Biological treatment of 2,4-dichlorophenol containing synthetic wastewater using a rotating brush biofilm reactor

Available online at www.sciencedirect.com

Bioresource Technology 99 (2008) 2319–2325

Biological treatment of 2,4-dichlorophenol containing synthetic wastewater using a rotating brush biofilm reactor Serkan Eker, Fikret Kargi

*

Department of Environmental Engineering, Dokuz Eylul University, Buca, Izmir, Turkey Received 4 January 2007; accepted 6 May 2007 Available online 26 June 2007

Abstract A newly developed rotating brush biofilm reactor was used for DCP, COD and toxicity removal from 2,4-dichlorophenol (DCP) containing synthetic wastewater at different feed COD, TCP concentrations and A/Q (biofilm surface area/feed flow rate) ratios. A Box– Wilson statistical experiment design was used by considering the feed DCP (50–500 mg l1), COD (2000–6000 mg l1) and A/Q ratio (73–293 m2 d m3) as the independent variables while percent DCP, COD, and toxicity removals were the objective functions. The experimental data were correlated by a quadratic response function and the coefficients were determined by regression analysis. Percent DCP, COD and toxicity removals calculated from the response functions were in good agreement with the experimental data. DCP, COD and toxicity removals increased with increasing A/Q ratio and decreasing feed DCP concentrations. The optimum A/Q ratio resulting in the highest COD (90%), DCP (100%) and toxicity (100%) removals with the highest feed COD (6000 mg l1) and DCP (500 mg l1) contents was nearly 210 m2 d m3. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Biological treatment; Rotating brush biofilm reactor (RBBR); Toxicity removal; 2,4-Dichlorophenol (DCP)

1. Introduction Chlorophenols present in some chemical industry wastewaters such as pulp and paper, pesticides, petrochemicals and plastics cause considerable environmental pollution problems when discharged to surface water sources without proper treatment. Conventional biological treatment processes such as the activated sludge are usually ineffective for removal of highly toxic and recalcitrant chlorophenols. Development of more effective processes for chlorophenol removal is needed for controlled discharge of such toxic compounds. Effluent toxicity levels should also be controlled along with the other conventional parameters. Different physical, chemical and biological methods such as activated carbon adsorption, chemical oxidation and aerobic/anaerobic biological treatment were used for removal of chlorophenols from industrial wastewater *

Corresponding author. Tel.: +90 232 4127109. E-mail address: fi[email protected] (F. Kargi).

0960-8524/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2007.05.016

(Armenante et al., 1999; Atuanya et al., 2000; Jung et al., 2001). Physicochemical methods are usually used to concentrate the chlorophenols and do not result in complete mineralization. Chemical or biological oxidation methods are much more effective for complete mineralization of chlorophenols usually in combination. Chemical oxidation methods usually result in formation of undesirable by products and also are expensive. Biodegradation of chlorophenols are more specific, relatively inexpensive and can be realized under aerobic and anaerobic conditions (Annachhatre and Gheewala, 1996; Armenante et al., 1999; Atuanya et al., 2000; Bali and Sengul, 2002). Mostly suspended pure cultures of different organisms were used for biodegradation of chlorophenols (Dapaah and Hill, 1992; Hill et al., 1996; Yee and Wood, 1997; Steinle et al., 1998; Fahr et al., 1999; Kim and Hao, 1999; Wang et al., 2000; Farrell and Quilty, 2002; Kargi and Eker, 2004, 2005a). A carbohydrate substrate was used as the primary metabolite and the chlorophenols as the cometabolite in

