Efficient azo dye removal in bioelectrochemical system and post-aerobic bioreactor: Optimization and characterization

Chemical Engineering Journal 243 (2014) 355–363

Contents lists available at ScienceDirect

Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej

Efficient azo dye removal in bioelectrochemical system and post-aerobic bioreactor: Optimization and characterization Dan Cui a, Yu-Qi Guo b, Hyung-Sool Lee c, Hao-Yi Cheng a, Bin Liang a, Fan-Ying Kong a, You-Zhao Wang a, Li-Ping Huang d, Mei-Ying Xu e, Ai-Jie Wang a,⇑ a

State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, No. 202 Haihe Road, Harbin 150090, PR China The Architecture Design and Research Institute of Harbin Institute of Technology, No.202 Haihe Road, Harbin, 150090, PR China Department of Civil and Environmental Engineering, University of Waterloo, 200 University Avenue West Waterloo, Ontario N2L 3G1, Canada d Key Laboratory of Industrial Ecology and Environmental Engineering, Ministry of Education (MOE), School of Environmental Science and Technology, Dalian University of Technology, Dalian 116024, PR China e State Key Laboratory of Applied Microbiology, Guangdong Institute of Microbiology, 100 Central Xianlie Road, Guangzhou 510070, PR China b c

h i g h l i g h t s  A new refractory wastewater treatment process (UBER–ABOR) was developed.  Alizarin Yellow R as the mode of azo dyes was efficiently removed.  The effect of cathode size on the performance of UBER was investigated.  The HRTs of UBER and ABOR were optimized.  The degradation mechanism of azo dye was discussed.

a r t i c l e

i n f o

Article history: Received 18 July 2013 Received in revised form 24 October 2013 Accepted 27 October 2013 Available online 4 November 2013 Keywords: Up-flow bio-electrocatalyzed electrolysis reactor (UBER) Cathode size Hydraulic retention time (HRT) AYR by-products Aerobic bio-contact oxidation reactor (ABOR)

a b s t r a c t A new process of an up-flow bio-electrocatalyzed electrolysis reactor (UBER) connected with an aerobic bio-contact oxidation reactor (ABOR) was developed for treating azo dye wastewater. Alizarin Yellow R (AYR), used as a model dye, was efficiently decolorized in UBERs (97.5 ± 1.0%) and further mineralized in the subsequent aerobic bio-contact oxidation reactor (ABOR). Decolorization efficiency was improved with increasing cathode size in UBERs, but AYR removal rate and current density were not increased in proportion to cathode size, mainly due to the limitation of anodic reactions. AYR decolorization rate was optimized at a cathode size of 90 cm3 in an UBER where the charge transfer resistance Rct (39.5 X) was minimal. We assessed the effect of hydraulic retention time (HRT: 6.5 h, 4.5 h, 3.5 h, 2.0 h) on the removal of residual by-products (p-phenylenediamine (PPD) and 5-aminosalicylic acid (5-ASA) in the ABOR. The concentrations of PPD and 5-ASA decreased down to 0.28 ± 0.01 and 0.27 ± 0.03 mg L1, respectively, in an optimum HRT 3.5 h. Decolorization efficiency and COD removal efficiency was 93.8 ± 0.7% and 93.0 ± 0.5% in the combined process of UBER and ABOR in overall HRT 6 h (HRT 2.5 h in UBER + HRT 3.5 h in ABOR). The Chroma in ABOR effluent was 80 times. Ó 2013 Elsevier B.V. All rights reserved.

1. Introduction Azo dyes are the largest chemical class of dyes and are frequently used for textile dying and paper printing industries due to cheap costs, firmness, and a variety of colors compared to natural dyes [1]. However, intensive color of dye-containing wastewater leads to severe aesthetic problems and obstructs light penetration and oxygen transfer into water bodies, adversely affecting aquatic life [2]. For these reasons, the color removal from dye-containing wastewater is one of major concerns in China ⇑ Corresponding author. Tel./fax: +86 451 86282195. E-mail address: [email protected] (A.-J. Wang). 1385-8947/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.cej.2013.10.082

where textile industry has grown exponentially in recent years [3]. Azo dyes are characterized by the number of azo groups (AN@NA), which was typically recalcitrant to microbial aerobic oxidation but could be reduced easily [4]. The azo group is substituted with benzene or naphthalene groups, which can contain many different substituents, such as chloro (ACl), methyl (ACH3), nitro (ANO2), amino (ANH2), hydroxyl (AOH), and carboxyl (ACOOH). Azo dyes typically remain in tail water after biological wastewater treatment processes because of their recalcitrance; by-products of azo dyes are even toxic and mutagenic [5]. Thus, the complete mineralization of azo dyes into carbon dioxide is desirable to protect ecosystem and human health. A number of removal technologies have been developed, which include dyestuff

