Bioresource Technology 200 (2016) 565571

Contents lists available at ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/locate /bior tech

Cooperative cathode electrode and in situ deposited copperfor subsequent enhanced Cd(II) removal and hydrogen evolutionin bioelectrochemical systems

http://dx.doi.org/10.1016/j.biortech.2015.10.0840960-8524/ 2015 Elsevier Ltd. All rights reserved.

Corresponding author. Tel./fax: +86 411 84708546.E-mail address: lipinghuang@dlut.edu.cn (L. Huang).

Qiang Wang a, Liping Huang a,, Yuzhen Pan b, Peng Zhou b, Xie Quan a, Bruce E. Logan c, Hongbo Chen baKey Laboratory of Industrial Ecology and Environmental Engineering, Ministry of Education (MOE), School of Environmental Science and Technology, Dalian University ofTechnology, Dalian 116024, ChinabCollege of Chemistry, Dalian University of Technology, Dalian 116024, ChinacDepartment of Civil and Environmental Engineering, The Pennsylvania State University, University Park, PA 16802, United States

h i g h l i g h t s

Sequential Cu(II) and Cd(II) reductionin BESs using a time electromagneticrelay.

Copper deposited in MFCs enhancedCd(II) reduction and hydrogenevolution in MECs.

Cd(II) reduction and hydrogenevolution in MECs were dependent oncathode material.

Ions of chloride and sulfate, andapplied voltage affected systemperformance.

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o

Article history:Received 18 September 2015Received in revised form 23 October 2015Accepted 24 October 2015Available online 10 November 2015

Keywords:Bioelectrochemical systemMicrobial fuel cellMicrobial electrolysis cellCu(II) and Cd(II) reductionCathode material

a b s t r a c t

Bioelectrochemical systems (BESs) were first operated in microbial fuel cell mode for recovering Cu(II),and then shifted to microbial electrolysis cells for Cd(II) reduction on the same cathodes of titanium sheet(TS), nickel foam (NF) or carbon cloth (CC). Cu(II) reduction was similar to all materials (4.794.88 mg/L h) whereas CC exhibited the best Cd(II) reduction (5.86 0.25 mg/L h) and hydrogen evolution(0.35 0.07 m3/m3 d), followed by TS (5.27 0.43 mg/L h and 0.15 0.02 m3/m3 d) and NF(4.96 0.48 mg/L h and 0.80 0.07 m3/m3 d). These values were higher than no copper controls by fac-tors of 2.0 and 5.0 (TS), 4.2 and 2.0 (NF), and 1.8 and 7.0 (CC). These results demonstrated cooperativecathode electrode and in situ deposited copper for subsequent enhanced Cd(II) reduction and hydrogenproduction in BESs, providing an alternative approach for efficiently remediating Cu(II) and Cd(II) co-contamination with simultaneous hydrogen production.

2015 Elsevier Ltd. All rights reserved.

1. Introduction

Heavy metals of Cu(II) and Cd(II) in electroplating wastewaters,if not removed, can severely contaminate surface waters, ground-

waters, and soils, and thus threaten human health (Martin-Laraet al., 2014). Conventional techniques including physical, chemi-cal/electrochemical, physicochemical, and biological methods canbe used to effectively treat these wastewaters (Martin-Lara et al.,2014). However, other approaches that avoid the high costs andenergy consumption are being sought to provide environmentalfriendly and cost-effective strategies for wastewater treatment.

http://crossmark.crossref.org/dialog/?doi=10.1016/j.biortech.2015.10.084&domain=pdfhttp://dx.doi.org/10.1016/j.biortech.2015.10.084mailto:lipinghuang@dlut.edu.cnhttp://dx.doi.org/10.1016/j.biortech.2015.10.084http://www.sciencedirect.com/science/journal/09608524http://www.elsevier.com/locate/biortech

566 Q. Wang et al. / Bioresource Technology 200 (2016) 565571

Bioelectrochemical systems (BESs) including microbial fuel cells(MFCs) and microbial electrolysis cells (MECs) have been shown tobe promising technologies for recovering metals from wastewatersin addition to diverse wastes and wastewaters treatment (Koket al., 2015; Li et al., 2014; Nancharaiah et al., 2015; Rivera et al.,2015; Wang and Ren, 2014). Cu(II) can be spontaneously removedin MFCs due to its favorable half-cell redox potential of 0.34 V (vs.standard hydrogen electrode, SHE) relative to that of organic mat-ter (ca. 0.30 V for acetate), whereas Cd(II) reduction requires amore negative potential of 0.52 V that can only be obtained inan MEC (An et al., 2014; Cheng et al., 2013; Heijne et al., 2010;Tao et al., 2011; Zhang et al., 2015). Mixed metals of either Cu(II), Pb(II), Zn(II) and Cd(II) or Zn(II), Pb(II) and Cu(II) have all beenindividually recovered in MECs through the change of applied volt-age (Modin et al., 2012; Tao et al., 2014) whereas Cu(II) and Ni(II)were simultaneously deposited on the same cathodes of MECs (Luoet al., 2014). MFCs operating with cathodic Cu(II) reduction wereshown to be able to provide the voltage needed for Cd(II) reductionin MECs without any external energy consumption (Zhang et al.,2015), somewhat similar to Cr(VI)-reduced MFCs coupled withCd(II)-reduced MFCs (Choi et al., 2014). However, only single Cu(II) or Cd(II) was individually fed in the cathodes of these MFCsor MECs, limiting the practical application of these systems. There-fore more approaches are needed to make BESs useful for Cu(II)and Cd(II) removal in the mixed solutions.

Considering the reduction of Cu(II) on the cathodes of singleMFCs (An et al., 2014; Cheng et al., 2013; Heijne et al., 2010; Taoet al., 2011, 2014; Zhang et al., 2015) and Cd(II) on the cathodesof sole MECs (Choi et al., 2014; Zhang et al., 2015), we hypothe-sized that Cu(II) and Cd(II) could be sequentially reduced on thesame cathodes of BESs through the use of a timed relay to switchoperational conditions (Fig. S1 in Supplementary Information(SI)). In this process, the BESs would first be operated in MFC modefor Cu(II) reduction, and then shifted to MEC mode for Cd(II) reduc-tion. Considering the catalysis of copper for CO2 and nitrate reduc-tion (Gattrell et al., 2006; Haque and Tariq, 2010), it is expected theprior deposition of copper might also be helpful in achievingimproved Cd(II) reduction in the MECs, which to our knowledgehas not been systematically investigated.