2320

S. Eker, F. Kargi / Bioresource Technology 99 (2008) 2319–2325

Nomenclature A total biofilm surface area of the brushes (m2) Q flow rate of wastewater (m3 h1) COD0 chemical oxygen demand in the feed wastewater (mg l1); DCP0 dichlorophenol (DCP) concentration in feed wastewater (mg l1) ECOD percent removal of COD (%) most of the chlorophenol degradation studies (Hill et al., 1996; Kim and Hao, 1999). Limited number of studies was reported on biological treatment of chlorophenol containing wastewaters (Sahinkaya and Dilek, 2002; Eker and Kargi, 2006a,b; Kargi and Eker, 2006). Pre-adaptation of the activated sludge cultures to the chlorophenols was reported to improve the rate and the extent of biodegradation of those compounds (Bali and Sengul, 2002; Sahinkaya and Dilek, 2002). Recent investigations on biodegradation of chlorophenols focused on the use of immobilized cells or biofilm reactors (Shieh et al., 1990; Radwan and Ramanujam, 1996; Swaminathan and Ramanujam, 1998; Shin et al., 1999; Kim et al., 2002; Kargi and Eker, 2005b; Eker and Kargi, 2006c). Suspended culture systems such as conventional activated sludge are easier to control, but not resistant to shock loadings of toxic compounds. Due to high biomass concentrations, biofilm reactors are more resistant to high chlorophenol concentrations and therefore are more effective than the suspended culture systems in treatment of such wastewaters (Eker and Kargi, 2006c; Kargi and Eker, 2005b). Biodegradability of chlorophenols decreases and toxicities increase with increasing number of chlorine groups (Annachhatre and Gheewala, 1996). Various biological tests were developed for toxicity assessment of toxic chemicals and effluents (Liu, 1986; Brouwer, 1991; Strotmann et al., 1993; Farre and Barcelo, 2003). One of the effective methods for toxicity measurement is the ‘Resazurin Assay’ which is relatively simple, inexpensive and rapid (Liu, 1986; Brouwer, 1991; Strotmann et al., 1993). The basic principle

EDCP percent removal of dichlorophenol (%) EToxicity percent removal of toxicity (%) LDCP DCP loading rate (Q Æ DCP0, g DCP h1) LCOD COD loading rate (Q Æ COD0, g COD h1) RBBR rotating brush biofilm reactor HH hydraulic residence time (HRT, V/Q) (h)

of the method is the measurement of percent inhibition on dehydrogenase activity of bacteria in the presence of toxic compounds. Toxicity values obtained with the Resazurin assay are comparable to those obtained with the more commonly used biological methods such as Daphnia magna, and Microtox TM (Farre and Barcelo, 2003). The major objective of this study is to investigate performance of a newly developed biofilm reactor namely the ‘rotating brush biofilm reactor (RBBR)’ for biological treatment of 2,4-dichlorophenol (DCP) containing synthetic wastewater and compare the performance with the rotating tubes biofilm reactor (RTBR) under the same experimental conditions. COD, dichlorophenol (DCP) and toxicity removals from the synthetic wastewater were investigated under different operating conditions such as the variable feed COD (2000–6000 mg l1), feed DCP (50–500 mg l1) and the A/Q ratio (73–293 m2 d m3). A Box–Wilson statistical experiment design method was used to investigate the effects of the operating parameters on percent COD, DCP and toxicity removals. 2. Methods 2.1. Experimental system Fig. 1 depicts a schematic diagram of the rotating brush biofilm reactor (RBBR). The experimental system consisted of a feed reservoir, wastewater tank containing a battery of brushes, driving motor, shaft and wastewater pump. The vertical discs containing the battery of brushes were

A

Shaft motor B

Air pump Vertical discs

A

Rotating Brush

Fig. 1. Schematic diagram of the rotating brush biofilm reactor used in experimental studies.

