- Email: [email protected]
Effect of in-situ immobilized anode on performance of the microbial fuel cell with high concentration of sodium acetate
Fuel 182 (2016) 732–739
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel
Full Length Article
Effect of in-situ immobilized anode on performance of the microbial fuel cell with high concentration of sodium acetate Haiping Luo, Shuxian Yu, Guangli Liu ⇑, Renduo Zhang, Wenkai Teng Guangdong Provincial Key Laboratory of Environmental Pollution Control and Remediation Technology, School of Environmental Science and Engineering, Sun Yat-sen University, Guangzhou 510275, China
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
An efficient in-situ immobilization
Maximum Power density Improved
method was developed for anode of the MFC. The MFC with immobilized anode generated higher voltages than the control MFC. The MFC with immobilized anode produced higher power densities than the control MFC. The Geobacter abundance in the MFC with immobilized anode was higher. Matured anode
i n f o
Article history: Received 18 February 2016 Received in revised form 5 June 2016 Accepted 7 June 2016
Keywords: Microbial fuel cell In-situ immobilization High concentration of sodium acetate Microbial community
R
vs. Anode
a r t i c l e
54%
R
In-situ immobilized anode by agarose gel
Cathode
MFC with immobilized anode (fed by 10g L-1 acetate)
Anode
Cathode
Controlled MFC
a b s t r a c t To improve performance of the microbial fuel cell (MFC) with high concentration of organics, a procedure for in-situ immobilized anode with agarose gel was proposed in this study. The performance of power generation, electrochemical activity, and microbial community of the MFC with immobilized anode (i-MFC) was investigated using different concentrations (i.e., 1, 5, 10, and 20 g/L) of sodium acetate. The i-MFC could generate electricity within one hour after refreshing substrate. With 5 g/L acetate, the maximum voltages were 560 and 460 mV in the i-MFC and controlled MFC (c-MFC), respectively. With 10 g/L acetate, the maximum voltage of the i-MFC was much higher than that of the c-MFC (500 vs. 300 mV). With the acetate concentrations of 5 and 10 g/L, the maximum power densities in the i-MFC were 610 ± 50 and 370 ± 40 mW/m2, respectively, while the maximum power densities in the c-MFC were 343 ± 30 and 240 ± 20 mW/m2, respectively. Cyclic voltammetry measurements indicated that microbial activity of the i-MFC anode was higher than that of the c-MFC anode under the high acetate concentration (3.00 vs. 2.19 mA). The relative abundance of Geobacter on the i-MFC anode was much higher than that on the c-MFC anode (62% vs. 40%). The in-situ immobilization strategy provided an easily performed and efficient way to keep the activity of exoelectrogens, resulting in the improvement of MFC performance. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction ⇑ Corresponding author. E-mail address: [email protected] (G. Liu). http://dx.doi.org/10.1016/j.fuel.2016.06.032 0016-2361/Ó 2016 Elsevier Ltd. All rights reserved.
High concentrations of organics were often detected in many types of industrial wastewater, such as wastewater from process-
H. Luo et al. / Fuel 182 (2016) 732–739
ing plants of brewery, wine, and potato. Values of chemical oxygen demand (COD) of such wastewater change from 700 to 8000 mg/L [1,2]. Microbial fuel cell (MFC) was a kind of bioelectrochemical system (BES) that can generate electricity via organic biodegradation by electrochemically active bacteria [1]. Therefore, the MFC could be a promising approach to treat wastewater containing high COD concentrations, which should consume little energy or even generate electricity output (i.e., energy harvesting) [3]. However, the performance of MFC was affected by many factors, such as anode and cathode materials, reactor structure, substrates, exoelectrogens, and others [1]. High concentrations of substrates could inhibit activity of exoelectrogens in the MFC. It had been shown that in a dual-chamber MFC constructed with Geobacter sulfurreducens, the maximum power densities decrease from 16 to 13 mW/m3, and coulumbic efficiency (CE) from 46% to 23% with acetate concentrations from 5 to 20 mM [4]. In an MFC inoculated with Shewanella putrefaciens, the maximum current could not increase further as the initial concentration of lactate exceeded 200 mM [5]. Sharma and Li [6] demonstrated that the maximum voltages generated by a single-chamber air-cathode MFC inoculated with domestic wastewater increased with concentrations of acetate, ethanol, and glucose from 0.5 to 20 mM, but significantly decreased with concentrations of 20–35 mM. Therefore, under high organic concentrations, it should be necessary to improve the MFC performance for its potential applications in industrial wastewater treatment. Cell immobilization methods, such as adsorption, covalent binding, and entrapment, had been shown to reduce the environmental pressure (e.g., high salinity, toxicity) to microbes and to keep high cell densities in the biofuel cells [7,8]. The entrapment method, in which polymers or inorganics were utilized to form a framework with cells, had the advantages of good mechanical strength and stability, tunable porosity, and others. Using the entrapment method to immobilize exoelectrogens on the anode could stabilize the MFC performance under severe environmental conditions. Shewanella oneidensis MR-1 had been immobilized with graphite and alginate granules in the MFC, resulting in 0.8– 1.7 times higher CE than the control [9]. A method of one-step vapor deposition of silica had been used to immobilize S. oneidensis to form an aqueous sol-gel in the anode of MFC [10], which improved the maximum power density of MFC. A culture mixture in an air-cathode MFC had been immobilized using the polyvinyl alcohol and powdered active carbon to treat distillery wastewater [11]. The immobilized MFC could produce electricity of 72 mW/m2 with the influent COD concentration of 3600 mg/L. The entrapment methods for the immobilized anode of MFC above included the following steps: mixing exoelectrogens with polymers and attaching the mixture to the electrode. Such methods may cause some problems, such as damage of exoelectrogens due to violently mixing between bacteria and polymers, and delay of electricity generation because of the time needed for the attachment. Therefore, an in-situ immobilization strategy should be more attractive in the MFC. As a natural organic polymer (i.e., a linear polyscaccharide), which can be produced from marine red aglae [12], agarose has higher biocompatibility with lower cost than some of artificial polymers, such as polyvinyl alcohol. Compared with alignate and silica, agarose is more sensitive to temperature, thus easier to form gel for immobilization [13,14]. The objectives of this study were to develop an in-situ agarose immobilized procedure and to investigate the effect of in-situ immobilized exoelectrogens on the MFC performance under high substrate (acetate) concentration. The characteristics of immobilized and control MFCs, including power density, internal resistance, CE, and bacteria community, were determined and compared.