Different cathode materials have been used for metals reduc-tion in BESs, with few comparisons made of different materialsunder the same operational conditions. Carbon-based cathodes,such as graphite plates, graphite foil, graphite rods, graphite disks,and carbon cloth have been used as cathodes in MFCs for Cu(II)reduction (Heijne et al., 2010; Luo et al., 2014; Tao et al., 2011).For Cd(II) reduction in MECs, titanium wire, titanium sheets, car-bon cloth, and carbon rods have been used (Choi et al., 2014;Modin et al., 2012; Zhang et al., 2015). In MECs, hydrogen gas pro-duction competes with metals reduction, and thus materials aredesired that have a good propensity for metals reduction but notH2 gas evolution (Kundu et al., 2013; Wang et al., 2015). Theimpact of the deposition of one metal on the subsequent deposi-tion of another metal, especially in the context of its impact onhydrogen evolution, has not been previously considered.

In this study, three different materials including titanium sheet(TS), nickel foam (NF), and carbon cloth (CC) were examined ascathodes in BESs first operated as MFCs, and then switched toMECs. In the MFCs, Cu(II) was first reduced and deposited on thecathode. The reactors were then operated as MECs for Cd(II) reduc-tion along with H2 gas evolution. Key factors influencing the reac-tion rates include the concentration of Cl and SO42 anions insolution, the impact of the deposited copper on Cd(II) reduction,and the applied voltage. The performance of the reactors was eval-uated in terms of rates of Cu(II) and Cd(II) reduction, hydrogen pro-duction, coulombic efficiency, energy efficiency, overall energyrecovery and cathode potential.

2. Methods

2.1. Reactor construction

Identical two-chamber BESs with cylindrical chambers 4.0 cmlong by 3.0 cm in diameter were used in all experiments. Theanode and cathode chambers were separated by a cation exchangemembrane (CEM) (CMI-7000 Membrane International, Glen Rock,NJ), with working liquid volumes of 15 mL in the anode, and25 mL in the cathode, due to the volume of the electrode materials.The anode was filled with graphite felt (3.0 cm 2.0 cm 1.0 cm,Sanye Co., Beijing, China) using a carbon rod as the current collec-tor. Cathodes of TS, NF or CC (2.0 cm 2.0 cm) were placed next tothe CEM, and connected to the circuit using titanium wire. Chem-ically inert titanium wire, rather than less-expensive stainless steelor copper wire, was used to avoid any corrosion during these tests(Logan, 2012; Zhu and Logan, 2014). These electrode materialswere cleaned prior to installment in the reactors (Wang et al.,2015). A reference electrode (Ag/AgCl, 195 mV vs. SHE) wasinstalled in the cathode chamber to measure cathode potential,with all potentials reported here vs SHE. All of the reactors werewrapped with aluminum foil to exclude light to avoid the growthof algae on the anodes and side reactions on the cathodes in thepresence of Cu(II) (Sun et al., 2015).

2.2. Start up and operation

The bioanodes were inoculated with suspended bacteria fromacetate-fed MFCs reactor and an equivalent volume of nutrientsolution, and acclimated in MFCs. The nutrient anode solution(pH 7.0) consisted of (g/L): acetate, 1.0; KH2PO4, 4.4; K2HPO4,3.4; NH4Cl, 1.3; KCl, 0.78; MgCl2, 0.2; CaCl2, 0.0146; NaCl, 0.5;and 1.0 mL of trace elements (Cheng et al., 2009). During the initialbioanode acclimation period, the catholyte was a 50 mM phos-phate buffer solution (KH2PO4K2HPO4). The anolyte and catholytewere periodically replaced (once every 35 days). The anode wasconsidered to be fully acclimated when the anode potential wasstable at 0.24 V (vs. SHE) for at least three batch cycles.

The catholyte used in metal reduction tests contained 150 mg/LCu(II) and 50 mg/L Cd(II), based on concentrations reported forthese metals in actual electroplating wastewaters (Prakash et al.,2012). After an operation period of 30 h as an MFC, which was suf-ficient for nearly all the Cu(II) to be deposited on the cathodes, thesystems were switched to MEC mode with an applied voltage of0.5 V (DC Power Supply PS-1502DD, Yihua, Guangzhou, China)using a timed electromagnetic relay switch (ST3PA-F, Fangfushi,Japan). A small resistor of 10Xwas connected in series with powersupply to allow calculation of the current. All potentials were col-lected using a data acquisition (PISO-813, Hongge Co., Taiwan).

In order to investigate the effect of amount of deposited copperon Cd(II) reduction, catholytes composed of 100500 mg/L Cu(II)and 50 mg/L Cd(II) were used to prepare various amounts of depos-ited copper in MFC mode for subsequent Cd(II) reduction in MECmode (applied voltage: 0.5 V). Considering the occurrence ofhydrogen evolution and the avoidance of metal precipitants abovecertain pHs, a buffer was used to sustain an appropriate pH for Cd(II) reduction. It has been shown that Cd(II) in an acetate buffer canform the complex Cd(CH3COO)2, which has a stability constant of2.19. This is lower than that of 2.937.0 for Cd(NH3)42+ and CdCl42

in NH4ClHCl buffer, and 9.7 for Cd[B(OH)4]2 in borate buffer(Bousher, 1995; Khallaf et al., 2008). Ions of Cd(II) were thereforeless tightly held by the Cd(CH3COO)2 complex than they wouldhave been in borate or NH4ClHCl buffers (Chen et al., 2014; Lvet al., 2013). CH3COONaCH3COOH was therefore used as a catho-lyte buffer as it was expected to be most beneficial for efficient Cd

Q. Wang et al. / Bioresource Technology 200 (2016) 565571 567

(II) reduction. A pH of 4.6 was not only optimal for the buffering ofCH3COONaCH3COOH (Chen et al., 2014; Lv et al., 2013), it wasalso unfavorable for hydrogen evolution and metal precipitant for-mation, both of which would benefit Cd(II) reduction. The value of4.6 was thus chosen as an initial catholyte pH. The catholyte had asolution conductivity of 5.8 mS/cm. Reactors were operated underopen circuit conditions (OCC) as controls for Cd(II) adsorption onthe electrodes in the absence of current generation.