A-A plan

S. Eker, F. Kargi / Bioresource Technology 99 (2008) 2319–2325

rotated by using a motor and a shaft passing through the central hole on the discs. Rods containing brushes were mounted on the discs through the holes on disc surfaces. The system was placed in a stainless steel reactor in the shape of half a barrel with dimensions of 60 cm length, 30 cm width (or diameter) and 20 cm depth open to atmosphere. The feed reservoir was placed in a deep refrigerator to keep the temperature below 5 °C to avoid any decomposition. The system had two sections mounted on the same shaft each having 12 brush rods with total of 24 brush rods of length L = 25 cm and diameter of D0 = 2.1 cm. Each brush rod contained 4200 bristles of length 0.9 cm and diameter of 0.6 mm yielding a total surface area of A = 2.11 m2 including the brush and rod surface areas on 24 tubes. Part of the brushes was completely immersed in the wastewater tank during rotation and part of them was in direct contact with air. Organisms grow in form of biofilm on the surfaces of the brushes and rod surfaces. Total liquid volume in the tank was VL = 12 l. Therefore, the biofilm area per unit wastewater volume in the tank was a = 176 m2 m3. Biomass concentration on the rod and brush surfaces in form of biofilm was approximately 42 ± 1 g m2 and the suspended biomass concentration in the tank was around 3.0 ± 0.2 g l1. 2.2. Wastewater composition Synthetic wastewater used throughout the studies was composed of diluted molasses, urea, KH2PO4 and MgSO4 resulting in COD/N/P = 100/8/1.5 in the feed wastewater. MgSO4 concentration in the feed was 50 mg l1 in all experiments. COD and TCP concentrations in the feed wastewater were adjusted to desired levels specified by the Box–Wilson experimental design method. COD content of the feed included COD content of DCP (1.18 g COD/g DCP) along with the COD content of diluted molasses. 2.3. Organisms The activated sludge culture used for inoculation was obtained from the Cigli municipal wastewater treatment plant in Izmir, Turkey. The culture was cultivated for several days in growth media containing diluted molasses, urea, KH2PO4, MgSO4 and 50 mg l1 DCP on a shaker at 200 rpm and 25 °C and was used for inoculation of the rotating brush biofilm reactor. 2.4. Experimental procedure About 10 l of the synthetic wastewater was placed in the treatment tank containing the battery of rotating brushes and was inoculated with 2 l of the sludge culture. The system was operated batch-wise for nearly two weeks by changing the wastewater media in every three days until a biofilm thickness of 1 mm was developed on the surfaces of the rods and brushes. Continuous operation was started

2321

after biofilm development. Feed wastewater was fed to the reactor with a desired flow rate between 0.3 and 1.2 l h1 resulting in hydraulic residence times (HRT) between 10 and 40 h or A/Q ratios between 73 and 293 m2 d m3 and removed with the same flow rate. A/Q ratio was changed by changing the feed flow rate while the biofilm surface area was constant (2.11 m2) throughout the experiments. Temperature and pH were approximately T = 23 ± 3 °C and pH = 7.0 ± 0.2 during operation. pH in the feed medium was nearly 6.9 which increased to pH > 7 due to ammonia release from urea biodegradation. pH of the reactor media was controlled around 7.0 by manual addition of dilute sulfuric acid to the reactor several times a day. Biofilm thickness was controlled around 1 mm by removing excess biofilm from the surfaces of the brushes. The liquid phase in the treatment tank was aerated using air pump and diffusors. Aeration was supplied to the biofilm by direct contact of the brushes with air during rotation. Dissolved oxygen (DO) concentration in the wastewater tank was above 2 mg l1 indicating no DO limitations. The brushes were rotated with a constant speed of 12 rpm. Experiments were performed in the order of increasing DCP loading rates (LDCP = Q Æ DCP0) to allow adaptation of the organisms to high concentrations of DCP. Every experiment was conducted until the system reached the steady-state with the same COD and TCP contents in the effluent for the last three days. Each experiment lasted about three to four weeks to reach the quasi steadystate. The samples collected from the feed and effluent wastewater at the steady-state were analyzed for COD, DCP contents and for the toxicity after centrifugation. 2.5. Analytical methods Samples were withdrawn everyday for analysis and centrifuged at 8000 rpm (7000 g) for 20 min to remove biomass from the liquid phase. Clear supernatants were analyzed for DCP contents by using 4-aminoantipyrene colorimetric method developed for determination of phenol and derivatives in form of phenol index as specified in the Standard Methods (Greenberg et al., 2005). COD was determined using the dichromate reflux method according to the Standard Methods (Greenberg et al., 2005). Biomass concentrations were determined by filtering the samples through 0.45 lm milipore filter and drying in an oven at 105 °C until constant weight. The samples were analyzed in triplicates for COD and DCP contents with less than 5% standard deviations from the average. Resazurin reduction method was used to determine the toxicity of the feed and effluent wastewater (Liu, 1986; Brouwer, 1991; Strotmann et al., 1993). The test organisms (washed activated sludge) to be subjected to the toxic feed and effluent wastewater were cultivated on nutrient broth and were used for determination of the toxicity of wastewater samples. The test cultures were transferred to a new medium every day to keep the sludge age constant throughout the toxicity measurements. In the presence of active