356

D. Cui et al. / Chemical Engineering Journal 243 (2014) 355–363

adsorption [6,7], electrochemical oxidation or reduction methods [8,9], electrochemical treatment combined with ultrasound technique [10], electrochemical coagulation [11], advanced oxidation [12] and membrane separation processes [13,14]. Each of these techniques offers advantages, but also has limitations, such as significant amounts of chemicals, high operating and maintenance costs, and large amounts of excessive sludge. Biological anaerobic treatment was proposed as an alternative approach to improve the economic efficiency in azo dye wastewaters treatment [15–17]. However, anaerobic treatment is usually very slow (large footprint) and would require exogenous electron donor for reducing azo dye, which can increase investment, and operating and maintenance costs [17,18]. In recent years, bio-electrochemical systems (BESs) have been explored to reduce refractory substances including azo dyes into stable, less toxic forms of aromatic amines in more economical ways because anodic bacteria can oxidize biodegradable fractions of wastewaters as electron donor, deliver them to the anode, and refractory matters could be electrochemically reduced to less toxic forms on the cathode in BESs. Based on the principle, anaerobic reductions of refractory substances using BESs have been studied extensively, which include nitrobenzene [19–21], copper (II) [22,23], chloroethenes [24,25], 2-chlorophenol [26], iodinated Xray contrast media [27], and azo dyes [28,29]. Previous reports demonstrated that Alizarin Yellow R (AYR) was reduced in a twochamber bio-cathode biocatalyzed electrolysis reactor, which was inoculated with the enriched inoculum [30]. Congo red and active brilliant red X-3B were decolorized in air-cathode single-chamber BESs, which have proton exchange membrane or microfiltration membrane as separator between electrodes [31,32]. Abiotic decolorization of acid orange 7 on the cathode was studied in a dual-chamber BES, where acetate was used as electron donor for exoelectrogens at the anode [19]. An efficient decolorization of the real dye wastewater and power generation was successfully achieved using a BES with granular carbon bioanode and biocathode [33]. However, architectures and configurations of previous dual-chamber BESs would not be ideal for full-scale applications, due to large footprint and expensive capital costs. For instance, the presence of membrane in dual-chamber BESs not only develops a pH gradient between two chambers, but also increases energy losses [34]. Thus, BESs produce low voltage in fuel cell mode or require high energy input in electrolysis mode [35]; such energy losses would be substantial in BESs fed with wastewater [36]. BES design should be optimized for wastewater treatment considering scale effects (capital costs and footprint). An up-flow biocatalyzed electrolysis reactor (UBER) lacking membrane proposed by Wang et al. [20] is scalable. Its investments are relatively cheap due to the lack of membrane and low-cost electrode materials (graphite granular and carbon brush without metal catalysts). Moreover, the flexible position and size of anode and cathode not only makes UBER flexible for treating diverse wastewaters, but also allow UBERs to be installed easily. Using UBER configuration nitrobenzene [20] and azo dye (AYR) [28] were efficiently reduced. Previous works have shown the accumulation of intermediates during anaerobic reduction of azo dye, such as p-phenylenediamine (PPD) and 5-aminosalicylic acid (5-ASA), which would be potentially toxic [32]. Thus, we employed the aerobic bio-contact oxidation reactor (ABOR) as posttreatment to the UBER in our study in order to further oxidize the intermediates into carbon dioxide. However, existing works do not provide information on design and operation parameters essential for scale-up of UBERs (e.g. characterization of resistances in UBERs, cathode size effect, HRT effect on fates of azo dyes in UBERs, and HRT effect on reductive by-products of azo dye in post-aerobic system). In our previous study, it was found that stacked granular graphite cathode and HRT were critical for

decolorizing azo dye in an UBER [29]. It confirmed that reductive by-products of azo groups (aromatic amines) were further oxidized to carbon dioxide in a post-aerobic bioreactor. While, operating conditions were not optimized in the post-system, and more information on fates of reductive by-products would be needed for field application. Hence, in this study, we first evaluated the effect of cathode size on decolorization efficiency of AYR in UBERs. Second, HRT was optimized for AYR decolorization in UBERs having different cathode sizes. Third, we assessed HRT effects on the oxidation of remaining AYR and its reductive byproducts in a post ABOR. Finally, we tracked the fates of AYR by-products in an UBER and the post-aerobic bioreactor and proposed metabolic pathways in the combined bioprocess. 2. Materials and methods 2.1. Construction of UBER–ABOR This study evaluated transformation of AYR in a continuous bioprocess that combines an UBER (Fig. 1A and B) with an ABOR (Fig. 1C). Six laboratory scale UBERs were manufactured and used for these experiments (Fig. 1B). Briefly describing, the UBER consists of a cathode zone at the bottom and an anode zone on the top without membrane. Graphite granules in a range of 3–5 mm were stacked as cathodes and a carbon brush with a diameter of 4.5 cm was used as the anode. The total volume of the UBERs was 250 mL, and the volume of anode zone was fixed at 45 cm3; refer to the literature [28,29] for details. Three different volumes of the cathode zone were set by adjusting a height of effluent ports in the UBERs, which were 45 cm3 (UBER-1), 90 cm3 (UBER-2) and 135 cm3 (UBER-3), respectively (the working volume of the UBER was 180 mL, 160 mL, and 140 mL in order). To improve experimental reliability, two identical UBERs for each cathodic configuration under the same operating conditions were operated. The distance between the anode and the cathode was fixed at 2 cm. The UBERs were operated with external resistance of 10 and applied voltage of 0.5 V using a DC power supply (IT6921, Itech Co., Ltd., USA). A saturated calomel electrode (SCE+247 mV vs. standard hydrogen electrode) (Shanghai Precision Scientific Instruments Co., Ltd., China) was placed between the anode and the cathode to measure half potentials; here the electrode potential was reported against SCE. Three samples ports (SPs) were installed at 1.0, 6.0, and 12.0 cm above the bottom (designated as SP1#, SP2#, and SP3#) to monitor the change of chemical compositions throughout the reactors. Effluent from the UBERs was pumped to the subsequent ABOR for further oxidizing the intermediates of AYR under aerobic conditions. 2.2. Inoculation and operational conditions The anode-respiring bacteria (ARB) were acclimated in a microbial fuel cell using carbon brush as the anode and acetate as carbon and electron source for 3 months. The carbon brushes were then transferred to the anode zone of the UBERs. The cathode zone was not inoculated with ARB to improve abiotic reduction of AYR on the cathode; a previous study showed that biofilm formation on the cathode deteriorated AYR reduction on the cathode [29]. The ABOR was inoculated with recycle activated sludge collected from Taiping municipal wastewater treatment plant in Harbin, China. The effect of cathode volume (i.e. cathode size) on AYR reduction was evaluated in the UBERs. Three groups of UBERs with the total cathode volume (TCV) of 45 cm3, 90 cm3 and 135 cm3 were operated so that the anode volume to cathode volume ratio was 1:1, 1:2, and 1:3 for UBER-1, UBER-2, and UBER-3,

