Kinetic parameters for 17α-ethinylestradiol removal by nitrifying activated sludge developed in a membrane bioreactor

Bioresource Technology 101 (2010) 6425–6431

Contents lists available at ScienceDirect

Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

Kinetic parameters for 17a-ethinylestradiol removal by nitrifying activated sludge developed in a membrane bioreactor L. Clouzot a, P. Doumenq b, N. Roche a, B. Marrot a,* a

Laboratoire de Modélisation, Mécanique et Procédés Propres (M2P2), UMR 6181, Université Paul Cézanne Aix-Marseille 3, Europôle de l’Arbois, Bât Laennec hall C BP 80, 13545 Aix-en-Provence cedex 4, France b Institut des Sciences Moléculaires de Marseille (ISM2), Université Paul Cézanne Aix-Marseille 3, Europôle de l’Arbois, Bât Villemin, BP 80, 13545 Aix-en-Provence cedex 4, France

a r t i c l e

i n f o

Article history: Received 20 January 2010 Received in revised form 5 March 2010 Accepted 11 March 2010 Available online 3 April 2010 Keywords: EE2 Nitrifying activity Biodegradation Sorption MBR

a b s t r a c t The synthetic hormone 17a-ethinylestradiol (EE2) is primarily removed in wastewater treatment plants (WWTPs) by sorption, and nitrifying biomass has been shown to be responsible for EE2 biodegradation. Membrane bioreactor (MBR) technology was chosen to develop a community of autotrophic, nitrifying micro-organisms and determine kinetic parameters for EE2 biodegradation. Biological inhibition by azide was applied to differentiate sorption from biodegradation. Activated sludge (AS) was acclimated in the MBR to a substrate specific to autotrophic biomass and resulted in an increase in nitrifying activity. Acclimated AS was used to successfully biodegrade EE2 (11% increase in EE2 removal), and the overall removal of EE2 was determined to be 99% (sorption + biodegradation). AS used directly from a WWTP without acclimation removed EE2 only through sorption (88% removal of EE2). Therefore, higher nitrifying activity developed by acclimating AS allowed almost complete removal of EE2. Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction The synthetic hormone 17a-ethinylestradiol (EE2) is an endocrine disrupter that has caused feminization in several species among fish (Mills and Chichester, 2005), molluscs (Ketata et al., 2008), amphibians (Gyllenhammar et al., 2009), birds (Brunström et al., in press) and mammals (Latendresse et al., in press), at concentrations as low as 0.1 ngEE2 L1 (Purdom et al., 1994). This xenobiotic was classified as at least R51/53 which means the compound is ‘‘toxic to aquatic organisms and may cause long-term effects in the aquatic environment” (Carlsson et al., 2006). EE2 is thus of great concern as a water contaminant. Pimephales promelas was exposed to EE2 (5–6 ngEE2 L1) during a whole-lake experiment and the exposure induced feminization in the males, followed by a near extinction of this fish species (Kidd et al., 2007). EE2 occurrence in the environment is due to insufficient removal of the compound by municipal wastewater treatment plants (WWTPs) (Clouzot et al., 2008). Contamination of receiving waters has been measured up to 100 km downstream from WWTPs, with concentrations sufficient to induce endocrine disruption (>1.5 ngEE2 L1) (Barel-Cohen et al., 2006). Sorption is the process that is primarily responsible for EE2 removal in WWTPs; the typical decrease in aqueous EE2 concentration due to sorption is between 60% and 80% (Andersen et al., 2005; * Corresponding author. Tel.: +33 4 42 90 85 11; fax: +33 4 42 90 85 15. E-mail address: [email protected] (B. Marrot). 0960-8524/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2010.03.039

Johnson and Sumpter, 2001). However, sorption is a pollutant transfer process, and a significant amount of the EE2 passes through the wastewater treatment processes; only biodegradation allows complete removal of the contaminant. Nitrifying biomass has been shown to biodegrade EE2 by co-metabolism of the enzyme ammonium monooxygenase (AMO) (Shi et al., 2004; Vader et al., 2000). A linear trend between nitrifying activity and EE2 removal has been established (Yi and Harper, 2007). How1 ever, for initial ammonium concentrations below 50 mgþ NH4 L , sorption was shown to be the predominant removal mechanism (up to 60%) because of low co-metabolic activity. For higher ammonium concentrations, biodegradation became more important (up to 50%) (Yi et al., 2006). Therefore, for relevant ammonium concentrations in WWTPs (in the low mg L1 range), sorption plays a more significant role than biodegradation (Johnson and Sumpter, 2001). Nitrifying micro-organisms are autotrophic and grow more slowly than heterotrophic ones. Therefore, heterotrophic microorganisms typically outnumber nitrifying micro-organisms. The development of nitrifying micro-organisms during wastewater treatment can be improved with high sludge retention times (SRT). In a conventional activated sludge (AS) system, low settling abilities of sludge generally result in low SRT (15–20 days). However, with a membrane bioreactor (MBR), complete biomass retention allows control of a higher SRT. Based on a biological model, EE2 removal was predicted to be more efficient with acclimated, nitrifying AS from a MBR (SRT = 30 days) than with AS from a