2322

S. Eker, F. Kargi / Bioresource Technology 99 (2008) 2319–2325

bacterial culture, as a result of dehydrogenase enzyme activity, resazurin changes color from blue to pink forming the reduced compound resorufin. Therefore, the color of the resazurin solution is an indicator of bacterial activity. A spectrometer was used at 610 nm for determination of the color intensity of the resazurin added samples.

response functions for COD, 2,4-DCP and toxicity removals were approximated by the following quadratic polynomial equation: Y ¼ b0 þ b1 X 1 þ b2 X 2 þ b3 X 3 þ b12 X 1 X 2 þ b13 X 1 X 3 þ b23 X 2 X 3 þ b11 X 21 þ b22 X 22 þ b33 X 23

ð2Þ

2.6. Box–Wilson statistical experiment design 3. Results and discussion Box–Wilson statistical experiment design method was used to determine the effects of operating parameters such as A/Q ratio, feed COD and TCP concentrations on percent COD, TCP and toxicity removals. Three important operating parameters; feed DCP0 (X1) and COD0 (X2) concentrations and A/Q ratio (X3) were considered as independent variables. Feed DCP0 (X1) was varied between 50 and 500 mg l1 while the feed COD0 (X2) was between 2000 and 6000 mg l1 and the A/Q ratio (X3) was between 73 and 293 m2 d m3 resulting in HRT values between 10 and 40 h. Response functions describing variations of dependent variables (percent COD, TCP and toxicity removals) with the independent variables (Xi) can be written as follows: X zfflfflffl}|fflfflffl{ X zfflfflfflfflfflfflfflffl}|fflfflfflfflfflfflfflffl{ X zfflfflffl}|fflfflffl{ bi  X i þ bij  X i  X j þ bii  X 2i ð1Þ Y ¼ b0 þ |fflfflffl{zfflfflffl} |fflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflffl} |fflfflffl{zfflfflffl} Linear

squared

interaction

where Y is the predicted response, b0 is offset term, bi is the linear effect while bii and bij are the square and the interaction effects, respectively. Experimental data points used in Box–Wilson statistical design are presented in Table 1. Some experiments were repeated twice and the center point was repeated three times. The response function coefficients were determined by correlating the experimental data with the response function using the Statistica 5.0 computer program. The Table 1 Experimental data points used in the Box–Wilson statistical design COD (mg l1)

HRT (h)

A/Q (m2 d m3)

4000 4000 2000 6000 4000 4000

25 25 25 25 10 40

183 183 183 183 73 293

Factorial points F1 405 F2 405 F3 405 F4 405 F5 145 F6 145 F7 145 F8 145

5156 5156 2844 2844 5156 5156 2844 2844

34 16 34 16 34 16 34 16

249 117 249 117 249 117 249 117

Center point C 275

4000

25

183

DCP (mg l Axial A1 A2 A3 A4 A5 A6

points 50 500 275 275 275 275

1

)

Experimental data was used for determination of the response function coefficients for each independent variable by iteration. The estimated coefficients of the response functions are presented in Table 2. Predicted values of the response functions using the estimated coefficients are compared with the experimental results in Table 3. Response function predictions were in good agreement with the experimental data. Variations of percent COD removal with the feed COD and DCP concentrations at constant A/Q ratio of 70 m2 d m3 are depicted in Fig. 2. Percent COD removal decreased with increasing feed DCP concentration from 50 to 500 mg l1 for all feed COD concentrations due to toxic effects of high DCP concentrations. Percent COD removal increased with increasing feed COD at low feed DCP contents (DCP < 300 mg l1) because of lack of DCP inhibition, but remained almost constant for all feed COD’s at high feed DCP contents above 300 mg l1. The highest COD removal (93%) was obtained with the feed COD and DCP of 6000 and 50 mg l1, respectively when A/Q ratio was 70 m2 d m3. Fig. 3 shows variations of percent COD removal with the A/Q ratio at different feed COD concentrations when the feed DCP was 225 mg l1. COD removal increased with increasing feed COD due to substrate limitation at low feed COD contents. Percent COD removal increased with increasing A/Q ratio and reached a maximum level at A/Q ratio of around 200 m2 d m3 indicating biofilm surface area (or biomass concentration) limitations at low A/Q ratios. Further increases in A/Q ratio resulted in decreases in percent COD removal probably as a result of low feed flow rates and insufficient COD loading rates to support high biomass concentrations. The optimal A/Q ratio was nearly 200 m2 d m3 below which COD removal was limited by biomass concentration (or biofilm surface area) and above which the limitation was due to insufficient substrate loading or low feed flow rates. The highest COD removal (97%) was obtained with the feed COD0 of 6000 mg l1 and A/Q ratio of 200 m2 d m3 when the feed DCP was 225 mg l1 yielding an optimal COD loading rate of 30 g COD m2 d1. Variations of percent DCP removal with the feed DCP concentration at different A/Q ratios and constant feed COD of 4000 mg l1 are depicted Fig. 4. Percent DCP removal increased with increasing A/Q ratio up to 200 m2d m3 due to biofim area limitations at low A/Q ratios and then decreased with further increases in A/Q

