Reductive Dechlorination of 2,3,4-Chlorobiphenyl by Biostimulation and Bioaugmentation

Characterization of sampling sites

Sediments and stream waters were collected from several natural sites around Bangkok and nearby Samuthpakarn Province in Thailand including (1) one site in HLP; (2) two sites along the canal receiving effluent discharge from the center wastewater treatment plant of the Bangplee Industrial Estate (BP1 and BP2) where several electronic-related factories are located; (3) two sites along the canal receiving discharge from small material recovery facilities (MF1 and MF2) that possibly received PCBs leakage during waste separation and cleaning; (4) one site in the canal near the South-Bangkok Power Plant (PWP) where a number of dumped transformers and capacitors were accumulated; and (5) one site from the Bangplakod Canal (BPK) which had good water quality and served as a habitat for many aquatic lives, because of which it was selected as a control sample for a non-PCB contaminated site. In the sampling of sediments, five centimeters of the surface sediment were carefully removed, and the bottom sediment was scraped and packed in a plastic bag. Canal waters were sampled and stored in the containers. Both sediment and water samples were stored at 4°C until they were used.

Media preparation

Sediment-water samples were prepared by mixing 5 g of sediment and 5 mL of stream water in a 20-mL serum bottle, shaken thoroughly by hand for 1 min, sealed with a butyl rubber stopper, and capped with an alumina cap until they were used (Anotai et al., 2010; Chen et al., 2010). Sediment slurry was prepared by mixing sediment and stream water in the ratio of 1:1 (v/v), shaken thoroughly by hand for 1 min, filtered to remove the particles larger than 0.7 mm, withdrawn by a 100-mL glass syringe with a 22G×2 hypodermic-needle (0.7 mm opening), and kept in a 100-mL alumina-capped serum bottle (Chen et al., 2010). A nutrient-strengthened sediment slurry was also prepared by adding 2.7 g of NH₄Cl, 0.1 g of MgCl₂.6H₂O, 0.1 g of CaCl₂.2H₂O, 0.02 g of FeCl₂.4H₂O, 0.27 g of K₂HPO₄, 0.35 g of KH₂PO₄, and 5 g of yeast extract to 1 L of fresh sediment slurry. Then, it was kept in an alumina-capped serum bottle until it was used.

Chemicals

All chemicals used in this research were reagent grade. Seven primary PCB congeners, that is, 234-, 22′5-, 24′5-, 22′35′-, 22′45-, 23′44′5-, and 22′34′5′6-CBps were purchased from AccuStandard, Inc. Acetate, lactate, and pyruvate in sodium form were obtained from Sigma-Aldrich, Inc. Stock PCB solutions in acetone were separated into several 1.5-mL vials, sealed with butyl rubber stoppers, capped with alumina caps, and kept refrigerated until they were used.

Preparation of dechlorination test

An appropriate amount of stock PCB solution was added into the serum bottles containing sediment slurries or sediment water. All serum bottles were kept in dark under room temperature and shaken by hand from time to time on a daily basis. All experiments were conducted in duplicate with a sterilized control (sterilized thrice in an autoclave). For certain intervals, the mixture in serum bottles from sediment slurry and nutrient-strengthened sediment slurry sets was withdrawn by a glass syringe with the 22G×2 hypodermic needle for extraction. In sediment-water sets, hexane-acetone mixture (9:1, v/v) was injected into the serum bottle, and the whole content in the bottle was extracted.

Biostimulation experiments

Acetate, lactate, and pyruvate at the concentration of 20 mM were separately introduced to the serum bottle in the dechlorination test as electron donors for anerobic microbes, with the purpose of studying their individual effects on 234-CBp dechlorination.

Bioaugmentation experiments

Three different of bioaugmentation tests were performed by mixed effective sediment slurry (HLP canal) with less-effective sediment slurry (six sediment slurries from other four streams), including 50% (in volume) effective+50% ineffective, 10% effective+90% ineffective, and 90% sterilized effective+10% ineffective. The experiment of 90% sterilized effective sediment slurry with 10% fresh ineffective sediment slurry was set to test the effect of non-bacterial components such as carbon sources, mineral salts, and growth factors in HLP sediment on 234-CBp dechlorination.

Extraction methods

After weeks of incubation, the residual PCBs and the dechlorination products in sediment-water serum bottles were extracted by solvent and ultrasonic treatment (EPA 3550) with some modifications. One milliliter of 6 N H₂SO₄ and 1 g of copper were added into the serum bottle and thoroughly mixed by a vortex mixer. After that, 5 mL of 9:1 hexane:acetone solvent were added into the serum bottle. The bottle was then shaken by a vortex mixer followed by ultrasonicating for 30 min. The mixture was centrifuged using a G5-15 centrifuge (Beckman Instruments) at 3000 rpm for 20 min, and the supernatant was transferred to a new vial. The mixture remaining in the serum bottle was repeatedly re-extracted following the same procedures. After the second extraction, the upper-layer liquid was pulled out and filled in the same vial that contained the first extracts. One milliliter of 6 N NaOH was added to the vial, and it was then shaken by a vortex mixer, removing the hexane layer to a new vial. A small amount of Na₂SO₄ was then added to this vial to remove the water residue. The extract was further cleaned up by a deactivated Florisil column. Finally, the elution was transferred to an analyzing vial for gas chromatograph analysis.