The impact of applied voltage on Cd(II) reduction on the cath-odes with or without deposited copper, was examined usingapplied voltages of 0.3, 0.5, 0.7, 0.9, 1.1 V. To examine the effectof anions on Cd(II) reduction, Cd(II) was added using two differentsalts (CdCl2 and CdSO4) at identical Cd(II) concentrations of 50 mg/L, and a fixed pH of 4.6. To examine the impact of copper oxide for-mation, cathodes containing copper from MFC operation wereexposed in the air for 48 h, and then examined using cyclic voltam-metry (CV). All catholytes in both MFC and MEC modes were bub-bled with ultrapure N2 at least 15 min before each batch cycleoperation. All experiments were conducted at a room temperatureof 25 3 C.

2.3. Analytical measurements

CV (CHI 650, Chenhua, Shanghai) was carried out in a three-electrode configuration of a working electrode (bare, copper-deposited or oxidized copper-deposited electrodes), platinum plateas a counter electrode, and a Ag/AgCl reference electrode. Perfor-mance of cathodes was evaluated by linear sweep voltammetry(LSV) (CHI 650, Chenhua, Shanghai). Both CV and LSV were con-ducted at a scan rate of 1.0 mV/s.

Concentrations of Cu(II) and Cd(II) in the catholyte were deter-mined by atomic absorption spectroscopy (AAnalyst 700, PerkinEl-mer). Hydrogen in the headspace of cathode in the MECs wasanalyzed using a gas chromatograph (GC7900, Tianmei, Shanghai)and a molecular sieve column (TDX-01, 6080, 4 mm 2 m) withnitrogen as the carrier gas.

A scanning electronic microscopy (SEM) (Nova NanoSEM450,FEI company, USA) equipped with an energy dispersive X ray spec-trometer (EDS) (X-MAX 20 mm2/50 mm2, Oxford Instruments, UK)was used to investigate the surface morphology of deposited cop-per and products as well as elemental composition. X-ray diffrac-tion measurements (XRD-6000, Shimadzu LabX, Japan) were usedto examine the morphology and crystal form of the products.

2.4. Calculations

Removal rates of Cu(II) and Cd(II) (mg/L h) were defined as thedecrease of Cu(II) or Cd(II) concentration within an operation per-iod (t). Net hydrogen production (m3 H2/m3 reactor d) based onrecovery in the MEC headspace was normalized by the volume ofthe cathode chamber and operation time. A glass tube with aninner diameter of 8 mm was glued to the top of the reactor to cre-ate a total headspace of 12 mL. A microsyringe (200 lL, Agilent)was used to sample the headspace. Hydrogen production (m3)was then calculated based on the hydrogen concentration (m3/m3) obtained by the gas chromatograph multiplied by the gaseousphase volume of 0.000012 m3. Energy efficiencies were calculatedas the ratios of either energy consumption for Cd(II) reduction (gE,Cd, %) or energy produced by heat of combustion of hydrogen toelectrical energy added (gE;H2 , %) (Eqs. (1) and (2)). Overall energyrecoveries of cadmium reduction (gE+S,Cd, %) and hydrogen evolu-tion (gE+S,H, %) were calculated based on both electrical energyadded and anodic substrate consumption (Call and Logan, 2008)(Eqs. (3) and (4)). Power and current densities were normalizedto the geometric surface of the cathode. One-way ANOVA in SPSS

19.0 was used to analyze the differences among the data, and allof the data indicated significance levels of p < 0.05.

gE;Cd Fb1DCCd2VcaECd2

103MCdEapPn

i1IiDti 100% 1

gE;H2 nH2DHs;H2

EapPn

i1IiDti 100% 2

gES;Cd Fb1DCCd2VcaECd2

103MCdEapPn

i1IiDti nsGs 100% 3

gES;H2 nH2DHs;H2

EapPn

i1IiDti nsGs 100% 4

where DCCd2 is the change of Cd(II) concentration in 4 h; b1(2 mol/mol) the number of electrons required for Cd(II) reduction;Vca the working volume of the cathode chamber (L); ECd2 the theo-retical electrode potential of Cd(II) reduction under the test condi-tion (V); Eap the applied voltage in MECs mode (V); t theoperation time of each batch cycle (s); nH the hydrogen produced(mol); ns the substrate (acetate) consumed (mol); Gs the Gibbs freeenergy of the corresponding reaction (J/mol); DHS;H2 the heat ofcombustion of hydrogen (J/mol); I the circuit current (A); MCd(112 g/mol) the molecular weight of Cd; F is Faradays constant(96485 C/mol e); and 103 is for a unit conversion (mg/g).

3. Results and discussion

3.1. Copper deposition in MFCs

Removal rates of Cu(II) in MFCs using catholytes containingboth Cu(II) and Cd (II) were similar for all three cathode materials,ranging from 4.79 to 4.88 mg/L h, with a complete Cu(II) removalfrom catholytes (Fig. 1A). No Cu(II) or Cd(II) was measured in theanolyte, although the existence of these ions in the cationexchange membrane cannot be excluded (Zhang et al., 2015). Verylittle Cu(II) was removed in the OCC controls (0.11 0.040.46 0.08 mg/L h; 2.2 0.99.1 1.5%). These Cu(II) removal rateswere much higher than previous reports using graphite plate cath-odes at similar initial Cu(II) concentrations (Tao et al., 2011), andslightly higher than rates obtained with graphite plate cathodes(Table S1) (An et al., 2014). MFC performance was similar for allmaterials, with current densities of 0.1110.122 A/m2 (Fig. 1A)and maximum power densities of 0.1250.128W/m2 (Fig. 1B).

3.2. Cd(II) reduction and hydrogen evolution in MECs

Following the MFC cycle, the reactors were switched to MECmode. Cd(II) removal was the highest with TS of 5.27 0.43 mg/L h (41.9 2.8%), followed by 4.96 0.48 mg/L h (39.4 3.2%) (NF)and 5.86 0.25 mg/L h (46.6 1.3%) (CC). These rates for Cd(II)removal are 1.84.2 times as large as the removal rates obtainedin MEC tests using cathodes without the previously deposited Cu(II) (Fig. 2A, Table S1). These results show that the in situ depositionof copper onto the electrode improved Cd(II) deposition to agreater extent than differences between the electrode materials.These Cd(II) removal rates with in situ deposited copper were 7.5times higher than those previously reported using similar cathodematerials and same initial Cd(II) concentration (Choi et al., 2014).These rates are even slightly higher than those obtained using amuch higher Cd(II) of 800 mg/L (Table S1) (Modin et al., 2012),and deposition rates are expected to increase in proportion to Cd(II) initial concentration. Cd(II) deposition rates were all muchhigher than rates obtained in both OCC controls and CCCs in the

Fig. 1. (A) Cu(II) removal rates and (B) power production in MFC mode (catholytecomposed of 150 mg/L Cu(II) and 50 mg/L Cd(II)).

