| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Earth and Environ. Sci., New Mexico Inst. of Mining and Technology, Socorro, NM 87801 (D.S. Oliver, current address: MWH, 10619 South Jordan Gateway, Salt Lake City, UT 84095)
b Pacific Northwest National Lab., Richland, WA 99352
c Dep. of Biology, New Mexico Inst. of Mining and Technology, Socorro, NM 87801
* Corresponding author (tkieft{at}nmt.edu)
Received for publication November 21, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Abbreviations: ANOVA, analysis of variance • BTC, breakthrough curve • XANES, x-ray absorption near edge structure • XRF, x-ray fluorescence
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Within the range of pH and Eh encountered in most natural waters, chromium occurs as Cr(III) and Cr(VI) (Richard and Bourg, 1991). The trivalent form is far less toxic than the hexavalent form, and is far less mobile in ground water, being found either as cationic species that sorb to solids or as relatively insoluble precipitates such as Cr(OH)3 (Rai et al., 1987). The hexavalent form typically occurs as anionic species such as chromate (CrO2-4) that tend to be mobile in ground water.
To remediate Cr-contaminated ground water and soil in situ, Cr(VI) can be reduced to Cr(III) (Palmer and Puls, 1994; Lovley, 1995; James, 1996). Many different bacteria have been shown to reduce Cr(VI) to Cr(III) under aerobic conditions (Bopp and Ehrlich, 1988; Ishibashi et al., 1990; Shen and Wang, 1993; Gopalan and Veeramani, 1994; Campos et al., 1995; Wang and Xiao, 1995; Chirwa and Wang, 1997; Garbisu et al., 1998) or anaerobic conditions (Wang et al., 1989; Shen and Wang, 1994; Chen and Hao, 1996; Turick et al., 1998). Bacteria used in these studies generally were isolated from sewage sludge or Cr-contaminated soil and water. Most studies of microbial Cr(VI) reduction have involved pure cultures, although a few have examined Cr(VI) reduction by native communities in soil and aquifer materials (Cifuentes et al., 1996; Shen et al., 1996; Bader et al., 1999; Turick et al., 1998). Cifuentes et al. (1996), Bader et al. (1999), and Turick et al. (1998) examined microbial reduction of Cr(VI) in surface soil samples by indigenous microorganisms with and without organic amendments. However, little research has been concerned specifically with microbial reduction of Cr(VI) in vadose zone materials.
The primary goal of this study was to assess the effect of nutrient addition on microbial reduction of Cr(VI) to Cr(III) using an indigenous microbial community under hydrologic conditions similar to those found in the unsaturated vadose zone. Batch microcosm experiments were performed to assess whether native populations were capable of Cr(VI) reduction and to evaluate treatments for enhancing Cr(VI) reduction. Treatments included addition of organic C (molasses) and/or nitrate (NO-3). Based on the results of the batch experiments, laboratory column experiments were used to assess Cr(VI) reduction under unsaturated flow conditions.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The sediment was collected beneath the root zone at a depth interval of 1 to 2 m below the ground surface. Overlying soil was removed with a backhoe. Samples were collected with flame-sterilized shovels. Samples were pooled in the field and sealed in 19-L (5-gallon) buckets that had been sterilized with a 1% NaOCl solution (10% bleach) and rinsed with sterile deionized water. Samples were stored for up to 2 yr at 5°C. Before use in the experiments, the sediment was homogenized using a sample splitter. The field gravimetric soil water content of the samples was 3.3%; drying was minimized during handling to sustain the bacteria and to minimize particle sorting by grain size during column packing. The sediment was primarily medium to coarse sand with minor amounts of silt (1–2%) and clay (5–6%) and traces of gravel; the gravel was removed before the experiments. The sand fraction was primarily quartz but also contained feldspar, volcanic glass, and minor amounts of mafic minerals including biotite and magnetite. The pH of the sediment was 8.04 (saturated paste). The total background concentration of chromium [Cr(III) and Cr(VI)] in the sediment, determined using x-ray fluorescence (XRF) spectroscopy, was 36 g kg-1. Of this total Cr, <5 g kg-1 was Cr(VI), as determined using x-ray absorption near edge structure (XANES) spectroscopy.
