Stimulation of reductive dechlorination of hexachlorobenzene in soil by inducing the native microbial activity

Chemosphere 55 (2004) 1477–1484 www.elsevier.com/locate/chemosphere

Stimulation of reductive dechlorination of hexachlorobenzene in soil by inducing the native microbial activity €rfler *, Reiner Schroll, Jean Charles Munch Ferdi Brahushi, Ulrike Do GSF––National Research Center for Environment and Health, Institute of Soil Ecology, Neuherberg 85764, Germany Received 16 April 2003; received in revised form 10 November 2003; accepted 10 January 2004

Abstract The reductive dechlorination and behaviour of 14 C-hexachlorobenzene (HCB) was investigated in an arable soil. The activity of the native anaerobic microbial communities could be induced by saturating the soil with water. Under these conditions high rates of dechlorination were observed. After 20 weeks of incubation only 1% of the applied 14 C-HCB could be detected in the fraction of extractable residues. Additional organic substances, like wheat straw and lucerne straw, however considerably delayed and reduced the dechlorination process in the soil. The decline of HCB was not only caused by dechlorination but also by the formation of non-extractable residues, whereby their amounts varied with time depending on the experimental conditions. Several dechlorination products were detected, indicating the following main HCB transformation pathway: HCB fi PCB fi 1,2,3,5-TeCB fi 1,3,5-TCB fi 1,3-DCB, with 1,3,5-TCB as main intermediate dechlorination product. The other TeCB-, TCB- and DCB-isomers were also detected in low amounts, showing the presence of more than one dechlorination pathway. Since the methane production rates were lowest when the dechlorination rates were highest, it can be assumed that methanogenic bacteria were not involved in the dechlorination process of HCB. The established 14 C-mass balances show, that with increasing dechlorination and incubation times, the 14 C-recoveries decreased.
2004 Elsevier Ltd. All rights reserved. Keywords: Hexachlorobenzene; Reductive dechlorination; Soil; Methane; Non-extractable residues

1. Introduction The commercial production of HCB was banned in Europe and the United States, but it continues to be produced as a byproduct during manufacturing of some solvents and pesticides (Bailey, 2001). Due to its physico-chemical properties and its persistence in the environment HCB is transported over far distances and bioaccumulated. It has adverse health effects on animals

*

Corresponding author. Tel.: +49-89-3187-3477; fax: +4989-3187-3376. E-mail address: doerfl[email protected] (U. D€ orfler).

and humans and is classified as a probable human carcinogen by the US Environmental Protection Agency (ATSDR, 1999). Elimination of HCB from the environment is therefore of high importance. Dechlorination is the key reaction for the degradation of chlorinated benzenes, but mechanisms for the cleavage of the C–Cl bond are not very wide spread in nature (Oliver and Nicol, 1982; Scheunert et al., 1983; Wang et al., 1995). Reductive dechlorination of chlorinated benzenes has been observed in different anaerobic habitats like soil (Rosenbrock et al., 1997), soil slurry (Ramanand et al., 1993), sediment (Bosma et al., 1988; Beurskens et al., 1993; Prytula and Pavlostathis, 1996; Susarla et al., 1996), sewage sludge (Fathepure et al.,

0045-6535/$ - see front matter
2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2004.01.022