S. Eker, F. Kargi / Bioresource Technology 99 (2008) 2319–2325

2323

Table 2 Coefficients of the response functions for COD, DCP and toxicity removals b1

b3

b11

b22

77.863466 96.635882 76.726095

0.0089574 0.0005358 0.1381144 6.246 * 105 1.572 * 107 0.0004803 3.904 * 106 0.0001918 1.213 * 106 94 0.0534779 0.0052714 0.1736781 6.68 * 105 5.871 * 107 0.0004184 2.868 * 106 1.524 * 105 1.591 * 107 90 5 0.0560679 0.0047273 0.3804252 2.041 * 10 2.985 * 107 0.0010748 1.403 * 105 0.0001111 2.188 * 107 59

Table 3 Observed and predicted COD, DCP and toxicity removal efficiencies EDCP

ECOD

EToxicity

Observed Predicted Observed Predicted Observed Predicted A1 A2 A3 A4 A5 A6 F1 F2 F3 F4 F5 F6 F7 F8 Center (avg)

93 89 93 96 86 90 92 89 93 89 91 93 89 92 94

92 90 94 95 87 89 92 88 93 88 92 94 90 92 94

100 100 99 99 87 96 99 97 97 95 99 97 99 97 97

100 100 99 100 89 95 99 96 98 94 99 95 99 96 97

100 97 96 96 70 97 99 97 86 96 97 96 97 96 98

95

-1

Percent COD removal

DCP (mg l ) 90

50

100 98 96 99 81 90 99 92 91 90 99 89 99 94 99

400 500 75 -3

A/Q=70 m .d m

6000

CODo (mgl-1) Fig. 2. Variation of percent COD removal with the feed COD at different feed DCP concentrations and constant A/Q ratio of 70 m2 d m3.

Percent COD removal

96 6000

93

5000 3000 4000

90

2000 87

-1

COD (mg l ) DCP= 225mg l

-1

84 70

110

150

190 2

230

-3

95

100

90

70 -1

COD= 4000 mg l

100

150

200

250

300

350

400

450

500

Fig. 4. Variations of percent DCP removal with the feed DCP concentration at different A/Q ratios. COD0 = 4000 mg l1.

80

5000

2

200 A/Q (m .d m ) 150 290

-1

300

4000

b23

DCPo (mg l )

85

3000

b13

100

50

100

2

b12

85

200

70 2000

b33

Percent DCP removal .

YCOD YDCP YToxicity

b2

R2

b0

270

-3

A/Q (m d.m ) Fig. 3. Variations of percent COD removal with the A/Q ratio at different feed COD concentrations. DCP0 = 225 mg l1.

ratio due to insufficient substrate loading to support high biomass concentrations. The optimal A/Q ratio resulting in the highest DCP removal was nearly 200 m2 d m3 with the feed COD of 4000 mg l1. Percent DCP removal decreased with increasing feed DCP contents up to 300 mg l1 due to toxic effects of high DCP. Further increases in the feed DCP above 300 mg l1 resulted in increases in DCP removal probably due to domination of DCP degrading organisms or adaptation of the organisms since the experiments were performed in the order of increasing DCP loading rates. Fig. 5 depicts variation of percent DCP removals with the A/Q ratio at different feed COD concentrations and a constant feed DCP of 225 mg l1. Percent DCP removal increased with increasing A/Q ratio up to 190– 200 m2 d m3 due to biofilm surface area (or biomass concentration) limitations at low A/Q ratios. Further increases in A/Q ratio resulted in decreases in percent DCP removal because of insufficient substrate loadings at low flow rates and high biomass concentrations. Increases in the feed COD from 2000 to 4000 mg l1 resulted in decreases in percent DCP removal due to preferential utilization of COD compounds (mainly sucrose) present in dilute molasses instead of DCP. Further increases in the feed COD above 4000 mg l1 resulted in higher percent DCP removals probably because of high biomass concentrations at high feed COD contents and also because of low feed DCP/COD ratios obtained at high feed COD contents. The highest percent DCP removal (100%) was obtained with the feed COD of 6000 mg l1 and an A/Q ratio of 200 m2 d m3 with a feed DCP of 225 mg l1 yielding the optimal DCP and COD loading rates of 30 g COD m2 d1 and 1.2 g DCP m2 d1 with an optimal feed DCP/COD ratio of 3.8%.

