A feasibility study of immobilized and free mixed culture bioaugmentation for treating atrazine in infiltrate

Journal of Hazardous Materials 168 (2009) 1373–1379

Contents lists available at ScienceDirect

Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat

A feasibility study of immobilized and free mixed culture bioaugmentation for treating atrazine in infiltrate Sumana Siripattanakul a,b , Wanpen Wirojanagud c , John M. McEvoy d , Francis X.M. Casey e , Eakalak Khan f,∗ a

National Center of Excellence for Environmental and Hazardous Waste Management, Chulalongkorn University, Bangkok 10330, Thailand Department of Chemical Engineering and National Center of Excellence for Environmental and Hazardous Waste Management, Ubon Ratchathani University, Ubon Ratchathani 34190, Thailand c Department of Environmental Engineering and Research Center for Environmental and Hazardous Substance Management, Khon Kaen University, Khon Kaen 40002, Thailand d Department of Veterinary and Microbiological Sciences, North Dakota State University, Fargo, ND 58105, USA e Department of Soil Sciences, North Dakota State University, Fargo, ND 58105, USA f Department of Civil Engineering, North Dakota State University, 1410 14th Avenue North, Civil and Industrial Engineering Building (Room 201), Fargo, ND 58105, USA b

a r t i c l e

i n f o

Article history: Received 15 December 2008 Received in revised form 5 March 2009 Accepted 5 March 2009 Available online 18 March 2009 Keywords: Atrazine biodegradation Bacterial community change Cell immobilization Cell loading Infiltration rate Single strand conformation polymorphism

a b s t r a c t A feasibility study of phosphorylated-polyvinyl alcohol immobilized and free mixed bacterial culture bioaugmentation for removing atrazine in agricultural infiltrate was conducted utilizing a sand column setup. The effects of bacterial cell loading and infiltration rate on atrazine degradation were investigated by short-term tests in which the amount of synthetic infiltrate fed through was five times of the void volume (five pore volumes) of the sand column. In addition, the loss of the inoculated atrazine-degrading cultures and the change of bacterial community were determined. Selected tests were continued for monitoring a long-term performance of the system (50 pore volumes of the sand column). The results indicated that the inoculated cells removed 42–80% of the atrazine. The infiltration rate and cell loading significantly affected the atrazine removal. In the short-term tests, the immobilized and free cells provided similar atrazine removal; however, leaching of the free cells was much greater than that of the immobilized cells. For the long-term performance, only the immobilized cells provided consistent atrazine removal efficiency throughout the test. Both immobilized and free cell systems exhibited a significant change in bacterial community structure during the atrazine degradation experiments. The infiltration rate was a significant factor for the change. © 2009 Elsevier B.V. All rights reserved.

1. Introduction There has been an increasing interest to develop new on-site remediation techniques. Biodegradation has been known as an effective method for removing organic contaminants. In some cases, biodegradation by indigenous species cannot cope with all contaminants or takes a long time. Cell bioaugmentation, an addition of sufficient contaminant-degrading microorganisms, can potentially be used to solve this problem. There are many factors influencing the survival of bioaugmented microorganisms and their contaminant degradation efficiencies, such as predation and competition with indigenous microorganisms, and unsuitable growth environments [1].

∗ Corresponding author. Tel.: +1 701 2317717; fax: +1 701 2316185. E-mail addresses: [email protected] (S. Siripattanakul), [email protected] (W. Wirojanagud), [email protected] (J.M. McEvoy), [email protected] (F.X.M. Casey), [email protected] (E. Khan). 0304-3894/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jhazmat.2009.03.025

Immobilized cells (IC) in a polymeric hydrogel are a potential alternative to address these concerns. Immobilization matrices can alleviate environmental stresses [2,3]. This technique also prevents cell wash-out off the contaminated sites resulting in high biomass concentrations and contaminant removal efficiencies. Several studies have reported successful applications of immobilized cell bioaugmentation for point source pollution control, especially wastewater treatment [4–8]. Thus far, no research has been conducted on its application for non-point source pollution control. Agricultural activities including the use of herbicides are one of the main contributors of non-point source pollution. Atrazine (6-chloro-N-ethyl-N-(1-methylethyl)-1,3,5-triazine2,4-diamine) is one of the most widely used herbicides and has been applied to control broad-leaf weeds for crop production. Atrazine detections in groundwater and surface water above the allowable contaminant levels for drinking water of 0.1 ␮g/L and 3.0 ␮g/L in Europe and the United States, respectively, have been frequently reported [9–11].