1478

F. Brahushi et al. / Chemosphere 55 (2004) 1477–1484

1988; Yuan et al., 1999) and in a variety of anaerobic cultures, that have been enriched from such habitats (Holliger et al., 1992; Masunaga et al., 1996; Nowak et al., 1996; Adrian et al., 1998, 2000; Pavlostathis and Prytula, 2000; Wu et al., 2002). The reductive dechlorination process is related to the presence of electron donors in the system. Fathepure and Vogel (1991) and Adrian et al. (2000) observed a stimulation of the reductive dechlorination of chlorobenzenes when acetate was added whereas Chang et al. (1998) found no effect of acetate on the dechlorination of HCB. Formate, lactate and pyruvate supported the dechlorination process (Nowak et al., 1996; Adrian et al., 1998; Chang et al., 1998). Addition of organic electron donors to the soil enhanced the dechlorination of HCB whereby wheat straw dust gave the best results (Rosenbrock et al., 1997). Substrates as sugars, short fatty acids or complex supplements did not support the growth of the strictly anaerobic bacterium Dehalococoides strain CBDB1 (Adrian et al., 2000; Adrian and G€ orisch, 2002). The influence of alternative electron acceptors on the reductive dechlorination of chlorobenzenes is also reported in the literature. In the presence of nitrate dechlorination of chlorobenzenes has not been observed (Adrian et al., 1998, 2000; Bosma et al., 1988; Yonezawa et al., 1994) or occurred in very low rates (Chang et al., 1998; Yuan et al., 1999). Under sulphate reducing conditions lower rates of dechlorination were detected than under methanogenic conditions (Chang et al., 1998; Yuan et al., 1999). Bosma et al. (1988) reported that the presence of sulphate did not affect the dechlorination process; in contrast Adrian et al. (1998) found that sulphate strongly inhibited the reductive dechlorination of trichlorobenzenes. Some efforts have been made to relate the dechlorination process with the activity of methanogenic bacteria. When the activity of methanogenic bacteria was inhibited Nowak et al. (1996) and Chang et al. (1998) found reduced dechlorination rates, Middeldorp et al. (1997) found an inhibition of some reactions within the dechlorination pathway, and Yuan et al. (1999) reported a significant inhibition of HCB dechlorination. But in literature there are also other findings: the dechlorination of 1,2,4- and 1,2,3-TCB was not affected by the inhibition of methanogenic bacteria activity, indicating, that this group of bacteria is not involved in the dechlorination process (Holliger et al., 1992; Adrian et al., 1998). The dechlorination of HCB in batch cultures, sewage sludge, sediment and soil slurry is well documented, but there exists only few information about the dechlorination of HCB in agricultural soils. Rosenbrock et al. (1997) investigated the dechlorination potential of three different arable soils by determining the dechlorination rates of HCB. The focus of our study was not only

on the estimation of dechlorination rates but also on the identification and quantification of dechlorination products to establish a dechlorination pathway. The use of 14 C-labelled HCB should enable an overall evaluation of the fate of HCB in an agricultural soil. Furthermore, the relation between dechlorination and methane production should be examined, as well as the influence of additional organic matter on the dechlorination process.

2. Materials and methods 2.1. Chemicals Uniformly 14 C-ring-labelled chlorobenzenes, hexachlorobenzene (HCB), 1,2,4-trichlorobenzene (1,2,4TCB) and monochlorobenzene (MCB), purity >98%, specific radioactivity 185 MBq mmol
1 were obtained from International Isotope (Munich, Germany). The non-labelled chlorobenzenes––including HCB, pentachlorobenzene (PCB), monochlorobenzene (MCB) and the isomers of tetrachlorobenzene (TeCB), trichlorobenzene (TCB) and dichlorobenzene (DCB), purity >99.5% were purchased from Dr. Ehrenstorfer (Augsburg, Germany). n-Hexane Picograde was obtained from Promochem (Wesel, Germany). All other solvents were of analytical grade and were purchased from Merck (Darmstadt, Germany). Sodium sulphate (Na2 SO4 ) and sea sand were also obtained from Merck (Darmstadt, Germany). The scintillation cocktails were purchased from Packard (Dreieich, Germany). 2.2. Soil and application procedure The model experiments were performed with soil material from the upper 20 cm of an arable soil from the FAM Research Station Scheyern (40 km north of Munich, Germany). Immediately after sampling, the soil material was sieved (<2 mm) and stored at room temperature for two days prior to starting the dechlorination experiments. The main soil characteristics are as follows: 24% clay, 60% silt, 16% sand, 1.68% organic carbon, 0.18% Ntotal , 15.6 mg kg
1 nitrate, 597 mg kg
1 Fe (EDTA), pH (CaCl2 ) 7.1. A preliminary test showed no detectable amount of HCB in the soil. 14 C-HCB, mixed with unlabelled HCB was dissolved in n-hexane Picograde and applied to an aliquot of 5 g dried and pulverised soil sample in a 50 ml beaker glass. After evaporation of n-hexane the soil aliquot was carefully stirred with a spatula and then transferred to a 200 ml beaker glass where it was mixed with 45 g (dry weight equivalent) of fresh soil. The spiked soil sample was transferred to the 130 ml incubation flask (Fig. 1), adjusted to 100% of water holding capacity (WHC) with an excess of 1 ml of distilled water and tightly closed. The initial amount of 14 C-HCB added to each soil

F. Brahushi et al. / Chemosphere 55 (2004) 1477–1484

1479

glycolmonomethylether and cooled to )10 C to absorb any volatile organic 14 C. The third and fourth trap was filled with 0.1 N NaOH to absorb 14 CO2 from total mineralization (for detailed description see Brahushi et al., 2002). This procedure should ensure that all 14 Cvolatile compounds as well as 14 CO2 , which were formed during the incubation time were removed from the gas phase before the flask was opened and the soil samples were extracted. 2.5. Extraction of the soil samples