2324

S. Eker, F. Kargi / Bioresource Technology 99 (2008) 2319–2325 100

98

3000 5000 4000

6000 2000

96

. Percent toxicity removal.

Percent DCP removal .

100

-1

COD (mg l )

94 92 90 88

DCP=225 mg l

-1

6000 5000 4000 3000

90

2000 -1

CO D (mg l ) 80 -1

DCP=225 mg l

70 70

110

150

190 2

230

70

270

110

A/Q (m d.m ) Fig. 5. Variations of percent DCP removal with the A/Q ratio at different feed COD concentrations. DCP0 = 225 mg l1.

Percent toxicity removal.

100

200 150

95 290

90

100

85

2

-3

A/Q (m .d m )

80

70 -1

COD= 4000 mg l 75 50

100

150

200

250

300

350

400

450

150

190

230

270

A/Q (m2d.m-3)

-3

500

-1

DCPo (mg l ) Fig. 6. Variation of percent toxicity removal with the feed DCP concentrations at different A/Q ratios and constant feed COD of 4000 mg l1.

Variations of percent toxicity removal with the feed DCP content at different A/Q ratios and a constant feed COD of 4000 mg l1 are depicted in Fig. 6. Percent toxicity removals increased with increasing A/Q ratio up to A/Q value of 200 m2 d m3 because of high biofilm surface area and therefore, high biomass concentrations at high A/Q ratios. Similar to DCP and COD removals further increases in A/Q ratio resulted in lower toxicity removals due to insufficient substrate loadings to support the biofilm organisms (i.e., high A and low Q). Percent toxicity removal decreased with increasing feed DCP at high A/Q ratios due to incomplete degradation of DCP at high loading rates. However, DCP removal was rather insensitive to the feed DCP contents at low A/Q’s below 150 m2 d m3 or high feed flow rates since the experiments were performed in the order of increasing DCP loading rates or A/Q ratios allowing domination of DCP degrading organisms or

Fig. 7. Variations of percent toxicity removal with the A/Q ratio at different feed COD concentrations and constant feed DCP of 225 mg l1.

adaptation to DCP. The system should be operated at A/Q ratio of 200 m2 d m3 in order to obtain high toxicity removal when the feed COD is 4000 mg l1. Variations of percent toxicity removal with A/Q ratio at different feed COD contents and constant feed DCP of 225 mg l1 are depicted in Fig. 7. Similar to the DCP removal, percent toxicity removal increased with increasing A/Q ratio up to 200 m2 d m3 due to high biomass concentrations at high A/Q ratios. Further increases in A/Q ratio resulted in reduced percent DCP removal because of insufficient substrate loadings at low flow rates. At high A/Q ratios above 200 m2 d m3 where the biofilm surface area or the biomass concentration is high, toxicity removal increased with increasing feed COD due to increasing COD loading rates (LCOD = Q Æ COD0) to support high biomass concentrations. A/Q ratio must be around 200 m2 d m3 when the feed COD is 6000 mg l1 in order to maximize the toxicity removal. Most of the literature studies on DCP biodegradation considered DCP concentrations lower than 200 mg l1 with DCP removals less than 80% (Yee and Wood, 1997; Steinle et al., 1998; Wang et al., 2000; Sahinkaya and Dilek, 2002). We were able to obtain almost complete DCP removals with a high feed DCP content of 500 mg l1 due to high biomass concentrations in our rotating brush biofilm reactor (RBBR). The results of this study were compared with our published data using the rotating perforated tubes biofilm reactor (RTBR) in Table 4 at the lowest, medium and the highest levels of the operating parameters (Kargi and Eker, 2005b). Although the COD and DCP removals are comparable for both reactors, toxicity removals are much better in the RBBR than RTBR. This is probably because of the difference in composition of the microbial commu-