Synthetic Pore Water
The synthetic pore water composition was based on water collected from Well 600-S3-25 at the Hanford Site (Kaplan et al., 2000). The synthetic pore water was comprised of 1.04 mM NaHCO3, 349 µM MgSO4·7H2O, 251 µM MgCl2·6H2O, 1.70 mM CaSO4·2H2O, and 188 µM KCl (pH 8.4). Chromium(VI) was added to the synthetic pore water in various amounts in the form of K2CrO4. Potassium nitrate (KNO3) and molasses (Grandma's Molasses, Mott's USA, Stamford, CT, or Brer Rabbit, Nabisco, East Hanover, NJ) were added to the synthetic pore water as sources of NO-3 and organic C. Nitrate was selected because it is a common co-contaminant with Cr(VI) at DOE sites, it is present in high concentrations (10 000–100 000 mg L-1, 1–10% wt/vol) in contaminant plumes at the Hanford Site (Riley and Zachara, 1992), it provides N for microbial growth, and it may be used as an electron acceptor in dissimilatory NO-3 reduction. Molasses was selected because it is an inexpensive source of organic C suitable for remediation schemes (Turick et al., 1997). All solutions were filter-sterilized (0.22-µm pore-size) before use.
Batch Microcosm Experiments
The batch microcosm experiment consisted of nine treatments, each in triplicate, representing all combinations of control, low, and high concentrations of NO-3 (0, 346, and 34 600 mg L-1) and molasses (0, 200, and 2000 mg L-1; Brer Rabbit). Sediment samples (225 g each, 3.3% gravimetric water content) were spread onto trays and sprayed with 22.5 mL of the appropriate solutions containing nutrients and 67 mg L-1 Cr(VI) until gravimetric water contents of approximately 13% were achieved. The sediment was then placed into sterilized, sealed glass bottles and incubated statically in the dark at 23 to 25°C for 35 d. Following incubation, samples were centrifuged in Maxi-Spin nylon centrifuge filters with a 0.45-µm pore size (Alltech Associates, Deerfield, IL) to extract water from the sediment for chemical analysis.
A separate batch microcosm experiment was conducted with sediment (containing its native microbial community) and abiotic controls to assess the relative contribution of microbial reduction and abiotic reduction on the loss of Cr(VI) from solution. Sterile synthetic pore water [2 mL containing 250 mg Cr(VI) L-1] was added to sediment samples of 20.6 g in sterile glass vials to achieve gravimetric water contents of 13.3%. Abiotic controls consisted of (i) an autoclaved set (autoclaved at 121°C for 20 min on 3 successive days) and (ii) a HgCl2–poisoned set (200 mg Hg L-1). A second set of samples was prepared in which 2000 mg L-1 molasses (Grandma's Molasses) was added to the Cr(VI) solution. Samples were prepared in duplicate. The samples were incubated statically at 23 to 25°C in the dark for 35 d followed by pore water extraction as described above.
Column Experiments
Plexiglas columns (15 cm long by 5 cm i.d.) from Soil Measurement Systems (Tucson, AZ) were used in these experiments. Each column had a nylon mesh screen at the top (inlet) of the column and a nylon filter membrane, with 1.2-µm pore size and bubbling (air entry) pressure of 60 kPa, located at the bottom (outlet) of the column. The nylon screen and membranes were held in place by perforated aluminum plates. Tensiometers with 1-cm diameter ceramic porous cups in contact with the sediment were installed through the column wall 4 cm from each end. Before packing, the empty columns were sterilized with a 1% NaOCl solution and rinsed with filter-sterilized, deionized water. The sediment was packed into the columns to a bulk density of approximately 1.5 g cm-3.
