the transport of certain metals may be enhanced by the biosolid colloids in some soils, but in others the transport may be impeded because of higher metal sorption affinity by the soil matrix.

Elution of Biosolid Colloids
Colloid breakthrough curves (BTC) through the three soils were generally irregular, ranging from 0 to 100% recoveries (Fig. 1) . The colloids showed a slower breakthrough than the conservative Cl– tracer, indicating considerable physical and chemical interaction in their flow path through the soil monoliths. This interaction is materialized through attachment–detachment processes with the soil matrix (Jacobsen et al., 1997), as well as sieving and physical entrapment of the colloids by the soil matrix. The transport of biosolid colloids in these experiments was generally erratic, exhibiting several maxima and minima depending on colloid and soil type. These results are consistent with the irregular colloid breakthrough reported by Jacobsen et al. (1997), indicating that colloids of like charge tend to be transported or deposited in clusters, rather than uniformly, as a result of macropore blockage followed by flushing due to water pressure buildup in the blocked pores. Evidence for preferential flow through soil macropores was provided by: (i) the early initial breakthrough of the conservative tracer before one pore volume was passed through the soil monoliths, and (ii) the nearly 4 to 5 pore volumes needed to reach C/Co = 1, suggesting considerable undersaturation of the soil matrix with the tracer during the first few pore volumes of leaching (Bouma, 1991).



View larger version (29K):
[in this window]
[in a new window]
  		Fig. 1. Breakthrough curves for a conservative tracer (Cl–) and LSB, ADB, and PMB colloids eluted from intact (a) Maury, (b) Woolper, and (c) Bruno soil monoliths (C, effluent concentration; Co, influent concentration; ADB, aerobically digested biosolid; LSB, lime-stabilized biosolid; PMB, poultry manure biosolid).

 
The ADB colloids proved to be the most mobile, particularly through the Maury and Bruno soils, with recoveries of 49 and 53%, respectively (Table 2). These colloids showed a rapid but irregular breakthrough after 3 to 4 pore volumes, reaching maxima of 1.0 C/Co at 9 and 16 (Maury) and 11 (Bruno) pore volumes, respectively (Fig. 1a and 1c). The BTCs through the Maury soil were particularly erratic, with minima of nearly 0.0 C/Co at 10 and between 17 and 21 pore volumes of leaching (Fig. 1a). This is indicative of preferential flow patterns with alternating macropore clogging and flushing cycles dominated by colloid straining and detachment events (Jacobsen et al., 1997). The erratic elution pattern through the Maury soil was associated with pH (6.5–4.0) and EC (6–21 µS/cm) fluctuations of the colloid suspensions, particularly during the first 10 pore volumes of leaching. The mobility of the ADB colloids through the Bruno soil may have been enhanced by the high negative surface charge of the colloids and the relatively high soil OM content (Table 1), which caused increased electrostatic repulsion with the soil matrix. The slow breakthrough of the ADB colloids through the Bruno soil during the first three to four pore volumes is probably the result of a corresponding pH drop from 6.0 to 4.0 in eluted colloid suspensions. A subsequent pH stabilization in the 6.5 to 7.0 range increased steadily colloid mobility (Fig. 1c). Contrary to the Maury and Bruno soils, the elution of ADB colloids through the Woolper soil provided <10% recoveries, peaking gradually to 0.16 C/Co after 14 pore volumes (Fig. 1b). The relative immobility of these colloids through the Woolper soil cannot be attributed to pH or EC effects, which were maintained relatively uniform throughout the leaching cycle. Apparently, physical straining processes overshadowed the influence of chemical factors in this experiment.


View this table:
[in this window]
[in a new window]
  		Table 2. Mass balance for colloid and metal recovery from the leaching experiments (applied mass for each colloid = 4200 mg; applied mass for each metal = 105 mg).

 
The elution of PMB colloids through the Woolper soil also provided significant recoveries (~27%), with an initial breakthrough after three pore volumes that steadily increased to a maximum of 0.45 C/Co at the 12th pore volume before tapering off to about 0.3 C/Co at the end of the leaching cycle (Table 2, Fig. 1b). In spite of the sizeable pH decrease in eluent suspensions after four pore volumes from about 8.0 to <5.0, there was no negative effect on colloid elution, suggesting that electrostatic repulsion between the small size colloids was adequate to maintain their mobility. The PMB colloids exhibited very low mobility through the Maury and Bruno soils, with total recoveries <10%, and maxima between 0.15 and 0.33 C/Co (Table 2, Fig. 1a and 1c). It is possible that the smaller size of the PMB colloids enhanced physical attachment and entrapment (DeGueldre et al., 1996) and their lower pH inhibited their mobility, particularly in the soils with the lowest pH and hydraulic conductivity (Table 1).