733
2. Materials and methods 2.1. Reactor construction and operation The base anode of MFC was made of carbon cloth (W0S 1002, CeTech., Taiwan, China). Characteristics of the carbon cloth included 125 g/m2, thickness of 360 lm, and effective area of 16 cm2 (L W = 8 cm 2 cm) [15]. The carbon cloth was treated with 450 °C in a muffle furnace for 30 min before use. Plain carbon cloth was rolled into a cylindrical shape (diameter of 1 cm and height of 2 cm) with a plastic mesh (grid: 3 mm 3 mm, thickness: 0.1 mm) as a separator to keep inner tunnels for bacterium growth. Platinum-catalyst air-cathode was made as previously described [16]. The perspex reactor had an effective volume of 28 mL with a diameter of 3 cm and a length of 4 cm. The anode and cathode were connected with an external resistance of 1000 X via titanium wire. The inoculum was from effluent of a grit chamber in a municipal wastewater treatment plant (Datansha, Guangzhou, China). During the startup stage, when the voltage was lower than 20 mV, the reactors were refreshed with a solution including 1 g/ L sodium acetate, 50 mM phosphate buffer, and vitamins [16]. After the reactors could generate the maximum voltage >500 mV for at least three cycles, the matured anodes in the MFCs were used for immobilization of exoelectrogens as follows. 2.2. Immobilization of exoelectrogens on the anode A schematic diagram of the procedure for anode immobilization was shown in Fig. 1. One gram of agar was dissolved into 100 mL phosphate buffer solution by heating at 100 °C. After the agar gel was cooled down to 40 °C, the matured anode was quickly put into the solution. A sterilized knife was used to cut the immobilized anode to a cubic size of 2.5 cm 1.5 cm (H d) after the agar gel was totally solidified at 30 °C. The immobilized anode was washed several times with distilled water and then refilled into the reactor (i-MFC). Reactors with the matured anodes without the immobilization procedure were used as the control (c-MFC). The reactors were operated with 1, 5, 10, and 20 g/L of sodium acetate as the substrate. All tests were duplicated at 30 ± 2 °C. 2.3. Analysis and calculations Voltages were measured across the external resistor every 15 min using a data electronic multimeter (DataTaker 85, Australia). The current in the circuit was calculated using Ohm’s law, which was normalized with the cathode projected surface area
Fig. 1. The schematic diagram of the procedure to the anode immobilization in the microbial fuel cell.
734
H. Luo et al. / Fuel 182 (2016) 732–739
(7 cm2) to estimate the current density. Polarization curve was obtained by changing external resistances from 100 to 3000 X [16]. CE was calculated based on the totally generated current and the COD removal [17]. COD was measured according to the standard method [18]. Biomass in the anode and cathode was determined based on the protein concentration using the Coomassie Brilliant Blue method [19]. The morphology of anode was examined using a scanning electron microscope (SEM) (Quanta 400, FEI, Netherlands) [20]. An electrochemical station (CHI660, Chenhua Co. Ltd, Shanghai, China) was used to carry out the cyclic voltammetry (CV) and linear sweep voltage (LSV) tests. For the CV test, a three-electrode (i.e., working electrode (the anode), counter electrode (the cathode), and reference electrode (Ag/AgCl)) arrangement was used with anode potentials from 0.8 to 0.2 V vs. Ag/AgCl at a scanning rate of 20 mV/s [21]. Each reactor was scanned for three cycles and the last cycle was plotted for discussion. For the LSV test, a used cathode (after at least three cycles of operation) was put into an abiotic MFC reactor filled with 50 mM phosphate buffer solution. The cathode was used as the working electrode, a platinum electrode as the counter electrode, and an Ag/AgCl electrode (closed to the working electrode) as the reference electrode. The LSV test was carried out with potentials from 0.5 to 0.2 V vs. Ag/AgCl at a scanning rate of 1 mV/s. Each cathode was tested for three times [22]. 2.4. Microbial community analysis After five operation cycles (about 1100 h) of the i-MFC and c-MFC fed with 10 g/L acetate, anode electrode samples were collected using a sterilized scissor. The deoxyribonucleic acid (DNA) kit (D5625-01, Omega Bio-Tek, USA) was used to extract the samples’ DNA. After examining the extracted DNA qualities with 1% agarose gel electrophoresis (100 V, 30 min), the universal primers 338F (ACTCCTACGGGAGGCAGCA) and 806R (GGACTACHVGGGTWTCTAAT) were used to amplify the V3-V4 region of 16S ribosomal DNA (rDNA) genes. The reverse primer with a 6-bp error-correcting barcode unique was assigned to each sample [23,24]. PCR amplification was carried out as follows: 3 min of denaturation at 95 °C, 27 cycles of 30 s at 95 °C, 30 s at 55 °C, 45 s at 72 °C, and then a final extension for 10 min at 72 °C. The PCR products were then purified and quantified. Sequencing was carried out on an Illumina Miseq platform by Majorbio (Shanghai, China). Pairs of reads were merged by FLASH and sequencing reads were assigned to each sample based on the unique barcode. QIIME software package (http://qiime.org) and UPARSE pipeline (http://drive5.com/upars/) were used to select high quality sequences and group them into operational taxonomic units (OTUs) at 97% similarity. The Silva database (http://www.arbsilva.de) was used to classify the sequences at the phylum and genus levels. Statistical indexes, including Chao 1 richness and Shannon diversity indexes, were calculated for each sample [25]. The nucleotide sequences reported in this study have been deposited in NCBI SRA with SRP065089. 3. Results 3.1. MFC performance with immobilized anode The i-MFC and c-MFC could generate electricity with 1.0, 5.0, and 10 g/L acetate as the substrate, respectively (Fig. 2). The i-MFC and c-MFC could generate electricity with 20 g/L acetate only in the first cycle operation. With 1 g/L acetate, the performance of i-MFC and c-MFC was similar. The maximum voltages reached 600 mV and the running time per cycle was about 22 h
Fig. 2. Electricity generation of the microbial fuel cell (MFC) with and without immobilized anode by agarose gel (i-MFC and c-MFC, respectively) with different concentrations of sodium acetate: (A) 1 g/L, (B) 5 g/L, and (C) 10 g/L.
