Bioresource Technology 249 (2018) 844–850

Contents lists available at ScienceDirect

Bioresource Technology

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

Accelerated removal of high concentration p-chloronitrobenzene using MARK bioelectrocatalysis process and its microbial communities analysis ⁎ Xinhong Penga,b, , Xianhui Pana, Xin Wangb, Dongyang Lia, Pengfei Huanga, Guanhua Qiua, Ke Shana, Xizhang Chua a Institute of Seawater Desalination and Multipurpose Utilization, State Oceanic Administration (SOA), Nankai District, Tianjin 300192, China b MOE Key Laboratory of Pollution Processes and Environmental Criteria, Tianjin Key Laboratory of Environmental Remediation and Pollution Control, Nankai University, No. 38 Tongyan Road, Jinnan District, Tianjin 300350, China

ARTICLE INFO ABSTRACT

Keywords: p-Chloronitrobenzene (p-CNB) is a persistent refractory and toxic pollutant with a concentration up to 200 mg/L Bio-electrocatalysis in industrial wastewater. Here, a super-fast removal rate was found at 0.2–0.8 V of external voltage over a p-CNB p-Chloronitrobenzene concentration of 40–120 mg/L when a bioelectrochemical technology is used comparing to the natural biode- Initial concentration gradation and electrochemical methods. The reduction kinetics (k) was fitted well according to pseudo-first Applied voltage order model with respect to the different initial concentration, indicating a 1.12-fold decrease from 1.80 to Microbial communities diversity − 0.85 h 1 within the experimental range. Meanwhile, the highest k was provided at 0.5 V with the characteristic of energy saving. It was revealed that the functional bacterial (Propionimicrobium, Desulfovibrio, Halanaerobium, Desulfobacterales) was selectively enriched under electro-stimulation, which possibly processed Cl-substituted nitro-aromatics reduction. The possible degradation pathway was also proposed. This work provides the ben- eficial choice on the rapid treatment of high-concentration p-CNB wastewater.

1. Introduction 2015). Therefore, it is absolutely necessary to seek new, green, and high-efficient approach to accelerate the reductive transformation of p- p-Chloronitrobenzene (p-CNB), a kind of important raw material for CNB to low-toxic compounds. industry synthesis of pesticides, medicines, lumber preservatives, sul- Bio-electrocatalysis conversion, noted as the association of bio-de- phur dyes, pharmaceuticals, gasoline additives and other chemicals gradation and electrochemical catalytic reduction, has recently been (Xia et al., 2011), has the characterization of high persistence in ex- used for successful bioremediation as an innovative, more effective and posed sediments and soil organic material, and poses a critical threat to less expensive wastewater treatment solution for emerging environ- the human health and social safety, due to their mutagenic, carcino- mental issue (Du et al., 2017), and has obvious advantages in con- genic and teratogenic effects resulting from the occurrences of elec- taminant detoxification accompanying with energy saving and con- trophilic nitro and chlorine groups (Li et al., 2015). It is reported that sumption reduction (Puyol et al., 2017), which is significantly against the highest concentration of p-CNB is up to 200 mg/L in industry from the conventional anaerobic and electrochemical processes. At wastewater (Linch, 1974). Thus, p-CNB is listed as precedence-con- present, two main bio-electrocatalytic processes are accepted: microbial trolled organic pollutant by US EPA and EC (Jones et al., 2006). The fuel cell and microbial electrolysis cell. The former is the direct oxi- traditional treatment processes address physical adsorption (Guo and dation/reduction of recalcitrant wastes by bio-catalysis with self-driven Zheng, 2009), advanced oxidization conversion (Li and Zhu, 2014; Ye electrostimulation, for example, Fan et al. successfully enhanced the et al., 2010; Yuan et al., 2012) and biological degradation (Alatraktchi removal of tetrabromobisphenol by anodic anaerobic co-metabolic et al., 2014; Xia et al., 2011). However, physicochemical methods are biodegradation using biocatalytic fuel cell (Fan et al., 2017). Miran usually marked as a pre-treatment to enhance the biodegradability of et al. achieved azo dye degradation by cathodic bio-reduction in dual recalcitrant wastewater with the disadvantage of high energy demands chamber microbial fuel cell (Miran et al., 2015). The other is to degrade and capital costs (Hurwitz et al., 2014), whose viability requires to be the contaminant on the base of the reduction of energy barrier by virtue further improved. The bio-treatment technology is usually extremely of external electrostimulation. It has been confirmed to be feasible time-consuming under anaerobic condition (Dareioti and Kornaros, to applied in the removal of substituted aromatic compounds by