6426

L. Clouzot et al. / Bioresource Technology 101 (2010) 6425–6431

conventional WWTP (SRT = 11 days) (Joss et al., 2004). Autotrophic nitrifying biomass requires specific conditions for growth, and the organisms have a high sensitivity to external parameters such as pH, dissolved oxygen concentration, and temperature. Typical required conditions include a pH of 7.5, a dissolved oxygen concentration above 4 mgO2 L1 and a temperature range of 30–36 °C. The MBR system configuration can also influence performance and degradation; in particular, immersed configurations are characterized by lower shear stress than external configurations. Thus, an immersed MBR was determined to be more appropriate for nitrifying micro-organism growth. However, operating conditions of an immersed MBR are known to induce higher membrane fouling. Recent work has resulted in the development of a new external-immersed MBR configuration (the membrane is immersed in an external carter), which allows improved fouling control in the membrane system by air sparging (Lesjean et al., 2002) while maintaining a lower shear stress than an external MBR. A review of relevant literature revealed few studies on the determination of kinetic constants for EE2 removal by AS (De Gusseme et al., 2009; Shi et al., 2004; Vader et al., 2000). Therefore, the aim of this research was to determine kinetic parameters for EE2 sorption and biodegradation by nitrifying AS. Biodegradation of the synthetic hormone was previously studied with concentrations between 1 and 1 lgEE2 L1 (Ternes et al., 1999). However, no significant decrease in EE2 concentrations was detected. Thus, experiments were performed with higher values, in the mgEE2 L1 range, to quantify both phenomena (sorption and biodegradation) involved in EE2 removal. To increase nitrifying activity, an external-immersed MBR was used with a specific influent and without any sludge waste to increase SRT and thus enhance autotrophic biomass growth. 2. Methods 2.1. Chemicals and solvents Carbon source and nutrients (C6H12O6, (NH4)2SO4, NaHCO3, KH2PO4, MgSO4, and CaCl2) were supplied by Chem-Lab (>99%, Zedlgem, Belgium). The reagent 1-allyl-2-thiourea (98%) was purchased from Aldrich (L’Isle d’Abeau, France) and sodium azide (99%) was purchased from Acros organics (Noisy-leGrand, France). 17a-ethynylestradiol (P98%) and b-estradiol17-acetate (P99.9%) were obtained from Sigma–Aldrich (L’Isle d’Abeau, France). b-Estradiol-17-acetate (P99.9%) was used as an internal standard. EE2 stock solutions were prepared in absolute ethanol (P99.8%, ACS reagent grade) from Carlo Erba (Val de Reuil, France). Distilled water was used for the analytical method. HPLC-grade methanol (P99.9%) was supplied by Sigma. Reagent kits used for spectrophotometric ammonium and nitrate analyzes were obtained from Merk (Darmstadt, Germany). 2.2. The MBR system The lab-scale MBR was composed of a 60-L bioreactor and polysulfone membranes submerged in a 250-mL external carter (130 hollow fibres, molecular weight cut-off 0.2 lm, total surface 0.2 m2, Polymem, France). Peristaltic pumps were used for the synthetic influent, permeate and sludge recycle back to the membrane module. Membrane fouling was limited by air sparging (6 L min1) in the external carter. The MBR was operated at a hydraulic retention time (HRT) of 50 h without wasting any sludge to maximize the SRT. The pH was maintained constant at seven by an automatic controller with a sodium bicarbonate (NaHCO3) solution (60 g L1).

2.3. Acclimation of autotrophic biomass The lab-scale MBR was started with activated sludge (AS) collected from a municipal WWTP operated with nitrification and denitrification tanks (165,000 population equivalent, Aix-en-Provence, France). The AS was concentrated in the MBR until 11 g L1 of mixed liquor volatile suspended solids (MLVSS) and a final volume of 50 L were achieved. The synthetic influent was prepared without any organic carbon source and with mass ratios of 1 (NH4)2SO4, 0.4 NaHCO3, 0.2 KH2PO4, 0.1 MgSO4, and 0.02 CaCl2. The ammonium 1 1 load was increased from 0.02 to 0:16 kgN—NH4 kgMLVSS d to enhance the growth of autotrophic micro-organisms.