in the MFCs. The operation time per cycle in both the MFCs was about 80 and 220 h with the acetate concentrations of 5 and 10 g/L, respectively. With 5 g/L acetate, the maximum voltages were 560 and 460 mV in the i-MFC and c-MFC, respectively. With 10 g/L acetate, the maximum voltages were 500 and 300 mV in the i-MFC and c-MFC, respectively. With the acetate concentrations of 1, 5, and 10 g/L, the maximum power densities in the i-MFC were 680 ± 55, 610 ± 50, and 370 ± 40 mW/m2, respectively, while the maximum power densities in the c-MFC were 610 ± 50, 343 ± 30, and 240 ± 20 mW/m2, respectively (Fig. 3). The reduction in power generation with the acetate concentrations was attributed to the increase of internal resistance in the systems. The internal resistances in the i-MFC and c-MFC increased from 255 ± 21 to 412 ± 35 X and from 291 ± 24 to 477 ± 39 X, respectively, with the acetate concentrations from 1 to 10 g/L. The COD removal reached about 90% within 96 h in the MFCs fed with 10 g/L acetate (Fig. 4), indicating that the exoelectrogens converted acetate into chemical energy first and then changed it
H. Luo et al. / Fuel 182 (2016) 732–739
735
for immobilizing exoelectrogens on the anode without inhibiting mass transfer between the substrate and biofilm. Values of CE decreased with increase of the acetate concentrations in the reactors. With 1 g/L acetate, the CE values in the i-MFC and c-MFC were almost the same (13.5 ± 2.1% vs. 14.3 ± 0.9%). With 10 g/L acetate, the CE values decreased to 10.7 ± 1.1% and 8.4 ± 2.0% in the i-MFC and c-MFC, respectively. 3.2. Electrochemical analysis on the MFCs With the acetate concentration of 1 g/L, the oxidation peaks in the forward and reverse scans were found at 0.31 V (vs Ag/AgCl) (3.43 mA) and 0.35 V (1.43 mA) for the i-MFC, respectively, and 0.37 V (2.83 mA) and 0.41 V (1.10 mA) for the c-MFC, respectively (Fig. 5A). With the acetate concentration of 10 g/L, the forward oxidation peaks were observed at 0.20 V (3.00 mA) and 0.32 V (2.19 mA) for the i-MFC and c-MFC, respectively. The reverse oxidation peaks were at 0.34 V (0.82 mA) and 0.39 V (0.89 mA) for the i-MFC and c-MFC, respectively (Fig. 5B). The potential of oxidation peak in the MFC was higher with 10 g/L than with 1 g/L acetate, which was likely attributed to the effect of substrate concentrations on the oxidation rate. The peak currents of the i-MFC were higher than those of the c-MFC, indicating that the agar immobilization enhanced the activity of exoelectrogens on the anode. The peak current of the i-MFC anode was lower with 10 g/L than with 1 g/L acetate, suggesting an adverse effect of high acetate concentration on the activity of immobilized exoelectrogens. As shown by the LSV results in Fig. 6, the cathode activity in the i-MFC was almost the same as that in the c-MFC. However, the
Fig. 3. (A) Power density-current curve and (B) voltage-current curve of the microbial fuel cell (MFC) with and without immobilized anode by agarose gel (i-MFC and c-MFC, respectively) fed by different concentrations of sodium acetate. i-MFC 1, i-MFC 5, i-MFC 10 represent i-MFCs fed by 1, 5, and 10 g/L of sodium acetate, respectively. c-MFC 1, c-MFC 5, c-MFC 10 represent c-MFCs fed by 1, 5, and 10 g/L of sodium acetate, respectively.
Fig. 4. Temporal change of COD removal in the microbial fuel cell (MFC) with and without immobilized anode by agarose gel (i-MFC and c-MFC, respectively) fed by 10 g/L of sodium acetate.
gradually to electricity. The COD removal between the i-MFC and c-MFC was not significantly different, while the total Coulombs were higher in the i-MFC than in the c-MFC (283 ± 20 C vs. 229 ± 27 C) (Fig. 4). These results suggested that agar was suitable
Fig. 5. The cyclic voltammetry (CV) curves of the anodes in the microbial fuel cell (MFC) with and without immobilized anode by agarose gel (i-MFC and c-MFC, respectively) fed by (A) 1.0 g/L and (B) 10 g/L of sodium acetate. i-MFC 1 and i-MFC 10 represent i-MFCs fed by 1 and 10 g/L of sodium acetate, respectively. c-MFC 1 and c-MFC 10 represent c-MFCs fed by 1 and 10 g/L of sodium acetate, respectively.
736
H. Luo et al. / Fuel 182 (2016) 732–739
Fig. 6. The linear sweep voltage (LSV) results of the cathodes in the microbial fuel cell (MFC) with and without immobilized anode by agarose gel (i-MFC and c-MFC, respectively) fed by different concentrations of sodium acetate. i-MFC 1 and i-MFC 10 represent i-MFCs fed by 1 and 10 g/L of sodium acetate, respectively. c-MFC 1 and c-MFC 10 represent c-MFCs fed by 1 and 10 g/L of sodium acetate, respectively.
was much less in the i-MFC than in the c-MFC (1.7% vs. 15%). Shannon diversity index was lower in the i-MFC than in the c-MFC (2.45 vs 2.69) (Table 1). Values of Chao 1 (i.e., the index of the species richness) were 187 and 200 in the i-MFC and c-MFC, respectively. The values of Shannon diversity index and Chao1 showed that the i-MFC had lower species diversity and richness than the c-MFC, indicating that the in-situ immobilization of the anode blocked the development of the microbial community on the anode. The biomass in the anode of c-MFC was much higher than that in the anode of i-MFC (5.3 ± 1.0 vs.2.9 ± 0.7 mg/g). The result was explained as follows. The estimated average pore size inside 1% agarose gel was smaller than 0.5 lm [12], thus it was more difficult for bacteria in the anolyte to move into the gel. On the contrary, the biomass in the anode of c-MFC increased with the operation, as the self-growth of the anode biofilm and the attachment of bacteria from the anolyte. Moreover, the SEM images showed that bacteria were not observed apparently to grow on the surface of agarose gel in the i-MFC (Fig. 8), suggesting that free bacteria in the anolyte might prefer to move onto the cathode instead of anode, which needs to be further studied.
4. Discussion activities of both i-MFC and c-MFC were greatly decreased with the increase of acetate concentrations. For example, at 0.2 V the cathode current decreased from 15.4 ± 0.2 to 1.5 ± 0.2 mA with the acetate concentrations from 1.0 to 10 g/L. Our LSV results of the cathodes in the i-MFC fed with 1 g/L acetate were comparable to other reports [22]. The cathodic biomass measurement showed that a thick biofilm layer was developed on the surface of the cathodes of i-MFC and c-MFC fed by 10 g/L acetate. The cathodic biomass in the i-MFC was almost the same as that in the c-MFC (6.1 ± 1.0 vs. 6.0 ± 1.0 mg/g). 3.3. Community analysis on the anodic biofilm The numbers of OTUs at 97% sequence similarity were 178 and 190 for the i-MFC and c-MFC, respectively (Table 1). The bacterial community distributions shown in Table 2 were based on the percentage of total sequences at the phylum level. The anode biofilm of i-MFC was mainly composed of Proteobacteria (68%), Bacteroidetes (16%), and Synergistetes (6.0%), while 49% of Proteobacteria, 15% of Bacteroidetes and 15% of Lentisphaerae were found in the anode biofilm of c-MFC. Geobacter was identified and dominated the microbial communities at the genus level in the MFCs (Fig. 7). The relative abundance of Geobacter in the i-MFC was much higher than that in the c-MFC (62% vs. 40%). Lentisphaerae in the anode
Table 1 Richness and diversity estimates of the anode bacterial communities in the i-MFC and c-MFC based on the OTUs at 97% similarity. Sample
Reads
OTU
Chao1
Shannon diversity index
i-MFC c-MFC
30489 34129
178 190
187 200
2.45 2.69
i-MFC and c-MFC represent MFCs with and without immobilized anode by agarose gel, respectively.