D. Cui et al. / Chemical Engineering Journal 243 (2014) 355–363

357

Fig. 1. Schematic diagram of an up-flow biocatalyzed electrolysis reactor (UBER) and aerobic bio-contact oxidation reactor (ABOR). (A) configuration of up-flow biocatalyzed electrolysis reactor (UBER). Influent flows from graphite granular cathodes (the bottom of UBER) to carbon brush anodes through a water distributor. UBER has three samples ports (SP #1 to #3) to better track AYR reductions at different positions in the UBER; (B) three sets of the six UBERs having different cathode volumes. Two UBERs were run under the same condition to improve replicability of experimental results; (C) ABOR having fixed biofilm carriers. Effluent from the UBER-2 was fed to the ABOR in experiments and (D) the combined process of UBER and ABOR.

respectively. Sodium acetate was the primary electron donor and carbon source for ARB in the anode. The composition of medium includes sodium acetate (1.0 g L1), AYR (100 mg L1, Shanghai Bioelectrochemical Industry Co., Ltd., China. Commercial purity), potassium phosphate (50 mM, pH 7.0), KCl (0.13 g L1), NH4Cl (0.31 g L1,), Wolfe’s trace element solution (0.5 mL L1) [37] and vitamin solution (0.5 mL L1). Medium pH was adjusted with HCl (0.1 mol L1) to 7.0 ± 0.2. The medium was fed to the UBERs with a peristaltic pump (BT 100 L, Longer Pump Co., Ltd., China). The hydraulic retention time (HRT) of the UBERs was decreased from 8 to 1.5 h, which corresponds to the AYR loading rate from 200 to 900 g m3 d1 in the UBERs. Electrode potentials and currents were monitored at every 10 min with a data acquisition system (Keithley 2700, Keithley Co., Ltd., USA). To further study aerobic degradation of AYR and its reductive intermediates, the effluent from the UBER was fed to the ABOR with a peristaltic pump (BT100-1L, Longer Pump Co., Ltd., China) (Fig. 1D). HRT was varied from 6.5 to 2.5 h in the ABOR to evaluate the oxidation of AYR intermediates. A secondary settling tank was connected with the ABOR, and then the effluent from the settling

tank was collected – the final effluent. All experiments were conducted at 25 °C. 2.3. Analytical methods 2.3.1. Chemical analysis Liquid samples taken from bioreactors were immediately filtered through a 0.22 lm filter (Xingya Material Co., Shanghai, China). AYR concentration was quantified using a UV–visible spectrophotometer (Shimadzu UV2550, Japan) at 387 nm. Reduced products of AYR were identified using a liquid chromatography– mass spectrometer (LC–MS, ThermoFinnigan LCQ Deca XP Max LC/MS, Germany), which was equipped with an electrospray ionization source and operated in the positive/negative polarity mode. Separation was performed with a C18 column (5 lm; 5 mm  250 mm, Waters Co., USA) and mobile phase was methanol/H2O (9:1) at a flow rate of 1 mL min1. The identified products by the LC–MS were then quantified with a high performance liquid chromatography (HPLC, model e2695, Waters Co., USA) under the same condition to the LC–MS using a UV detector (288 nm);

358

D. Cui et al. / Chemical Engineering Journal 243 (2014) 355–363

Fig. 2. Performance comparison of different UBERs: (A) Variation of AYR loading rate (AYR LR) and decolorization efficiency; (B) AYR removal rate based on the total volume (TV); (C) AYR removal rate based on the total cathode volumes (TCV); (D) current density based on the total cathode volumes (TCV) and (E) the current density based on the anode volumes (TAV) in the UBERs.

p-phenylenediamine (PPD) and 5-aminosalicylic acid (5-ASA) were main reductive products of AYR. Their retention times were 2.1 and 3.2 min, respectively, in the HPLC. AYR decolorization efficiency (DE), AYR removal rate of the entire volume of UBERs (RR: g m3 TV d1) and AYR removal rate in the cathode zone (RRcathode: g m3 TCV d1) were calculated based on the difference between influent and effluent concentrations and reactor empty total volume (TV, cm3) as well as total cathode volumes (TCV, cm3);