1374

S. Siripattanakul et al. / Journal of Hazardous Materials 168 (2009) 1373–1379

Fig. 1. A diagram of potential field-scale application of immobilized cell bioaugmentation for atrazine removal from agricultural infiltrate.

In our previous work, biodegradation of atrazine in water by a phosphorylated-polyvinyl alcohol (PPVA) immobilized mixed culture (MC) was successfully performed [12]. The aim of this research was to examine the possibility of the use of the immobilized cell bioaugmentation for remediating agricultural infiltrate contaminated with atrazine, which could lead to a novel groundwater pollution prevention scheme as depicted in Fig. 1. In practice, the immobilized cells can be applied to topsoil during the soil preparation period before the growing season. To validate this potential bioremediation concept, bench-scale sand column experiments were conducted. The effects of cell loading and infiltration rate on atrazine removal were tested. Inoculation by free cells (FC) was evaluated against immobilized cells for their atrazine removal capacity. During the test, the loss of the inoculated atrazine-degrading culture and the change of bacterial community were determined. Long-term performance of atrazine removal was also monitored. 2. Materials and methods 2.1. Bacterial strain and cultural condition A stable atrazine-degrading MC was enriched from atrazine contaminated soil collected from a field site in Oakes, North Dakota, USA following the procedure of Siripattanakul et al. [12]. The culture was previously characterized and found to contain mainly four bacterial strains with two of them being atrazine-degraders (Klebsiella ornithinolytica ND2 and Agrobacterium tumefaciens ND4) [13]. The MC was routinely subcultured every seven days in a sterile buffered bacterial medium (20 mM sodium phosphate buffer at pH 6.8) which contained (per liter) 20 mg of atrazine, 1.0 g of glucose, 0.5 g of K2 HPO4 , 0.5 g of MgSO4 ·7H2 O, 10 mg of FeCl3 ·H2 O, 10 mg of CaCl2 ·H2 O, 0.1 mg of MnCl2 , and 0.01 mg of ZnSO4 . For its use in atrazine biodegradation tests, MC was subcultured in nutrient broth spiked with 20 mg/L of atrazine for 24 h and then centrifuged at 4500 × g for 10 min. The concentrated cells were immediately used for immobilization and/or the biodegradation tests. For the preparation of immobilized dead cells (ID), the concentrated cells were autoclaved at 121 ◦ C prior to the immobilization.

sodium phosphate solution (pH 7) for 60 min for hardening. The gel beads were washed in de-ionized (DI) water and then stored in a 20 mM sodium phosphate solution (pH 6.8) at 4 ◦ C. The final PVA concentration and cell-to-matrix ratio were 10% (w/v) and 3.5 mg dry cells/ml matrix, respectively. 2.3. Synthetic agricultural infiltrate, sand, and column preparations 2.3.1. Synthetic agricultural infiltrate and sand preparations Synthetic agricultural infiltrate was prepared in the same manner as the bacterial medium (described above) except the addition of 1.5 mg/L atrazine. Silica-quartz sand from Le Sueur, MN, USA (Unimin Corporation, CT, USA) was used. The sand was washed with tap water and dried at 105 ◦ C for 24 h. The cleaned sand was sieved to obtain the grain sizes between 0.25 mm and 0.42 mm (US standard sieves number 60 and 40). The sieved sand was autoclaved at 121 ◦ C for 30 min three times within three consecutive days. The void ratio (v/v) of sieved sand loosely packed in a 400 ml graduate cylinder was 0.30 (a void volume of 120 ml). 2.3.2. Column setup A sand column was 6.35 cm in diameter and 23 cm in length. It had an effluent sampling port at the bottom (Fig. 2). All columns were rinsed with 70% isopropanol and autoclaved DI water, respectively before used. The sterile sand and the cells were aseptically mixed and then used to fill the columns. Three sets of sand columns including set ID, set IC, and set FC, were packed as described in Table 1. The packing depth was 14 cm. Each set of the columns comprised three columns at the cell loadings of 300 mg, 600 mg, and 900 mg dry cells/L empty bed volume. Note that the test set ID was conducted as a control to determine whether there was atrazine adsorption by the immobilization matrix and/or the cells.