⁎ Corresponding author at: Institute of Seawater Desalination and Multipurpose Utilization, State Oceanic Administration (SOA), Nankai District, Tianjin 300192, China. E-mail address: [email protected] (X. Peng). http://dx.doi.org/10.1016/j.biortech.2017.10.068 Received 13 July 2017; Received in revised form 9 October 2017; Accepted 18 October 2017 Available online 19 October 2017 0960-8524/ © 2017 Elsevier Ltd. All rights reserved. X. Peng et al. Bioresource Technology 249 (2018) 844–850 biocathode electrolysis technique, such as p-fluoronitrobenzene (Shen ionization (EI) ion source of 70 eV and a HP-5MS capillary column et al., 2014), chloramphenicol (Kong et al., 2014a), p-nitrophenol (Tao (30 m × 250 μm × 0.25 μm). The GC injection and interface ports were et al., 2013), azo dye (Wang et al., 2016) and so on. It was studied that set at 250 °C and 280 °C, respectively, with an injection volume of 1 μL the bio-electrocatalysis conversion process could facilitate the p-CNB at a constant flow rate of 1 mL/min. The GC oven temperature program removal (Guo et al., 2015). However, its reduction rate was relatively was as follows: held at 35 °C for 2 min; increased up to 150 °C for 1 min − low with the value of 0.2291 h 1 only for 20 mg/L p-CNB. at 10 °C/min; and then increased up to 280 °C for 5 min at 20 °C/min. In this work, p-CNB with higher concentration up to 120 mg/L had According to the results of qualitative analysis by GC, the pollutant been rapidly degraded with higher reduction kinetics by the bio-elec- and its transformation products (p-chloroaniline (p-CAN) and aniline trocatalysis process using an adjusting architecture. The effect of initial (AN)) were quantified by the calibration curve of the pure compounds pollutant concentrations (40–120 mg/L) and applied external voltage relative to external standards at 275 nm for p-CNB, 240 nm for p-CAN (0–0.8 V) on p-CNB reduction rates was taken into account. At the and 231 nm for AN using high performance liquid chromatography meantime, it is known that the bio-electrocatalytic process is extra- (HPLC, Agilent 1200 series, Agilent, USA), respectively, which complicated. Based on the analysis of microbial communities structure, equipped with a diode array detector (DAD) and a Zorbax Eclipse XDB- the possible degradation pathway and bio-electrochemical analysis C18 reversed-phase column (4.6 mm × 150 mm × 5 μm). The sample were also explored. In addition, the water quality of p-CNB wastewater was mixed (v/v, 1:1) with methanol, and then filter by 0.22 μmmi- before and after treatment was subjected to chemical oxygen demand croporous membrane. The mobile phase was water-methanol (v/v, 4:6) (COD) determination. It is noteworthy that the optimization of opera- with the flow of 1.0 mL/min. The injection volume was 10 μL at the tion condition can lead to cost cutting and high-efficient removal of column temperature of 25 °C. Chemical oxygen demand (COD, TNT target contaminant, thus providing a reference for practical application plus COD Reagent, HACH Company, US) and MLVSS were made ac- of CNBs wastewater treatment. cording to APHA standard methods (APHA, 2005).

2. Materials and methods 2.2.2. Bioelectrochemical technique The voltage across resistor was monitored online every 30 min using 2.1. Experimental construction a date acquisition system (PISO-813, ICP DAS Co. Ltd, China). All the

electrode potentials were displayed according to Ag|AgCl|KClsat re- All the rectangular dual chamber reactors, made of the ference electrode (+197 mV vs. standard hydrogen electrode; SHE). polymethyl methacrylate with approximately 4 × 5 × 5 cm3 The bioelectrocatalytical test of cyclic voltammetry (CV) was swept in

(height × width × length) were run at 30 ± 0.5 °C temperature-con- situ from 0.6 to −1.0 V (vs. Ag|AgCl|KClsat) at a scan rate of 0.1 mV/s trolled biochemical incubator in fed-batch mode under 10 Ω with DC by a multi-channel potentiostat (CHI1000C, Shanghai CH Instrument power source of 0.5 V except as noted otherwise. Two carbon mesh Company, China) in a three-electrode mode for the configuration re- brush electrodes (3 cm in diameter and 3 cm long) were divided by actor, with the cathode as the working electrode, the anode as the 4×4cm2 (width × length) cation exchange membrane (CEM, Ultrex counter electrode, and an Ag/AgCl close to the cathode as the reference CMI7000, Membranes International Inc., USA) with vertical deploy- electrode. ment. Prior to use, the CEM was equilibrated in 1 M NaCl solution for 24 h. The electrode was subjected to a 24 h acetone soak followed by a 2.2.3. Molecular biotechnique 30 min heat treatment at 450 °C in a muffle furnace, and then adsorbed Bacterial biofilms, sampled by scraping carbon brush carrier using until p-CNB saturation. sterile forceps (Wu et al., 2017), were designed as S0, S1 and S2, which The anode chamber, in the bio-electrocatalytical reactor (BER), was were collected from source inoculum, anaerobic biocarrier and anae- inoculated by the activated sludge from a local sewage treatment plant robic biocathode, respectively. Genome DNA was extracted according in Tianjin City with the mixed liquor volatile suspended solids (MLVSS) to CTAB/SDS method, and confirmed its purity by 1% agarose gel of 3.7 g/L. The cathode one was bio-acclimated with improving p-CNB- electrophoresis with diluting to 1 ng/μL using sterile water. Bacterial resistant properties through p-CNB gradient cultivation. The inoculum 16S rRNA gene fragments were PCR-amplified using specific primers was mixed with the ratio of 1:5 (v/v) to nutrient medium: 50 mM (GTGCCAGCMGCCGCGGTAA,GGACTACHVGGGTWTCTAAT, 16S V4: phosphate buffer solution (PBS, g/L, Na2HPO4 4.576, NaH2PO4 2.132, 515F-806R) with Phusion®High-Fidelity PCR Master Mix (New England NH4Cl 0.31, KCl 0.13) with the pH of 7.0, trace minerals (12.5 mL/L) Biolabs). After detected by 2% agarose gel electrophoresis, mixture of and vitamins (5 mL/L) as described previously (Peng et al., 2012, PCR products with equidensity ratios was purified with Qiagen Gel 2013). The same nutrient medium was used in each chamber except for Extraction Kit (Qiagen, Germany). The library prepared during this the supporting substrate of 1 g/L glucose in the anode and 0.5 g/L study was constructed by TruSeq®DNA PCR-Free Sample Preparation glucose with initial p-CNB concentration of 40 mg/L in the cathode. All Kit (Illumina, USA) following manufacturer’s recommendations, and chemical reagents were used in analytical grade without further pur- quantified on the [email protected] Fluorometer (Thermo Scientific) and ification. The bulk solution was refreshed every three days to make sure Agilent Bioanalyzer 2100 system, and then sequenced on an Illumina the organics to degrade completely. In order to retain anaerobic en- HiSeq2500 PE250 platform. vironment, nitrogen exposure was kept for 15 min before injection into the reactor. Meanwhile, the anaerobic bioreactor (ABR) without ex- 2.3. Computation methods ternal voltage and electrocatalytical reactor (ECR) without bacteria attachment were run in parallel for control comparison under the same p-CNB concentration variation as a function of time was used to condition with BER. explain the kinetics of p-CNB elimination. According to the reference (Guo et al., 2015), it is known that p-CNB degradation describes 2.2. Analytical methods pseudo-first order decay. The constant of kinetic degradation (k) was