DE ¼

Cin-AYR  Cef-AYR  100% Cin-AYR

ð1Þ

RR ¼

ðCin-AYR  Cef-AYR Þ  Q A  106 TV

RRcathode ¼

ðCin-AYR  Cef-cathode-AYR Þ  Q A  106 TCV

ð2Þ

ð3Þ

where Cin-AYR is the influent AYR concentration, mg L1, Cef-AYR is the effluent AYR concentration, mg L1. Cef-cathode-AYR is the effluent AYR concentration of cathode zone, mg L1. QA is the influent flow rate, m3 d1. COD was measured according to standard method (APHA, 1998). Chroma was measured according to Chinese standard

D. Cui et al. / Chemical Engineering Journal 243 (2014) 355–363

359

Fig. 3. (A) Nyquist plots of UBERs with different electrode ratios; (B) equivalent circuits for UBERs, Rs – solution resistance (ohmic resistance), Rct – charge transfer resistance, Zw – infinite size element for transport across a boundary layer, CPE – constant phase element.

method (GB1190-85, China). All samples were filtered through a filter paper prior to COD and Chroma measurements. 2.3.2. Electrochemical monitoring and calculation The current was calculated from the external resistance using Ohm’s law. The volumetric current density was reported based on the total cathode volume (A m3 TCV) and total anode volume (A m3 TAV), respectively. In order to characterize ohmic, charge, and the whole cell resistances, electrochemical impedance spectroscopy (EIS) measurements were carried out in a frequency range of 105–0.01 Hz with an AC signal of 10 mV amplitude using an electrochemical workstation (model-660D, CH Instruments Inc., US) equipped with two-electrode system. To measure the whole cell resistance, the cathode was served as the working electrode and the anode was served as the counter electrode. The Nyquist plot expresses the impedance with a real part (plotted on the X-axis) and an imaginary part (plotted on the Y-axis that is negative). Each point on the Nyquist plot represents the impedance at a certain frequency [38]. Based on the experimental EIS data, a Randles circuit model was used for quantifying ohmic resistance (Rs), charge transfer resistance (Rct), and whole cell resistance (RUBER); the infinite size element for transport across a boundary layer (Zw) and a parallel capacitance (CPE) were included in the Randles circuit model [38]. The model describing our EIS results is a generalized Randles circuit with an infinite length Warburg element for the diffusion transport of reactant to the reacting surface. We used the software of Zsimpwin to simulate EIS data for the Randles model. The type of simulation was complex (both the real and imaginary impedance values were fit) and the maximum number of iterations for convergence was 100. The error estimation for each parameter was always below 5% for the diffusive components and below 1% for the pure resistive and capacitive elements. 3. Results and discussion 3.1. AYR decolorization in the UBERs with different cathode sizes Decolorization efficiency of AYR was enhanced with increasing cathode size. At HRT 8 h (AYR loading rate 205.9 ± 10.8 g m3 d1), the decolorization efficiencies in UBER-1, UBER-2 and UBER-3 were 90.6 ± 3.9%, 94.5 ± 1.7% and 97.5 ± 1.0%, respectively. However, as HRT decreased and AYR loading rate increased, cathode volumes became significant for the decolorization efficiency in the UBERs.

As shown in Fig. 2A, the decolorization efficiency was decreased down to 81.3 ± 5.9% and 87.1 ± 5.7%, respectively, in UBER-2 and UBER-3 at a short HRT of 2 h. UBER-1 (the smallest cathode volume) showed significant HRT impact on the decolorization efficiency; it was reduced to 77.3 ± 6.3% at HRT 3 h (AYR loading rate of 678.5 ± 15.2 g m3 d1), and seriously dropped down to 44.0 ± 6.1% and 43.0 ± 1.6% at HRT 2 and 1.5 h (AYR loading rate of 86.6 ± 10.6 and 908.4 ± 10.3 g m3 d1), respectively. These results indicated that AYR decolorization on the cathode would be kinetically limited in UBERs. Fig. 2B (AYR removal rate and AYR loading rate expressed with total volume (250 cm3)) showed that AYR removal rates of AYR in the three series UBERs were similar at different AYR loading rates; linear equations for UBER-1, UBER-2, and UBER-3 were y = 0.7446x, y = 0.8111x, and y = 0.9151x (All R2s > 0.95) where y is decolorization rate of AYR and x is AYR loading rate. As shown in Fig. 2B, the maximum decolorization rate in UBER-1, UBER-2 and UBER-3 were 601.1 ± 25.3, 650.0 ± 50.9 and 748.8 ± 51.6 g m3 TV d1, respectively. Apparently, decolorization capability did not increase in proportional to cathode size in the UBERs. A large cathode substantially increased AYR reduction efficiency, but would not improve AYR decolorization rate much. When AYR removal rate was plotted against its loading rate, based on cathode size, it was observed that the decolorization rate based on cathode volume decreased with the cathode size magnification. The fastest decolorization rates for UBER-1, UBER-2 and UBER-3 were 3333.3 ± 12.7, 1880.9 ± 77.1 and 1386.6 ± 44.6 g m3 TCV d1, respectively (Fig. 2C). The highest slope was obtained for UBER-1 in linear equations for AYR removal rate and AYR loading rate; the linear equations for UBER-1, UBER-2 and UBER-3 were y = 4.1369x, y = 2.3897x, and y = 1.6494x (All R2s > 0.97), respectively. This result implied that small cathode volume in UBER-1 was efficiently used for AYR reduction, while parts of larger cathode volumes wound not work in the other UBERs. Fig. 2D supports inefficient utilization of cathodes in UBER-2 and UBER-3. It is suspected that poor hydrodynamic conditions (i.e. advection) in UBERs would create dead zones of cathodes in UBER-2 and UBER-3. Volumetric current density (based on total cathode volume) in UBER-1 was two to three times higher than that in the other UBERs. For example, the current density in UBER-1 at HRT of 2.5 h was 62.6 ± 4.6 A m3 TCV, while those of UBER-2 and UBER-3 were 28.2 ± 2.9 A m3 TCV and 17.2 ± 2.1 A m3 TCV, respectively. The current density ratio of UBER-1, UBER-2 and UBER-3 was 3:2:1 approximately, which is reciprocal of the cathode size ratio (1:2:3). Different anodic reactions (acetate oxidation) in three UBERs might cause this