2.2. PPVA cell immobilization procedure The MC was immobilized using a PPVA technique following the procedure described elsewhere [12]. Polyvinyl alcohol (PVA) was chosen as an immobilization matrix since it is durable and has no negative effect on microorganisms and the environment [4,14]. The concentrated cells (20 mg dry cells/ml) were mixed with a PVA solution. The mixture was slowly dropped into a saturated boric acid solution in a 1 L cylinder for 30 min to form 6 mm spherical beads. The formed hydrogel beads were then submerged in a 1 M

Fig. 2. A diagram of sand column setup.

S. Siripattanakul et al. / Journal of Hazardous Materials 168 (2009) 1373–1379

1375

Table 1 Descriptions of sand columns and their components. Column set

Column number

Column description (cell type)

Cell loading (mg dry cells/L empty bed volume)

Cell mass (mg dry cells)

Bulk inoculated volume* (ml)

Bulk dry sand volume (ml)

Total empty bed volume (ml)

ID

ID1

Immobilized dead cells Immobilized dead cells Immobilized dead cells Immobilized cells Immobilized cells Immobilized cells Free cells Free cells Free cells

300

120

40

360

400

600

240

80

320

400

900

360

120

280

400

300

120

40

360

400

600

240

80

320

400

900

360

120

280

400

300 600 900

120 240 360

N/A** N/A** N/A**

400 400 400

400 400 400

ID2 ID3 IC1

IC

IC2 IC3 FC

* **

FC1 FC2 FC3

Volume of the cells and matrices. Volume of the cells was negligible.

2.4. Atrazine biodegradation test in column system Duplicate atrazine biodegradation experiments in sand columns were performed. The columns were operated as follows. They were filled up with a 20 mM sodium phosphate solution (pH 6.8). The synthetic agricultural infiltrate was continuously applied as a step input at an 8 h interval (three times/d) under gravity-flow condition. The flow rates studied were 30 ml/d, 90 ml/d, and 180 ml/d corresponding to the infiltration rates of 1 cm/d, 3 cm/d, and 6 cm/d for obtaining the actual, high, and critical (extremely high) infiltration rates, respectively. The test was run for five pore volumes (PV) (five times of the void volume of the column or total pore water volume of 600 ml), which was selected because it was the period that the breakthrough was reached for all test conditions. During the test, the effluent was sampled every 0.25 PV to measure atrazine and intermediate metabolite concentrations for the infiltration rates of 1 cm/d and 3 cm/d. However, for the infiltration rate of 6 cm/d, this effluent sampling was conducted at a 0.5 PV frequency. The samples at every 1 PV were determined for the number of viable MC cells using a plate count technique. At 5 PV, the effluent samples from all viable cell columns (column sets IC and FC) were taken for detecting the change of bacterial community structure using a single strand conformation polymorphism (SSCP) technique. After 5 PV, flow was continued for selected columns for long-term monitoring, where the effluent was monitored for atrazine and intermediate metabolite concentrations every 5 PV between 5 PV and 50 PV. 2.5. Analytical methods 2.5.1. Atrazine and metabolite analysis The analytical methods used for atrazine, desethylatrazine (DEA), deisopropylatrazine (DIA), and hydroxyatrazine (HA) were modified from D’Archivio et al. [15]. A solid phase extraction technique using 200 mg of polymeric sorbent in a 6 ml cartridge (StrataX, Phenomenex, CA, USA) was applied for atrazine and metabolite extraction. The cartridge was prewashed and conditioned using 6 ml of ethyl acetate and 6 ml of methanol, respectively. The cartridge was then washed with 6 ml of DI water. After loading a sample (2 ml) and drying the cartridge under vacuum condition, the cartridge was eluted with 6 ml of acetonitrile: methanol (1:1, v/v) through gravity flow. Then, the extract was evaporated to dryness under a gentle stream of dry nitrogen. The dry residue was dissolved in 0.5 ml of water: acetonitrile (1:1, v/v). The extract was analyzed for atrazine and its metabolites on a Hewlett Packard 1100