Fig. 2. (A) Removals of Cd(II) in MEC mode and controls of OCC in the presence ofdeposited copper and CCC in the absence of deposited copper, and (B) hydrogenproduction and current densities in MEC mode with or without the presence ofdeposited copper.

568 Q. Wang et al. / Bioresource Technology 200 (2016) 565571

absence of deposited copper, implying the critical roles of both cir-cuit current and deposited copper in Cd(II) removal. More Cd(II)adsorption in the presence of deposited copper (1.772.30 mg/L h) than the controls with no deposited copper (0.181.20 mg/L h), indicated that Cu(II) deposition also enhanced Cd(II)adsorption.

Hydrogen production was competitive with metals reduction,although there was no consistent trend in these rates relative tometal reduction rates (Fig. 2B). The H2 production rate was thehighest for the NF material (0.80 0.07 m3/m3 d) that had the low-est rate of Cd(II) reduction. However, the next lowest H2 produc-tion rate (0.35 0.07 m3/m3 d) was obtained using the CCmaterial, which had the highest Cd(II) reduction rate, with the TSmaterial having the lowest H2 production rate (0.15 0.02 m3/m3 d). These H2 production rates were 2.0 (NF) to 7.0 (CC) timesthose obtained in tests without previously deposited copper.

The ratio of charge consumed for Cd(II) reduction to hydrogenevolution (shown as Cd/H2 in Table S2) was further used to clarifythe relationship between Cd(II) reduction and hydrogen evolutionon the various cathode materials in the absence or presence of cop-per. Despite the consistent increase in net charge for both Cd(II)reduction and hydrogen evolution on all the material cathodes inthe presence of copper, more charge was diverted for Cd(II) reduc-tion than hydrogen evolution on the NF cathodes, as shown by ahigher Cd/H2 ratio of 0.02 0.000.03 0.00, compared to0.01 0.00 in the absence of copper (Table S2 and Fig. S2). Con-versely, lower Cd/H2 ratios of 0.18 0.020.33 0.03 (TS) and0.08 0.000.16 0.01 (CC) were obtained in the presence ofdeposited copper, compared to 0.56 0.00 (TS) and 0.35 0.03(CC) in the absence of copper. These results clearly demonstratedthe competitive relationship between Cd(II) reduction and

hydrogen evolution on all cathode materials, and that NF was bet-ter than TS or CC for Cd(II) reduction compared to hydrogen evolu-tion due to the presence of deposited copper.

While the intrinsic NF material was superior for higherexchange current densities and lower overpotentials for hydrogenevolution than TS and CC in the absence of copper (Kundu et al.,2013), a better resulting morphology of the copper deposited onthe NF cathodes may have provided a more effective catalytic sur-face area, and thus better system performance in the present sys-tem as shown in the later Fig. S4B. The morphology of themetallic deposits has previously been shown to play a critical rolein the effective catalytic surface area, and thus system performanceby others (Habibi et al., 2008; Liu et al., 2014).

Increasing the amount of copper in solution in the MFC tests,and thus the amount of copper deposited on the cathode, did notnecessarily improve Cd(II) deposition rates or H2 production.Higher initial Cu(II) concentrations of 300 mg/L or 500 mg/L dimin-ished subsequent Cd(II) reduction rates on all cathodes (Fig. 3A),presumably due to the aggregation of large metallic Cu(0) particlesthat would decrease the active surface area (Habibi et al., 2008).Higher Cd(II) reduction rates were consistently obtained usingthe CC cathodes in both the absence and the presence of copper(Fig. 3A), and deposition rates were substantially lower in theOCC controls (Fig. 3B). Hydrogen production (Fig. 3C) and currentdensities (Fig. 3D) were similarly decreased using the two highestinitial copper concentrations. These results show improvements inCd(II) reduction rate and hydrogen production in MEC mode thatchanged with the amount of copper deposited in MFC mode. Con-sidering the fluctuation of these ratios in actual wastewaters, morecontrolled ratios of deposited copper and reduced Cd(II) would be

Fig. 3. Cd(II) removal rates in (A) MECs and (B) OCC controls, and (C) hydrogen production and (D) current densities for MEC tests with initial copper concentrations rangingfrom 0 to 500 mg/L in the catholyte (Eap = 0.5 V, initial Cd(II) concentration = 50 mg/L).

Q. Wang et al. / Bioresource Technology 200 (2016) 565571 569

needed for the subsequent efficient Cd(II) reduction in MEC mode.Further investigations in this direction are warranted.

3.3. Energy recovery in MECs

Energy produced in the MFC was 0.8 kWh/kg Cu for all threeelectrode materials (Table S1). The deposition of copper on thecathode in the MFC improved both Cd(II) reduction and hydrogenevolution in the MEC (Table S2). This finding is in agreement withother reports which have shown that copper reduction facilitateshydrogen evolution or oxygen reduction (Heijne et al., 2010;Modin et al., 2012; Luo et al., 2014). The reduced overpotentialsin the presence of copper resulted in good energy recoveries, withgE,H2 > 100% in most cases (Table S2). The energy required for Cd(II)reduction in MEC mode ranged from 4.1 kWh/kg Cd (CC) to8.5 kWh/kg Cd (NF) (Table S1). The energy production in theMFC, combined with energy in the hydrogen gas that was pro-duced (0.514.8 kWh/kg Cd), resulted in a net energy input of0.3 kWh for the TS cathode, whereas the use of the other materialcathodes produced net energies of 7.7 kWh (NF) and 0.5 kWh (CC)based on per kilogram copper and cadmium. On the basis of thisenergy demand, the TS material cathode was the least suitablefor use in these reactors and the NF cathodes provided a good com-promise in terms of performance and energy recovery.

3.4. Electrode morphologies and product characterization

Compared to the bare cathodes of TS (Fig. S3A), NF (Fig. S3E) andCC (Fig. S3I), a layer of typical reddish-brown color was observedon the electrodes with deposited copper (Fig. S3B, F and J), imply-ing the successful occurrence of Cu(II) reduction on these variouselectrodes (An et al., 2014; Tao et al., 2011; Wu et al., 2015b). Withsubsequent Cd(II) reduction, the color was reddish-brown, result-ing in the blend as shown in Fig. S3C, G and K. Compared to thereddish-brown color of freshly deposited copper, dark black was

observed on the various material cathodes after exposed in theair for 48 h (Fig. S3D, H and L), implying the oxidation of pure cop-per and reflecting the transformation of pure copper to cuprousand cupric oxides.