determined from the slope of ln(Ct/C0) vs. t (min) with the following 2.2.1. Physicochemical techniques equation: The intermediates of p-CNB degradation were identified qualita- lnlCktC=− + n tively by gas chromatography coupled with mass spectrometry (GC-MS, t 0

Agilent 7890A-5975C, Agilent, USA), which equipped with an electron where “t” was the reaction time, “lnCt” and “lnC0” were natural

845 X. Peng et al. Bioresource Technology 249 (2018) 844–850 logarithms of target pollutant concentration at “t” and 0 time, respec- Table 1 tively. Thus the half-time (τ)ofp-CNB was calculated by the following The fitting results of pseudo-first order kinetics on p-CNB disappearance. equation: − Reaction system Fitting equation k (h 1)R2 τ = ln2/k BER lnC0/Ct = 1.80 t 1.80 0.98 ffi Δ and the removal e ciency of p-CNB ( Rp-CNB) can be calculated as: ABR lnC0/Ct = 0.74 t 0.74 0.98 ECR lnC0/Ct = 0.40 t 0.40 0.98 ΔR()/pt-CNB=−CCC 0 0 correspondingly, the yield rate of p-CAN (ΔY ) was calculated as: p-CAN pollutant degradation demonstrated good linear fitting profile using the t 0 2 ΔY/p-CAN = CC pseudo-first order model with the corresponding R value (ca. 0.98) under different operating conditions (Fig. 1a). It was observed that the “ t” “ ” “ 0” where C was the tested concentration of p-CAN at t time, and C p-CNB removal rate could achieve at 98% above after different treated was the theoretic concentration of p-CAN which was derived from the periods, which were followed the order: bio-electrocatalytic reactors stoichiometric transformation of p-CNB. (BER, 3.0 h) < anaerobic reactors (ABR, 4.0 h) < electrocatalytical reactors (ECR, 5.5 h). However, the p-CNB degradation rates (k) among 3. Results and discussion those were very obviously different (Table 1). The k in BER was the −1 highest (kBER = 1.80 h ) followed by those in ABR and ECR 3.1. Degradation performance of p-CNB wastewater −1 −1 (kABR = 0.74 h and kECR = 0.40 h ), indicating that the particular pollutant-reducing consortium were established under electric power. 3.1.1. Degradation kinetics comparison with other studies What’s more, it was 6.83 times higher by the amending bio-electro- During the startup period, the anodic and cathodic potentials ar- catalysis architecture than that in previous report by Guo et al. (2015), − − ∣ ∣ rived at 338 mV and 837 mV vs. Ag AgCl KClsat, respectively, sug- possibly attributed to vertical electrode deployment with declined in- gesting that the bio-electrocatalysis reactors (BER) went into the steady ternal resistance (Kong et al., 2014b) and the difference of seeding state with similar voltage outputs for several consecutive cycles at the microbial communities. initial p-CNB concentration of 40 mg/L. The experimental data of target The test lasted from 0 to 52 h under the condition of 0.5 V with initial concentration of 40 mg/L for p-CNB degradation. The con- centration curves of pollutant and its corresponding productions in p- CAN and AN were demonstrated in Fig. 1b, where the concentration conversion was plotted against the elapsed time, respectively. A re- markable feature was the higher slope of p-CNB concentration decline

(dCp-CNB/dt), which indicated that the disappearance of pollutant was at a short delay time only by 3 h, thus, resulting in the subsequently quick rise in the p-CAN concentration at the same 3 h, possibly due to the rapid reduction of electron-withdrawing nitro substituent. As seen in Fig. 1b, the molar ratio of produced p-CAN to removed p-CNB was not 1.0 but 0.89, suggesting that not all of p-CNB was converted to p- CAN. However, subsequently, the slopes in the p-CAN concentration

versus time (dCp-CAN/dt) appeared to become lower and lower at the range of 3.0–7.0 h comparing with those at 7–31 h and 31–52 h until 3.88 mg/L for p-CAN, which was identified as a time-dependent p-CAN decay, indicating that the dechlorination of p-CAN was the rate-limiting step for p-CNB degradation. Interestingly, the concentration of pro- duced AN was slowly increased up to 6.48 mg/L during the experiment.