series high-performance liquid chromatograph equipped with a C18 reverse phase column (Jupiter, Phenomenex, CA, USA) at an ultraviolet wavelength of 220 nm. The isocratic mobile phase of water:acetonitrile (1:1) at a flow rate of 1 ml/min was used. 2.5.2. Viable plate count The bacterial loss from the column system was determined by the number of viable bacteria in the effluent samples. Each effluent sample was serially diluted and spread onto a selective bacterial medium agar. The agar formulation was the same as the synthetic agricultural infiltrate but added with agar of 2% (w/v). Bacterial colonies were counted after 48 h incubation at 30 ◦ C. 2.6. Bacterial community change detection using SSCP technique 2.6.1. Deoxyribonucleic acid (DNA) extraction Total genomic DNA was extracted from both immobilized and free cells which were collected before and after the 5 PV atrazine biodegradation test. For free cells, the concentrated cells (as described previously in Section 2.1) were used to represent the cells before the test while the column effluent samples at 5 PV were the sources of cells after the test. Similarly, the cells immediately after immobilization and the effluent samples from the immobilized cell column after the 5 PV test were collected. The DNA extraction procedure for immobilized cells began with separating the cells from the matrix. Ten beads containing immobilized cells were cut in half and mixed thoroughly in 10 ml of 20 mM phosphate buffer (pH 6.8) at 3200 rpm for 2 min using a vortex mixer (VWR International, PA, USA). The immobilized cell samples (10 ml of the supernatant from the cell separation procedure), concentrated free cell samples, and effluent samples were centrifuged at 16,000 × g for 2 min and used for extracting DNA. The genomic DNA extraction procedure followed the instruction from the DNA extraction kit (Wizard Genomic DNA Purification Kit, Promega, CA, USA). 2.6.2. DNA amplification Polymerase chain reaction (PCR)-SSCP procedure was modified from Lin et al. [16]. The V3 region of the 16S ribosomal ribonucleic acid (rRNA) gene (nucleotide positions 334–514 of Esherichia coli) was amplified with primers EUB1 (5 -CAG ACT CCT ACG GGA GGC AGC AG 3 ) and UNV2 (5 -GTA TTA CCG CGG CTG CTG GCA C 3 ). A 25 ␮L PCR reaction contained 1.5 mM of MgCl2 , 200 ␮M of dNTP, 5.0 ␮L of Taq polymerase buffer 5 × (Promega, CA, USA), 50 ␮M of each primer, 1.25 U of Taq Polymerase (Promega, CA, USA), and 2 ␮L

1376

S. Siripattanakul et al. / Journal of Hazardous Materials 168 (2009) 1373–1379

Fig. 3. Breakthrough curves of atrazine at the infiltration rates of: (A) 1 cm/d, (B) 3 cm/d, and (C) 6 cm/d. ID, IC, and FC were the columns of the immobilized dead cells, immobilized cells, and free cells, respectively. The numbers 1, 2, and 3 represent the results of the columns at the cell loadings of 300 mg/L, 600 mg/L, and 900 mg/L, respectively.