The fluid delivery system was comprised of a Harvard Apparatus PHD 2000 programmable syringe pump (Holliston, MA), Tygon tubing (formulation B-44-4X), 5-mL sterile plastic syringes, and plastic check valves. A drip chamber, which provided free fall of feed solution, was installed immediately above each column to reduce the likelihood of microbial contamination of the reservoirs. The fluid delivery system was sterilized with 1% NaOCl and rinsed with filter-sterilized deionized water before the experiments. The column outlets were connected to vacuum chambers to maintain unsaturated conditions. Chamber vacuums were regulated with Moore model 43 subatmospheric pressure regulators (Spring House, PA) connected to vacuum and pressure sources. Effluent samples were collected in 20-mL glass vials with ISCO Retriever II fraction collectors (Lincoln, NE) situated in the vacuum chambers. Vials in the fraction collectors were removed and weighed periodically to determine flow rates. Sample evaporation within the vacuum chambers was determined by monitoring water loss from test vials. Columns were weighed daily to monitor gravimetric water contents. Soil water tensions were measured daily at the tensiometers to monitor hydraulic gradients.
The columns were initially leached for 5 d with synthetic pore water, during which time they attained unit hydraulic gradients (indicated by tensiometer measurements), near-steady discharge rates, and near-constant water contents. The water input rate to each column was approximately 50 mL d-1. Following the equilibration period, the feed solutions were spiked with tritiated water (740 Bq mL-1) to determine hydraulic properties of each column, and collection of effluent fractions commenced. Following this initial tracer test, the inlet solutions were switched to solutions of synthetic pore water containing Cr(VI) and nutrient additives (Table 1). The experiment ran for 45 d. Treatment columns received solutions of synthetic pore water that contained approximately 65 mg Cr(VI) L-1 and additions of organic C (Grandma's Molasses) and/or NO-3. Control columns received only Cr(VI) solution. Each column treatment was run in duplicate. The temperature throughout the column experiments was 23 to 25°C. Effluent samples were analyzed for total Cr concentrations, which were then corrected for evaporation. A second tritium tracer test was performed after 38 d of Cr(VI)/nutrient addition to assess changes in hydraulic properties. Immediately following the column experiment, sediment from the top (0–5 cm), middle (5–10 cm), and bottom (10–15 cm) of each column was collected into separate Ziploc bags and homogenized.
|
Sediment from the columns was analyzed for total Cr using x-ray fluorescence (XRF) spectroscopy. Samples (unsieved and dry-sieved fractions) were ground to a fine powder with an aluminum oxide mortar and pestle, pressed into a self-supporting disc, and analyzed in triplicate on 2 successive days. The six measurements were averaged. Sieved fractions (in mm) were <0.125, 0.125 to 0.25, 0.25 to 0.5, 0.5 to 1.0, and >1.0. To determine solid phase Cr speciation, air-dried sediment was analyzed with x-ray absorption near edge structure (XANES) spectroscopy at the Advanced Photon Source at Argonne National Laboratory. Scans were made from 20 eV below the Cr K-edge (5989 eV) to 100 eV above the edge at a resolution of 0.5 eV at ±20 eV of the edge and at 1 eV from 20 to 100 eV above the edge. Fluorescence data were corrected for changes in the incident intensity during the scan. X-ray energy was selected using a Si (111) double-crystal flat-plate monochromator of the BESSRC type with 3 by 8 mm beam. Chromium fluorescence was measured using a Canberra Ge detector. Samples were prepared by packing sediment into a slotted sample holder (5 by 5 by 22 mm) sealed with 25-µm thick Kapton film. A Cr standard was prepared with a physical mixture of Cr(III) and Cr(VI) diluted in quartz sand.