Steadily periodical electricity was generated in the i-MFC for more than 1100 h (5 cycles), indicating that the performance of i-MFC was kept stable with the agarose gel. Compared with other methods, such as polyvinyl alcohol with powered active carbon gel immobilization [11] and alginate with graphite granules immobilization [26], our immobilization strategy resulted in better performance of MFC (Table 3). With 5 g/L acetate, the maximum power density and voltage in the i-MFC reached 610 mW/m2 and 560 mV, which were higher than those reported in the literature. The in-situ agarose gel immobilization method also made the startup time much shorter. The i-MFC in our test could generate electricity within one hour after refreshing substrate. Compared to other immobilization methods, such as adsorption and binding, our in-situ immobilization had the shortest startup time as far as we known. With those immobilization methods, several to hundreds hours were needed for the exoelectrogens to locate in the anode surface and to form a stable biofilm. For example, the startup time of MFC with the carbon nanoparticles immobilized anode was about 40 h, which was used for extracting bacteria from the matured biofilm (1 year operation), acclimating (12 h), and entrapping bacteria [26]. A startup time of 120 h was taken in the MFC with layer-by-layer assembled gold nanoparticles modified anode [27]. The anode immobilization through adsorption of the electron shuttles, such as methylene blue and neutral red, required more than 6 days to startup [28]. The performance improvement in the i-MFC were explained as follows. The exoelectrogens had high biocompatibility with the natural polymer of agarose. The high relative abundance of Geobacter in the anode biofilm indicated that the main mechanism in the MFC should be a direct electron transfer from Geobacter to the anode with ‘‘nanowires” [29]. The direct electron transfer between Geobacter and the anode reduced the start-up time compared to other immobilization methods. In addition, the appropriate pore size of agarose gel can minimize
Table 2 Microbial community distributions in percentage of total sequences at the phylum level for the anodes of the microbial fuel cells (MFCs).
i-MFC c-MFC
Proteobacteria
Bacteroidetes
Synergistetes
Firmicutes
Lentisphaerae
Actinobacteria
Verrucomicrobia
Acidobacteria
Others
67.8 49.2
16.0 14.5
6.0 6.5
3.4 3.5
1.7 15.2
1.6 4.3
1.0 4.7
0.2 0.3
2.3 1.8
i-MFC and c-MFC represent MFCs with and without immobilized anode by agarose gel, respectively.
H. Luo et al. / Fuel 182 (2016) 732–739
A
737
Geobacter Aminiphilus Victivallis Spirochaeta
26.4%
Christensenellaceae Gordonia Proteiniphilum Azospirillum
1.0% 1.0%
Stappia
Synergistaceae
1.2% 1.4% 1.6% 1.7% 3.9%
Aminivibrio
61.8%
Comamonadaceae Azoarcus Others
B 31.2% 39.8%
1.6% 1.2% 1.1% 3.4% 1.4%
1.0%
15.1%
4.3%
Fig. 7. The relative abundance of major bacterial orders based 16S rDNA sequences in (A). the microbial fuel cell (MFC) with immobilized anode by agarose gel (i-MFC); (B) the control MFC without immobilized anode by agarose gel (c-MFC) fed by 10 g/ L of sodium acetate.
the concentration gradient of substances exchanging between bacteria and solutions. Similar to Blanchet et al. [25], the bacterial community of i-MFC demonstrated that more than 90% of the microbial community distributions was composed by Proteobacteria, Bacteroidetes, and Synergistetes. Geobacter as one of typical exoelectrogens has been widely reported in the acetate-fed MFC and the MFC performance can be related to the dominance of Geobacter species [25]. Therefore, higher percentage of Geobacter in the anode of i-MFC (62%) should result in better performance of the i-MFC than that of the c-MFC. Lentisphaerae has been reported in the MFC fed with glucose and pentachlorophenol [30]. The difference of the microbial communities between the i-MFC and c-MFC further confirmed that high acetate concentration negatively influenced the anode biofilm, resulting low performance of MFCs. Under high acetate concentrations, more electrons were consumed for bacterial growth instead of electricity generation, and some enzymes may bind with the outer membrane of cells, which hinders the electron transfer from exoelectrogens to the anode [4,6]. With the anode immobilization, exoelectrogens were surrounded with agar, which retarded the diffusion of high acetate concentration, thus improved the MFC performance. On the other hand, the growth of biofilm on the cathode could suppress the catalyst activity and increase the internal resistance [31]. Therefore, the deterioration of MFC performance with increase of substrate concentrations was attributable to the increase of cathodic biofilm, which reduced the catalytic activity in the cathode. The cathode deterioration with bacterial growth may be eliminated by incorporating antimicrobial, such as silver into the catalyst layer [31]. The advantages of the entrapment methods in the MFC have not been fully recognized so far. Compared with the c-MFC, inhibition
Fig. 8. The SEM images of the anode surfaces in the microbial fuel cell (MFC) with and without immobilized anode by agarose gel (i-MFC and c-MFC, respectively). (A) i-MFC 230; (B) i-MFC 3500; (C) c-MFC 10000.
of electron transfer and acetate adsorption were not observed in the i-MFC (Figs. 3 and 4). Yuan et al. [26] demonstrated that the maximum power density of 1947 mW/m2 in the MFC could be achieved via the anode entrapped exoelectrogens into carbon nanoparticles matrix. The results indicated that a suitable entrapment could enhance the performance of MFC instead of retarding electron transfer, waste removal, and substrate absorption. The
738
H. Luo et al. / Fuel 182 (2016) 732–739
Table 3 Comparison of different MFC with anode immobilization on power density and maximum voltage. Type of substrates
Concentration
Type of MFC
Immobilization method
Power density (mW/m2)
Maximum voltage (mV)
Reference
Lactate Acetate Acetate Simultaneous distillery wastewater Lactate Acetate
2.24 g/L (20 mM) 1.64 g/L (20 mM) 1.39 g/L (17 mM) 800–3600 mg/L (COD)
Dual-chamber Dual-chamber Single-chamber Single-chamber
129 35 – 36.8–72.9
386a 395a 520 544–598a
[10] [10] [32] [11]
2.02 g/L (18 mM) 1 g/L
Dual-chamber Single-chamber
34 1947
259 mV 500 mV
[9] [27]
Acetate
1 g/L 5 g/L 10 g/L 1 g/L 5 g/L 10 g/L
Single-chamber Single-chamber Single-chamber Single-chamber Single-chamber Single-chamber
Silica coatings Silica coatings Latex coatings Polyvinyl alcohol + powered active carbon gel Alginate/graphite granules Carbon nanoparticle and Teflon entrapment In-situ agarose gel immobilization
680 ± 55 610 ± 50 370 ± 40 610 ± 50 343 ± 30 240 ± 20
600 560 500 600 460 300
This study
Acetate
a
Unmodified anode
This study
Open circuit voltage.