360

D. Cui et al. / Chemical Engineering Journal 243 (2014) 355–363

Fig. 4. Concentrations of AYR, COD, PPD, and 5-ASA in the combined process of UBER and ABOR. (A) AYR concentration; (B) COD concentration; (C) PPD concentration and (D) 5-ASA concentration.

reverse pattern, but they would not mainly affect current density due to the same inoculum used for the three UBERs. Fig. 2E compared anode volumetric current density in three UBERs. There were relatively large differences in the volumetric current density at long HRT, while the volumetric current density became close with decreasing HRT (increasing ARY loading rate). This trend of anode volumetric current density against HRT and ARY loading rate supports that anodic reactions did not mainly cause higher ARY removal rate and current density in UBER-1. Fig. 3A summarized the experimental and simulated data of the three UBERs at HRT 8 h with AYR loading rate of 205.9 ± 10.8 g m3 d1 using the Randles circuit model in Fig. 3B (see materials and methods for details). Experimental data well accords to simulated ones. The impedance at the high frequency limit is the ohmic resistance Rs. The results showed that ohmic (electrolyte) resistance (Rs) in the three UBERs did not significantly change with the variation of cathode size, which were 40.8 X, 41.6 X and 41.8 X, respectively, for UBER-1, UBER-2 and UBER-3. Ohmic resistance should be linearly increased with increasing cathode volumes given that entire cathode volumes were used for AYR reduction in the UBERs. However, slight increases of ohmic resistance 0.8 X in UBER-2 and 1.0 X in UBER-3 were observed, as compared to UBER-1. These results support that only part of cathode volumes in UBER-2 and UBER-3 reduced AYR, which is consistent to faster AYR removal rate and higher current density in UBER-1 (Fig. 2C and D). A main variation in resistance was observed in charge transfer resistance (Rct), which was 59.9 X, 39.5 X and 45.0 X, respectively, for UBER-1, UBER-2 and UBER-3. Charge transfer resistance (Rct) is affected by the kinetics of electrode reactions. In the UBERs, two parameters can affect Rct values: (i) the

electron transfer between the cathode and electron acceptor (AYR), and (ii) the anode and electron donor (acetate). Lower charge transfer resistances (Rct) in UBER-2 and UBER-3 than UBER-1 indicated that a large cathode size could increase the contact surface area between the cathode and AYR and thus improve the charge transfer between the cathode and AYR. In comparison, slightly higher Rct was found in UBER-3, although UBER-3 has a large cathode size over UBER-2. This result implies that the charge transfer resistance on the anodes would be dominant in UBER-3, which is supported by less current density (Fig. 2D and E). 3.2. Effect of HRT on the AYR decolorized products removal in aerobic bio-contact oxidation reactor (ABOR) To further remove the residual AYR by-products, they were polished using the ABOR connected with the UBER-2. The effluent from the UBER-2 run at HRT 2.5 h was introduced into the ABOR that was operated at different HRTs: 6.5 h, 4.5 h, 3.5 h and 2.5 h. Experimental results showed that two reductive products, PPD and 5-ASA, were formed in the effluent from UBER-2. The formation efficiencies of PPD and 5-ASA in the UBER were nearly 100% and 80% at HRT 2.5 h, respectively. As shown in Fig. 4A, most of AYR was removed in the UBER. The AYR concentration was decreased from 100 to below 10 mg L1 in the UBER effluent (decolorization efficiency > 90%), and a small part of AYR was further removed in the ABOR. Total decolorization efficiency in the combined process of UBER and ABOR was up to 98.1 ± 2.0% when HRT of ABOR was 6.5 h; however, it gradually decreased to 90.6 ± 0.5% at HRT 2.5 h (see Table S1). This result indicated that the cleavage of azo bond in AYR is resistant to aerobic

361

D. Cui et al. / Chemical Engineering Journal 243 (2014) 355–363

Relative Abundance

(A)

100 92.03

80 60

92.03

H 2N

40

NH 2

20 108.92

0

50

55

60

65

70

75

80

85

90

95

100

105

110

115

m/z

Relative Abundance

(B)

100 136.00

80

COOH 137

60

136

H 2N

OH

40 20 135.05

0

137.04 154.00

50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 135 140 145 150 155

m/z

Relative Abundance

(C)

100 93.03

80 93

60

H 2N

OH

40 20 110.06

0

50

55

60

65

70

75

80

85

90

95

100

105

110

115

m/z

Relative Abundance

(D)