The mobility of LSB colloids was the lowest among the biosolid colloids studied, with total recoveries below 11% and maxima never exceeding 0.3 C/Co in any soil monolith (Fig. 1). The relative immobility of the LSB compared with other colloids may be due to carbonate dissolution during the leaching cycle, since the pH of the eluted suspensions dropped from 11.3 (initial colloid pH) to <7.0. It is likely, that the increased levels of the released Ca in solution induced colloid coagulation and reduced their mobility. The higher ionic strength of the eluted LSB suspensions (35–52 µS/cm) compared with other colloids supports the potential negative influence of Ca as an inhibitor of colloid mobility (Degueldre et al., 1996).

Metal Elution in Association with Biosolid Colloids
Figures 2 to 10 Go Go Go Go Go Go Go Go depict average metal elution of replicated breakthrough curves (BTCs) in the absence (control) and presence of biosolid colloids. Variability between replicated BTCs ranged from 3 to 17%, with an average value of about 8%. Metal elution in the presence of colloids was plotted separately for the soluble metal fraction and the sum of the colloid-bound (sorbed) and soluble fraction, which comprised the total metal fraction. Mass-balance for metal forms recovered during the leaching experiments are shown in Table 2. The presence of colloids generally enhanced the elution of total, sorbed, and in some cases even the soluble metal fraction, exhibiting high correlations with colloid elution (r = 0.61**–0.91**).



View larger version (29K):
[in this window]
[in a new window]
  		Fig. 2. Breakthrough curves for Cu eluted in the absence (control) and presence of biosolid colloids from intact Maury soil monoliths (C, effluent concentration; Co, influent concentration; ADB, aerobically digested biosolid; LSB, lime-stabilized biosolid; PMB, poultry manure biosolid).

 


View larger version (27K):
[in this window]
[in a new window]
  		Fig. 3. Breakthrough curves for Cu eluted in the absence (control) and presence of biosolid colloids from intact Woolper soil monoliths (C, effluent concentration; Co, influent concentration; ADB, aerobically digested biosolid; LSB, lime-stabilized biosolid; PMB, poultry manure biosolid).

 


View larger version (28K):
[in this window]
[in a new window]
  		Fig. 4. Breakthrough curves for Cu eluted in the absence (control) and presence of biosolid colloids from intact Bruno soil monoliths (C, effluent concentration; Co, influent concentration; ADB, aerobically digested biosolid; LSB, lime-stabilized biosolid; PMB, poultry manure biosolid).

 


View larger version (29K):
[in this window]
[in a new window]
  		Fig. 5. Breakthrough curves for Zn eluted in the absence (control) and presence of biosolid colloids from intact Maury soil monoliths (C, effluent concentration; Co, influent concentration; ADB, aerobically digested biosolid; LSB, lime-stabilized biosolid; PMB, poultry manure biosolid).

 


View larger version (29K):
[in this window]
[in a new window]
  		Fig. 6. Breakthrough curves for Zn eluted in the absence (control) and presence of biosolid colloids from intact Woolper soil monoliths (C, effluent concentration; Co, influent concentration; ADB, aerobically digested biosolid; LSB, lime-stabilized biosolid; PMB, poultry manure biosolid).

 


View larger version (28K):
[in this window]
[in a new window]
  		Fig. 7. Breakthrough curves for Zn eluted in the absence (control) and presence of biosolid colloids from intact Bruno soil monoliths (C, effluent concentration; Co, influent concentration; ADB, aerobically digested biosolid; LSB, lime-stabilized biosolid; PMB, poultry manure biosolid).

 


View larger version (30K):
[in this window]
[in a new window]
  		Fig. 8. Breakthrough curves for Pb eluted in the absence (control) and presence of biosolid colloids from intact Maury soil monoliths (C, effluent concentration; Co, influent concentration; ADB, aerobically digested biosolid; LSB, lime-stabilized biosolid; PMB, poultry manure biosolid).

 


View larger version (28K):
[in this window]
[in a new window]
  		Fig. 9. Breakthrough curves for Pb eluted in the absence (control) and presence of biosolid colloids from intact Woolper soil monoliths (C, effluent concentration; Co, influent concentration; ADB, aerobically digested biosolid; LSB, lime-stabilized biosolid; PMB, poultry manure biosolid).