of DNA template. DNase/RNase-free water was used for making up the volume of samples. The PCR conditions consisted of an initial denaturation at 94 ◦ C for 5 min, 30 cycles at 94 ◦ C for 30 s, 55 ◦ C for 30 s, and 72 ◦ C for 30 s, and a final extension at 72 ◦ C for 5 min. The presence of PCR products (approximately 200 bp) was confirmed by 1.5% agarose gel electrophoresis. 2.6.3. SSCP gel electrophoresis The SSCP test was performed in a horizontal electrophoresis setup (Origins, Elchrom Scientific, Switzerland). The SSCP procedure followed the instruction from the manufacturer. Three microliters of PCR products were mixed with 7 ␮L of a denaturing solution (1 ml of formamide, 10 ␮L of 1 M NaOH, and 20 ␮L of 0.02% (w/v) bromphenol blue). The mixtures were heated at 95 ◦ C for 5 min and immediately placed into ice until loading to the SSCP gel. The 10 ␮L denatured PCR products were loaded into a pre-cast polyacrylamide gel (GMATM , Elchrom Scientific, Switzerland). The gel was run at a constant voltage of 72 V and 9 ◦ C for 10 h. The gel then was visualized with SYBR® Gold staining (Molecular probes, OR, USA). 2.6.4. SSCP gel data analysis The images of DNA profiles were analyzed using Bionumerics version 5 (Applied Maths, TX, USA). The pair-wise similarity among the samples was calculated using Dice index and an unweighted pair-group method with arithmetic average. 2.7. Statistical analysis For atrazine biodegradation data analysis, mass of atrazine in the effluent samples was statistically analyzed using JMP IN® 5.1.2 (SAS, NC, USA). The data were analyzed using a multiple regression model and examined for the significance of the atrazine mass difference between test conditions with analysis of variance and t-test.

3. Results and discussion 3.1. Atrazine biodegradation using immobilized dead, immobilized, and free cells Fig. 3 shows the breakthrough curves of relative atrazine concentration (C/C0 ) of the ID, IC, and FC within 5 PV. In all curves, C/C0 increased after 0.5–1.0 PV and reached a plateau at approximately 2.5–4.0 PV. The immobilized dead cells did not have degradation ability resulting in no atrazine removal (C/C0 of 1.0) after 3.0–4.0 PV while the breakthrough curves of immobilized and free cell columns were stable at C/C0 of 0.2–0.6. At the same cell loading and infiltration rate, atrazine was detected in the effluent of the free cell columns earlier than the immobilized cell and immobilized dead cell columns, in which atrazine sorption to the immobilization matrix likely occurred [12]. Based on the breakthrough curves of the immobilized dead cells (stable at C/C0 of 1.0), the PPVA matrix only retarded the atrazine breakthrough but did not permanently remove it. The immobilized and free cell columns at the infiltration rates of 1 cm/d, 3 cm/d, and 6 cm/d (Fig. 3) removed atrazine at 65–80%, 50–73%, and 42–58%, respectively. The atrazine removal efficiency decreased with increasing infiltration rate whereas higher cell loading rates provided greater removal efficiency. The statistical analysis showed that both infiltration rate (p < 0.0001) and cell loading (p = 0.0002) significantly influenced atrazine removal. However, for all experiments, there were no significant differences between the immobilized and free cell systems for atrazine removal (p = 0.4493). Therefore, the atrazine-degrading mixed culture used in this study in either immobilized or free cell forms was efficient for atrazine removal. During the atrazine biodegradation test, the atrazine primary intermediate metabolites (HA, DEA, and DIA) were not detected in any effluent samples. The result suggests that atrazine

S. Siripattanakul et al. / Journal of Hazardous Materials 168 (2009) 1373–1379

1377

from the free cell columns were very close to the initial number of cells. This could be because the sand columns could not effectively retain the free cells. Yolcubal et al. [17] reported 90% loss of bioaugmented free cells treating salicylate and polycyclic aromatic hydrocarbons in a sand column setup. Even though microbial growth might take place during the atrazine biodegradation test, it is likely that the growth was not as much as the microbial loss from the column leading to less numbers of cells left in the columns. This inference is supported by less numbers of the leaching cells at later PVs. At 0 PV and 1 PV, the numbers of cell losing from the immobilized cell columns were about hundred times less compared to those of the free cell columns. This indicates that the immobilization matrix could effectively protect the cells from leaching. 3.3. Bacterial community change during atrazine biodegradation test