For sediment slurry sets, at each predetermined time, 2 mL of homogeneous sediment slurry were withdrawn from the serum bottle by a glass syringe with a 22G×2 hypodermic needle and extracted for remaining PCBs following the procedures described by Chen et al. (2010).

Gas chromatography analysis

Analysis of PCBs followed the EPA 8082A method for gas chromatograph with micro-electron capture detector (quantification) and EPA 680 method for gas chromatograph and mass spectrometer (qualification). The 6890N Network GC System equipped with a μECD (Agilent Technologies) and a 30-m×0.25-mm ID fused-silica capillary column DB-5 (Agilent Technologies) was used to quantitatively analyze PCB congeners and their intermediates. All intermediate products were qualitatively confirmed by the 6890N Network GC System (Agilent Technologies) equipped with a 30-m×0.25-mm ID fused-silica capillary column DB-5MS (Agilent Technologies) and connected with the 5973 Network Mass Selective Detector (Agilent Technologies). Both GCs were operated under the following conditions: the oven temperature was maintained at 120°C for 2 min, raised to 225°C at 3°C/min, maintained for 3 min, and then raised again at 5°C/min to the final temperature of 270°C, which was held for 11 min. The temperature of the injector and the detector were set at 280°C and 300°C, respectively. Nitrogen and helium gases were used as the makeup (20 mL/min) and carrier (1 mL/min) gases, respectively, and the split ratio was kept at 10:1.

Methane analysis

Quantification of headspace methane gas in the serum bottles was checked by the GC-TCD installed in a pack column with two analytical ports (GC-8A; Shimadzu): port 1 equipped with the UnibeadsC 80/100 mesh (4 mm OD, 3 mm ID, 3 m long); port 2 equipped with the 3% unisole 30T on flusinP 30/60 mesh (4 mm OD, 3 mm ID, 3 m long). Temperature of the injector and column was maintained at 160°C and 110°C, respectively. Nitrogen and helium (300 kPa) were employed as make-up and carrier gases, respectively.

Characterization of experimental media

The physical and chemical properties of the sediment samples were shown in Table 1, indicating that organic substrate and primary nutrients should be sufficient for anerobic digestion. Table 2 shows the characteristics of prepared sediment slurries. Low volatile suspended solids to suspended solids (VSS:SS) ratio implied that the amount of viable microbes had not reached an outbreak condition in the sediment. Nevertheless, the VSS values were higher than those of typical biological treatment systems, thus suggesting the sufficiency of viable microbes. The chemical oxygen demand (COD) values were quite high (all greater than 2000 mg/L), which also suggested a possible carbon sources-rich environment to support the activities of heterotrophic bacteria; archaea and fermentators that might play important roles in dehalogenation processes. Moreover, the ratios of COD:N:P were in the range of 62–401:3–11:1, which showed that there were adequate nutrients for microbial requirements (Droste, 1997).

PCBs dechlorination ability of microbial consortium in natural stream sediments

To evaluate the PCB-dechlorination ability of native microbes, seven CBp congeners were separately inoculated into the sediment water and sediment slurries without any acclimation and enrichment. Results from the sediment slurry experiments as shown in Table 3 revealed that, of all tested PCB congeners, only 234-CBp which contains three chlorine atoms located at three consecutive positions on the same phenyl ring could be dechlorinated within 30 weeks of incubation. The results indicated that all indigenous microorganisms from five sampling streams (seven sampling sites) could dechlorinate 234-CBp, although the microbes from the HLP Canal with sporadic HCB contamination (Chen et al., 2010) provided the best performance. This result is different from the study of Chen et al. (1997), who found that the sediment slurries from Erh-Jen River in Kaohsiung City of Taiwan could not deliver the dechlorination of 234-CBp within the incubation period of 150 days. The diversity of microbial PCB-dechlorination between these two countries might be from the different climate, that is, tropical Thailand versus sub-tropical Taiwan as noticed by Chen et al. (2010). For the sediment-water sets, 234-CBp dechlorination could only occur at sites HLP, MF1, and PWP within the study period of 24 weeks. Sediment in the sediment-water sets was a more complicated solid-like matrix than the liquid-like sediment slurry; hence, adsorption of PCB congeners onto the solid particles might inhibit the dechlorinators in accessing 234-CBp. The results of all other congeners except 234-CBp, which could not be dechlorinated, agreed with the finding of VanDort et al. (1997) that the potential of steric effect of bulky substitutions on the opposite ring was significant in the PCB dechlorination.