Fig. 1. Incubation flask.

sample was 30 lg g
1 soil (dry weight), corresponding to 105.2 nmol g
1 soil, corresponding to 7.56 kBq. 2.3. Experimental design After application of 14 C-HCB, the soil samples in the flasks were incubated at 30 C in the dark. The incubation flasks were equipped with a gas tight inlet allowing gas sampling to measure methane production. To study the influence of additional electron donors on the dechlorination of HCB, the experiments were conducted in three different treatments: (1) soil (without amendment), (2) soil + wheat straw and (3) soil + wheat straw + lucerne straw. Wheat and lucerne straw (<2 mm) were added at a rate of 10 mg g
1 soil, respectively. Each treatment was investigated in seven tests. The total incubation time of the tests was 20 weeks. After 4, 8, 12 and 16 weeks of incubation one test was investigated and after 20 weeks of incubation three tests were used for analysis, thus enabling to study the time course of the dechlorination process. At the end of each incubation time interval methane production, 14 C-volatile compounds, 14 CO2 , extractable 14 C-residues, dechlorination products and non-extractable 14 C-residues were measured. 2.4. Methane production, volatilisation and mineralisation The production of methane was quantified at the end of each incubation interval by sampling 1 cm3 of gas in the headspace above the soil surface with a gas tight needle which was introduced through the inlet of the incubation flask. The gas samples were analysed by a GC-FID (GC 14A, Shimadzu, Duisburg, Germany), according to the analytical procedure published by Loftfield et al. (1997). Then, the incubation flasks were connected to special traps and aerated for 1 h (1l h
1 ) to quantify separately the volatile 14 C-compounds and 14 CO2 . The first two traps were filled with ethylene-

Accelerated solvent extraction (ASE) was used for the extraction of chlorobenzenes from soil, since this method gave the best results compared with soxhlet and column extraction, as tested in preliminary experiments. Prior to extraction the soil samples (50 g dry weight) were mixed with 100 g Na2 SO4 and 50 g sea sand and homogenized in a porcelain mortar. Na2 SO4 and sea sand were purified at 400 C for 4 h. The extraction was performed in an ASE-200 (Dionex, Idstein, Germany) at a temperature of 90 C and a pressure of 10 MPa. Hexane/acetone (3:1 v/v) of analytical grade, were used as extraction solvents. For determination of radioactivity in liquid samples, aliquots of the sample were mixed with a scintillation cocktail (Ultima Gold XR) and measured in a liquid scintillation counter (Tricarb 1900 TR, Packard, Dreieich, Germany). 2.6. Preparation of the samples for GC-analysis For elimination of water from the soil extracts, a liquid–liquid separation step followed the extraction procedure. After the first separation step the acetone– water phase was washed with n-hexane. The combined hexane–acetone phases were filtered through a paper filter (595 12, £ 270 mm, Schleicher & Schuell, Dassel, Germany), which was filled with 10 g Na2 SO4 . The extracts were reduced to a volume of 3–4 ml in a Kuderna–Danish apparatus at a temperature of 70 C. The concentrated extracts were cleaned up by passing them through SPE columns containing 2 g of Florisil (Varian, Darmstadt, Germany). n-Hexane Picograde was used for elution and the first 15 ml were collected in a graduated vial for further analysis. For each chlorobenzene the loss during extraction and clean up procedure was measured and the results were corrected by the respective recovery correction factors. The recoveries varied between 84.1% for HCB and 9.8% for MCB. 2.7. Identification and quantification The analysis of the chlorobenzenes was carried out on a GC-ECD system (Trace GC, 2000 Series, ThermoQuest, Egelsbach, Germany) which was equipped with a DB-5 capillary column (30 m length, 0.32 mm ID