Tritium was quantified in Scintiverse universal liquid scintillation cocktail (Fisher) with a Beckman LS6500 scintillation counter (Beckman Instruments, Palo Alto, CA). The resulting breakthrough curves were analyzed with the transport parameter estimation software package CXTFIT2 (Toride et al., 1995) and fit using a two-region nonequilibrium deterministic convection-dispersion model (MODE 2 in CXTFIT2) with a pulse input (MODB 3 in CXTFIT2).
Statistical Analyses
The effects of NO-3 and organic C concentrations on aqueous Cr concentrations in the batch microcosm experiments were assessed with single-factor and two-factor analyses of variance (ANOVA) methods with replicates. Two-factor ANOVA was used to test the effects of molasses and nitrate concentrations and the interactions of these factors. Single-factor ANOVA followed by Tukey's procedure were used to identify significant differences between treatments (Devore, 1991). Single-factor ANOVA was used to assess differences between sterile and live treatments that were performed to assess microbial reduction and abiotic reduction. One-factor ANOVA and the T method using the Studentized range distribution were used to assess the differences in effluent Cr(VI) concentrations from the column experiments (Devore, 1991). Results were considered statistically significance if P < 0.05 for these tests.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
Effluent Chromium Concentrations through Time
Column effluent Cr concentrations were normalized to influent concentrations, adjusted for evaporation, and plotted against time (Fig. 3)
. The average effluent Cr concentration for columns NC1 and NC2 (NO-3 plus organic C treatment), was approximately 90% of the influent Cr concentration for the duration of the test, indicating that 10% of added Cr was retained in the sediment. The other three treatments (organic C only, NO-3 only, and the control) exhibited no significant Cr loss, i.e., effluent concentrations were not statistically different from influent concentrations.
|
|
21.5%), considering the spatial variability of background Cr levels in replicate untreated subsamples measured by XRF. Total sediment Cr concentrations were measured at the top, middle, and bottom of each column. However, no systematic pattern was evident in Cr distribution with respect to location in the column, so a mean concentration was calculated. The XRF analysis of total Cr by grain size fraction showed that the highest concentrations of native Cr were in the smallest grain size fractions and that the bulk of the added Cr that was immobilized by the organic C and N amendments was in the coarser-grained fractions (Table 5). The >0.125-mm fractions consisted primarily of individual particles, while the <0.125-mm fraction contained aggregates where most of the microporosity would occur. The unamended B2 column sediment and the original untreated sediment had similar concentrations, except in the <0.25 mm fractions, where small amounts of Cr were immobilized. In the molasses- and nitrate-amended NC2 column, the larger size fractions (>0.25 mm), which accounted for approximately 90% of the sediment, contained two- to fourfold more Cr than in the unamended column (B2).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In both the batch and column experiments, Cr(VI) reduction and loss from the aqueous phase were enhanced when high levels of both NO-3 and organic C were added to the aqueous phase. In the batch experiments, additions of a high level of either NO-3 or organic C alone led to significantly greater amounts of aqueous Cr loss than were observed in the control and in samples with additions of low levels of these nutrients. In the column experiment, however, simultaneous addition of both NO-3 and organic C was required for significant Cr reduction. Neither significant aqueous Cr loss nor insoluble Cr accumulation was observed, either in the column receiving organic C alone or in the columns receiving NO-3 alone. However, note that the nitrate level in the column experiment was intermediate to the high and low levels of nitrate in the batch experiment. Also, the lack of Cr reduction in the nitrate-alone columns was similar to the low levels of Cr reduction observed in the zero-molasses-low-nitrate batch experiment treatment (Table 2).