lower CEs in the i-MFC were attributable to low exoelectrogen acclimation in the anode and the cathode deterioration. However, the CE values were still comparable to those reported previously [17]. The maximum power density and CE of the i-MFC can be further improved via other novel anode materials and optimal potential for the exoelectrogen acclimation in the anode [33,34]. Moreover, the long-term voltage output in the i-MFC fed with 10 g/L acetate was comparable with that reported previously [16] (500 vs. 480 mV). Nevertheless, the effect of immobilization process on the exoelectrogens should be further investigated using pure culture, such as Geobacter species. Some bacteria identified in different MFCs, such as Azoarcus sp. [30], Gordonia sp. [35], and Stappia sp. [36], have not been considered as exoelectrogens in the literature. Victivallis as fermentative bacterium has been identified in the anode of MFCs [37,38], while with 2 g/L acetate, the relative abundance of Victivallis was less than 5% [37]. The high abundance of Victivallis in the anode of c-MFC (15%) indicated that 10 g/L acetate was still suitable for the fermentative bacterium growth, resulting in competition with the exoelectrogens and decrease of electricity generation in the MFC. Further research is needed to study the relationship between these nonexolectrogens and Geobacter. In addition, agarose degradation and corresponding changes in anode biofilm during long-term operation for real wastewater treatment should be further investigated. 5. Conclusion In this study, an easily performed in-situ immobilization method was developed for anode of the MFC using agarose gel. The MFC with immobilized anode showed better performance than the control MFC. With the high concentration of acetate (10 g/L), the maximum voltage of the i-MFC was much higher than that of the c-MFC (500 vs. 300 mV). With the acetate concentrations of 5 and 10 g/L, the maximum power densities in the i-MFC were 78% and 54% higher than those in the c-MFC, respectively. The performance improvement was mainly attributable the much higher relative abundance of Geobacter in the i-MFC than in the c-MFC (62% vs 40%). The immobilization of anode biofilm using agarose gel should be a promising method to improve the performance of MFC to treat wastewater containing high concentration organics. Acknowledgments This work was partly supported by Grants from the National Natural Science Foundation of China (Nos. 51278500, 41471181, and 51308557), the Natural Science Foundation of Guangdong Province (S2013010012984, 2015A030313102), and the Program
of Guangzhou Science & Technology Department (Nos. 2015100 10125, 201604010043).
References [1] Cusick RD, Bryan B, Parker DS, Merrill MD, Mehanna M, Kiely PD, et al. Performance of a pilot-scale continuous flow microbial electrolysis cell fed winery wastewater. Appl Microbiol Biot 2011;89:2053–63. [2] Bakheet B, Yuan S, Li ZX, Wang HJ, Zuo JN, Komarneni S, et al. Electro-peroxone treatment of Orange II dye wastewater. Water Res 2013;47:6234–43. [3] Das R, Abd Hamid SB, Ali ME, Ismail AF, Annuar MSM, Ramakrishna S. Multifunctional carbon nanotubes in water treatment: the present, past and future. Desalination 2014;354:160–79. [4] Kim MS, Cha J, Kim DH. Enhancing factors of electricity generation in a microbial fuel cell using Geobacter sulfurreducens. J Microbiol Biotechn 2012;22:1395–400. [5] Park DH, Zeikus JG. Impact of electrode composition on electricity generation in a single-compartment fuel cell using Shewanella putrefaciens. Appl Microbiol Biot 2002;59:58–61. [6] Sharma Y, Li BK. The variation of power generation with organic substrates in single-chamber microbial fuel cells (SCMFCs). Bioresour Technol 2010;101:1844–50. [7] Yang XY, Tian G, Jiang N, Su BL. Immobilization technology: a sustainable solution for biofuel cell design. Energ Environ Sci 2012;5:5540–63. [8] Bai X, Gu HX, Chen W, Shi HC, Yang B, Huang X, et al. Immobilized Laccase on Activated Poly(Vinyl Alcohol) Microspheres for enzyme thermistor application. Appl Biochem Biotech 2014;173:1097–107. [9] Yong YC, Liao ZH, Sun JZ, Zheng T, Jiang RR, Song H. Enhancement of coulombic efficiency and salt tolerance in microbial fuel cells by graphite/alginate granules immobilization of Shewanella oneidensis MR-1. Process Biochem 2013;48:1947–51. [10] Luckarift HR, Sizemore SR, Roy J, Lau C, Gupta G, Atanassov P, et al. Standardized microbial fuel cell anodes of silica-immobilized Shewanella oneidensis. Chem Commun 2010;46:6048–50. [11] Lin CW, Wu CH, Huang WT, Tsai SL. Evaluation of different cell-immobilization strategies for simultaneous distillery wastewater treatment and electricity generation in microbial fuel cells. Fuel 2015;144:1–8. [12] Maaloum M, Pernodet N, Tinland B. Pore size of agarose gels by atomic force microscopy. Electrophoresis 1997;18:55–8. [13] Nussinovitch A, Nussinovitch M, Shapira R, Gershon Z. Influence of immobilization of bacteria, yeasts and fungal spores on the mechanical properties of agar and alginate gels. Food Hydrocolloid 1994;3:361–72. [14] Cassidy MB, Lee H, Trevors JT. Environmental applications of immobilized microbial cells: a review. J Ind Microbiol 1996;2:79–101. [15] Jiang S, Shi T, Zhan X, Long H, Xi S, Hu H, et al. High-performance all-solid-state flexible supercapacitors based on two-step activated carbon cloth. J Power Sources 2014;272:16–23. [16] Kiely PD, Rader G, Regan JM, Logan BE. Long-term cathode performance and the microbial communities that develop in microbial fuel cells fed different fermentation endproducts. Bioresour Technol 2011;102:361–6. [17] Liu H, Cheng SA, Logan BE. Production of electricity from acetate or butyrate using a single-chamber microbial fuel cell. Environ Sci Technol 2005;39:658–62. [18] Clesceri LS, Greenberg AE, Eaton AD. Standard methods for the examination of water and wastewater. Washington, DC: APHA; 1998. [19] Zhang XY, Cheng SA, Wang X, Huang X, Logan BE. Separator characteristics for increasing performance of microbial fuel cells. Environ Sci Technol 2009;43:8456–61. [20] Zhang CP, Li MC, Liu GL, Luo HP, Zhang RD. Pyridine degradation in the microbial fuel cells. J Hazard Mater 2009;172:465–71.