Fig. 4. Number of viable cells in the effluent samples from the columns of the immobilized (IC1, IC2, and IC3) and free (FC1, FC2, and FC3) cells at the infiltration rates of (A) 1 cm/d, (B) 3 cm/d, and (C) 6 cm/d.

was quickly degraded to some other intermediate products beyond the primary intermediate metabolites. 3.2. Cell leaching during atrazine biodegradation The results of cell losses are presented in Fig. 4. The bacteria leaching from the immobilized and free cell columns at all cell loading was about 6–7 log and 6–9 log CFU/ml, respectively. The trends of the cell losses in the immobilized and free cell columns were different. The cell losses from the immobilized cell columns (columns IC1 to IC3) were stable while those from all free cell columns (columns FC1 to FC3) declined with duration of the experiment. The initial number of cells inoculated to all columns was approximately 9 log CFU/ml. At 0 PV and 1 PV, the numbers of cells leaching

Fig. 5 presents the SSCP profiles of the 16S rRNA gene fragment of the atrazine-degrading mixed culture. The DNA band numbers 3 and 6, 4 and 6, 1 and 2, and 5 represented Alcaligenes faecalis ND1 (ND1) (EU075145), K. ornithinolytica ND2 (ND2) (EU075144), Bacillus megaterium ND3 (ND3) (EU075147), and A. tumefaciens ND4 (ND4) (EU075146) as previously identified by Siripattanakul et al. [13]. The samples (initial FC and initial IC) of atrazine-degrading mixed culture before and after immobilization were analyzed for determining the effect that the immobilization process had on the bacterial community structure. The cluster analysis showed that the similarity of the culture before and after the immobilization process was approximately 60%. This suggests that the immobilization process affected the bacterial community structure to some extent. Chemical or physical stresses during the immobilization process might be the cause. It was reported that the immobilization process affected the bacterial viability [12]. Some bacterial species might be less tolerant than the others and were killed in the immobilization procedure leading to the difference in the bacterial community structure. However, ND2 and ND4 which are atrazinedegrading strains [13], are present in the SSCP profiles of both initial free and immobilized cell samples. This indicates that the immobilization process affected some bacterial species but not the atrazine-degrading strains resulting in no influence on atrazine biodegradation. This inference is supported by the biodegradation result presented earlier, which both free and immobilized cells biodegraded atrazine successfully. For the pair-wise comparison between before and after the atrazine biodegradation test, the immobilized and free cell systems provided comparable results. The similarity of the bacterial community structure before and after the test was approximately 40–50% (Fig. 5). The effluent samples at the same infiltration rate but different cell loadings had the similarity of the bacterial community structure of 80–100%. On the contrary, the samples at the same cell loading but different infiltration rates had the similarity of the bacterial community structure of 50–60%. This suggests that the infiltration rate influenced the bacterial community structure more than the cell loading. The DNA band number 5 represented ND4, is present in all samples after the 5 PV test while ND2 (band number 4) was detected only from the columns tested at the infiltration rate of 1 cm/d. This could be because ND2 did not retain and/or grow well in the columns at the higher infiltration rates (3 cm/d and 6 cm/d). Consequently, at these infiltration rates, lower atrazine removal was observed. As shown in the previous section, the infiltration rate also affected the atrazine removal more than the cell loading. This coincident seemed reasonable because intuitively atrazine removal and bacterial community structure should be related.

1378

S. Siripattanakul et al. / Journal of Hazardous Materials 168 (2009) 1373–1379

Fig. 5. SSCP profiles and cluster analysis of PCR-amplified 16S rRNA gene fragments of the atrazine-degrading mixed culture. Initial IC and FC are the immobilized and free cell cultures before the biodegradation test. IC1 and FC1, IC2 and FC2, and IC3 and FC3 represent the effluent samples at 5 PV of the columns and the cell loadings of 300 mg/L, 600 mg/L, and 900 mg/L, respectively.