The exact mechanisms of microbially mediated Cr reduction–immobilization in our unsaturated batch and column experiments are difficult to discern. Chemical analysis showed the presence of nitrite, indicating use of nitrate as an electron acceptor for dissimilatory NO-3 reduction, in the two batch treatments containing nitrate without molasses (Table 2; nitrate and nitrite data not shown). Thus, it is probable that some dissimilatory NO-3 reduction occurred in anoxic microsites in the NC1 and NC2 columns. However, assuming that there was no remobilization and transport of reduced Cr, Cr (VI) reduction occurred primarily in the coarse-grained fractions of the sediment rather than in the finer-grained fractions (Table 5), where longer residence times would favor oxygen depletion. Chromium(VI) reduction under anaerobic conditions has been found to be induced by NO-3 when linked with benzoate catabolism (Shen et al., 1996), but nitrate has also been shown to inhibit Cr(VI) reduction in an anaerobic aquifer sediment (Marsh et al., 2000). Under aerobic conditions, Cr(VI) reduction by Pseudomonas putida and Bacillus subtilis was shown to be unaffected by additions of nitrate (Ishibashi et al., 1990; Garbisu et al., 1998). Alleviation of N limitation could partially explain the beneficial effects of high amounts of nitrate in our experiments; however, one would expect the 346 mg/L nitrate in our low nitrate treatment to provide sufficient N. Stimulation of dissimilatory nitrate reduction could also be a factor; however, the effect was surprisingly great in the batch experiment even in the absence of added organic C. The column experiment more closely approximated field conditions, and here a combination of both organic C and NO-3 amendments was required to achieve significant Cr reduction–immobilization. Although the exact metabolic pathways stimulated by molasses amendment are unknown, clearly molasses provided sugars and other substrates for respiration and fermentation. Microbial respiration (aerobic or anaerobic) has been linked to reduction of Cr(VI) (Bopp and Ehrlich, 1988; Ishibashi et al., 1990; Shen and Wang, 1993). Also, metabolites from microbial fermentation (e.g., lactate) are known to reduce Cr, especially in the presence of mineral surfaces as catalysts (Deng and Stone, 1996); and this may have contributed to Cr reduction in our study.
The total aqueous Cr loss in the batch microcosm and column experiments can be attributed to the combined effects of microbial reduction, abiotic reduction, and sorption. Batch sorption experiments showed that <3 mg kg-1 of Cr(VI) sorbs to the sediment at the aqueous concentrations used in the experiments (data not shown), below the detection limit for XANES analysis. The abiotic batch microcosm experiment demonstrated that abiotic reduction of Cr(VI) was minor relative to microbial reduction. The addition of organic C led to increased microbial Cr(VI) reduction, but did little to stimulate abiotic reduction (Table 3). The XRF analyses of sieved fractions from columns NC2 and B2 indicate that abiotic reduction was primarily associated with the <0.25 mm fractions (Table 5). The amounts of Cr measured in the <0.25 mm fractions in the NC2 and B2 columns were the same or similar and were considerably more than the amount measured in the same fractions of the untreated sediment. The <0.125 mm fraction contained a much higher proportion of mafic minerals, many of which contain ferrous iron, which is capable of reducing Cr(VI) (Eary and Rai, 1991). Abiotic reduction in the bulk sediment was minimal in all experiments because the <0.125 mm fraction was only 1 to 3% of the total sediment mass. Elevated Cr concentrations in the fractions from 0.25 to >1 mm in column NC2 compared with column B2 were due to microbial reduction stimulated by the additions of organic C and NO-3.
The differences in Cr(VI) reduction between the two batch experiments (Tables 2 and 3) were likely due to increased microbial inhibition encountered at the higher initial Cr(VI) concentration. Chromium loss for the live samples with no additives was only 9% for the batch microcosm experiment with initial Cr(VI) concentration of 250 mg L-1 compared with 26% in the other batch microcosm experiment that had an initial Cr(VI) concentration of 67 mg L-1. Likewise, reduction in the live sample with organic C additives was only 21% for the batch microcosm experiment with initial Cr(VI) concentration of 250 mg L-1 compared with 69% in the other batch microcosm experiment. A number of studies have examined Cr(VI) tolerance and concentrations inhibitory to microbial growth. Chirwa and Wang (1997) found that microbial reduction by a Cr-tolerant Bacillus sp. was not inhibited by Cr(VI) concentrations of 200 mg L-1, but was nearly completely inhibited at 500 mg L-1. Garbisu et al. (1998) found that Cr(VI) reduction by a Cr-tolerant Bacillus subtilis was partially inhibited at a concentration of 52 mg L-1 Cr(VI) and completely inhibited at 104 mg L-1 Cr(VI). Ross et al. (1981) studied the Cr tolerance of bacteria in soil and found that aqueous Cr(VI) concentrations of 12 mg L-1 were inhibitory to most bacteria. Bacterial growth was not inhibited when similar concentrations of Cr(III) were added, indicating that Cr(III) is less toxic than Cr(VI) to bacteria.