Table 4 Comparison of COD, DCP and toxicity removal efficiencies of rotating brush (RBBR) and rotating tubes biofilm (RTBR) reactors DCP0 (mg l1) 50 225 500

COD0 (mg l1) 2000 4000 6000

A/Q (m2 d m3)

70 180 290

ECOD (%)

EDCP (%)

EToxicity (%)

RBBR

RTBR

RBBR

RTBR

RBBR

RTBR

87 94 89

89 93 85

96 97 100

98 96 100

89 100 100

10 73 45

S. Eker, F. Kargi / Bioresource Technology 99 (2008) 2319–2325

nity in both reactors. Apparently, the organisms in the RBBR completely degraded TCP resulting in high toxicity removals while the organisms in RTBR produced some intermediate products with some levels of toxicity. 4. Conclusions 2,4-Dichlorophenol containing synthetic wastewater was biologically treated using a newly developed rotating brush biofilm reactor (RBBR). Effects of major operating variables such as the feed DCP, COD and A/Q ratio on percent COD, DCP and toxicity removals were investigated by using a Box–Wilson statistical experiment design approach. Quadratic response functions were correlated with the experimental data using the Statistica 5.0 computer program to determine the coefficients. Response function predictions were in good agreement with the experimental results. Percent COD removals decreased with increasing feed DCP due to toxic effects of DCP on the organisms, but increased with increasing A/Q ratio due to high concentrations of biofilm organisms at high A/Q ratios. Percent DCP and toxicity removals also increased with increasing A/Q ratio and decreasing feed DCP concentrations. The optimal A/Q ratio maximizing COD (90%), DCP (100%) and toxicity (100%) removals for the feed COD and DCP of 6000 mg l1 and 500 mg l1, respectively was nearly 210 m2 d m3. Acknowledgements This study was supported by the research funds of the State Planning Organization and Dokuz Eylul University, Izmir, Turkey. References Annachhatre, A.P., Gheewala, S.H., 1996. Biodegradation of chlorinated phenolic compounds. Biotechnol. Adv. 14, 35–56. Armenante, P.M., Kafkewitz, D., Lewandowski, G.A., Jou, C.J., 1999. Anaerobic–aerobic treatment of halogenated phenolic compounds. Water Res. 33, 681–692. Atuanya, E.I., Purohit, H.J., Chakrabarti, T., 2000. Anaerobic and aerobic biodegradation of chlorophenols using UASB and ASG bioreactors. World J. Microb. Biot. 16, 95–98. Bali, U., Sengul, F., 2002. Performance of a fed-batch reactor treating a wastewater containing 4-chlorophenol. Process Biochem. 37, 1317– 1323. Brouwer, H., 1991. Testing for chemical toxicity using bacteria. J. Chem. Educ. 68, 695–697. Dapaah, S.Y., Hill, G.A., 1992. Biodegradation of chlorophenol mixtures by Pseudomonas putida. Biotechnol. Bioeng. 40, 1353–1358. Eker, S., Kargi, F., 2006a. Kinetic modelling and parameter estimation for an activated sludge unit treating 2,4-dichlorophenol containing synthetic wastewater. Environ. Eng. Sci. 23 (2), 263–271. Eker, S., Kargi, F., 2006b. Hydraulic residence time effects on performance of an activated sludge unit treating dichlorophenol containing wastewater. Water Environ. Res. 78 (7), 686–690. Eker, S., Kargi, F., 2006c. Impacts of COD and DCP loading rates on biological treatment of 2,4-dichlorophenol (DCP) containing waste-