 


View larger version (27K):
[in this window]
[in a new window]
  		Fig. 10. Breakthrough curves for Pb eluted in the absence (control) and presence of biosolid colloids from intact Bruno soil monoliths (C, effluent concentration; Co, influent concentration; ADB, aerobically digested biosolid; LSB, lime-stabilized biosolid; PMB, poultry manure biosolid).

 
Copper Elution
Total Cu elution was highest in association with the ADB colloids through the Maury and Bruno soils and the PMB colloids through the Woolper soil, with recoveries ranging from 53 to 65% (Table 2, Fig. 2–4). Elutions through the Woolper and Bruno soil monoliths were associated with fairly symmetrical BTCs that reached maxima of ~0.80 C/Co, while the Maury BTCs were erratic, exhibiting several maxima (up to 1.4 C/Co) and minima (near 0). In all cases, Cu elution correlated well with colloid elution (r = 0.88**), corroborating their association in the transport process. The >1.0 C/Co observed in the Maury BTC is probably the result of detachment and flushing of previously strained colloids carrying an extra Cu load in the eluent. This pattern is typical in soils with considerable preferential transport and/or numerous clogging and flushing cycles due to water pressure buildup (Jacobsen et al., 1997). The high organic matter and fairly low suspension pH associated with the eluted colloids (Table 1) accounted for 56% of the total variability in total Cu elution as indicated by multiple regression analysis. The higher colloid surface charge and surface area may have also contributed to greater total Cu elution, showing correlation coefficients (r) of 0.50* and 0.55*, respectively. In addition, the low ionic strength of the colloid suspensions, particularly of the ADB and PMB colloids, would tend to promote double layer expansion and limit colloid attachment to the soil matrices. Soluble Cu accounted for 15 to 45% of the total Cu eluted with ADB and PMB colloids through the above soils, showing a correlation of r = 0.61** with colloid elution. This high elution of the soluble Cu fraction is attributed to the fairly low affinities of these colloids for Cu (Table 1) and the high organic matter status of colloids and soils, which induced the formation of soluble organic–Cu complexes (Han and Thompson, 1999; Li and Shuman, 1997). This result may have been accentuated by the higher molecular weight and hydrophilic nature of the organic complexes associated with the ADB colloids, which significantly affects Cu binding and enhances their mobility (Han and Thompson, 1999). The formation of similar organic complexes in the Woolper soil, combined with its high hydraulic conductivity also facilitated soluble Cu elution. These organo–metal complexes are fairly nonadsorptive within the soil matrix and may explain the sizeable soluble metal fraction mobilized in this study (McBride et al., 1997).

Total Cu elution associated with the LSB colloids through the Woolper and Bruno soils, the PMB colloids through the Maury soil, and the ADB colloids through the Woolper soil was moderate (14–22%), with maxima in the range of 0.2 and 0.5 C/Co (Table 2, Fig. 2–4). These trends were consistent with lower colloid elution and greater Cu attenuation by the soil matrices. Generally, total Cu elution associated with the above colloids increased gradually toward the end of the leaching stage, suggesting that metal elution increased as sorption sites in the soil matrix became more saturated or as a result of colloid–metal flushing cycles. This explanation is consistent with the higher affinity for Cu exhibited by the soils than the colloid suspensions (Table 1). The remaining colloids (LSB through Maury and PMB through Bruno) exhibited little potential for Cu cotransport (Table 2, Fig. 2 and 4), with Cu recoveries <3% and elution maxima below 0.05 C/Co. The relative immobility of these colloids through the soils inhibited Cu cotransport and resulted in nearly complete attenuation by the soil matrices. The small amounts of soluble Cu eluted in association with the LSB colloids may be attributed to their high pH (r = –0.56*) and their lower organic matter, surface charge, and surface area contents (Table 1). Soluble Cu associated with the control treatments (absence of colloids) was negligible through all soils, suggesting complete attenuation by the soil matrices.