2.4. Analytical methods Mixed liquor suspended solids (MLSS) were determined from AS centrifugation (15,892g, 30 min) followed by 24-h drying at 105 °C. MLVSS were then determined from heating the MLSS at 550 °C for 2 h. Ammonium, nitrate and polysaccharide (PS) were analyzed in the supernatant by spectrophotometry (Spectro Aquamate, Thermo spectronic, Cambridge, UK) with reagent kits for ammonium (2–150 mg L1 N–NH4, R.S.D. 2%) and nitrate (0–20 mg L1 N–NO3, R.S.D. 3%) and based on the method of Dubois et al. (1956) for PS (R.S.D. 9%).

2.5. Biological methods Ammonium removal rates were determined during the acclimation period from 4-h batch experiments. Aerated reactors were filled with 1 L of the acclimated AS and 0:02 gN—NH4 g1 MLVSS was added. The ammonium concentration was analyzed 4 h after nutrient addition. Additionally, respirometry tests were used to differentiate autotrophic and heterotrophic micro-organisms. An aerated reactor was filled with 1 L of the acclimated AS and every 2 min, 50 mL was sampled and injected into another reactor without oxygenation to measure the oxygen uptake rate (OUR) with a continuous dissolved oxygen probe (HQ 40d, Hach LDO, Düsseldorf, Germany). Then, specific nutrients or inhibitor were successively added to the aerated reactor. The respirometry method was divided into four steps. Endogenous respirations of autotrophic and heterotrophic micro-organisms were first measured over a period of 1 h. Secondly, ammonium was added (0:02 gN—NH4 g1 MLVSS ) to measure the maximum activity of autotrophic micro-organisms (ammonium removal was confirmed to be correlated with nitrate production). Thirdly, autotrophic micro-organisms were inhibited with allylthiourea (inhibitor of the nitrifying enzyme AMO) to isolate heterotrophic endogenous respiration. A concentration of 0:1 gallylthiourea g1 MLVSS was shown to inhibit the respirometry activity, and nitrates were not produced after ammonium addition ð0:02 gN—NH4 g1 MLVSS Þ, which validated the autotrophic inhibition. Lastly, maximum activity of heterotrophic micro-organisms was quantified after glucose addition (equivalent to 0:3 gCOD g1 MLVSS ).

2.6. EE2 removal in the MBR After 53 days of acclimation, EE2 was added into the synthetic MBR influent to reach a final concentration of 1 mgEE2 L1. EE2 concentrations in the permeate were followed for 60 h. The bioreactor volume was reduced to 10 L, and the HRT was reduced to 19 h. The MLVSS concentration was verified to be constant at 7 g L1. The same continuous MBR operation and addition of EE2 was performed with AS directly sampled from the WWTP.

L. Clouzot et al. / Bioresource Technology 101 (2010) 6425–6431

2.7. Determination of kinetic parameters In parallel with the continuously-operated MBR, kinetic parameters were determined from batch experiments performed at 1 and 0.5 mgEE2 L1 with acclimated AS and AS taken directly from WWTP (4 gMLVSS L1). The AS samples were split into amber glass reactors (2 L of sludge into each reactor) and aerated until oxygen saturation with aquarium air diffusers. After EE2 addition, samples ranging from 100–400 mL were collected during a total incubation time of 24 h (between 5 and 7 samples). Then, the mixed liquor was filtered with a 1.2 lm glass fibre filter (Whatman GF/C, Maidstone, UK) followed by a 0.45 lm PTFE membrane filter (Millipore, Ireland) and the permeate was stored at 10 °C until analysis. Two reactors were used to measure the overall removal of EE2, including both biodegradation and sorption effects. To quantify EE2 sorption onto AS, biological activity was inhibited with sodium azide (inhibition of the terminal enzyme of the electron transport system) in two other reactors. A concentration of 0:9 gazide g1 MLVSS was necessary to inhibit the respirometry activity, and neither glucose 1 (equivalent to 0:3 gCOD g1 MLVSS ) nor ammonium ð0:02 gN—NH4 gMLVSS Þ could be removed after this biological inhibition. Consequently, the EE2 decrease measured in the dissolved matrix was due to the sorption mechanism. Finally, water controls highlighted that EE2 was not lost by stripping or sorption on the inner surfaces of the reactors.