100 162.25

80 60

102.83 NH 2 H2 H2 HOOC C C C H

145.30 H2 C

COOH

40 20 102.83

0

145.30

50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 135 140 145 150 155 160 165

m/z Fig. 5. HPLC–MS chromatograms of effluents from UBER 2 and the ABOR. (A) Polarity scan for PPD identify; (B) polarity scan for 5-ASA identify; (C) polarity scan for 4aminophenol; and (D) polarity scan for 3-amino-1, 6-adipic acid identify.

oxidation [39]. In comparison, COD concentration sharply declined in the ABOR (see Fig. 4B). The COD removal in the UBER was due to the consumption of acetate by anodic bacteria. While in the ABOR, residual acetate and the reductive products of AYR were simultaneously oxidized by the abundant aerobic bacteria on the filler carrier surface (Fig. S1). Total COD removal efficiency in the combined process varied from 94.2 ± 4.0% to 86.5 ± 2.7%. As discussed above, AYR was reduced to PPD and 5-ASA in the UBER, which were

resistant to further degradation under anaerobic condition, but could be oxidized under aerobic conditions. Fig. 4C and D clearly showed that 5-ASA and PPD were efficiently oxidized in the ABOR. The concentration of 5-ASA ranged from 0.05 ± 0.01 to 0.27 ± 0.03 mg L1 and the concentration of PPD was 0.50 ± 0.01 to 0.28 ± 0.01 mg L1 in effluent from the ABOR run under HRT 3.5 to 6.5 h. However, it found that the concentrations of PPD and 5-ASA sharply increased to 13.6 mg L1 and 19.7 mg L1 at

362

D. Cui et al. / Chemical Engineering Journal 243 (2014) 355–363

from UBER-2. The product with m/z 110 [M + H]+ was 4-aminophenol and the last one with m/z 162 [M + H]+ was 3-amino-1, 6-adipic acid. Both of them were also identified in the polarity scan using the corresponding fragment products as shown in Fig. 5C and D. Based on degradation products identified with LC–MS, the possible transformation pathway of AYR in the UBER coupled with the ABOR is suggested in Fig. 6. Individual reactions for AYR reduction on the cathode in the UBER are described in Eqs. (4)–(6) [32], in which the AN@NA double bond could be reduced to hydrazo (4) or amine (5), via the consumption of two or four electrons. ANO2 could be reduced to ANH2 according to Eq. (6), via the consumption of six electrons.

AN@NA þ 2e þ 2Hþ ! ANHANHA

ð4Þ

AN@NA þ 4e þ 4Hþ ! ANH2 þ NH2

ð5Þ

ANO2 þ 6e þ 6Hþ ! ANH2 þ 2H2 O

ð6Þ

AYR got electrons from the cathode in the UBER, resulting in the formation of colorless products PPD and 5-ASA. The AYR by-products could be further oxidized to simple acids via 4-aminophenol and 3-amino-catechol in the ABOR. Fig. 6 shows more detailed information on proposed metabolic pathways. 4-aminopheno can be oxidized and 3-amino-catechol would be formed in aerobic conditions; then, benzene ring is cleaved by oxygenases catalysis. The double bonds of AC@CA add HA and are cleaved to form 3-amino1, 6-adipic acid, which would be further oxidized to 3-oxoadipate. Finally, 3-oxoadipate can be converted to succinate and acetate acid. These simple acids could be easily oxidized to CO2 and H2O in citric acid cycle. 4. Conclusions

Fig. 6. A proposed metabolic pathway of AYR in the UBER and ABOR, based on LC– MS results.

HRT 2.5 h, which were much higher than the limit concentrations in the textile wastewater discharge standard II in China. Therefore, it was concluded that an optimum HRT for removing 5-ASA and PPD would be close to 3.5 h in the ABOR considering the footprint of the ABOR against effluent quality.

AYR was successfully discolored and mineralized in the integrated process of UBER and ABOR. Cathode size was significant for AYR decolorization efficiency, which reached at 97.5 ± 1.0% (in UBER-3) along with product formation efficiencies of 90% for PPD and 80% for 5-ASA. AYR decolorization was increased with increasing cathode size, but AYR removal rate and current density were not improved in proportion to cathode size, due to limitation in anodic reaction rate. When the volumes of anode and cathode zones were 45 cm3 and 90 cm3, UBERs presented the best performance with the lowest electron transfer resistance (Rct) of 39.5 X. Reductive by-products of AYR were further mineralized in the ABOR. Main by-products PPD and 5-ASA were aerobically degraded to 0.28 ± 0.01 mg L1 and 0.27 ± 0.03 mg L1, respectively, in an optimized HRT of 3.5 h in the ABOR. Overall decolorization and COD removal efficiencies were 93.8 ± 0.7% and 93.0 ± 0.5%, respectively, in the combined process, together with low Chroma (80 times) and negligible accumulations of PPD and 5-ASA.