3.4. Atrazine biodegradation: long-term performance (50 PV) The long-term performance of the atrazine biodegradation was examined to determine the feasibility of the bioaugmentation scheme for the real world application. The infiltration rate of 6 cm/d (extremely high value) was selected to represent the worst case scenario. Fig. 6 shows that the trend of C/C0 from the immobilized cell columns was stable whereas the trend from the free cell columns gradually increased after testing for 15 PV. At the end of the test (50 PV), the atrazine removal efficiencies from the immobilized and free cell columns were 40–60% and 10–15%, respectively. Similar to the result of the short-term test (5 PV), the cell leaching from the immobilized cell columns was consistent at about 5–7 log CFU/ml whereas the cell loss from the free cell

system decreased with increasing pore water volumes (data not shown). At 10 PV, the cell leaching from the free cell columns was less than 4 log CFU/ml, which was much less than the beginning (9 log CFU/ml). Also, 10 PV was where atrazine biodegradation by the free cells started to be less efficient than that by the immobilized cells. Based on the results of the free cell system, cell loss was a main factor influencing the atrazine removal. In the short-term tests, the atrazine biodegradation performance by the free cells was not significantly different from that by the immobilized cells since the number of the free cells left in the column from the inoculation and growth could cope with atrazine at the tested concentration and retention time. However, the long-term results apparently showed that after sometimes the free cell loss was so

Fig. 6. Long-term performance of the immobilized (IC1, IC2, and IC3) and free (FC1, FC2, and FC3) cells at the infiltration rate of 6 cm/d and the cell loadings of (A) 300 mg/L, (B) 600 mg/L, and (C) 900 mg/L.

S. Siripattanakul et al. / Journal of Hazardous Materials 168 (2009) 1373–1379

much and the number of the cells left in the column was not enough to remove atrazine effectively. Therefore, the immobilized cell bioaugmentation should be a better approach for long-term atrazine bioremediation. In the cases of field-scale applications, possible attachment of bioaugmented cells onto soil surface could play a role in contaminant biodegradation. The magnitude of the cell attachment depends on soil characteristics, interaction with indigenous microorganisms, and other environmental factors [1,18]. These attributes are uncontrollable or difficult to control. The immobilized cell bioaugmentation, which is more immune to these attributes, is therefore a promising bioremediation alternative. 4. Conclusions The cell inoculation provided successful atrazine removal in the infiltrate. The infiltration rate influenced the atrazine biodegradation in terms of efficiency and bacterial community structure. The cell loading affected atrazine removal efficiency. For the long-term use, the immobilized cell inoculation overcame the cell leaching and maintained consistent atrazine degradation. Therefore, the immobilized cell bioaugmentation is a potential alternative for remediating organic contaminant in infiltrate and in turn preventing groundwater contamination. For future work, the effects of soil characteristics such as organic content, percent of clay, and pH value on atrazine degradation and a pilot-scale study are recommended. Acknowledgments This material is based upon work partially supported by the National Science Foundation under Grant No. 0449125. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation. The work is also partially funded by the Commission on Higher Education, Ministry of Education, Thailand and Research Center for Environmental and Hazardous Substance Management, Khon Kaen University, Thailand. The authors thank Thunyalux Ratpukdi and Christopher Hill, Department of Civil Engineering, North Dakota State University (NDSU) for their assistance on the sand column setup. The authors acknowledge Dr. Mohamed Fakhr, Department of Veterinary and Microbiological Sciences, NDSU for his help on SSCP data analysis. The authors also thank Catherine Giddings, Department of Vet-