6427

20 Pa. The hysteresis area was also determined by reversing the second stress ramp (from 20 to 0 Pa).

3. Results and discussion 3.1. Evaluation of nitrification efficiency Despite the increase in the influent ammonium concentration, MLVSS decreased continuously during the acclimation (Fig. 1A). Organic carbon deficiency was unfavourable for the growth of heterotrophic micro-organisms, which are predominant in AS biomass. Therefore, autotrophic growth was probably masked by the decrease in heterotrophic organisms. This result is in agreement with the ammonium removal efficiency, which was maintained at approximately 100% (Fig. 1B). Thus, experimental conditions

2.8. EE2 analysis The analytical method for measuring EE2 was previously optimized and validated (Clouzot, 2009). Samples were spiked with b-estradiol 17-acetate as the internal standard. Solid phase extraction (SPE) was performed on 500 mg Strata-X columns (6 mL) (Phenomenex, Le Pecq, France) conditioned with methanol and distilled water. After the extraction, columns were washed with methanol/water (0.15:1 v:v) and dried by vacuum aspiration. Analytes were then eluted with methanol (15 mL) and evaporated at 50 °C under a gentle stream of N2 to a final volume of 1 mL. Analyte separation was carried out on an Agilent 7890A gas chromatograph (Agilent Technologies, Massy, France) equipped with an HP-5MS fused silica capillary column (30 m  0.25 mm I.D.  0.25 lm film thickness). The oven temperature was programmed from 50 (isothermal 1 min) to 100 °C with a ramp of 50 °C min1, then from 100 (isothermal 1 min) to 300 °C with a ramp of 10 °C min1 and held at 300 °C for 5 min. The gas chromatograph (GC) was coupled to a 5975C Agilent quadrupole mass spectrometer (MS) (Agilent Technologies, Massy, France) operated with electron impact ionisation (EI-70 eV). Mass spectra were recorded in selected ion monitoring (SIM) mode (m/z 213, 228, 296 for EE2 and 172, 225 for the internal standard). 2.9. Sludge characterization During the acclimation and to validate sorption kinetic, sludge structure was characterized by rheological measurements. In the literature, physico-chemical parameters of AS were correlated to their rheological properties (Seyssiecq et al., 2003). A rotational and computer-controlled stress rheometer (AR550, TA Instrument, France) coupled with a helical ribbon impeller (impeller Ø 14 mm, stator Ø 15 mm, immersed height 31 mm) was used (the method was detailed in Mori et al., 2006). The method of the Couette analogy was applied to determine constants of shear rate (6.85 rad1) and shear stress (27,103 Pa N1 m1). With MLVSS concentrations below 26 g L1, the Ostwald model was chosen to calculate rheological parameters from two continuous stress ramps (from 0 to 20 Pa with 240 points in 40 min) separated by a 1 min isobar at

Fig. 1. Evolution of biomass and its purification efficiencies during the acclimation (A) MLVSS (d) versus F/M (}) (B) N–NH4 removal (d) versus N–NO3 (4) and (C) versus PS (4).

6428

L. Clouzot et al. / Bioresource Technology 101 (2010) 6425–6431

seemed to be adapted to efficient nitrifying activity. One of the consequences of nitrification was nitrate accumulation in the bioreactor. Oxygenation cycles could not be applied to reduce nitrate concentrations because of the organic carbon deficiency. Therefore, the bioreactor had to be washed out regularly (which explains the F/M ratios equal to 0 in Fig. 1A). Following washout, nitrates were reduced, and complete ammonium removal was recovered. At the end of the acclimation, ammonium removal appeared to be independent of the nitrate concentration, which is one indication of biomass adaptation. At first glance, nitrification inhibition could be deduced from the correlation between the decrease in ammonium removal efficiency and nitrate accumulation. However, this hypothesis was eliminated because it is well known that nitrates do not inhibit nitrification. The same correlation was obtained for polysaccharides (PS) and nitrification inhibition (Fig. 1C). Inhibition may be caused by other soluble microbial product (SMP) molecules not analyzed but also linked to biomass metabolism.