The copper-deposited cathodes were observed by SEM(Fig. S4AC), EDS (Fig. S4DF) and XRD (Fig. S4GI) after Cd(II)reduction with 3 batch cycles operation. The three different mate-rials showed distinctive morphologies, with comparatively largercluster crystals observed on the NF cathodes (Fig. S4B) comparedto the TS (Fig. S4A) and CC (Fig. S4C) cathodes. This result mayexplain the better performance of NF than the others (Fig. 2). UsingEDS to examine the composition of the precipitates, Cu and Cd sig-nals were detected on all the precipitates of TS (Fig. S4D), NF(Fig. S4E) and CC (Fig. S4F) cathodes. Similarly, peaks of coppermatched those of Cu0 at 43.3, 50.4, and 74.1 in 2h (Fig. S4GI),consistent with other reports on electrodeposited copper (Luoet al., 2014; Wu et al., 2015a,b; Zhang et al., 2015). This resultdemonstrates the independency of copper crystal form on thesecathode materials. The absence of Cd signals in XRD (Fig. S4GI),but appearance in the EDS analysis (Fig. S4DF), implies an amor-phous state of the reduced cadmium on these materials. The inten-sity of copper peak declined after Cd(II) reduction (Fig. S4GI),which is expected due to the deposition of the cadmium.

3.5. Cyclic voltammetry analysis

Cd(II) reduction peaks were observed at potentials of 0.592 V(TS) (Fig. 4A) and 0.576 V (CC) (Fig. 4C) with deposited copper,which were more positive than those of the controls in the absenceof copper (0.621 V, TS and 0.613 V, CC) (Fig. 4A and C), demon-strating the similar catalysis role of deposited copper in Cd(II)reduction on these electrodes. The different Cd(II) reductive peakpotentials for TS and CC without the presence of deposited copperwas mainly ascribed to the intrinsic characters of these materials.No apparent Cd(II) reduction peaks appeared on the NF cathodes

Fig. 4. Comparison of bare, copper-deposited and oxidative copper-depositedelectrodes with (A) TS, (B) NF, and (C) CC materials and analyzed by CVs with orwithout the presence of Cd(II) (50 mg/L) (deposited copper: 150 mg/L, oxidativecopper-deposited means copper-deposited cathodes were exposed in the air for48 h).

570 Q. Wang et al. / Bioresource Technology 200 (2016) 565571

regardless of the presence or absence of deposited copper (Fig. 4B),presumably due to very high hydrogen gas production under theseconditions (Fig. 2 and Table S2) (Kundu et al., 2013). With enlarge-ment of parts of the curves (Inserts in Fig. 4AC), however, thesame reductive peak at 0.417 V was found on all the three mate-rials with deposited copper, but this peak was not present in theabsence of copper, implying the contribution of deposited copperto Cd(II) reduction. This value of 0.417 V was more positive thanthe theoretical Cd(II) redox potential of 0.520 V under theseexperimental conditions, clearly demonstrating the role of depos-ited copper in decreasing the overpotential for Cd(II) reduction,consistent with a previous report under more conventional elec-trodeposition conditions (Lovell and Roy, 1998). Accordingly,reductive peak currents on these electrodes with deposited copper

were always higher than the controls with no copper. Higherreductive peak currents were observed on the CC electrodes thanNF, followed by TS with deposited copper on the surface. The dif-ferent reductive peak currents and reductive peak potentials onthese materials can be also explained by the diverse morphologiesof deposited copper (Fig. S4AC), which may have resulted in thedifferent effective catalysis surface area, and thus affected systemperformance (Habibi et al., 2008; Liu et al., 2014). All three mate-rials with oxidative copper-deposited on the surface exhibited sim-ilar abatements in reductive peak currents (Fig. 4AC),demonstrating the adverse effects of oxidative copper on Cd(II)reduction and reflecting the role of pure copper instead of oxida-tive copper in high reductive peak currents. Hydrogen evolutionwas observed based on the formation of gas bubbles on the surfaceof all material electrodes with deposited copper at a potential morenegative than 0.6 V for NF, compared to 0.7 V for TS and CC,illustrating that NF had more superior catalytic properties forhydrogen evolution, consistent with higher hydrogen productionrates in MEC tests (Fig. 2B) and the lowest ratio of charges con-sumed for Cd(II) reduction and hydrogen evolution (Table S2).

3.6. Effect of anions of Cl and SO42

For all the three material cathodes with copper deposited on thesurface, Cd(II) reduction (Fig. S5A), hydrogen production rates(Fig. S5B) and current densities (Fig. S5C) were higher using theCl catholyte compared to catholytes with SO42. The improve-ments of 59% in Cd(II) reduction, 74650% in hydrogen produc-tion, and 28195% in current density were achieved. In theabsence of copper on the cathodes, there was very little differencein Cd(II) reduction (Fig. S5A) or hydrogen production (Fig. S5B) dueto the use of Cl or SO42 in the catholyte. However, Cl ions in thecatholyte resulted in copper corrosion during Cd(II) reduction, withthe release of 0.2% (CC), 0.8% (TS) and 5.3% (NF) of the depositedcopper into the catholyte, compared to 0.0% (CC), 0.3% (TS) and0.3% (NF) using the CdSO4 catholyte. These results were consistentwith previous reports using conventional electrochemical pro-cesses (Tansug et al., 2014), where a copper electrode corrodedin chloride media.

3.7. Effect of applied voltage

Cd(II) reduction on all the three material cathodes increasedwith the applied voltages up to 0.7 V, approaching the maximalrates of 5.65 (NF)6.59 (CC) mg/L h, and decreased thereafter(Fig. S6A). Hydrogen evolution on all the cathodes, however,always increased with the increase in applied voltage (Fig. S6B).Although the NF cathodes had the highest current densities atthe different applied voltages (Fig. S6C), more electrons werediverted to hydrogen evolution (Fig. S6B) than Cd(II) reduction athigher applied voltages (Fig. S6A and Table S3). CC cathodes weremore favorable for Cd(II) reduction (Fig. S6A) than hydrogen evolu-tion (Fig. S6B and Table S3) despite a low current density (Fig. S6C).These results demonstrate the dependency of both Cd(II) reductionand hydrogen evolution on cathode material with depositedcopper.