1379

erinary and Microbiological Sciences, NDSU and Mirjana Kozulic, Elchrom Scientific, Switzerland for helpful discussion on the SSCP procedure. References [1] J.A. van Veen, L.S. van Overbeek, J.D. van Elsas, Fate and activity of microorganism introduced into soil, Microbiol. Mol. Biol. Rev. 61 (1997) 121–135. [2] N. Cochet, J.M. Lebeault, M.A. Vijayalakshmi, Physicochemical aspects of cell adsorption, in: R.D. Tyagi, K. Vembu (Eds.), Wastewater treatment by immobilized cells, CRC Press, Florida, 1990, pp. 1–28. [3] G.A. Dervakos, C. Webb, On the merits of viable-cell immobilization, Biotechnol. Adv. 9 (1991) 559–612. [4] K.C. Chen, Y.F. Lin, Immobilization of microorganisms with phosphorylated polyvinyl alcohol (PVA) gel, Enzyme Microb. Technol. 16 (1994) 79–83. [5] I.S. Chang, C.I. Kim, B.U. Nam, The influence of poly-vinyl-alcohol (PVA) characteristics on the physical stability of encapsulated immobilization media for advanced wastewater treatment, Process Biochem. 40 (2005) 3050–3054. [6] R. Jittawattanarat, K. Kostarelos, E. Khan, Immobilized cell augmented activated sludge process for treating wastewater containing hazardous compounds, Water Environ. Res. 79 (2007) 461–471. [7] R. Jittawattanarat, K. Kostarelos, E. Khan, Immobilized cell augmented activated sludge process for enhanced nitrogen removal from wastewater, Water Environ. Res. 79 (2007) 2325–2335. [8] L.S. Zhang, W.Z. Wu, J.L. Wang, Immobilization of activated sludge using improved polyvinyl alcohol (PVA) gel, J. Environ. Sci. 19 (2007) 1293–1297. [9] E. Gannon, Site Remediation: Environmental Clean-up of Fertilizer and Agrichemical Dealer Sites, vol. 28. Iowa Case Studies, Iowa Natural Heritage Foundation, Iowa, 1992. [10] W.C. Koskinen, S.A. Clay, Factors affecting atrazine fate in North Central U.S. soils, Rev. Environ. Contam. Toxicol. 151 (1997) 117–165. [11] S. Rousseaux, A. Hartmann, B. Lagacherie, S. Piutti, F. Andreux, G. Soulas, Inoculation of an atrazine-degrading strain, Chelatobacter heintzii Cit1, in four different soils: effects of different inoculum densities, Chemosphere 51 (2003) 569–576. [12] S. Siripattanakul, W. Wirojanagud, J. McEvoy, E. Khan, Effect of cell-to-matrix ratio in polyvinyl alcohol immobilized pure and mixed cultures for atrazine degradation, Water Air Soil Poll.: Focus 8 (2008) 257–266. [13] S. Siripattanakul, W. Wirojanagud, J. McEvoy, T. Limpiyakorn, E. Khan, Atrazine degradation by stable mixed cultures enriched from agricultural soil and their characterization, J. Appl. Microbiol. 106 (2009) 986–992. [14] T.J. Gentry, C. Rensing, I.L. Pepper, New approach for bioaugmentation as a remediation technology, Crit. Rev. Env. Sci. Technol. 34 (2004) 447–494. [15] A.A. D’Archivio, M. Fanelli, P. Mazzeo, F. Ruggieri, Comparison of different sorbents for multiresidue solid-phase extraction of 16 pesticides from groundwater coupled with high-performance liquid chromatography, Talanta 71 (2007) 25–30. [16] C.W. Lin, C.Y. Lai, L.H. Chen, W.F. Chiang, Microbial community structure during oxygen-stimulated bioremediation in phenol-contaminated groundwater, J. Hazard. Mater. 140 (2007) 221–229. [17] I. Yolcubal, S.A. Pierce, R.M. Maier, M.L. Brusseau, Biodegradation during contaminant transport in porous media. V. the influence of growth and cell elution on microbial distribution, J. Environ. Qual. 31 (2002) 1824–1830. [18] E.M. Godsy, D.F. Goerlitz, D. Grbic-Galic, Methanogenic biodegradation of creosote contaminants in natural and simulated ground-water ecosystems, Ground Water 30 (1992) 232–242.