3.2. Characterization of MBR fouling For 20 days, permeate flux was kept constant at approximately 4 L h1 m2, with a constant frequency of back-washes (Fig. 2A). The frequency of back-washes had to be increased until the 30th day, when high membrane fouling required the installation of another membrane module (same membrane surface as the first module, 0.2 m2) to recover process stability. Once the second module was installed, permeate flux was divided by two to achieve an identical volumetric flow. SMP accumulation was measured in the bioreactor and can be correlated to membrane fouling. In the literature, SMP released after substrate deficiency have been shown to have a higher membrane fouling potential than the SMP that appeared during substrate uptake (Drews et al., 2006; Rosenberger et al., 2006). Membrane fouling associated with SMP release was linked to notable modifications in the sludge characteristics. To characterize this change, rheological parameters were calculated with the Ostwald model. Rheological parameters are linked to shear thinning of bacterial flocs through their ability to store or release water (Seyssiecq et al., 2008). During flocculation, water is stored whereas deflocculation induces a release of water responsible for decreasing viscosity. Rheological measurements did not demonstrate any apparent differences in viscosity (l) between acclimated AS and AS directly sampled from the WWTP (Fig. 2B). In addition, the flow index (n) and the consistency index (K) were not significantly different between the two AS (n: 0.12 ± 0.01 and K: 6.2 ± 0.5 Pa sn for acclimated AS/n: 0.13 ± 0.01 and K: 5.7 ± 0.5 Pa sn for WWTP AS). Thus, changes in the sludge characteristics could not be explained by AS deflocculation. An additional hypothesis for membrane fouling is the development of filamentous bacteria due to the organic carbon deficiency. A high area of hysteresis calculated from the viscosity curve has been previously used as a valuable tool for the detection of filamentous bacteria (Tixier et al., 2003). The hysteresis area was 1195 Pa s1 for AS taken from the WWTP and 43 Pa s1 for acclimated AS (Fig. 2C). The low value obtained for acclimated AS can be explained by the presence of homogenous pre-sheared sludge, which is produced as a result of the high shear stress of the MBR. Contrary to the result obtained for acclimated AS, sludge from the conventional WWTP was much less sheared, which resulted in a more heterogeneous structure and size and thus a higher hysteresis area. Therefore, in the MBR, it was not possible to deduce the occurrence of filamentous bacteria from measuring the area of hysteresis.

Fig. 2. Characterization of membrane fouling during the acclimation in the MBR: (A) Evolution of permeate flux with the back-washes; (B) viscosity and (C) hysteresis area for acclimated AS and sludge from WWTP.

3.3. Development of autotrophic biomass Autotrophic micro-organism activity, measured by respirometry, confirmed the development of this biomass at the end of the acclimation (Fig. 3). The two horizontal lines represent maximum activity of autotrophic and heterotrophic bacteria measured in the AS that was sampled from the WWTP. The organic carbon source deficiency induced an irreversible decrease of heterotrophic biomass activity, which explained the loss of MLVSS (Fig. 1A). The initial decrease in autotrophic micro-organisms was due to biomass adaptation to a reduced and specific influent. However, the increase in ammonium load led to the growth of nitrifying biomass. The same results were obtained with endogenous activity of autotrophic and heterotrophic bacteria. The development of autotrophic micro-organisms was validated by an ammonium removal rate that increased by a factor of four between the beginning

L. Clouzot et al. / Bioresource Technology 101 (2010) 6425–6431

Fig. 3. Maximal activity of autotrophic and heterotrophic micro-organisms during the acclimation (the two lines represent the activities measured in AS taken from WWTP).

1

and the end of the acclimation (2.2 ± 0.2 mgN—NH4 g1 for MLVSS h 1 for acclimated AS). WWTP AS and 8.0 ± 0.6 mgN—NH4 g1 MLVSS h Therefore, the external-immersed MBR and the acclimation period resulted in optimal nitrifying activity. 3.4. Kinetic parameters for EE2 removal Removal kinetics were first determined in batch reactors with AS directly taken from the WWTP (Fig. 4A). For each batch reactor, after 10 min, a dramatic decrease in EE2 concentration (approximately 88%) was observed, followed by a plateau. Superimposition of the overall removal and sorption kinetic demonstrated that EE2 removal by AS from the WWTP was only due to sorption. However, with the acclimated AS, the EE2 concentration continued to decrease after the first 10 min of operation in the batch reactor (Fig. 4B). Therefore, the acclimated AS developed the ability to bio-

Fig. 4. Kinetic for EE2 overall removal (full symbols) and sorption (empty symbols) from (A) WWTP AS and (B) acclimated AS.