Effect of biostimulation on PCB dechlorination

Effect of organic substrate on microbial ability to reductively dechlorinate PCBs was investigated to evaluate the possibility of dechlorination enhancement by external supplement. Sediment slurry (with or without being nutrient-strengthened) were used in this part. Pyruvate and lactate were chosen as the fermentation products that can support the growth of acetogens and sulfate-reducing bacteria. Acetate was selected as a direct substrate used by the methanogens and sulfate-reducing bacteria. The results showed that the supplement of external substrate and nutrients did not have any significant impact on 234-CBp dechlorination (Table 4). The dechlorination product occurrence time (lag phase) and the dechlorination completion time were even longer than the nonstimulated set. It revealed that HLP sediment might contain sufficient substrates and nutrients for dechlorination consortium to degrade 234-CBp. This finding was quite different from the results of Nies and Vogel (1990), who found that organic supplement was the most important factor controlling the dechlorination of Aroclor 1242 in soil.

From the concurrence of 24-CBp and methane production in the electron donor amended set as shown in Fig. 1, it was likely that the methanogens might involve in 234-CBp dechlorination as discussed by Ye et al. (1995) and Chang et al. (2001). However, it was contrary to the findings of Cutter et al. (2001), who worked with 2356-CBp and found that the methanogens were not responsible for the dechlorination. From the results of methane production of Fig. 1, it was found that the production in the lactate-amended set was the lowest. The reason might be because the sulfate-reducing bacteria (SRB) could use lactate directly and enriched as a dominant bacteria than methanogen and somewhat inhibited the methanogenesis. Under this condition, the methanogens were less enriched when compared with the SRB. Therefore, the role of fermentators who provided essential carbon sources and electron donors was important in methanogenesis and, possibly, in the dechlorination process as well.

OPEN IN VIEWER

FIG. 1.

234-CBp dechlorination profiles in the Hua Lum Poo sediment slurry amended with various electron donors. (a–c) Nutrient-strengthened sediment slurry amended with pyruvate (a), lactate (b), and acetate (c); (d) fresh sediment slurry amended with pyruvate. 234-CBp, 2,3,4-chlorobiphenyl.

The 24-CBp was the only observed dechlorination product of 234-CBp (by removing the middle chlorine in the three consecutive chlorine atoms) rather than 23-CBp and 34-CBp (by removing the edging chlorine atom). This was in agreement with the thermodynamic (ΔH) and extrathermodynamic (ΔlnRRT) favorable reactions proposed by Chen et al. (2001a, 2001b). However, due to the insensibility of the GC-ECD used in this research to 2-CBp, 4-CBp, and biphenyl, the sequential dechlorination of 24-CBp to less chlorinated products could not be quantified.

Effect of bioaugmentation on PCB dechlorination

Although there is a growing recognition that to determine the possibility of bioagumentation of PCBs dechlorination is essential in the environmental remediation of PCB-contamination, the bioaugmentation by the un-enriched, un-acclimated, and un-treated active indigenous microbes to the less-active sediment slurry samples has not been explored. It was previously found that only HLP indigenous microorganisms could dechlorinate 234-CBp effectively (Table 3). Therefore, untreated sediment microbes from HLP were used as seeds to inoculate into the sediment slurries or sediments from other sites that were identified as less effective in PCBs dechlorination in Table 3. Different quantities of fresh effective sediment slurry (HLP) were mixed with the less-dechlorination-effective samples to test the bioaugmentation potential (Table 5). The result revealed that 50% and 10% of effective sediment slurry (HLP) could dechlorinate 234-CBp, but 90% of sterilized HLP slurry set did not show any movement in 234-dechlorination. Entire loss of dechlorination activity for 90% sterilized HLP + 10% less-effective slurry set revealed that ten times dilution of less-effective slurry cultures would completely inhibit 234-CBp dechlorination potential. The reason should be the sparseness of active dechlorination consortium in the diluted less-effective slurry in those serum bottles, even though the unknown factors from the sterilized effective slurry might develop a positive effect on the dechlorination.

Comparing the results from different ratios of active cultures with inoculated sets of 10% and 50%, it showed that lag phase and completion time were similar, regardless of augmentation portion. The results showed that the active consortium from HLP site could activate the dechlorination without any artificial supplements, that is, the lag phase and dechlorination completion time were shortened. Moreover, it also suggested that only 10% of active consortia could perform the enhancement of PCBs dechlorination. This is the solid evidence that the promotion of dechlorination activity for the less-effective sites could derive from bioaugmentation procedures rather than amending chemicals in the contaminated sites.

A possible explanation that the dechlorination could not happen in the less-effective sites might be not only from the inactiveness of the main dechlorinators such as methanogens and SRB but also from the lack of supporting bacteria (e.g., acetogens, fermentators, and acidogens). After the bioaugmentation by mixing active microbes, including supporting bacteria and dechlorinators, sooner or later, the dechlorination was initiated. Hence, nutrient and carbon sources were possibly already adequate to support the dechlorinating microbes in these sites. This result implied the potential of introducing the active microorganisms from the remote site to enhance PCBs dechlorination of the ineffective sites. It was also significantly different from the study of Bedard et al. (1997), which introduced the enriched PCB-dechlorinating cultures to treat the contaminated sites, and found that the dechlorination could not be accomplished.