H. Luo et al. / Fuel 182 (2016) 732–739 [21] Zhang F, Liu J, Ivanov I, Hatzell MC, Yang WL, Ahn Y, et al. Reference and counter electrode positions affect electrochemical characterization of bioanodes in different bioelectrochemical systems. Biotechnol Bioeng 2014;111:1931–9. [22] Cheng SA, Wu JC. Air-cathode preparation with activated carbon as catalyst, PTFE as binder and nickel foam as current collector for microbial fuel cells. Bioelectrochemistry 2013;92:22–6. [23] Gao CY, Wang AJ, Wu WM, Yin YL, Zhao YG. Enrichment of anodic biofilm inoculated with anaerobic or aerobic sludge in single chambered air-cathode microbial fuel cells. Bioresour Technol 2014;167:124–32. [24] Xing YP, Liu C, Zhou XH, Shi HC. Label-free detection of kanamycin based on a G-quadruplex DNA aptamer-based fluorescent intercalator displacement assay. Sci Rep-UK 2015;5:8125. [25] Blanchet E, Desmond E, Erable B, Bridier A, Bouchez T, Bergel A. Comparison of synthetic medium and wastewater used as dilution medium to design scalable microbial anodes: application to food waste treatment. Bioresour Technol 2015;185:106–15. [26] Yuan Y, Zhou SG, Xu N, Zhuang L. Microorganism-immobilized carbon nanoparticle anode for microbial fuel cells based on direct electron transfer. Appl Microbiol Biot 2011;89:1629–35. [27] Guo W, Pi YQ, Song H, Tang W, Sun JH. Layer-by-layer assembled gold nanoparticles modified anode and its application in microbial fuel cells. Colloid Surface A 2012;415:105–11. [28] Popov AL, Kim JR, Dinsdale RM, Esteves SR, Guwy AJ, Premier GC. The effect of physico-chemically immobilized methylene blue and neutral red on the anode of microbial fuel cell. Biotechnol Bioproc E 2012;17:361–70. [29] Logan BE. Exoelectrogenic bacteria that power microbial fuel cells. Nat Rev Microbiol 2009;5:375–81.
739
[30] Kim JR, Jung SH, Regan JM, Logan BE. Electricity generation and microbial community analysis of alcohol powered microbial fuel cells. Bioresour Technol 2007;98:2568–77. [31] Liu WF, Cheng SA, Sun D, Huang HB, Chen J, Cen KF. Inhibition of microbial growth on air cathodes of single chamber microbial fuel cells by incorporating enrofloxacin into the catalyst layer. Biosens Bioelectron 2015;72:44–50. [32] Wagner R, Porter-Gill S, Logan B. Immobilization of anode-attached microbes in a microbial fuel cell. AMB Expr 2012;2:1–6. [33] Guo K, Soeriyadi AH, Feng HJ, Prevoteau A, Patil SA, Gooding JJ, et al. Heattreated stainless steel felt as scalable anode material for bioelectrochemical systems. Bioresour Technol 2015;195:46–50. [34] Torres CI, Krajmalnik-Brown R, Parameswaran P, Marcus AK, Wanger G, Gorby YA, et al. Selecting anode-respiring bacteria based on anode potential: phylogenetic, electrochemical, and microscopic characterization. Environ Sci Technol 2009;43:9519–24. [35] Huang LP, Chai XL, Quan X, Logan BE, Chen GH. Reductive dechlorination and mineralization of pentachlorophenol in biocathode microbial fuel cells. Bioresour Technol 2012;111:167–74. [36] Cristiani P, Franzetti A, Gandolfi I, Guerrini E, Bestetti G. Bacterial DGGE fingerprints of biofilms on electrodes of membraneless microbial fuel cells. Int Biodeter Biodegr 2013;84:211–9. [37] Liu LH, Tsyganova O, Lee DJ, Su A, Chang JS, Wang AJ, et al. Anodic biofilm in single-chamber microbial fuel cells cultivated under different temperatures. Int J Hydrogen Energ 2012;37:15792–800. [38] Wang SS, Huang LP, Gan LL, Quan X, Li N, Chen GH, et al. Combined effects of enrichment procedure and non-fermentable or fermentable co-substrate on performance and bacterial community for pentachlorophenol degradation in microbial fuel cells. Bioresour Technol 2012;120:120–6.
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel
Full Length Article
Effect of in-situ immobilized anode on performance of the microbial fuel cell with high concentration of sodium acetate Haiping Luo, Shuxian Yu, Guangli Liu ⇑, Renduo Zhang, Wenkai Teng Guangdong Provincial Key Laboratory of Environmental Pollution Control and Remediation Technology, School of Environmental Science and Engineering, Sun Yat-sen University, Guangzhou 510275, China
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
An efficient in-situ immobilization
Maximum Power density Improved
method was developed for anode of the MFC. The MFC with immobilized anode generated higher voltages than the control MFC. The MFC with immobilized anode produced higher power densities than the control MFC. The Geobacter abundance in the MFC with immobilized anode was higher. Matured anode
i n f o
Article history: Received 18 February 2016 Received in revised form 5 June 2016 Accepted 7 June 2016
Keywords: Microbial fuel cell In-situ immobilization High concentration of sodium acetate Microbial community
R
vs. Anode
a r t i c l e
54%
R
In-situ immobilized anode by agarose gel
Cathode
MFC with immobilized anode (fed by 10g L-1 acetate)
Anode
Cathode
Controlled MFC
a b s t r a c t To improve performance of the microbial fuel cell (MFC) with high concentration of organics, a procedure for in-situ immobilized anode with agarose gel was proposed in this study. The performance of power generation, electrochemical activity, and microbial community of the MFC with immobilized anode (i-MFC) was investigated using different concentrations (i.e., 1, 5, 10, and 20 g/L) of sodium acetate. The i-MFC could generate electricity within one hour after refreshing substrate. With 5 g/L acetate, the maximum voltages were 560 and 460 mV in the i-MFC and controlled MFC (c-MFC), respectively. With 10 g/L acetate, the maximum voltage of the i-MFC was much higher than that of the c-MFC (500 vs. 300 mV). With the acetate concentrations of 5 and 10 g/L, the maximum power densities in the i-MFC were 610 ± 50 and 370 ± 40 mW/m2, respectively, while the maximum power densities in the c-MFC were 343 ± 30 and 240 ± 20 mW/m2, respectively. Cyclic voltammetry measurements indicated that microbial activity of the i-MFC anode was higher than that of the c-MFC anode under the high acetate concentration (3.00 vs. 2.19 mA). The relative abundance of Geobacter on the i-MFC anode was much higher than that on the c-MFC anode (62% vs. 40%). The in-situ immobilization strategy provided an easily performed and efficient way to keep the activity of exoelectrogens, resulting in the improvement of MFC performance. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction ⇑ Corresponding author. E-mail address: [email protected] (G. Liu). http://dx.doi.org/10.1016/j.fuel.2016.06.032 0016-2361/Ó 2016 Elsevier Ltd. All rights reserved.