2325

water in a perforated tubes biofilm reactor. Chemosphere 64 (9), 1609– 1617. Fahr, K., Wetzstein, H.G., Grey, R., Schlosser, D., 1999. Degradation of 2,4-dichlorophenol and pentachlorophenol by two brown rot fungi. FEMS Microbiol. Lett. 175, 127–162. Farre, M., Barcelo, D., 2003. Toxicity testing of wastewater and sewage sludge by biosensors, bioassays and chemical analysis. Trends Anal. Chem. 22, 299–310. Farrell, A., Quilty, B., 2002. Substrate-dependent autoaggregation of Psedomonas putida CP1 during the degradation of mono-chlorophenols and phenol. J. Ind. Microbiol. Biotechnol. 28, 316–324. Greenberg, A.E., Clesceri, L.S., Eaton, A.D. (Eds.), 2005. Standard Methods for the Examination of Water and Wastewater, 21st ed. American Public Health Association (APHA), Washington DC, USA. Hill, G.A., Milne, B.J., Nawrocki, P.A., 1996. Cometabolic degradation of 4-chlorophenol by Alcaligenes eutrophus. Appl. Microbiol. Biotechnol. 46, 163–168. Jung, M.W., Ahn, K.H., Lee, Y., Kim, K.P., Rhee, J.S., Park, J.T., Paeng, K.J., 2001. Adsorption characteristics of phenol and chlorophenols on granular activated carbons. Microchem. J. 70, 123–131. Kargi, F., Eker, S., 2004. Toxicity and batch biodegradation kinetics of 2,4-dichlorophenol by pure Pseudomonas putida. Enzyme Microb. Technol. 35, 424–428. Kargi, F., Eker, S., 2005a. Kinetics of 2,4-dichlorophenol degradation by Pseudomonas putida CP1 in batch culture. Inter. Biodet. Biodeg. 55, 25–28. Kargi, F., Eker, S., 2005b. Removal of 2,4-dichlorophenol and toxicity from synthetic wastewater in a rotating perforated tube biofilm reactor. Process Biochem. 40, 2105–2111. Kargi, F., Eker, S., 2006. Effect of sludge age on performance of an activated sludge unit treating 2,4-dichlorophenol containing synthetic wastewater. Enzyme Microb. Technol. 37, 60–64. Kim, M.H., Hao, O.J., 1999. Cometabolic degradation of chlorophenols by Acinetobacter species. Water Res. 33, 562–574. Kim, J.H., Oh, K.K., Lee, S.T., Kim, S.W., Hong, S.I., 2002. Biodegradation of phenol and chlorophenols with defined mixed culture in shake-flasks and packed bed reactor. Process Biochem. 37, 1367–1373. Liu, D., 1986. Resazurin reduction method for toxicity assessment of water soluble and insoluble chemicals. Toxic. Asses. 1, 253–258. Radwan, K.H., Ramanujam, T.K., 1996. Studies on organic removal of 2,4-dichlorophenol wastewaters using a modified RBC. Bioprocess Eng. 16, 219–223. Sahinkaya, E., Dilek, F.B., 2002. Effects of 2,4-dichlorophenol on activated sludge. Appl. Microbiol. Biotechnol. 59, 361–367. Shieh, B.W.K., Puhakka, J.A., Melin, E., Tuhkannen, T., 1990. Immobilized cell degradation of chlorophenols. J. Environ. Eng. ASCE 116, 683–697. Shin, H.S., Yoo, K.S., Park, J.K., 1999. Removal of polychlorinated phenols in a sequential anaerobic–aerobic biofilm reactors packed with tire chips. Water Environ. Res. 71, 363–367. Steinle, P., Stucki, G., Stettler, R., Hanselmann, K.W., 1998. Aerobic mineralization of 2,6-dichloro-phenol by Ralstonia sp. strain RK1. Appl. Environ. Microb. 64, 2566–2571. Strotmann, U.J., Butz, B., Bias, W.R., 1993. The dehydrogenase assay with resazurin: Practical performance as a monitoring system and pH dependent toxicity of phenolic compounds. Ecotox. Environ. Safety. 25, 79–89. Swaminathan, G., Ramanujam, T.K., 1998. Effect of substrate concentration on biodegradation of 2,4-dichlorophenol using modified rotating biological contactors. Bioprocess Eng. 18, 169–173. Wang, C.C., Lee, C.M., Kuan, C.H., 2000. Removal of 2,4-dichlorophenol by suspended and immobilized Bacillus insolitus. Chemosphere 41, 447–452. Yee, D.C., Wood, T.K., 1997. 2,4-Dichlorophenol degradation using Streptomyces viridosporus T7A lignin peroxidase. Biotechnol. Progr. 13, 53–59.