3.1.2. Effect of initial p-CNB concentration In order to study the influence of initial p-CNB concentration on the reaction rate, variable target contaminant concentrations (40, 60, 80, 100 and 120 mg/L) were adjusted in parallel comparison. The con- centration magnitude of target contaminant did not have a direct effect

on the p-CNB percentage removal (ΔRp-CNB), which almost achieved at about 98% within equilibrium time (Fig. S1 in Supporting information). With respect to the increase of initial p-CNB concentrations, the p-CAN

yields (ΔYp-CAN) declined from the highest level (89.1%, 40 mg/L p- CNB) to the lowest one (72.8%, 120 mg/L p-CNB), indicating that the higher the pollutant concentration, the more inhibited the conversion of p-CNB to less-toxic p-CAN. In addition, it was shown a shorter decay time at mere low p-CNB concentration (3.0 h, 40 mg/L), due to the immediate reduction of low-concentration target molecule on the bio- electrode, which was in accordance with the reduction rate constant (k). Table 2 described that the initial concentration had an obvious negative effect on the p-CNB degradation kinetics (k). It was a factor of −1 Fig. 1. (a) Rate constants in different reactors (BER: bio-electro-catalytical reactor, ABR: 1.12 from 1.80 to 0.85 h with p-CNB concentration ranging from 40 anaerobic bioreactor, and ECR: electrocatalytical reactor) and (b) the concentration of to 120 mg/L. And the corresponding half-times for p-CNB degradation potential transformation intermediates for 40 mg/L p-CNB degradation in BER. Error were from 0.385 to 0.815 h. The possible reason is that, the available bars ± SD were based on the averages measured in duplicate. electron donors are constant and biocatalytic activity is weaken by the

846 X. Peng et al. Bioresource Technology 249 (2018) 844–850

Table 2 p-CNB removal kinetics for different initial concentration in BER.

− Initial p-CNB concentrations Fitting equation k (h 1) Half-time R2 (mg/L) (h)

40 lnC0/Ct = 1.80 t 1.80 0.385 0.98

60 lnC0/Ct = 1.52 t 1.52 0.456 0.97

80 lnC0/Ct = 1.35 t 1.35 0.513 0.98

100 lnC0/Ct = 1.16 t 1.16 0.597 0.99

120 lnC0/Ct = 0.85 t 0.85 0.815 0.99 toxicity of target pollutant, hence, the pollutant reduction is limited with an increase in its initial concentration. Therefore, p-CNB con- centration affects the reduction rate of pollutant removal. A higher p- CNB concentration may depress the microbial electrocatalytic process.

3.1.3. Effect of applied voltage In this text, the reactors were set under variable different potential conditions (Ev = 0, 0.2, 0.5, 0.8 V), this moment, the normal con- Fig. 3. Effect of initial COD concentration on COD removal of p-CNB wastewater. centration of p-CNB in the cathode chamber was 80 mg/L rather than 40 mg/L in order for better experiment observation. For each external 3.1.4. Effect of chemical oxygen demand (COD) voltage establishment, the remaining p-CNB was all scarce with the The chemical oxygen demand profile is used as an effective reflec- approximate p-CAN yield above 70% excep for the condition of no tion to characterize the organic content of wastewater on account of the power load. It has been proved that the faster the contaminant trans- total quantity of oxygen required for the oxidation of the organic sub- port, the faster the contaminant degradation (Sun et al., 2012). strate into CO2 and water. Fig. 3 illustrated the COD of p-CNB waste- Therefore, it is a good approach for contaminant detoxicity to mitigate water before and after the treatment. It was observed that degree of it into the overlying water. Considering that the reduction kinetics is COD decrement was largely dependent on the extent of different initial the rate-controlling step, the applied voltage can be used as external pollutant concentration. At lower initial concentrations, the ratio of the thermodynamic reactive drive to accelerate the pollutant detoxication. initial strength of target molecule to the available electron donor and Fig. 2 explained the tendency of p-CNB reduction rate constants (k) with bioactivity was higher, and thus the COD removal decreased sig- fi the progressive application of external electric eld. It was noted that it nificantly from 65.7% to 45.7% with the increase of initial p-CNB −1 was increased by 1.6 folds from the control of 0.52 h (Ev = 0 V) to concentrations from 40 to 120 mg/L. The residual COD expressed that −1 1.35 h when the applied voltage of 0.5 V was supplied. Decrease or the contaminant was not thoroughly mineralized. This was in good increase of the external voltage to 0.2 V or 0.8 V could result in the consistence with the profile of p-CAN yields (ΔYp-CAN). In short, the −1 −1 decrease in k to 0.76 h and 1.23 h . The electrode potential, which reduction in COD values of treated p-CNB wastewater verifies the is dependent on the voltage, is the major determinant for reaction practicability of bio-electrocatalysis process with efficient degradation proceeding. With the applied voltage increasing, the cathode potential at high concentration. decreased steadily and the anode potential increased sharply (Fig. 2). The relative low cathode potential was conducive to p-CNB reduction kinetics accelerating, while the dramatically increasing anode potential 3.2. Microbial communities structure