High concentrations of organics were often detected in many types of industrial wastewater, such as wastewater from process-
H. Luo et al. / Fuel 182 (2016) 732–739
ing plants of brewery, wine, and potato. Values of chemical oxygen demand (COD) of such wastewater change from 700 to 8000 mg/L [1,2]. Microbial fuel cell (MFC) was a kind of bioelectrochemical system (BES) that can generate electricity via organic biodegradation by electrochemically active bacteria [1]. Therefore, the MFC could be a promising approach to treat wastewater containing high COD concentrations, which should consume little energy or even generate electricity output (i.e., energy harvesting) [3]. However, the performance of MFC was affected by many factors, such as anode and cathode materials, reactor structure, substrates, exoelectrogens, and others [1]. High concentrations of substrates could inhibit activity of exoelectrogens in the MFC. It had been shown that in a dual-chamber MFC constructed with Geobacter sulfurreducens, the maximum power densities decrease from 16 to 13 mW/m3, and coulumbic efficiency (CE) from 46% to 23% with acetate concentrations from 5 to 20 mM [4]. In an MFC inoculated with Shewanella putrefaciens, the maximum current could not increase further as the initial concentration of lactate exceeded 200 mM [5]. Sharma and Li [6] demonstrated that the maximum voltages generated by a single-chamber air-cathode MFC inoculated with domestic wastewater increased with concentrations of acetate, ethanol, and glucose from 0.5 to 20 mM, but significantly decreased with concentrations of 20–35 mM. Therefore, under high organic concentrations, it should be necessary to improve the MFC performance for its potential applications in industrial wastewater treatment. Cell immobilization methods, such as adsorption, covalent binding, and entrapment, had been shown to reduce the environmental pressure (e.g., high salinity, toxicity) to microbes and to keep high cell densities in the biofuel cells [7,8]. The entrapment method, in which polymers or inorganics were utilized to form a framework with cells, had the advantages of good mechanical strength and stability, tunable porosity, and others. Using the entrapment method to immobilize exoelectrogens on the anode could stabilize the MFC performance under severe environmental conditions. Shewanella oneidensis MR-1 had been immobilized with graphite and alginate granules in the MFC, resulting in 0.8– 1.7 times higher CE than the control [9]. A method of one-step vapor deposition of silica had been used to immobilize S. oneidensis to form an aqueous sol-gel in the anode of MFC [10], which improved the maximum power density of MFC. A culture mixture in an air-cathode MFC had been immobilized using the polyvinyl alcohol and powdered active carbon to treat distillery wastewater [11]. The immobilized MFC could produce electricity of 72 mW/m2 with the influent COD concentration of 3600 mg/L. The entrapment methods for the immobilized anode of MFC above included the following steps: mixing exoelectrogens with polymers and attaching the mixture to the electrode. Such methods may cause some problems, such as damage of exoelectrogens due to violently mixing between bacteria and polymers, and delay of electricity generation because of the time needed for the attachment. Therefore, an in-situ immobilization strategy should be more attractive in the MFC. As a natural organic polymer (i.e., a linear polyscaccharide), which can be produced from marine red aglae [12], agarose has higher biocompatibility with lower cost than some of artificial polymers, such as polyvinyl alcohol. Compared with alignate and silica, agarose is more sensitive to temperature, thus easier to form gel for immobilization [13,14]. The objectives of this study were to develop an in-situ agarose immobilized procedure and to investigate the effect of in-situ immobilized exoelectrogens on the MFC performance under high substrate (acetate) concentration. The characteristics of immobilized and control MFCs, including power density, internal resistance, CE, and bacteria community, were determined and compared.
733
2. Materials and methods 2.1. Reactor construction and operation The base anode of MFC was made of carbon cloth (W0S 1002, CeTech., Taiwan, China). Characteristics of the carbon cloth included 125 g/m2, thickness of 360 lm, and effective area of 16 cm2 (L W = 8 cm 2 cm) [15]. The carbon cloth was treated with 450 °C in a muffle furnace for 30 min before use. Plain carbon cloth was rolled into a cylindrical shape (diameter of 1 cm and height of 2 cm) with a plastic mesh (grid: 3 mm 3 mm, thickness: 0.1 mm) as a separator to keep inner tunnels for bacterium growth. Platinum-catalyst air-cathode was made as previously described [16]. The perspex reactor had an effective volume of 28 mL with a diameter of 3 cm and a length of 4 cm. The anode and cathode were connected with an external resistance of 1000 X via titanium wire. The inoculum was from effluent of a grit chamber in a municipal wastewater treatment plant (Datansha, Guangzhou, China). During the startup stage, when the voltage was lower than 20 mV, the reactors were refreshed with a solution including 1 g/ L sodium acetate, 50 mM phosphate buffer, and vitamins [16]. After the reactors could generate the maximum voltage >500 mV for at least three cycles, the matured anodes in the MFCs were used for immobilization of exoelectrogens as follows. 2.2. Immobilization of exoelectrogens on the anode A schematic diagram of the procedure for anode immobilization was shown in Fig. 1. One gram of agar was dissolved into 100 mL phosphate buffer solution by heating at 100 °C. After the agar gel was cooled down to 40 °C, the matured anode was quickly put into the solution. A sterilized knife was used to cut the immobilized anode to a cubic size of 2.5 cm 1.5 cm (H d) after the agar gel was totally solidified at 30 °C. The immobilized anode was washed several times with distilled water and then refilled into the reactor (i-MFC). Reactors with the matured anodes without the immobilization procedure were used as the control (c-MFC). The reactors were operated with 1, 5, 10, and 20 g/L of sodium acetate as the substrate. All tests were duplicated at 30 ± 2 °C. 2.3. Analysis and calculations Voltages were measured across the external resistor every 15 min using a data electronic multimeter (DataTaker 85, Australia). The current in the circuit was calculated using Ohm’s law, which was normalized with the cathode projected surface area
Fig. 1. The schematic diagram of the procedure to the anode immobilization in the microbial fuel cell.