1480

F. Brahushi et al. / Chemosphere 55 (2004) 1477–1484

and 0.25 lm film thickness, J&W Scientific, USA) and an AS 2000 autosampler. On the DB-5 column all chlorobenzenes could be separated with the exception of 1,2,3,5- and 1,2,4,5-TeCB. Temperature program: 60 C for 2 min, 5 C min
1 to 190 C, 20 C min
1 to 280 C, hold for 7 min. Injector temperature: 240 C; detector temperature: 290 C. Helium was used as carrier gas and nitrogen as make-up gas. The injection volume was 1 ll in the splitless mode (75 s). The chlorobenzenes were identified by comparing their retention times with reference substances. The quantification was performed by using linear calibration curves (r2 ¼ 0:99) of the individual chlorobenzenes. The method detection limits were as follows: 765 lg kg
1 soil (d.w.) for MCB, 5–10 lg kg
1 for DCBs; 1.3–2.0 lg kg
1 for TCBs; 0.4–0.5 lg kg
1 for TeCBs; 0.23 lg kg
1 for PCB and 0.14 lg kg
1 for HCB. 2.8. Non-extractable residues and mass balance The residual radioactivity in the soil after extraction (non-extractable residues) was determined by combusting aliquots of the soil samples (Oxidizer 306, Packard, Dreieich, Germany), trapping the evolved 14 CO2 in Carbo-Sorb E (Packard, Dreieich, Germany) and mixing it with Permafluor E (Packard, Dreieich, Germany) prior to liquid scintillation counting. Addition of radioactivity of the 14 C-volatile compounds, the evolved 14 CO2 , the extractable and non-extractable 14 C-residues allowed to establish a 14 C-mass balance.

3. Results 3.1. Volatilisation, mineralisation and methane production Negligible amounts of 14 C-volatile substances and CO2 were trapped at the end of each incubation time in all treatments. Methane was detected after four weeks of incubation in all three treatments (Table 1). Since production of methane occurs only under strictly anaerobic conditions (Tate, 1994) its occurrence is an indicator for the establishment of anaerobic conditions in the incubation system. The results further show that production of methane is related strongly with the amount of carbon sources in the soil. The highest production rates 14

were observed when both, wheat and lucerne straw were added to the soil. In soil without addition of substrates methane production was very low within the first eight weeks, but increased with time up to 58 lg CH4 g
1 soil after 16 weeks, then it slightly decreased until the 20th week of incubation (Table 1). A similar time course could be observed in the treatments soil + wheat straw and soil + wheat straw + lucerne straw whereby the decrease of methane production started already earlier, i.e. within 12 weeks of incubation (Table 1). 3.2. Dechlorination kinetics and dechlorination products The quantity of HCB and the detected dechlorination products for each sampling time and treatment are shown in Table 2. Lower chlorinated benzenes were detected after four weeks of incubation for all treatments; their amounts increased considerably after this time in unamended soil and after eight weeks for the treatment soil + wheat straw. In the treatment soil + wheat straw + lucerne straw their amounts increased after 12 weeks, but in lower rates. These results show that addition of wheat and lucerne straw into the soil was accompanied by a decrease of dechlorination. Between the two amendments there was also a clear difference in the dechlorination rates of HCB: the simultaneous addition of wheat and lucerne straw yielded the lowest dechlorination rates and even after 20 weeks of incubation only a low amount of dechlorination products were detected. The time courses of HCB, 1,3,5-TCB and nonextractable residues are presented in Fig. 2. In unamended soil no significant dechlorination was observed within four weeks of incubation as could be evaluated from the low amount of lower chlorinated products (Table 2). The decline of HCB after four weeks is mostly due to the formation of non-extractable residues (Fig. 2A). After four weeks the dechlorination process occurred very fast resulting in the formation of high amounts of 1,3,5-TCB, which increased from 8 to 12 weeks of incubation and then decreased again. PCB, the TeCB isomers, 1,2,3-TCB, 1,2,4-TCB and the DCB isomers were found in lower amounts (Table 2). The remaining amount of HCB after 20 weeks of incubation in the fraction of extractable residues was about 1% of the applied amount. By addition of wheat straw into the

Table 1 Amounts of methane evolved from HCB spiked soil under anaerobic conditions within different incubation times Treatments Soil, unamended Soil + wheat straw Soil + wheat straw + lucerne straw

CH4 production (lg CH4 g
1 soil) 4 weeks

8 weeks

12 weeks

16 weeks

20 weeks

0.008 17.2 190

0.516 292 570

4.6 413 1163

58 361 1019

52 ± 13.4 317 ± 29.3 881 ± 40.3

Table 2 Formation of dechlorination products (nmol) during anaerobic incubation of an arable soil spiked with HCBa Chlorobenzenes