Anode potentials stabilized at 0.250 V at applied voltagesbelow 0.7 V, whereas cathode potentials always decreased inver-sely with the applied voltage (Fig. S6D), showing the importanceof cathode rather than anode potential. The lowest energy con-sumption for Cd(II) reduction was obtained at 0.5 V, reaching1.6 0.1 kWh/kg Cd (TS), 2.1 0.0 kWh/kg Cd (CC) and7.8 0.2 kWh/kg Cd (NF) (Table S3). Net energy production wasonly obtained at 0.3 and 0.5 V for the TS and CC materials.

Q. Wang et al. / Bioresource Technology 200 (2016) 565571 571

3.8. Long-term stability of copper deposited cathodes

In order to assess the stability of Cu-deposited cathodes, thecathodes with deposited copper of 150 mg/L were subsequentlyused in multiple batch cycles for Cd(II) reduction. A similardecreased trend in Cd(II) reduction rates were observed for allthree materials over the 20 cycles, where there was a decrease inthe rate of 31% (NF) to 37% (CC) compared to cycle 1 (Fig. S7A).These results show the instability of the copper on the surface overrepeated cycles. Hydrogen production showed a steeper sharpdecline over time, with a reduction of 100% (TS), 59% (NF) and91% (CC) after 20 cycles (Fig. S7B). This indicates that hydrogenevolution was affected to a greater extent than Cd(II) reductionover time. Current density was decreased over time in all cases(Fig. S7C). LSVs on all material cathodes were consistent with theperformances (Fig. S7DF), where the conductive and catalysisdeposited copper resulted in higher current densities and lowervoltages needed to initiate substantial currents while the un-conductive cadmium deposited within an 8 cycles operationadversely affected current densities and voltages needed to initiatesubstantial currents.

4. Conclusions

Cathodes with in situ deposited copper fromMFC operation suc-cessfully enhanced subsequent Cd(II) removal and hydrogen evolu-tion in MEC mode using the cathode materials TS, NF and CC.Despite similar Cu(II) reduction rates in MFC mode, reduction ofCd(II) competed with hydrogen evolution in the MEC mode forthe produced cathodic charge. This competitive relationship wasalways found on all the material cathodes, but it changed withcathode material, anions of Cl versus SO42, and applied voltage.These results show this BES can be used for efficient remediationof Cu(II) and Cd(II) contaminated waters with simultaneous energyrecovery.

Acknowledgements

The authors gratefully acknowledge financial support from theNational Natural Science Foundation of China (Nos. 21377019 and51578104), Specialized Research Fund for the Doctoral Program ofHigher Education SRFDP (No. 20120041110026), the Program forChangjiang Scholars and Innovative Research Team in University(IRT_13R05) and the Programme of Introducing Talents of Disci-pline to Universities (B13012).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.biortech.2015.10.084.

References

An, Z., Zhang, H., Wen, Q., Chen, Z., Du, M., 2014. Desalination combined with copper(II) removal in a novel microbial desalination cell. Desalination 346, 115121.

Bousher, A., 1995. Unidentate complexes involving borate. J. Coord. Chem. 34, 111.Call, D., Logan, B.E., 2008. Hydrogen production in a single chamber microbial

electrolysis cell lacking a membrane. Environ. Sci. Technol. 42, 34013406.Chen, M., Chao, M., Ma, X., 2014. Poly(crystal violet)/graphene-modified electrode

for the simultaneous determination of trace lead and cadmium ions in watersamples. J. Appl. Electrochem. 44, 337344.

Cheng, S., Wang, B., Wang, Y., 2013. Increasing efficiencies of microbial fuel cells forcollaborative treatment of copper and organic wastewater by designing reactorand selecting operating parameters. Bioresour. Technol. 147, 332337.

Cheng, S., Xing, D., Call, D.F., Logan, B.E., 2009. Direct biological conversion ofelectrical current into methane by electromethanogenesis. Environ. Sci.Technol. 43, 39533958.

Choi, C., Hu, N., Lim, B., 2014. Cadmium recovery by coupling double microbial fuelcells. Bioresour. Technol. 170, 361369.

Gattrell, M., Gupta, N., Co, A., 2006. A review of the aqueous electrochemicalreduction of CO2 to hydrocarbons at copper. J. Electroanal. Chem. 594, 119.

Habibi, B., Pournaghi-Azar, M.H., Razmi, H., Abdolmohammad-Zadeh, H., 2008.Electrochemical preparation of a novel, effective and low cast catalytic surfacefor hydrogen evolution reaction. Int. J. Hydrogen Energy 33, 26682678.

Haque, I.U., Tariq, M., 2010. Electrochemical reduction of nitrate: a review. J. Chem.Soc. Pak. 32, 396418.

Heijne, A.T., Liu, F., Weijden, R., Weijma, J., Buisman, C.J.N., Hamelers, H.V.M., 2010.Copper recovery combined with electricity production in a microbial fuel cell.Environ. Sci. Technol. 44, 43764381.

Khallaf, H., Oladeji, I.O., Chai, G., Chow, L., 2008. Characterization of CdS thin filmsgrown by chemical bath deposition using four different cadmium sources. ThinSolid Films 516, 73067312.

Kok, L., Rzsenberszki, T., Nemestthy, N., Blafi-Bak, K., Bakonyi, P., 2015.Bioelectrochemical treatment of municipal waste liquor in microbial fuel cellsfor energy valorization. J. Clean. Prod. http://dx.doi.org/10.1016/j.jclepro.2015.06.116 (in press).

Kundu, A., Sahu, J.N., Redzwan, G., Hashim, M.A., 2013. An overview of cathodematerial and catalysts suitable for generating hydrogen in microbial electrolysiscell. Int. J. Hydrogen Energy 38, 17451757.

Li, W., Yu, H., He, Z., 2014. Towards sustainable wastewater treatment by usingmicrobial fuel cells-centered technologies. Energy Environ. Sci. 7, 911924.

Liu, X., Li, W., Yu, H., 2014. Cathodic catalysts in bioelectrochemical systems forenergy recovery from wastewater. Chem. Soc. Rev. 43, 77187745.

Logan, B.E., 2012. Essential data and techniques for conducting microbial fuel celland other types of bioelectrochemical system experiments. ChemSusChem 5,988994.