6429

degrade EE2. Removal percentages and kinetic parameters are presented in Table 1. Overall EE2 removal was approximately 88% with WWTP AS whereas biodegradation of EE2 by acclimated AS resulted in almost complete removal of the synthetic hormone (99%). For both types of AS, the primary removal mechanism for EE2 was sorption; this result has been previously demonstrated in the literature (Hashimoto and Murakami, 2009; Ren et al., 2007). Similar sorption efficiencies were obtained for each AS and each EE2 initial concentration (Table 1). Removal rates doubled when the initial concentration of EE2 doubled. Therefore, EE2 sorption was a first order reaction for the tested range of concentrations (<1 mgEE2 L1) with a reaction rate constant around 5 h1. Sorption reactions are commonly defined by a first order rate for the lowest concentrations and towards zero order for higher values, due to saturation. In literature, AS taken from a sequencing batch reactor fed with a wastewater of a hog farm was characterized by a sorption constant of 7 h1 (Ren et al., 2007). Despite the different origins of AS and considering experimental errors, the constants are relatively similar and are in the same order of magnitude. By increasing nitrifying activity, 11% removal of EE2 due to biodegradation was observed (Table 1). Biodegradation was a first order reaction for the EE2 concentrations tested, similar to the sorption phenomenon. Respirometry allowed determination of the dissolved oxygen concentration that was required by autotrophic biomass to biodegrade EE2 (around 1 lgEE2 mg1 O2 ). If the EE2 concentrations typically found in the environment (in the ngEE2 L1 range) are compared to the typical dissolved oxygen concentrations found in WWTPs (concentrations between 2 and 4 mgO2 L1 ), oxygen is found to not be a limiting factor for EE2 removal during wastewater treatment. 3.5. Methods comparison for biodegradation constants Only a few reaction rate constants for EE2 biodegradation can be found in the literature. However, the different methods used to measure the constants make reliable comparison difficult. In this study, the rate constant for EE2 biodegradation was 32 s1, and the 1 nitrifying activity was 8 mgN—NH4 g1 MVLSS h . A biodegradation rate 1 constant of 58 s was obtained from nitrifying AS after removal of suspended solids by sedimentation (Vader et al., 2000) with a 1 nitrifying activity of 50 mgN—NH4 g1 MVLSS h . Therefore, the linear trend between EE2 biodegradation and nitrifying activity that has been shown previously (Yi and Harper, 2007) is not confirmed. Nevertheless, this indirect method raises questions about the complete removal of EE2 sorbed onto the sludge. Other authors chose liquid–liquid extraction to remove the sorption fraction before EE2 analysis (Shi et al., 2004). They demonstrated a biodegradation 1 constant of 90 s1 from nitrifying activity of 3 mgN—NH4 g1 MVLSS h . This high constant compared to the low activity can be explained by partial removal of EE2 sorbed or by co-extraction of molecules that cause analysis interferences. Other indirect methods inhibit biological activity to only measure the sorption phenomenon. Thermal inactivation (120 °C, 30 min) resulted in a constant in the same range as the one determined in this study (25 s1) (De Gusseme et al., 2009). However, the authors did not verify that any sludge structure change occurred following the inactivation. Biological inactivation with sodium azide was recently used to inhibit biological activity (based on respirometry measurement) without any cell lysis or any change in the sludge hydrophobicity (Yi and Harper, in press). The impact of azide inhibition on sludge structure was determined from PS analysis and rheological characterization. Biological inhibition did not induce any PS release but viscosity was reduced (around 40%). After azide addition, the flow index increased (n: from 0.13 ± 0.01 without azide to 0.18 ± 0.01 with azide), and the consistency index decreased (K: from

6430

L. Clouzot et al. / Bioresource Technology 101 (2010) 6425–6431

Table 1 Efficiencies and kinetic parameters for EE2 removal by sorption and biodegradation for initial concentrations of 1 and 0.5 mgEE2 L1. EE2 (mg L1)

Sorption

WWTP AS

Removal (%) Rate ðmgEE2 g1 MLVSS h Constant (h1)

Biodegradation

1

Þ

Removal (%) 1

Rate ðlgEE2 g1 Þ MLVSS h Constant (h1) Activity ðlgEE2 mgO2 Þ Overall removal (%)

Acclimated AS

1

0.5

87.1 ± 1.5 1.6 ± 0.3

86.7 ± 0.9 0.8 ± 0.2

1

5.2 ± 1.0

5.1 ± 1.3

5.3 ± 0.7

5.3 ± 0.5

0.6 ± 2 –

0.0 ± 1.9 –

10.3 ± 1.3 2.7 ± 0.3

11.0 ± 1.1 1.4 ± 0.1

0.009 ± 0.001 0.15 ± 0.04

0.009 ± 0.001 0.08 ± 0.02

– –

– –

87.8 ± 0.5

86.7 ± 1.0

3.70 ± 0.01 without azide to 1.88 ± 0.01 with azide). Biological inhibition by azide seemed to induce AS deflocculation, which could increase the specific surface area of sludge and thus overestimate its sorption abilities. However, kinetic graphs determined for overall removal and sorption for similar EE2 initial concentrations for the WWTP AS were nearly identical when superimposed (Fig. 4A). Therefore, deflocculation induced by azide did not modify AS sorption abilities. Chemical extractions commonly used to measure analytes sorbed onto AS are subject to many interferences in such a complex matrix. Therefore, this indirect method appeared to be a useful and reliable alternative.