Soil, unamended

Soil + wheat straw

Soil + wheat straw + lucerne straw

Incubation time (week) 4

a

8

– – – – – –

12

16

20

4

8

12

16

– – 3037.3 12.6 – 4.8

44.2 ± 12.8 5.2 ± 9.0 2.3 ± 3.9 2901.4 ± 74.8 7.9 ± 4.2 0.7 ± 1.2 2.4 ± 0.6

– – – –

– – –

– – –

– – –

1.0

6.2

28.0

64.6

– – 1798.6 ± 246.8 4.7 ± 4.1 – 53.7 ± 17.6

1.0 4.9 3115.7 283.5

2.3 ± 1.2 1.3 ± 0.3 2967.5 ± 73.5 49.5 ± 4.5

0.7 11.7 14.3 4004.0

1.4 20.5 49.0 3926.4

0.4 30.0 350.9 3301.1

1.4 96.5 302.1 3290.7

0.7 ± 1.2 86.7 ± 23.4 1996.3 ± 239.0 1624.4 ± 331.8

19.9

89.4

55.0

1.0

– – 1347.2 1.8 – 70.0

– – 3941.4 15.0 1.0 4.3

0.4 7.8 9.2 3940.0

1.3 87.7 1527.9 2606.7

0.4 5.5 4057.0 611.5

20.8 0.9

291.7 0.9

– –

–

20 51.9 ± 47.7

139.6 – –

–

4

8

12

– – – – – –

– – – – – –

– – –

16

20 86.1

55.2 ± 25.1

– – 12.5

– –

– 3.5 ± 6.0 2.2 ± 3.8 0.9 ± 1.6

11.9 – –

–

1.2

1.3

1.2

2.5

0.8 11.7 13.6 3944.7

1.7 25.6 28.6 4437.7

0.4 13.2 27.3 2235.6

6.3 46.8 153.5 2757.9

4.5 ± 0.8 1.3 ± 1.3 55.6 ± 23.8 123.2 ± 60.7 4235.8 ± 178.1

Initial amount of applied HCB was 5260 nmol. DPs ¼ sum of dechlorination products.

bP

Residues (% of appl. amount) 100

90

80

70

60

50

40

30

20

10

0

0

0

0

90

80

70

60

50

40

30

20

10

100

(A)

Residues (% of appl. amount) 0

90

80

70

60

50

40

30

20

10

0

100

(B)

(C)

4

4

4

8 16

HCB 1,3,5 TCB non extr. residues

12

16

HCB 1,3,5 TCB non extr. residues

12

Time (week)

8

16

HCB non extr. residues 1,3,5 TCB

12

Time (week)

8

Time (week)

1481

20

20

20

Fig. 2. Residues of HCB, 1,3,5-TCB and non-extractable residues during dechlorination of HCB under anaerobic conditions: (A) unamended soil, (B) soil + wheat straw and (C) soil + wheat straw + lucerne straw.

Residues (% of appl. amount)

F. Brahushi et al. / Chemosphere 55 (2004) 1477–1484

1,3-DCB 1,4-DCB 1,2-DCB 1,3,5-TCB 1,2,4-TCB 1,2,3-TCB 1,2,3,5- + 1,2,4,5-TeCB 1,2,3,4-TeCB PCB P DPsb HCB

1482

F. Brahushi et al. / Chemosphere 55 (2004) 1477–1484

soil the dechlorination process was delayed, only after 12 weeks of incubation appreciable amounts of dechlorination products were detected, mainly 1,3,5-TCB, with highest amounts after 20 weeks. At the end of incubation time HCB was found in an amount of 31% of the applied amount in the fraction of extractable residues (Fig. 2B). Dechlorination of HCB was considerably impeded in the presence of both, wheat and lucerne straw. Only after 16 and 20 weeks of incubation appreciable amounts of dechlorination products could be detected. PCB and 1,3-DCB were observed as main metabolites (Table 2). The time course of HCB was different from that in the other treatments and was strongly related with the formation of non-extractable residues (Fig. 2C). Increasing amounts of non-extractable residues during the first 12 weeks resulted in a decrease of HCB residues in the extracts. But after 12 weeks the amount of parent HCB increased obviously due to release of non-extractable residues. After 20 weeks up to 81% of applied HCB remained in the fraction of extractable residues. 3.3. Non-extractable residues and mass balance The non-extractable residues in unamended soil were 24% of the applied amount after four weeks of incubation and decreased continuously with incubation time till to 5% after 20 weeks (Fig. 2A). In the treatment soil + wheat straw the non-extractable residues were about 25% during the first 16 weeks and then decreased to 7% (Fig. 2B). In soil + wheat straw + lucerne straw the non-extractable residues increased during the first 12 weeks up to 47% and then decreased to 10% after 20 weeks (Fig. 2C). 14 C-mass balances were different

Soil, unamended Soil + wheat straw Soil + wheat straw + lucerne

Total residues (% of applied radioactivity)

120 100 80 60 40 20 0 4

8

12

16

20

Time (week)

Fig. 3. 14 C-mass balances of applied 14 C-HCB after different treatments and at different incubation times.

depending on the incubation times and the treatments (Fig. 3). In general, with higher extent of dechlorination and longer incubation times the 14 C-mass balances decreased.