3.3. Possible metabolic pathway of AYR in UBER–ABOR Acknowledgments To understand AYR structure changes in anaerobic and aerobic degradation, the chemical structures of AYR by-products were identified using LC–MS. The full scan and polarity scan for each identified products were shown in Fig. S2(A–D) and Fig. 5(A–D), respectively. Four main peaks emerged with m/z ratio of 109, 154, 110, 162 in positive polarity mode, respectively. The product with m/z of 109 [M + H]+ (Fig. 5A) was in accordance with PPD. The fragment product in the polarity scan with m/z 92.03 corresponds to PPD without ANH2. The product with m/z 154 [M + H]+ (Fig. 5B) was 5-ASA. When 5-ASA loses ANH2 or AOH in the polarity scan, two fragment products with m/z 136 and 137 were formed. PPD and 5-ASA were the major products in the effluent

This research was supported by the National Natural Science Foundation of China (NSFC, Grant No. 51078100), by National Science Foundation for Distinguished Young Scholars (Grant No. 51225802), by Science Fund for Creative Research Groups of the National Natural Science Foundation of China (Grant No. 51121062), by the National High-tech R&D Program of China (863 Program, Grant No. 2011AA060904), by The Ph.D. Programs Foundation of Ministry of Education of China (20102302110055), by ‘‘Hundred Talents Program’’ of the Chinese Academy of Sciences, and by Heilongjiang Science Foundation for Distinguished Young Scholars (Grant No. JC201003).

D. Cui et al. / Chemical Engineering Journal 243 (2014) 355–363

Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cej.2013.10.082. References [1] A. Stolz, Basic and applied aspects in the microbial degradation of azo dyes, Appl. Microbiol. Biotechnol. 56 (2001) 69–80. [2] J. Sun, Y.Y. Hu, Z. Bi, Y.Q. Cao, Simultaneous decolorization of azo dye and bioelectricity generation using a microfiltration membrane air-cathode singlechamber microbial fuel cell, Biores. Technol. 100 (2009) 3185–3192. [3] X.J. Lu, L. Liu, R.R. Liu, J.H. Chen, Textile wastewater reuse as an alternative water source for dyeing and finishing processes: a case study, Desalination 258 (2010) 229–232. [4] G.M. Shaul, T.J. Holdsworth, C.R. Dempsey, K.A. Dostal, Fate of water-soluble azo dyes in the activated-sludge process, Chemosphere 22 (1991) 107–119. [5] K. Selvam, K. Swaminathan, K.S. Chae, Microbial decolorization of azo dyes and dye industry effluent by Fomes lividus, World J. Microbiol. Biotechnol. 19 (2003) 591–593. [6] R. Gong, M. Li, C. Yang, Y.Z. Sun, J. Chen, Removal of cationic dyes from aqueous solution by adsorption on peanut hull, J. Hazard. Mater. 121 (2005) 247–250. [7] V.K. Gupta, Suhas, Application of low-cost adsorbents for dye removal-a review, J. Environ. Manage. 90 (2009) 2313–2342. [8] A.G. Alberto, D. Pletcher, The removal of low level organics via hydrogen peroxide formed in a reticulated vitreous carbon cathode cell. Part 2: the removal of phenols and related compounds from aqueous effluents, Electrochim. Acta. 44 (1999) 2483–2492. [9] Y. Yoshida, S. Ogata, S. Nakamatsu, T. Shimamune, K. Kikawa, H. Inoue, C. Iwakura, Decoloration of azo dye using atomic hydrogen permeating through a Pt-modified palladized Pd sheet electrode, Electrochim. Acta. 45 (1999) 409– 414. [10] M. Rivera, M. Pazos, M.A. Sanromán, Improvement of dye electrochemical treatment by combination with ultrasound technique, J. Chem. Technol. Biotechnol. 84 (2009) 1118–1124. [11] C.L. Yang, J. McGarrahan, Electrochemical coagulation for textile effluent decolorization, J. Hazard. Mater. 127 (2005) 40–47. [12] B. Liang, Q. Yao, H.Y. Cheng, S.H. Gao, F.Y. Kong, D. Cui, Y.Q. Guo, N.Q. Ren, A.J. Wang, Enhanced degradation of azo dye Alizarin Yellow R in a combined process of iron–carbon microelectrolysis and aerobic bio-contact oxidation, Environ. Sci. Pollut. Res. 19 (2012) 1385–1391. [13] C. Diaper, V.M. Correia, S.J. Judd, The use of membranes for the recycling of water and chemicals form dye house effluents: an economic assessment, J. Soc. Dyers Color. 112 (1996) 273–281. [14] S. Sachdeva, A. Kumar, Preparation of nanoporous composite carbon membrane for separation of rhodamine B dye, J. Membr. Sci. 329 (2009) 2–10. [15] S.W. FitzGerald, P.L. Bishop, Two stage anaerobic/aerobic treatment of sulfonated azo dyes, J. Environ. Sci. Health. Part A: Environ. Sci. Eng. Toxicol. 30 (1995) 1251–1276. [16] P. Reyes, M. Pickard, V.D. Rafael, Hydroxybenzotriazole increases the range of textile dyes decolorized by immobilized laccase, Biotechnol. Lett. 21 (1999) 875–880. [17] F.P. Van der Zee, S. Villaverde, Combined anaerobic–aerobic treatment of azo dyes-A short review of bioreactor studies, Water Res. 39 (2005) 1425–1440. [18] A.B. dos Santos, J. Traverse, F.J. Cervantes, J.B. van Lier, Enhancing the electron transfer capacity and subsequent color removal in bioreactors by applying thermophilic anaerobic treatment and redox mediators, Biotechnol. Bioeng. 89 (2005) 42–52. [19] Y. Mu, R.A. Rozendal, K. Rabaey, J. Keller, Nitrobenzene removal in bioelectrochemical systems, Environ. Sci. Technol. 43 (2009) 8690–8695.