(up to −82 mV vs. Ag|AgCl|KClsat) undesirably made the exoelectro- genic bacteria inactive, thus preventing the k from further increase. The In order to better understand the effect of microbial communities on experimental result implies that the 0.5 V of external voltage can be target contaminant degradation under electric field, Illumina ideal for quick removal of p-CNB with a relative low electric energy HiSeq2500 high-throughput sequencing analysis was used to analyze consumption. the bacterial samples among the anaerobic bio-electrode (S2), anae- robic biocarrier (S1) and source inoculum (S0). The species diversity index (Chao1/ACE) basing on operational taxonomic units (OTUs) is a kind of statistics tool to evaluate the diversity of microbial community (Xu et al., 2016). As shown in Table 3, the microbial community in S2 system was the most diversified, and the one in the system of S0 was the least diversified, which was in good agreement with the report by Xu et al. (2016), suggesting that the biological diversity would be im- proved through electric induction after the selective attachment of microorganism on the carbon brush carrier. It is reported that the high bacterial biodiversity is conducive to increase degradation performance of hydrocarbons (Dell'Anno et al., 2012). Fig. 4 elucidated the apparent differences of microbial communities

Table 3 Microbial communities diversity indices among the samples of S0, S1 and S2.

Sample name Chao1 ACE Chao1/ACE

S0 581.451 592.53 0.981302 S1 404.873 410.554 0.986163 Fig. 2. Effects of applied voltage on p-CNB reduction rate constant and the electrode S2 592.316 600.332 0.986647 potential.

847 X. Peng et al. Bioresource Technology 249 (2018) 844–850

Fig. 4. Microbial communities identified between source inoculum (S0), anaerobic biofilm (S1) and anaerobic biocathode (S2) at the phylum level.

structure between S0, S1 and S2 at the phylum level. The relative of biological degradation (Table 4). Propionimicrobium (Actinobacteria, abundance data presented that Proteobacteria and Bacteroidetes firstly unidentified Actinobacteria) was frequently detected with higher abun- increased from 14% and 9% (S0) to 44% and 20% (S1), and then de- dance (0.8%) in the S2 communities than that in S1 (0.2%), which was creased to 27% and 14% (S2). Similarly, Actinobacteria and Firmicutes reputed to reduce the nitro-group as stated by Williams (2015). Geo- decreased from 7.0% and 9.9% (S0) to 2.3% and 4.3% (S1) followed by bacter (Proteobacteria, Deltaproteobacteria), detected with the relative the reverse increase to 4.8% and 24.7% (S2). Whereas, Nitrospirae and abundance of 0.19% (S2), was more identified than that in S1 system Chloroflexi remarkably increased by 9 times (0.07%) and 26 times (0.05%), which not only releases power energy but also reduces the (1.15%) in S2 sample than those of S0 inoculation (0.007% and 0.04%), nitro-aromatics with the capability of accelerating electron transport respectively. In contrast, there was only an increase of 6 folds (0.05%) (Jiang et al., 2016). In addition, it was exciting that there was a and 23 folds (1.02%) in S1 sample, indicating that the electrical sti- markedly increased proportion for Desulfovibrio (Proteobacteria, Delta- mulation could boost the oriented enrichment of the specific functional proteobacteria) from 0.21% (S1) to 1.30% (S2), which is known to bacteria. multiply with the electrode as electron donor to produce hydrogen (Yun Further analysis based on the dominant species at the genus level et al., 2017) and accelerate extracellular electron transfer from elec- was carried out to reveal more information on the possible mechanism trode to contaminant (Croese et al., 2011). So did the Halanaerobium

Table 4 Phylogenetic classification of the 16S rRNA gene sequences in S1 and S2 based on the relative abundance > 0.1% at genus level.

Phylum Class Family Genus (%) S2 S1

Actinobacteria Unidentified Actinobacteria Propionibacteriaceae Propionimicrobium 0.8 0.2 Proteobacteria Deltaproteobacteria Geobacteraceae Geobacter 0.19 0.05 Proteobacteria Deltaproteobacteria Desulfovibrionaceae Desulfovibrio 1.3 0.21 Firmicutes Clostridia Halanaerobiaceae Halanaerobium 7.2 0 Proteobacteria Deltaproteobacteria Desulfobulbaceae Desulfobacterales 4.56 3.1