4. Discussion In unamended soil the dechlorination rates of HCB were highest when methane was produced only in small amounts; when methane production was intensive, most of the HCB had already been dechlorinated. This means that the dechlorination of HCB already occurred before the optimal redox conditions for methane production were reached. Thus, indicating the participation of nitrate reducing or sulphate reducing microorganisms or fermentative processes on the degradation of HCB rather than methanogenic bacteria. This assumption is confirmed by the fact, that in the presence of methanogenic bacteria the HCB dechlorination was significantly reduced, as was observed in the soils with additional organic substrates. Hence, our results support the findings of Holliger et al. (1992), Rosenbrock et al. (1997) and Adrian et al. (1998) which indicated that methanogenic bacteria are not involved in the dechlorination process. The reductive dechlorination of HCB in soil started after an acclimation period of at least four weeks. In literature different acclimation periods for the dechlorination of HCB are reported. In organic materials as sewage sludge or in a methanogenic mixed culture enriched from river sediment lower lag phases of two days or one week for the dechlorination of HCB were observed (Fathepure et al., 1988; Nowak et al., 1996; Yuan et al., 1999), whereas in soil slurry dechlorination of HCB started after an acclimation period of two months (Ramanand et al., 1993). The observed lag time in our study could be explained with the fact that we used an arable soil with no detectable amount of HCB and therefore dechlorinating bacteria needed some time for adaptation to reach a required density for dechlorinating HCB in an extent that the dechlorination products could be observed in measurable amounts. Additional carbon sources and electron donors in the soil resulted in reduced dechlorination rates. This negative effect of wheat straw and lucerne straw on the dechlorination process could be explained on one side with the increased activity of methanogenic bacteria. Fresh added organic substances as wheat and lucerne straw appear to be used easily by methanogenic bacteria resulting in the observed high production rates of methane. On the other side, the transformation of inherent organic matter and fresh added organic matter can result in different intermediate organic substances which can serve as electron donors or electron acceptors. Furthermore, inorganic substances which are formed during

F. Brahushi et al. / Chemosphere 55 (2004) 1477–1484

the anaerobic degradation of organic matter can also serve as electron acceptors. When different groups of electron acceptors––organic substances, inorganic substances, chlorobenzenes––are present in the soil, it will depend on the dominant anaerobic microbes, which substance will preferably be used as electron acceptor. In literature it has also been reported that additional organic substances had no or adverse effects on the dechlorination of chlorobenzenes. Substrates as sugars, short fatty acids or complex supplements did not support the growth of the dechlorinating bacterium Dehalococoides strain CBDB1 (Adrian et al., 2000; Adrian and G€ orisch, 2002) and propionate inhibited the dechlorination of HCB (Nowak et al., 1996). According to the detected amounts of dechlorination products in our study, the predominant transformation pathway was as follows: the reductive dechlorination of HCB occurred via PCB and 1,2,3,5-TeCB resulting in 1,3,5-TCB as the main dechlorination product and 1,3DCB. Further dechlorination of DCB isomers to MCB as has been reported (Ramanand et al., 1993; Masunaga et al., 1996; Nowak et al., 1996) was not observed. However, this is not necessarily a prove for the absence of this dechlorination step since the detection limit of our method for MCB was very high. From thermodynamical reasons the removal of chlorine substituents which have on both sides adjacent chlorine atoms is more favourable than the removal of a chlorine substituent which has only one neighbouring chlorine. The main pathways for HCB dechlorination that are reported in literature (Fathepure et al., 1988; Holliger et al., 1992; Beurskens et al., 1993; Ramanand et al., 1993; Masunaga et al., 1996; Nowak et al., 1996; Susarla et al., 1996; Chang et al., 1998; Yuan et al., 1999; Pavlostathis and Prytula, 2000; Wu et al., 2002) show therefore removal of chlorine atoms located between two others or in ortho position to another chlorine substituent. Reductive dechlorination of a chlorine substituent in meta position yields the lowest energy. Therefore, 1,3,5TCB is considered by several authors as a dead end product of the reductive HCB dechlorination (Fathepure et al., 1988; Ramanand et al., 1993). But Chang et al. (1998) and Nowak et al. (1996) found a further dechlorination of 1,3,5-TCB to 1,3-DCB. Yuan et al. (1999) also reported the removal of ortho-, meta- and para-chlorines during HCB transformation. The decrease of the amount of 1,3,5-TCB with longer incubation times and the presence of dichlorobenzenes in our study supports the theory, that 1,3,5-TCB can further be dechlorinated. Furthermore, the occurrence of all TeCB-, TCBand DCB-isomers indicates the presence of more than one dechlorination pathway for HCB, which is in accordance with the findings in literature (Fathepure et al., 1988; Ramanand et al., 1993; Masunaga et al., 1996). Since we used 14 C-labelled HCB in our dechlorination study, we were able to determine the non-extract-