3.6. EE2 removal in MBR Continuous removal of EE2 was performed in the MBR with acclimated AS and AS directly sampled from the WWTP (Fig. 5). EE2 removal was maintained around 100% with acclimated AS whereas a 10% decrease in EE2 removal was observed with WWTP 1 AS. The EE2 load applied in the MBR ð2:5 mgEE2 g1 MLVSS d Þ was eight times higher than the load used for kinetics studies 1 ð0:3 mgEE2 g1 MLVSS d Þ. Despite these high EE2 concentrations, the acclimated AS resulted in a more efficient EE2 removal than the AS from the WWTP. EE2 removal by the WWTP AS was shown to be due only to sorption onto sludge. Therefore, the efficiency decrease of EE2 removal indicated that sorption sites started to be saturated. However, complete saturation was not reached. Therefore, the EE2 total load applied during this continuous purification ð6 mgEE2 g1 MLVSS Þ was lower than maximum sorption ability of the AS. Furthermore, the biodegradation developed with the acclimation of autotrophic

Fig. 5. Continuous purification of EE2 in the MBR with a HRT of 19 h and a mass 1 . load of 2:5 mgEE2 g1 MLVSS d

0.5

88.9 ± 1.2 1.6 ± 0.2

99.3 ± 0.1

87.2 ± 0.9 0.8 ± 0.1

98.2 ± 0.2

micro-organisms compensated for the observed decrease in sorption ability. 4. Conclusion Organic carbon deficiency caused a decrease in heterotrophic micro-organisms with SMP release. The consequences of SMP release were membrane fouling and nitrification inhibition. However, this specific substrate was necessary for the growth of autotrophic biomass in the MBR. Sorption was confirmed to be the predominant mechanism for EE2 removal (88%). However, biodegradation by the acclimated AS resulted in maximum removal (99%). The increase in nitrifying activity appeared to be the key factor for EE2 removal, and the MBR system used in this research appears to be a promising process for this purpose. Acknowledgements The financial support of this work was partly from the Research Federation ECCOREV. The authors thank Ph.D. student L. Devesvre and I. Seyssiecq, Ph.D., from the laboratory M2P2 for the rheological measurements and P. Vanloot, Ph.D., from the laboratory ISM2AD2EM for EE2 analysis. References Andersen, H.R., Hansen, M., Kjlholt, J., Stuer-Lauridsen, F., Ternes, T., HallingSrensen, B., 2005. Assessment of the importance of sorption for steroid estrogens removal during activated sludge treatment. Chemosphere 61, 139– 146. Barel-Cohen, K., Shore, L.S., Shemesh, M., Wenzel, A., Mueller, J., Kronfeld-Schor, N., 2006. Monitoring of natural and synthetic hormones in a polluted river. J. Environ. Manage. 78, 16–23. Brunström, B., Axelsson, J., Mattsson, A., Halldin, K., in press. Effects of estrogens on sex differentiation in Japanese quail and chicken. Gen. Comp. Endocrinol. (Corrected proof). Carlsson, C., Johansson, A.-K., Alvan, G., Bergman, K., Kuhler, T., 2006. Are pharmaceuticals potent environmental pollutants? Part I. Environmental risk assessments of selected active pharmaceutical ingredients. Sci. Total Environ. 364, 67–87. Clouzot, L., 2009. Membrane bioreactor to eliminate the synthetic hormone 17(alpha)-ethinylestradiol. Thesis, M2P2 Aix-Marseille. Clouzot, L., Marrot, B., Doumenq, P., Roche, N., 2008. 17a-ethinylestradiol: an endocrine disrupter of great concern. Analytical methods and removal processes applied to water purification. A review. Environ. Prog. 27, 383–396. De Gusseme, B., Pycke, B., Hennebel, T., Marcoen, A., Vlaeminck, S.E., Noppe, H., Boon, N., Verstraete, W., 2009. Biological removal of 17(alpha)-ethinylestradiol by a nitrifier enrichment culture in a membrane bioreactor. Water Res. 43, 2493–2503. Drews, A., Lee, C.-H., Kraume, M., 2006. Membrane fouling – a review on the role of EPS. Desalination 200, 186–188. Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F., 1956. Colorimetric method for determination of sugars and related substances. Anal. Chem. 28, 350–356. Gyllenhammar, I., Holm, L., Eklund, R., Berg, C., 2009. Reproductive toxicity in Xenopus tropicalis after developmental exposure to environmental concentrations of ethynylestradiol. Aquat. Toxicol. 91, 171–178.