363

[20] A.J. Wang, H.Y. Cheng, B. Liang, N.Q. Ren, D. Cui, N. Lin, B.H. Kim, K. Rabaey, Efficient reduction of nitrobenzene to aniline with a biocatalyzed cathode, Environ. Sci. Technol. 45 (2011) 10186–10193. [21] A.J. Wang, D. Cui, H.Y. Cheng, Y.Q. Guo, F.Y. Kong, N.Q. Ren, W.M. Wu, A membrane-free, continuously feeding, single chamber up-flow biocatalyzed electrolysis reactor for nitrobenzene reduction, J. Hazard. Mater. 199–200 (2012) 401–409. [22] H.C. Tao, W. Li, M. Liang, N. Xu, J.R. Ni, W.M. Wu, A membrane-free baffled microbial fuel cell for cathodic reduction of Cu (II) with electricity generation, Biores. Technol. 102 (2011) 4774–4778. [23] H.C. Tao, M. Liang, W. Li, L.J. Zhang, J.R. Ni, W.M. Wu, Removal of copper from aqueous solution by electrodeposition in cathode chamber of microbial fuel cell, J. Hazard. Mater. 189 (2011) 186–192. [24] S.M. Strycharz, T.L. Woodard, J.P. Johnson, K.P. Nevin, R.A. Sanford, F.E. Loffler, D.R. Lovley, Graphite electrode as a sole electron donor for reductive dechlorination of tetrachlorethene by Geobacter lovleyi, Appl. Environ. Microbiol. 74 (2008) 5943–5947. [25] S.T. Lohner, D. Becker, K.M. Mangold, A. Tiehm, Sequential reductive and oxidative biodegradation of chloroethenes stimulated in a coupled bioelectroprocess, Environ. Sci. Technol. 45 (2011) 6491–6497. [26] S.M. Strycharz, S.M. Gannon, A.R. Boles, A.E. Franks, K.P. Nevin, D.R. Lovley, Reductive dechlorination of 2-chlorophenol by Anaeromyxobacter dehalogenans with an electrode serving as the electron donor, Environ. Microbiol. Rep. 2 (2010) 289–294. [27] Y. Mu, J. Radjenovic, J.Y. Shen, R.A. Rozendal, K. Rabaey, J. Keller, Dehalogenation of iodinated X-ray contrast media in a bioelectrochemical process, Environ. Sci. Technol. 45 (2011) 782–788. [28] Y. Mu, K. Rabaey, R.A. Rozendal, Z. Yuan Jr., Keller, Decolorization of azo dyes in bioelectrochemical process, Environ. Sci. Technol. 43 (2009) 5137–5143. [29] D. Cui, Y.Q. Guo, H.Y. Cheng, B. Liang, F.Y. Kong, H.S. Lee, A.J. Wang, Azo dye removal in a membrane-free up-flow biocatalyzed electrolysis reactor coupled with an aerobic bio-contact oxidation reactor, J. Hazard. Mater. 239–240 (2012) 257–264. [30] D. Cui, F.Y. Kong, B. Liang, D. Liu, Q. Sun, A.J. Wang, Decolorization of azo dyes in a dual-chamber biocatalyzed electrolysis system seeding with enhanced inoculum, J. Environ. Analytic. Toxicol. S3-001 (2011), http://dx.doi.org/ 10.4172/2161- 0525. [31] Y.Q. Cao, Y.Y. Hu, J. Sun, B. Hou, Explore various co-substrates for simultaneous electricity generation and Congo red degradation in air-cathode singlechamber microbial fuel cell, Bioelectrochemistry 79 (2010) 71–76. [32] J. Sun, Z. Bi, B. Hou, Y.Q. Cao, Y.Y. Hu, Further treatment of decolorization liquid of azo dye coupled with increased power production using microbial fuel cell equipped with an aerobic biocathode, Water Res. 45 (2011) 283–291. [33] J. Kanagaraj, A.B. Mandal, Combined biodegradation and ozonation for removal of tannins and dyes for the reduction of pollution loads, Environ. Sci. Pollut. Res. 19 (2012) 42–52. [34] H. Liu, B.E. Logan, Electricity generation using an air-cathode single chamber microbial fuel cell in the presence and absence of a proton exchange membrane, Environ. Sci. Technol. 38 (2004) 4040–4046. [35] P.A. Selembo, M.D. Merrill, B.E. Logan, The use of stainless steel and nickel alloys as low-cost cathodes in microbial electrolysis cells, J. Power Sources 190 (2009) 271–278. [36] H.V. Hamelers, A. Ter Heijne, T.H. Sleutels, A.W. Jeremiasse, D.P. Strik, C.J. Buisman, New applications and performance of bioelectrochemical systems, Appl. Microbiol. Biotechnol. 85 (2010) 1673–1685. [37] E.A. Wolin, M.J. Wolin, R.S. Wolfe, Formation of methane by bacterial extracts, J. Biol. Chem. 238 (1963) 2882–2886. [38] Z. He, M. Florian, Exploring the use of electrochemical impedance spectroscopy (EIS) in microbial fuel cell studies, Energy Environ. Sci. 2 (2009) 215–219. [39] U. Pagga, D. Brown, The degradation of dyestuffs: Part II behaviour of dyestuffs in aerobic biodegradation test, Chemosphere 15 (1986) 479–491.