Zinc Elution
Breakthrough curves of total Zn eluted in the presence of biosolid colloids (Fig. 5–7) showed only subtle differences compared with Cu BTCs, showing a strong correlation to colloid elution patterns (r = 0.88**). The ADB colloids through the Maury and Bruno soils and the PMB colloids through the Woolper soil appeared to be the most effective Zn carriers, showing recoveries of 42 to73% (Table 2, Fig. 5–7). As was the case with Cu, the Woolper and Bruno BTCs were relatively symmetrical with C/Co maxima of 0.80–0.90 being reached after about 6 pore volumes of leaching (Fig. 6 and 7). The Maury BTC was fairly irregular, with two elution C/Co maxima > 1.0, apparently due to detachment and flushing of Cu-carrying colloids previously retained within the soil monolith matrix (Fig. 5). Although the PMB colloids in this study did not show high affinity for Zn (Table 1), the presence of PMB materials has been reported to greatly enhance Zn mobility in the subsurface environment (Li and Shuman, 1997). Organic matter, surface area, CEC, and carbonate colloid content in conjunction with the pH of ADB and PMB colloid suspensions were the most influential variables, accounting for 70% of the variation in total Zn elution.

Total Zn elution in association with PMB colloids through the Maury soil, the LSB and ADB colloids through the Woolper soil, and the LSB colloids through the Bruno soil was moderate (Table 2), with total metal recoveries in the range of 12-25%, and C/Co maxima ranging between 0.25 and 0.60 (Fig. 5–7). Decreased Zn elution associated with these colloids is related to lower colloid elution, which was influenced by the presence of carbonates and the decreased organic matter and surface charge of the LSB colloids (Table 1). The remaining colloids (LSB through Maury, ADB through Woolper, and PMB through Bruno) were less effective Zn carriers, with total metal recoveries < 7% and elution maxima below 0.25 C/Co (Table 2, Fig. 5–7). Significant amounts of soluble Zn were eluted only in the presence of PMB colloids through the Woolper soil and the ADB colloids through the Bruno soil, averaging recoveries of 20-35%, and amounting to about 50% of the total Zn elution. The above trend is attributed to the high organic matter of these colloids and soils, which induces soluble metal-organic complexation. Soluble Zn elution in association with other colloids and soils was limited to < 0.1 C/Co, being similar to that of the control (no colloids) treatments.

Lead Elution
As was the case with Cu and Zn, total Pb breakthrough was highest for the PMB colloids through the Woolper soil and the ADB colloids through the Bruno soil, with total recoveries in the range of 46 to 50% (Table 2, Fig. 8–10). Total Pb eluted in association with PMB colloids through the Woolper soil peaked between the 6th and the 14th pore volume at about 0.75 C/Co, when it dropped to about 0.40 before resurging to 0.60 at the end of the leaching cycle (Fig. 9). Total Pb eluted with the ADB colloids through the Bruno soil gradually increased throughout the leaching cycle, culminating at ~0.80 C/Co at the end of the leaching cycle (Fig. 10). Total Pb elution in association with ADB colloids through the Maury soil was also significant, showing a recovery of 33% and erratic BTCs closely following colloid elution patterns, as it was the case for Cu and Zn (Table 2, Fig. 8). The observed maxima of about 1.6 C/Co at the 9th and the 16th pore volume in this BTC are attributed to detachment and flushing of initially strained colloids within the soil matrix. The increased Pb mobility in association with the above colloids is consistent with their high organic matter content and relatively high affinity for Pb (Table 1) (Amrhein et al., 1993; Denaix et al., 2001). The strong affinity of the colloid surfaces for Pb is related to the low pKa (~5.0) of organic colloid–OH groups, and the significant covalency of the Pb–carboxylic or Pb–phenolic groups (Logan et al., 1997; McBride, 1989).

The PMB colloids through the Maury soil and the LSB colloids through the Woolper and Bruno soils also exhibited irregular but moderate Pb co-transport, with recoveries ranging from 12 to 19% and maxima not exceeding 0.60 C/Co. The moderate transport of these colloids may have been affected either by lower colloid mobility, or adverse colloid and soil properties that compromised their high Pb sorption affinity (Table 1). The remaining colloids did not exhibit any meaningful Pb transportability. The soluble Pb fraction transported in the presence of biosolid colloids was much smaller than the soluble Cu or Zn fraction, being similar to or at times even lower than that in control treatments. This is corroborated by the contrasting correlation coefficients of colloid elution with sorbed Pb (r = 0.87**) vs. soluble Pb (r = 0.34).