Lovell, M.A., Roy, D., 1998. Effects of sub-surface oxygen on electrodeposition ofcadmium on copper. Electrochim. Acta 43, 21172130.

Luo, H., Liu, G., Zhang, R., Bai, Y., Fu, S., Hou, Y., 2014. Heavy metal recoverycombined with H2 production from artificial acid mine drainage using themicrobial electrolysis cell. J. Hazard. Mater. 270, 153159.

Lv, M., Wang, X., Li, J., Yang, X., Zhang, C., Yang, J., Hu, H., 2013. Cyclodextrin-reduced graphene oxide hybrid nanosheets for the simultaneous determinationof lead(II) and cadmium(II) using square wave anodic stripping voltammetry.Electrochim. Acta 108, 412420.

Martin-Lara, M.A., Blazquez, G., Trujillo, M.C., Perez, A., Calero, M., 2014. Newtreatment of real electroplating wastewater containing heavy metal ions byadsorption onto olive stone. J. Clean. Prod. 81, 120129.

Modin, O., Wang, X., Wu, X., Rauch, S., Fedje, K.K., 2012. Bioelectrochemical recoveryof Cu, Pb, Cd, and Zn from dilute solutions. J. Hazard. Mater. 235, 291297.

Nancharaiah, Y.V., Mohan, S.V., Lens, P.N.L., 2015. Metals removal and recovery inbioelectrochemical systems: A review. Bioresour. Technol. 195, 102114.

Prakash, N., Sudha, P.N., Renganathan, N.G., 2012. Copper and cadmium removalfrom synthetic industrial wastewater using chitosan and nylon 6. Environ. Sci.Pollut. R. 19, 29302941.

Rivera, I., Buitrn, G., Bakonyi, P., Nemestthy, N., Blafi-Bak, K., 2015. Hydrogenproduction in a microbial electrolysis cell fed with a dark fermentation effluent.J. Appl. Electrochem. 45, 12231229.

Sun, Z., Cao, R.X., Huang, M.H., Chen, D.H., Zheng, W., Lin, L., 2015. Effect of lightirradiation on the photoelectricity performance of microbial fuel cell with acopper oxide nanowire photocathode. J. Photochem. Photobiol. A Chem. 300,3843.

Tansug, G., Tken, T., Giray, E.S., Fndkkran, G., Sgrck, G., Demirkol, O., Erbil, M.,2014. A new corrosion inhibitor for copper protection. Corros. Sci. 84, 2129.

Tao, H.C., Li, W., Liang, M., Xu, N., Ni, J.R., Wu, W.M., 2011. A membrane-free baffledmicrobial fuel cell for cathodic reduction of Cu(II) with electricity generation.Bioresour. Technol. 102, 47744778.

Tao, H.C., Lei, T., Shi, G., Sun, X.N., Wei, X.Y., Zhang, L.J., Wu, W.M., 2014. Removal ofheavy metals from fly ash leachate using combined bioelectrochemical systemsand electrolysis. J. Hazard. Mater. 264, 17.

Wang, H., Ren, Z.J., 2014. Bioelectrochemical metal recovery from wastewater: areview. Water Res. 66, 219232.

Wang, Q., Huang, L., Yu, H., Quan, X., Li, Y., Fan, G., Li, L., 2015. Assessment of fivedifferent cathode materials for Co(II) reduction with simultaneous hydrogenevolution in microbial electrolysis cells. Int. J Hydrogen Energy 40, 184196.

Wu, D., Pan, Y., Huang, L., Zhou, P., Quan, X., Chen, H., 2015a. Complete separation ofCu(II), Co(II) and Li(I) using self-driven MFCs-MECs with stainless steel meshcathodes under continuous flow conditions. Sep. Purif. Technol. 147, 114124.

Wu, D., Pan, Y., Huang, L., Quan, X., Yang, J., 2015b. Comparison of Co(II) reductionon three different cathodes of microbial electrolysis cells driven by Cu(II)-reduced microbial fuel cells under various cathode volume conditions. Chem.Eng. J. 266, 121132.

Zhang, Y., Yu, L., Wu, D., Huang, L., Zhou, P., Quan, X., Chen, G., 2015. Dependency ofsimultaneous Cr(VI), Cu(II) and Cd(II) reduction on the cathodes of microbialelectrolysis cells self-driven by microbial fuel cells. J. Power Sources 273, 11031113.

Zhu, X., Logan, B.E., 2014. Copper anode corrosion affects power generation inmicrobial fuel cells. J. Chem. Technol. Biotechnol. 89, 471474.