848 X. Peng et al. Bioresource Technology 249 (2018) 844–850

(Firmicutes, Clostridia)(Kivisto et al., 2013), the relative abundance of electron mediator between electrodes and microbes to accelerate the which greatly increased to 7.2% in S2, comparing with the nearly un- bio-electrochemical reduction of specific pollutant (Li et al., 2010). In detected S1. Herein, Desulfovibrio and Halanaerobium might play an contrast with the PBS blank control, the peak appearing at around important role on the reduction transformation of p-CNB based on the −0.54 to −0.76 V corresponded to the reduction reaction of p-CNB. generated hydrogen as the electron donors (Li et al., 2014). Other The reductive peak current remarkably increased with the increase significantly enriched microorganism such as Dehalococcoidia, Acineto- of target pollutant concentration at a regression equation of −2 −1 2 bacter, Aeromonas, Pseudomonas and Methanobacterium, was small in ip = 1.2 × 10 C – 0.6 × 10 (R = 0.945, C in mg/L, ip in mA), relative content but had an enormous effect with the advantages to which could be explained by the “concentration-polarization” phe- reduce chlorinated aromatic hydrocarbon and nitro-aromatics nomena. A higher initial concentration can provide a more significant (Fagervold et al., 2007; Yun et al., 2017). Here, what deserved to be driving force to overcome the mass transfer resistances from con- mentioned the most was Desulfobacterales (Proteobacteria, Deltaproteo- centration-polarization. However, the decrease in COD removal de- bacteria) accounting for 4.56% (S2), which was 47% and 5.5 times monstrates that the p-CNB reduction reaction is confined to the electron higher than those in S1 (3.1%) and S0 (0.7%), respectively. It was re- donors deficiency and bioactivity inhibition at high concentration. ported that Desulfobacterales could further metabolized AN into benzoic acid via reductive deamination based on the para-carboxylated trans- form to 4-aminobenzoate (Londry and Fedorak, 1992; Schnell and 3.3.2. Degradation pathway for p-CNB removal Schink, 1991). This was in accordance with the result of GC-MS. In Various transformation intermediates for p-CNB degradation were essence, it is highly presumed that a synergistic cooperation among qualitatively detected according to the GC-MS chromatograms, and it those bacteria is responsible for p-CNB degradation under the applied showed that the SPE sample from bio-electrocatalytical cultures con- electric voltage. tained the following potential intermediates. p-CAN (11.917 min) was analysed by mass spectra with a highly relative abundance, and a small 3.3. Synergistic degradation pathway analysis for p-CNB quantity of AN (8.192 min) was also confirmed. This was in good ac- cordance with the quantitative results by HPLC. Interestingly, the peak 3.3.1. Catalytic activity of bio-electrochemical reduction of p-CNB of phthalic acid was observed at 22.311 min for GC-MS monitoring (Fig. Fig. 5 showed the i-V profiles of bio-cathodes to describe the effect S2 in Supporting information), which was similar with the statement by of p-CNB concentration on its bio-electrotransformation behavior in Feng et al. (2017). Thus, based on the microbial communities function 50 mM phosphate buffer (pH 7.0). As the results shown, the peak at analysis above, it is inferred that p-CNB is quickly reduced into p-CAN about −0.8 to −1.0 V vs. Ag|AgCl|KClsat might be ascribed to the re- as the major intermediate, and then possibly performed by two se- duction reaction of hydrogen evolution, which could be performed as quential pathways. First, it is converted into AN through dechlorination reaction by functional Proteobacteria. And then is further transformed into phthalic acid derivative, which was speculated to have some connection with the carboxylation and deamination by Desulfobacterales (Duan et al., 2015; Londry and Fedorak, 1992; Schnell and Schink, 1991; Yang et al., 2017). However, further investigation should be done to provide sufficient evidence to identify the possible degradation pathway of p-CNB in detail.

4. Conclusion

The result reveals that high-concentration p-CNB can be quickly degraded using the bio-electrocatalysis process in an adjusting config- uration with vertical electrode deployment. The optimization studies demonstrate that its degradation is dependent on the initial pollutant concentration and operating applied voltage. The degradation rate follows the pseudo first-order kinetics, and has an obviously negative correlation with the initial concentration that the higher the con- centration, the lower the degradation rate. The optimal external voltage is about 0.5 V with the highest rate constant to satisfy the energy re- quirement. The decrease of COD demonstrated a great application prospect for high-concentration p-CNB removal.

Acknowledgements

This work was supported by National Natural Science Foundation of China (NSFC, No. 51409052), National Key R & D Program of China (No. 2017YFC0403502), the Scientific Research Project of the Marine Public Welfare Industry of China (No. 201505006) and the Central Public Interest Scientific Institution Basal Research Fund (Nos. K-JBYWF-2016-T17, K-JBYWF-2015-T17, K-JBYWF-2015-T16 and K- JBYWF-2015-G34). We also thank the Shanghai Tongji Gao Tingyao Environmental Science & Technology Development Foundation (STGEF). What’ more, special thanks go to Prof. Liang Zhu of Fig. 5. Cyclic voltammetry (a) and first derivative cyclic voltammetry (b) with different Department of Environmental Engineering, Zhejiang University for his p-CNB concentration (0, 40, 80 and 120 mg/L). kind helps.