Effect of Colloid and Soil Type on Metal Elution
Colloid-mediated metal transport was variable between colloid and soil types. For the most part, increased OM (r = 0.56*), colloid surface area (r = 0. 55*), and surface charge (r = 0.50*), as well as decreased eluent pH (r = –0.56*) facilitated metal transport, while larger colloid size (r = 0.41*) and the presence of carbonates (r = –0.39*) were inhibiting. High levels of OM increased surface charge and promoted colloidal stability, while contributing significant amounts of soluble organic ligands that enhanced metal co-transport. Increased colloid surface area promoted co-transport by providing physical bonding sites for metal sorption, while relatively low elution pH levels enhanced metal solubility and promoted colloidal stability by inducing hydrolysis of functional groups on the organic colloids (Stumm, 1992). Carbonates derived from the LSB colloids, and to a lesser degree, ADB colloids negatively affected metal elution, since colloid and metal elution were adversely affected by increases in eluent pH (Scancar et al., 2001). No significant effects on metal elution by Fe- and Al-hydroxides were observed in the range encountered in this study. The fact that metal elution generally increased at lower pH levels suggests that increased metal solubility and organic complexation processes overcame the surface charge deficit experienced by the biosolid colloid as a result of the pH drop. Further, this pH drop may not have been significant enough for the organic matter content of the biosolid colloids to limit their role in metal-facilitated transport. Since eluent suspension pH levels were generally higher than the range where metal solubility is greatly enhanced, increases in soluble metal elution may have been also caused by hydrolysis of metal cations previously retained by the soil matrices (Martinez and Motto, 2000). Overall, the soil matrix consistently retained a much greater metal load than colloids moving through it, particularly for Cu and Zn for which it generally showed higher affinity than the colloids. Therefore, a large number of metal sorption sites on the soil matrix are exposed to interactions with migrating colloids, including metal desorption that could significantly contribute to increased soluble and sorbed metal fluxes through soil macropores.

Soil type also had a significant effect on colloid-mediated metal mobility. Although there was considerable metal mobility associated with biosolid colloids, the soils consistently attenuated considerable amounts of the metals applied. The Woolper soil monoliths induced considerable metal mobility mainly due to greater hydraulic conductivity and organic matter content, which promoted formation of soluble metal–organic complexes (Karathanasis, 1999). The Maury soil in spite of its high bulk density exhibited extensive macroporosity that facilitated moderate colloid-bound metal transport (Karathanasis and Ming, 2002). Metal BTCs through the Maury soil exhibited several maxima and minima, suggesting macropore blockage and flushing cycles (Jacobsen et al., 1997). Metal elution through the Bruno soil was only half that of the other two soils, apparently due to limited preferential flow. This contradicts findings by Gove et al. (2001), which reported that sandy soils with relatively low organic matter content (<5%) may pose worst-case scenarios in heavy metal leaching. Even though the Bruno soil had nearly as much organic matter as the Woolper soil, formation of soluble metal–organic complexes did not play as important a role in metal transport processes. The low soluble metal elution through this soil may be due to greater interaction with the soil matrix (more uniform porosity, lack of channels) that may increase solute residence time.


   SUMMARY AND CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
REFERENCES
 
The results of this study clearly demonstrate the potential of colloids derived from biosolid wastes to co-transport heavy metals in subsurface soil environments. Based on mass balance calculations, metal elution in the presence of biosolid colloids was enhanced up to 77 times over that of control treatments, depending on colloid and soil type. Metal elution potential through the three soils followed the order Zn = Cu > Pb. Elution potential of metals by the biosolid colloids followed the sequence ADB > PMB > LSB, with the ADB colloids transporting nearly twice as much metal load as the PMB colloids and more than four times the load of LSB colloids. Although the majority of the eluted metal load was colloid-bound, about 39% of Cu, 34% of Zn, and 4% of Pb was soluble. Colloid and metal mobility, was enhanced by decreases in solution pH and colloid size, and increases in OM, which resulted in higher elution of sorbed and soluble metal loads through metal–organic complex formation. Metal sorption affinity differences between soils and biosolid colloids were inadequate in predicting metal elution potential, emphasizing the flaws of batch experiments alone in assessing the dynamic nature of contaminant transport processes. The findings also explain the important role of mobile biosolid colloids in explaining metal mass balance losses and the necessity for considering their contribution as transport agents in contaminant transport modeling approaches. This is particularly critical in areas where heavy applications of biosolids as soil amendments may increase the risk for soil and ground water contamination.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
Technical article no. 04-06-159.


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


Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2005 34: 0. [Full Text]  




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
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 (5)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Karathanasis, A. D.
Right arrow Articles by Matocha, C. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Karathanasis, A. D.
Right arrow Articles by Matocha, C. J.
Agricola
Right arrow Articles by Karathanasis, A. D.
Right arrow Articles by Matocha, C. J.
Related Collections
Right arrow Ground Water Quality
Right arrow Mixed Wastes
Right arrow Colloid-Facilitated Transport
Right arrow Heavy Metals
Right arrow Preferen