http://dx.doi.org/10.1016/j.biortech.2015.10.084http://dx.doi.org/10.1016/j.biortech.2015.10.084http://refhub.elsevier.com/S0960-8524(15)01487-X/h0005http://refhub.elsevier.com/S0960-8524(15)01487-X/h0005http://refhub.elsevier.com/S0960-8524(15)01487-X/h0010http://refhub.elsevier.com/S0960-8524(15)01487-X/h0015http://refhub.elsevier.com/S0960-8524(15)01487-X/h0015http://refhub.elsevier.com/S0960-8524(15)01487-X/h0020http://refhub.elsevier.com/S0960-8524(15)01487-X/h0020http://refhub.elsevier.com/S0960-8524(15)01487-X/h0020http://refhub.elsevier.com/S0960-8524(15)01487-X/h0025http://refhub.elsevier.com/S0960-8524(15)01487-X/h0025http://refhub.elsevier.com/S0960-8524(15)01487-X/h0025http://refhub.elsevier.com/S0960-8524(15)01487-X/h0030http://refhub.elsevier.com/S0960-8524(15)01487-X/h0030http://refhub.elsevier.com/S0960-8524(15)01487-X/h0030http://refhub.elsevier.com/S0960-8524(15)01487-X/h0035http://refhub.elsevier.com/S0960-8524(15)01487-X/h0035http://refhub.elsevier.com/S0960-8524(15)01487-X/h0040http://refhub.elsevier.com/S0960-8524(15)01487-X/h0040http://refhub.elsevier.com/S0960-8524(15)01487-X/h0040http://refhub.elsevier.com/S0960-8524(15)01487-X/h0045http://refhub.elsevier.com/S0960-8524(15)01487-X/h0045http://refhub.elsevier.com/S0960-8524(15)01487-X/h0045http://refhub.elsevier.com/S0960-8524(15)01487-X/h0050http://refhub.elsevier.com/S0960-8524(15)01487-X/h0050http://refhub.elsevier.com/S0960-8524(15)01487-X/h0055http://refhub.elsevier.com/S0960-8524(15)01487-X/h0055http://refhub.elsevier.com/S0960-8524(15)01487-X/h0055http://refhub.elsevier.com/S0960-8524(15)01487-X/h0060http://refhub.elsevier.com/S0960-8524(15)01487-X/h0060http://refhub.elsevier.com/S0960-8524(15)01487-X/h0060http://dx.doi.org/10.1016/j.jclepro.2015.06.116http://dx.doi.org/10.1016/j.jclepro.2015.06.116http://refhub.elsevier.com/S0960-8524(15)01487-X/h0070http://refhub.elsevier.com/S0960-8524(15)01487-X/h0070http://refhub.elsevier.com/S0960-8524(15)01487-X/h0070http://refhub.elsevier.com/S0960-8524(15)01487-X/h0075http://refhub.elsevier.com/S0960-8524(15)01487-X/h0075http://refhub.elsevier.com/S0960-8524(15)01487-X/h0080http://refhub.elsevier.com/S0960-8524(15)01487-X/h0080http://refhub.elsevier.com/S0960-8524(15)01487-X/h0085http://refhub.elsevier.com/S0960-8524(15)01487-X/h0085http://refhub.elsevier.com/S0960-8524(15)01487-X/h0085http://refhub.elsevier.com/S0960-8524(15)01487-X/h0090http://refhub.elsevier.com/S0960-8524(15)01487-X/h0090http://refhub.elsevier.com/S0960-8524(15)01487-X/h0095http://refhub.elsevier.com/S0960-8524(15)01487-X/h0095http://refhub.elsevier.com/S0960-8524(15)01487-X/h0095http://refhub.elsevier.com/S0960-8524(15)01487-X/h0095http://refhub.elsevier.com/S0960-8524(15)01487-X/h0100http://refhub.elsevier.com/S0960-8524(15)01487-X/h0100http://refhub.elsevier.com/S0960-8524(15)01487-X/h0100http://refhub.elsevier.com/S0960-8524(15)01487-X/h0100http://refhub.elsevier.com/S0960-8524(15)01487-X/h0105http://refhub.elsevier.com/S0960-8524(15)01487-X/h0105http://refhub.elsevier.com/S0960-8524(15)01487-X/h0105http://refhub.elsevier.com/S0960-8524(15)01487-X/h0110http://refhub.elsevier.com/S0960-8524(15)01487-X/h0110http://refhub.elsevier.com/S0960-8524(15)01487-X/h0115http://refhub.elsevier.com/S0960-8524(15)01487-X/h0115http://refhub.elsevier.com/S0960-8524(15)01487-X/h0120http://refhub.elsevier.com/S0960-8524(15)01487-X/h0120http://refhub.elsevier.com/S0960-8524(15)01487-X/h0120http://refhub.elsevier.com/S0960-8524(15)01487-X/h0125http://refhub.elsevier.com/S0960-8524(15)01487-X/h0125http://refhub.elsevier.com/S0960-8524(15)01487-X/h0125http://refhub.elsevier.com/S0960-8524(15)01487-X/h0130http://refhub.elsevier.com/S0960-8524(15)01487-X/h0130http://refhub.elsevier.com/S0960-8524(15)01487-X/h0130http://refhub.elsevier.com/S0960-8524(15)01487-X/h0130http://refhub.elsevier.com/S0960-8524(15)01487-X/h0135http://refhub.elsevier.com/S0960-8524(15)01487-X/h0135http://refhub.elsevier.com/S0960-8524(15)01487-X/h0135http://refhub.elsevier.com/S0960-8524(15)01487-X/h0135http://refhub.elsevier.com/S0960-8524(15)01487-X/h0135http://refhub.elsevier.com/S0960-8524(15)01487-X/h0135http://refhub.elsevier.com/S0960-8524(15)01487-X/h0135http://refhub.elsevier.com/S0960-8524(15)01487-X/h0135http://refhub.elsevier.com/S0960-8524(15)01487-X/h0140http://refhub.elsevier.com/S0960-8524(15)01487-X/h0140http://refhub.elsevier.com/S0960-8524(15)01487-X/h0140http://refhub.elsevier.com/S0960-8524(15)01487-X/h0145http://refhub.elsevier.com/S0960-8524(15)01487-X/h0145http://refhub.elsevier.com/S0960-8524(15)01487-X/h0145http://refhub.elsevier.com/S0960-8524(15)01487-X/h0150http://refhub.elsevier.com/S0960-8524(15)01487-X/h0150http://refhub.elsevier.com/S0960-8524(15)01487-X/h0155http://refhub.elsevier.com/S0960-8524(15)01487-X/h0155http://refhub.elsevier.com/S0960-8524(15)01487-X/h0155http://refhub.elsevier.com/S0960-8524(15)01487-X/h0160http://refhub.elsevier.com/S0960-8524(15)01487-X/h0160http://refhub.elsevier.com/S0960-8524(15)01487-X/h0160http://refhub.elsevier.com/S0960-8524(15)01487-X/h0165http://refhub.elsevier.com/S0960-8524(15)01487-X/h0165http://refhub.elsevier.com/S0960-8524(15)01487-X/h0165http://refhub.elsevier.com/S0960-8524(15)01487-X/h0165http://refhub.elsevier.com/S0960-8524(15)01487-X/h0170http://refhub.elsevier.com/S0960-8524(15)01487-X/h0170http://refhub.elsevier.com/S0960-8524(15)01487-X/h0170http://refhub.elsevier.com/S0960-8524(15)01487-X/h0170http://refhub.elsevier.com/S0960-8524(15)01487-X/h0175http://refhub.elsevier.com/S0960-8524(15)01487-X/h0175

Cooperative cathode electrode and in&blank;situ deposited copper for subsequent enhanced Cd\(II\) removal and hydrogen evolution in bioelectrochemical systems1 Introduction2 Methods2.1 Reactor construction2.2 Start up and operation2.3 Analytical measurements2.4 Calculations

3 Results and discussion3.1 Copper deposition in MFCs3.2 Cd\(II\) reduction and hydrogen evolution in MECs3.3 Energy recovery in MECs3.4 Electrode morphologies and product characterization3.5 Cyclic voltammetry analysis3.6 Effect of anions of Cl and SO423.7 Effect of applied voltage3.8 Long-term stability of copper deposited cathodes

4 ConclusionsAcknowledgementsAppendix A Supplementary dataReferences