Differences in hydraulic parameters between the two tracer tests are attributable to microbial growth. The in situ volumetric water content of the sediment was 5%; during the experiments, volumetric water contents increased to >10% in all columns. Microbial growth occurred in all columns as a result of these increased water contents. The lack of added phosphate likely minimized growth. The addition of organic C (columns NC1, NC2, and C1) led to the greatest changes in hydraulic parameters and also caused a 9- to 45-fold greater biomass (as indicated by total membrane phospholipid fatty acids) than in the N1, N2, and B2 columns that did not receive organic C (data not shown). The column that received high levels of C alone (C1) became clogged, probably due to extracellular polysaccharide production (slime was observed at the termination of the experiment) caused by excess C relative to other nutrients (unbalanced growth). The 2:1 C/N ratio in the NC1 and NC2 columns provided N at 10 times the level required for balanced growth.
Environmental Significance
The columns receiving additions of both NO-3 and organic C showed a 10% Cr reduction and immobilization, even though the travel distance (15 cm) and the residence time (30–33 h) were very short. The removal of Cr(VI) could be very significant in thick vadose zones where flow paths may be tens to hundreds of meters to the water table and where residence times may be years rather than hours. Microbially mediated Cr(VI) reduction in thick vadose zones of arid and semiarid regions could approach 100%. Once it has been reduced to Cr(III), nearly all of the Cr should remain stable in this form indefinitely, since there are relatively few mechanisms for its reoxidation. The only redox couple in natural systems that can oxidize Cr(III) to Cr(VI) is Mn(IV)/Mn(II) (Eary and Rai, 1987; Palmer and Puls, 1994; James, 1996). The sediment in this study had 513 mg kg-1 total Mn by XRF analysis. If one assumes that all of the Mn occurs as Mn(IV) and that 0.1% of this is available at particle surfaces for redox reactions with Cr(III), then only 0.3 mg kg-1 Cr would be reoxidized from Cr(III) to Cr(VI). This corresponds to 1% of the Cr(VI) that was reduced to Cr(III) in the columns with nitrate and organic C.
For enhanced in situ bioremediation strategies, the addition of organic C alone may be sufficient for enhancing microbial reduction of Cr(VI) if NO-3 is present in adequate concentrations as a co-contaminant. Otherwise, nitrate amendment may also be required for enhancing microbial reduction of Cr(VI). Proper nutrient balance (20:1 C/N ratio) should limit microbial production of extracellular polysaccharides that can reduce hydraulic conductivity. Enhanced microbial reduction appears to be a promising technique for remediating Cr(VI)-contaminated sites.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. H. Priester, S. G. Olson, S. M. Webb, M. P. Neu, L. E. Hersman, and P. A. Holden Enhanced Exopolymer Production and Chromium Stabilization in Pseudomonas putida Unsaturated Biofilms. Appl. Envir. Microbiol., March 1, 2006; 72(3): 1988 - 1996. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. A. Holden and N. Fierer Microbial Processes in the Vadose Zone Vadose Zone J., February 1, 2005; 4(1): 1 - 21. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Coombs and T. Barkay Molecular Evidence for the Evolution of Metal Homeostasis Genes by Lateral Gene Transfer in Bacteria from the Deep Terrestrial Subsurface Appl. Envir. Microbiol., March 1, 2004; 70(3): 1698 - 1707. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||