849 X. Peng et al. Bioresource Technology 249 (2018) 844–850

Appendix A. Supplementary data Li, B., Zhu, J., 2014. Removal of p-chloronitrobenzene from groundwater: effectiveness and degradation mechanism of a heterogeneous nanoparticulate zero-valent iron (NZVI)-induced Fenton process. Chem. Eng. J. 255, 225–232. Supplementary data associated with this article can be found, in the Li, H., Zhang, Z., Xu, X., Liang, J., Xia, S., 2014. Bioreduction of para-chloronitrobenzene online version, at http://dx.doi.org/10.1016/j.biortech.2017.10.068. in a hydrogen-based hollow-fiber membrane biofilm reactor: effects of nitrate and sulfate. Biodegradation 25 (2), 205–215. Li, J., Liu, G., Zhang, R., Luo, Y., Zhang, C., Li, M., 2010. Electricity generation by two References types of microbial fuel cells using nitrobenzene as the anodic or cathodic reactants. Bioresour. Technol. 101 (11), 4013–4020. Li, X., Huang, Y., Li, C., Shen, J., Deng, Y., 2015. Degradation of p-CNB by Fenton like Alatraktchi, F.A.A., Zhang, Y., Angelidaki, I., 2014. Nanomodification of the electrodes in process using α-FeOOH. Chem. Eng. J. 260, 28–36. microbial fuel cell: impact of nanoparticle density on electricity production and Linch, A.L., 1974. Biological monitoring for industrial exposure to cyanogenic aromatic microbial community. Appl. Energy 116, 216–222. nitro and amino compounds. Am. Ind. Hygiene Assoc. J. 35 (7), 426–432. APHA, 2005. Standard Methods for the Examination of Water & Wastewater, 20th ed. Londry, K.L., Fedorak, P.M., 1992. Benzoic acid intermediates in the anaerobic biode- American Public Health Association. Washington D.C., USA. gradation of phenols. Can. J. Microbiol. 38 (1), 1–11. Croese, E., Pereira, M.A., Euverink, G.-J.W., Stams, A.J.M., Geelhoed, J.S., 2011. Analysis Miran, W., Nawaz, M., Kadam, A., Shin, S., Heo, J., Jang, J., Lee, D.S., 2015. Microbial of the microbial community of the biocathode of a hydrogen-producing microbial community structure in a dual chamber microbial fuel cell fed with brewery waste for electrolysis cell. Appl. Microbiol. Biotechnol. 92 (5), 1083–1093. azo dye degradation and electricity generation. Environ. Sci. Pollut. Res. Int. 22 (17), Dareioti, M.A., Kornaros, M., 2015. Anaerobic mesophilic co-digestion of ensiled sor- 13477–13485. ghum, cheese whey and liquid cow manure in a two-stage CSTR system: effect of Peng, X., Yu, H., Wang, X., Zhou, Q., Zhang, S., Geng, L., Sun, J., Cai, Z., 2012. Enhanced hydraulic retention time. Bioresour. Technol. 175, 553–562. performance and capacitance behavior of anode by rolling Fe O into activated Dell'Anno, A., Beolchini, F., Rocchetti, L., Luna, G.M., Danavaro, R., 2012. High bacterial 3 4 carbon in microbial fuel cells. Bioresour. Technol. 121, 450–453. biodiversity increases degradation performance of hydrocarbons during bior- Peng, X., Yu, H., Yu, H., Wang, X., 2013. Lack of anodic capacitance causes power emediation of contaminated harbor marine sediments. Environ. Pollut. 167, 85–92. overshoot in microbial fuel cells. Bioresour. Technol. 138, 353–358. Du, Q., Li, T., Li, N., Wang, X., 2017. Protection of electroactive biofilm from extreme acid Puyol, D., Batstone, D.J., Hulsen, T., Astals, S., Peces, M., Kromer, J.O., 2017. Resource shock by polydopamine encapsulation. Environ. Sci. Technol. Lett. 4 (8), 345–349. recovery from wastewater by biological technologies: opportunities, challenges, and Duan, F., Li, Y., Cao, H., Wang, Y., Crittenden, J.C., Zhang, Y., 2015. Activated carbon prospects. Front. Microbiol. 7. electrodes: electrochemical oxidation coupled with desalination for wastewater Schnell, S., Schink, B., 1991. Anaerobic aniline degradation via reductive deamination of treatment. Chemosphere 125, 205–211. 4-aminobenzoyl-CoA in Desulfobacterium anilini. Arch. Microbiol. 155 (2), 183–190. Fagervold, S.K., May, H.D., Sowers, K.R., 2007. Microbial reductive dechlorination of Shen, D., Zhang, X., Feng, H., Zhang, K., Wang, K., Long, Y., Wang, M., Wang, Y., 2014. aroclor 1260 in Baltimore Harbor sediment microcosms is catalyzed by three phy- Stimulative mineralization of p-fluoronitrobenzene in biocathode microbial electro- lotypes within the phylum Chloroflexi. Appl. Environ. Microbiol. 73 (9), 3009–3018. lysis cell with an oxygen-limited environment. Bioresour. Technol. 