L. Clouzot et al. / Bioresource Technology 101 (2010) 6425–6431 Hashimoto, T., Murakami, T., 2009. Removal and degradation characteristics of natural and synthetic estrogens by activated sludge in batch experiments. Water Res. 43, 573–582. Johnson, A.C., Sumpter, J., 2001. Removal of endocrine-disrupting chemicals in activated sludge treatment works. Environ. Sci. Technol. 35, 4697–4703. Joss, A., Andersen, H., Ternes, T., Richle, P.R., Siegrist, H., 2004. Removal of estrogens in municipal wastewater treatment under aerobic and anaerobic conditions: consequences for plant optimization. Environ. Sci. Technol. 38, 3047–3055. Ketata, I., Denier, X., Hamza-Chaffai, A., Minier, C., 2008. Endocrine-related reproductive effects in molluscs. Comp. Biochem. Physiol. C: Toxicol. Pharmacol. 147, 261–270. Kidd, K.A., Blanchfield, P.J., Mills, K.H., Palace, V.P., Evans, R.E., Lazorchak, J.M., Flick, R.W., 2007. Collapse of fish population after exposure to a synthetic estrogen. Proc. Natl. Acad. Sci. 104, 8897–8901. Latendresse, J.R., Bucci, T.J., Olson, G., Mellick, P., Weis, C.C., Thorn, B., Newbold, R.R., Delclos, K.B., in press. Genistein and ethinyl estradiol dietary exposure in multigenerational and chronic studies induce similar proliferative lesions in mammary gland of male Sprague-Dawley rats. Reprod. Toxicol. (Corrected proof). Lesjean, B., Gnirss, R., Adam, C., 2002. Process configurations adapted to membrane bioreactors for enhanced biological phosphorous and nitrogen removal. Desalination 149, 217–224. Mills, L.J., Chichester, C., 2005. Review of evidence. Are endocrine-disrupting chemicals in the aquatic environment impacting fish populations? Sci. Total Environ. 343, 1–34. Mori, M., Seyssiecq, I., Roche, N., 2006. Rheological measurements of sewage sludge for various solids concentrations and geometry. Process Biochem. 41, 1656– 1662. Purdom, C.E., Hardiman, P.A., Bye, V.J., Eno, C.R., Tyler, C.R., Sumpter, J., 1994. Estrogenic effects of effluents from sewage treatment works. Chem. Ecol. 8, 275–285.

6431

Ren, Y.-X., Nakano, K., Nomura, M., Chiba, N., Nishimura, O., 2007. A thermodynamic analysis on adsorption of estrogens in activated sludge process. Water Res. 41, 2341–2348. Rosenberger, S., Laabs, C., Lesjean, B., Gnirss, R., Amy, G., Jekel, M., Schrotter, J.C., 2006. Impact of colloidal and soluble organic material on membrane performance in membrane bioreactors for municipal wastewater treatment. Water Res. 40, 710–720. Seyssiecq, I., Ferrasse, J.-H., Roche, N., 2003. State-of-the-art: rheological characterisation of wastewater treatment sludge. Biochem. Eng. J. 16, 41–56. Seyssiecq, I., Marrot, B., Djerroud, D., Roche, N., 2008. In situ triphasic rheological characterisation of activated sludge, in an aerated bioreactor. Chem. Eng. J. 142, 40–47. Shi, J., Fujisawa, S., Nakai, S., Hosomi, M., 2004. Biodegradation of natural and synthetic estrogens by nitrifying activated sludge and ammonia-oxidizing bacterium Nitrosomonas europaea. Water Res. 38, 2323–2330. Ternes, T.A., Kreckel, P., Mueller, J., 1999. Behaviour and occurrence of estrogens in municipal sewage treatment plants – II. Aerobic batch experiments with activated sludge. Sci. Total Environ. 225, 91–99. Tixier, N., Guibaud, G., Baudu, M., 2003. Determination of some rheological parameters for the characterization of activated sludge. Bioresour. Technol. 90, 215–220. Vader, J.S., van Ginkel, C.G., Sperling, F.M.G.M., de Jong, J., de Boer, W., de Graaf, J.S., van der Most, M., Stokman, P.G.W., 2000. Degradation of ethinyl estradiol by nitrifying activated sludge. Chemosphere 41, 1239–1243. Yi, T., Harper Jr., W.F., 2007. The link between nitrification and biotransformation of 17(alpha)-ethinylestradiol. Environ. Sci. Technol. 41, 4311–4316. Yi, T., Harper, J.W.F., in press. The effect of biomass characteristics on the partitioning and sorption hysteresis of 17(alpha)-ethinylestradiol. Water Res. (Corrected proof). Yi, T., Harper, J.W., Holbrook, R., Love, N., 2006. Role of particle size and ammonium oxidation in removal of 17a-ethinyl estradiol in bioreactors. J. Environ. Eng. 132 (11), 1527–1529.