1483

able residues as well as to establish a mass balance. The term non-extractable residues includes HCB and its degradation products tightly bound to or incorporated into the soil matrix or the microbial biomass. These kind of residues are considered to be not bioavailable for microorganisms, in this case not available for the dechlorination process. Our results show that the nonextractable residues represent a significant part in the fate of HCB in soil under anaerobic conditions. But the portion of the non-extractable residues was not stable and increased and decreased with time depending on the experimental conditions, indicating that these residues are released during the normal transformation processes in soil and could be again subject for dechlorination. The nature of these residues is yet not known, since we had no possibility to characterise them; further research is therefore needed to elucidate these processes. The established mass balances showed that the increase of the dechlorination process resulted in lower 14 C-recoveries. This could clearly be seen in unamended soil where high dechlorination rates were observed. Though the dechlorination rates were high over the whole incubation period, the 14 C-recoveries were quite good until the 12th week. After that time no further accumulation of the main metabolite 1,3,5-TCB occurred, but the concentration of this dechlorination product was reduced, accompanied by a drastic decrease of the 14 C-recoveries. Therefore it can be assumed, that 1,3,5-TCB was further dechlorinated to dichlorobenzenes and monochlorobenzene, as was already supposed above. These low chlorinated benzenes are very volatile and could be lost from the system during the incubation time––as no radioactivity was found in the traps at the end of the incubation intervals––thus resulting in poor 14 C-recoveries. Prytula and Pavlostathis (1996) also found low mass balances of chlorobenzenes with increasing incubation times. Even during a short incubation time of 14 days about 15% of low chlorinated benzenes, especially dichlorobenzenes were lost (Pavlostathis and Prytula, 2000). Nevertheless, the establishment of 14 C-mass balances gave in a first approach valuable information about the fate of HCB in an arable soil but it also revealed the enervations of the experimental set up that can be improved in future experiments, to verify the above mentioned assumptions. Previous studies reported that indigenous microorganisms originating from contaminated sites could dechlorinate HCB (Holliger et al., 1992; Beurskens et al., 1993; Prytula and Pavlostathis, 1996). Our results show that a native microbial population of an arable soil was also able to transform HCB in soil via reductive dechlorination, as was already indicated by the study of Rosenbrock et al. (1997). This means that microorganisms with capability for dechlorination were present in the investigated agricultural soil and their activity could be induced by establishing anaerobic conditions.

1484

F. Brahushi et al. / Chemosphere 55 (2004) 1477–1484

Moreover, the dechlorination pathway was elucidated and the use of 14 C technique gave valuable information about the overall behaviour of HCB in an arable soil. Since the nature of the organic matter in the soil strongly influenced the dechlorination capacity, further research is necessary to investigate the interrelationship of different soil types, different additional carbon sources and the dechlorination process.

References Adrian, L., G€ orisch, H., 2002. Microbial transformation of chlorinated benzenes under anaerobic conditions. Res. Microbiol. 153, 131–137. Adrian, L., Manz, W., Szewzyk, U., G€ orisch, H., 1998. Physiological characterization of a bacterial consortium reductively dechlorinating 1,2,3- and 1,2,4-trichlorobenzene. Appl. Environ. Microbiol. 64, 496–503. Adrian, L., Manz, W., Szewzyk, U., G€ orisch, H., 2000. Bacterial dehalorespiration with chlorinated benzenes. Nature 408, 580–583. ATSDR, 1999. ATSDR Public Health Statement: Hexachlorobenzene. Agency for Toxic Substances and Disease Registry, Division Toxicology, Atlanta, GA. Bailey, R.E., 2001. Global hexachlorobenzene emissions. Chemosphere 43, 167–182. Beurskens, J.E.M., Dekker, C.G.C., Jonkhoff, J., Pompstra, L., 1993. Microbial dechlorination of hexachlorobenzene in a sedimentation area of Rhine River. Biogeochemistry 19, 61– 81. Bosma, T.N.P., van der Meer, J.R., Schraa, G., Tros, M.E., Zehnder, A.J.B., 1988. Reductive dechlorination of all trichloro- and dichlorobenzene isomers. FEMS Microbiol. Ecol. 53, 223–229. Brahushi, F., D€ orfler, U., Schroll, R., Feicht, E., Munch, J.C., 2002. Environmental behavior of monochlorobenzene in an arable soil. Fresen. Environ. Bull. 11, 599–604. Chang, B.-V., Su, C.-J., Yan, S.-Y., 1998. Microbial hexachlorobenzene dechlorination under three reducing conditions. Chemosphere 13, 2721–2730. Fathepure, B.Z., Vogel, T.M., 1991. Complete degradation of polychlorinated hydrocarbons by a two-stage biofilm reactor. Appl. Environ. Microbiol. 55, 3418–3422. Fathepure, B.Z., Tiedje, J.M., Boyd, S.A., 1988. Reductive dechlorination of hexachlorobenzene to tri- and dichlorobenzenes in anaerobic sewage sludge. Appl. Environ. Microbiol. 54, 327–330. Holliger, H., Schraa, G., Stams, A.J.M., Zehnder, A.J.B., 1992. Enrichment and properties of an anaerobic mixed culture reductively dechlorinating 1,2,3-trichlorobenzene to 1,3dichlorobenzene. Appl. Environ. Microbiol. 58, 1636–1644. Loftfield, N., Flessa, H., Augustin, J., Beese, F., 1997. Automated gas chromatographic system for rapid analysis