172, 104–111. Fan, M., Zhou, N., Li, P., Chen, L., Chen, Y., Shen, S., Zhu, S., 2017. Anaerobic co-me- Sun, M., Reible, D.D., Lowry, G.V., Gregory, K.B., 2012. Effect of applied voltage, initial tabolic biodegradation of tetrabromobisphenol A using a bioelectrochemical system. concentration, and natural organic matter on sequential reduction/oxidation of ni- J. Hazard. Mater. 321, 791–800. trobenzene by graphite electrodes. Environ. Sci. Technol. 46 (11), 6174–6181. Feng, H., Wang, Y., Zhang, X., Shen, D., Li, N., Chen, W., Huang, B., Liang, Y., Zhou, Y., Tao, H.-C., Wei, X.-Y., Zhang, L.-J., Lei, T., Xu, N., 2013. Degradation of p-nitrophenol in a 2017. Degradation of p-fluoronitrobenzene in biological and bioelectrochemical BES-Fenton system based on limonite. J. Hazard. Mater. 254, 236–241. systems: differences in kinetics, pathways, and microbial community evolutions. Wang, H.-C., Cheng, H.-Y., Wang, S.-S., Cui, D., Han, J.-L., Hu, Y.-P., Su, S.-G., Wang, A.- Chem. Eng. J. 314, 232–239. J., 2016. Efficient treatment of azo dye containing wastewater in a hybrid acidogenic Guo, W.-Q., Guo, S., Yin, R.-L., Yuan, Y., Ren, N.-Q., Wang, A.-J., Qu, D.-X., 2015. bioreactor stimulated by biocatalyzed electrolysis. J. Environ. Sci. 39, 198–207. Reduction of 4-chloronitrobenzene in a bioelectrochemical reactor with biocathode Williams, G.D., 2015. Two cases of urinary tract infection caused by Propionimicrobium at ambient temperature for a long-term operation. J. Taiwan Inst. Chem. Eng. 46, lymphophilum. J. Clin. Microbiol. 53 (9), 3077–3080. 119–124. Wu, J., Chen, W., Yan, Y., Gao, K., Liao, C., Li, Q., Wang, X., 2017. Enhanced oxygen Guo, Z., Zheng, S., 2009. Separation of p-chloronitrobenzene and o-chloronitrobenzene by reducing biocathode electroactivity by using sediment extract as inoculum. selective adsorption using Silicalite-1 zeolite. Chem. Eng. J. 155 (3), 654–659. Bioelectrochemistry 117, 9–14 (Amsterdam, Netherlands). Hurwitz, G., Pornwongthong, P., Mahendra, S., Hoek, E.M.V., 2014. Degradation of Xia, S., Li, H., Zhang, Z., Zhang, Y., Yang, X., Jia, R., Xie, K., Xu, X., 2011. Bioreduction of phenol by synergistic chlorine-enhanced photo-assisted electrochemical oxidation. para-chloronitrobenzene in drinking water using a continuous stirred hydrogen-based Chem. Eng. J. 240, 235–243. hollow fiber membrane biofilm reactor. J. Hazard. Mater. 192 (2), 593–598. Jiang, X., Shen, J., Lou, S., Mu, Y., Wang, N., Han, W., Sun, X., Li, J., Wang, L., 2016. Xu, X., Shao, J., Li, M., Gao, K., Jin, J., Zhu, L., 2016. Reductive transformation of p- Comprehensive comparison of bacterial communities in a membrane-free bioelec- chloronitrobenzene in the upflow anaerobic sludge blanket reactor coupled with trochemical system for removing different mononitrophenols from wastewater. microbial electrolysis cell: performance and microbial community. Bioresour. Bioresour. Technol. 216, 645–652. Technol. 218, 1037–1045. Jones, C.R., Liu, Y.-Y., Sepai, O., Yan, H., Sabbioni, G., 2006. Internal exposure, health Yang, B.-R., Sun, Z.-Q., Wang, L.-P., Li, Z.-X., Ding, C., 2017. Kinetic analysis and de- effects, and cancer risk of humans exposed to chloronitrobenzene. Environ. Sci. gradation pathway for m-dichlorobenzene removal by Brevibacillus agri DH-1 and its Technol. 40 (1), 387–394. performance in a biotrickling filter. Bioresour. Technol. 231, 19–25. Kivisto, A., Larjo, A., Ciranna, A., Santala, V., Roos, C., Karp, M., 2013. Genome sequence Ye, M., Chen, Z., Wang, W., Shen, J., Ma, J., 2010. Hydrothermal synthesis of TiO hollow of Halanaerobium saccharolyticum subsp. saccharolyticum strain DSM 6643T, a halo- 2 microspheres for the photocatalytic degradation of 4-chloronitrobenzene. J. Hazard. philic hydrogen-producing bacterium. Genome Announc. 1 (2) e00187-13. Mater. 184 (1–3), 612–619. Kong, D., Liang, B., Lee, D.-J., Wang, A., Ren, N., 2014a. Effect of temperature switchover Yuan, L., Shen, J., Chen, Z., Liu, Y., 2012. Pumice-catalyzed ozonation degradation of p- on the degradation of antibiotic chloramphenicol by biocathode bioelectrochemical chloronitrobenzene in aqueous solution. Appl. Catal. B 117, 414–419. system. J. Environ. Sci. 26 (8), 1689–1697. Yun, H., Liang, B., Kong, D.-Y., Cheng, H.-Y., Li, Z.-L., Gu, Y.-B., Yin, H.-Q., Wang, A.-J., Kong, F., Wang, A., Cheng, H., Liang, B., 2014b. Accelerated decolorization of azo dye 2017. Polarity inversion of bioanode for biocathodic reduction of aromatic pollu- Congo red in a combined bioanode-biocathode bioelectrochemical system with tants. J. Hazard. Mater. 331, 280–288. modified electrodes deployment. Bioresour. Technol. 151, 332–339.

850