of the atmospheric trace gases methane, carbon dioxide, and nitrous oxide. J. Environ. Qual. 26, 560–564. Masunaga, S., Susarla, S., Yonezawa, Y., 1996. Dechlorination of chlorobenzenes in anaerobic estuarine sediment. Water Sci. Technol. 33, 173–180. Middeldorp, P.J.M., de Wolf, J., Zehnder, A.J.B., Schraa, G., 1997. Enrichment and properties of a 1,2,4-trichlorobenzene-dechlorinating methanogenic microbial consortium. Appl. Environ. Microbiol. 63, 1225–1229. Nowak, J., Kirsch, N.H., Hegemann, W., Stan, H.J., 1996. Total reductive dechlorination of chlorobenzenes to benzene by methanogenic mixed culture enriched from Saale river sediment. Appl. Microbiol. Biotechnol. 45, 700– 709. Oliver, B.G., Nicol, K.D., 1982. Chlorobenzenes in sediments, water, and selected fish from Lakes Superior, Huron, Erie and Ontario. Environ. Sci. Technol. 16, 532–536. Pavlostathis, S.G., Prytula, M.T., 2000. Kinetics of the sequential microbial reductive dechlorination of hexachlorobenzene. Environ. Sci. Technol. 34, 4001–4009. Prytula, M.T., Pavlostathis, S.G., 1996. Effect of contaminant and organic matter bioavailability on the microbial dehalogenation of sediment-bound chlorobenzenes. Water Res. 30, 2669–2680. Ramanand, K., Balba, M.T., Duffy, J., 1993. Reductive dehalogenation of chlorinated benzenes and toluenes under methanogenic conditions. Appl. Environ. Microbiol. 59, 3266–3272. Rosenbrock, P., Martens, R., Buscot, F., Munch, J.C., 1997. Initiation of [36 Cl]hexachlorobenzene dechlorination in three different soils under artificially induced anaerobic conditions. Appl. Microbiol. Biotechnol. 48, 115–120. Scheunert, I., Marra, C., Viswanathan, R., Klein, W., Korte, F., 1983. Fate of hexachlorobenzene–14 C in wheat plants and soil under outdoor conditions. Chemosphere 12, 843– 858. Susarla, S., Masunaga, S., Yonezawa, Y., 1996. Reductive dechlorination pathways of chloro organics under anaerobic conditions. Water Sci. Technol. 34, 489–494. Tate III, R.L., 1994. Soil Microbiology, second ed. John Wiley and Sons, New York. Wang, M.-J., McGrath, S.P., Jones, K.C., 1995. Chlorobenzenes in field soil with a history of multiple sewage sludge application. Environ. Sci. Technol. 29, 356–362. Wu, Q., Milliken, C.E., Meier, G.P., Watts, J.E.M., Sowers, K.R., May, H.D., 2002. Dechlorination of chlorobenzenes by a culture containing bacterium DF-1, a PCB dechlorinating microorganism. Environ. Sci. Technol. 36, 3290– 3294. Yonezawa, Y., Fukui, M., Masunaga, S., Urushigawa, Y., 1994. Dechlorination of 1,2,4-trichlorobenzene in the sediment of Ise bay. Chemosphere 28, 2179–2184. Yuan, S.Y., Su, C.J., Chang, B.V., 1999. Microbial dechlorination of hexachlorobenzene in anaerobic sewage sludge. Chemosphere 38, 1015–1023.