734
H. Luo et al. / Fuel 182 (2016) 732–739
(7 cm2) to estimate the current density. Polarization curve was obtained by changing external resistances from 100 to 3000 X [16]. CE was calculated based on the totally generated current and the COD removal [17]. COD was measured according to the standard method [18]. Biomass in the anode and cathode was determined based on the protein concentration using the Coomassie Brilliant Blue method [19]. The morphology of anode was examined using a scanning electron microscope (SEM) (Quanta 400, FEI, Netherlands) [20]. An electrochemical station (CHI660, Chenhua Co. Ltd, Shanghai, China) was used to carry out the cyclic voltammetry (CV) and linear sweep voltage (LSV) tests. For the CV test, a three-electrode (i.e., working electrode (the anode), counter electrode (the cathode), and reference electrode (Ag/AgCl)) arrangement was used with anode potentials from 0.8 to 0.2 V vs. Ag/AgCl at a scanning rate of 20 mV/s [21]. Each reactor was scanned for three cycles and the last cycle was plotted for discussion. For the LSV test, a used cathode (after at least three cycles of operation) was put into an abiotic MFC reactor filled with 50 mM phosphate buffer solution. The cathode was used as the working electrode, a platinum electrode as the counter electrode, and an Ag/AgCl electrode (closed to the working electrode) as the reference electrode. The LSV test was carried out with potentials from 0.5 to 0.2 V vs. Ag/AgCl at a scanning rate of 1 mV/s. Each cathode was tested for three times [22]. 2.4. Microbial community analysis After five operation cycles (about 1100 h) of the i-MFC and c-MFC fed with 10 g/L acetate, anode electrode samples were collected using a sterilized scissor. The deoxyribonucleic acid (DNA) kit (D5625-01, Omega Bio-Tek, USA) was used to extract the samples’ DNA. After examining the extracted DNA qualities with 1% agarose gel electrophoresis (100 V, 30 min), the universal primers 338F (ACTCCTACGGGAGGCAGCA) and 806R (GGACTACHVGGGTWTCTAAT) were used to amplify the V3-V4 region of 16S ribosomal DNA (rDNA) genes. The reverse primer with a 6-bp error-correcting barcode unique was assigned to each sample [23,24]. PCR amplification was carried out as follows: 3 min of denaturation at 95 °C, 27 cycles of 30 s at 95 °C, 30 s at 55 °C, 45 s at 72 °C, and then a final extension for 10 min at 72 °C. The PCR products were then purified and quantified. Sequencing was carried out on an Illumina Miseq platform by Majorbio (Shanghai, China). Pairs of reads were merged by FLASH and sequencing reads were assigned to each sample based on the unique barcode. QIIME software package (http://qiime.org) and UPARSE pipeline (http://drive5.com/upars/) were used to select high quality sequences and group them into operational taxonomic units (OTUs) at 97% similarity. The Silva database (http://www.arbsilva.de) was used to classify the sequences at the phylum and genus levels. Statistical indexes, including Chao 1 richness and Shannon diversity indexes, were calculated for each sample [25]. The nucleotide sequences reported in this study have been deposited in NCBI SRA with SRP065089. 3. Results 3.1. MFC performance with immobilized anode The i-MFC and c-MFC could generate electricity with 1.0, 5.0, and 10 g/L acetate as the substrate, respectively (Fig. 2). The i-MFC and c-MFC could generate electricity with 20 g/L acetate only in the first cycle operation. With 1 g/L acetate, the performance of i-MFC and c-MFC was similar. The maximum voltages reached 600 mV and the running time per cycle was about 22 h
Fig. 2. Electricity generation of the microbial fuel cell (MFC) with and without immobilized anode by agarose gel (i-MFC and c-MFC, respectively) with different concentrations of sodium acetate: (A) 1 g/L, (B) 5 g/L, and (C) 10 g/L.
in the MFCs. The operation time per cycle in both the MFCs was about 80 and 220 h with the acetate concentrations of 5 and 10 g/L, respectively. With 5 g/L acetate, the maximum voltages were 560 and 460 mV in the i-MFC and c-MFC, respectively. With 10 g/L acetate, the maximum voltages were 500 and 300 mV in the i-MFC and c-MFC, respectively. With the acetate concentrations of 1, 5, and 10 g/L, the maximum power densities in the i-MFC were 680 ± 55, 610 ± 50, and 370 ± 40 mW/m2, respectively, while the maximum power densities in the c-MFC were 610 ± 50, 343 ± 30, and 240 ± 20 mW/m2, respectively (Fig. 3). The reduction in power generation with the acetate concentrations was attributed to the increase of internal resistance in the systems. The internal resistances in the i-MFC and c-MFC increased from 255 ± 21 to 412 ± 35 X and from 291 ± 24 to 477 ± 39 X, respectively, with the acetate concentrations from 1 to 10 g/L. The COD removal reached about 90% within 96 h in the MFCs fed with 10 g/L acetate (Fig. 4), indicating that the exoelectrogens converted acetate into chemical energy first and then changed it
H. Luo et al. / Fuel 182 (2016) 732–739
735
for immobilizing exoelectrogens on the anode without inhibiting mass transfer between the substrate and biofilm. Values of CE decreased with increase of the acetate concentrations in the reactors. With 1 g/L acetate, the CE values in the i-MFC and c-MFC were almost the same (13.5 ± 2.1% vs. 14.3 ± 0.9%). With 10 g/L acetate, the CE values decreased to 10.7 ± 1.1% and 8.4 ± 2.0% in the i-MFC and c-MFC, respectively. 3.2. Electrochemical analysis on the MFCs With the acetate concentration of 1 g/L, the oxidation peaks in the forward and reverse scans were found at 0.31 V (vs Ag/AgCl) (3.43 mA) and 0.35 V (1.43 mA) for the i-MFC, respectively, and 0.37 V (2.83 mA) and 0.41 V (1.10 mA) for the c-MFC, respectively (Fig. 5A). With the acetate concentration of 10 g/L, the forward oxidation peaks were observed at 0.20 V (3.00 mA) and 0.32 V (2.19 mA) for the i-MFC and c-MFC, respectively. The reverse oxidation peaks were at 0.34 V (0.82 mA) and 0.39 V (0.89 mA) for the i-MFC and c-MFC, respectively (Fig. 5B). The potential of oxidation peak in the MFC was higher with 10 g/L than with 1 g/L acetate, which was likely attributed to the effect of substrate concentrations on the oxidation rate. The peak currents of the i-MFC were higher than those of the c-MFC, indicating that the agar immobilization enhanced the activity of exoelectrogens on the anode. The peak current of the i-MFC anode was lower with 10 g/L than with 1 g/L acetate, suggesting an adverse effect of high acetate concentration on the activity of immobilized exoelectrogens. As shown by the LSV results in Fig. 6, the cathode activity in the i-MFC was almost the same as that in the c-MFC. However, the
Fig. 3. (A) Power density-current curve and (B) voltage-current curve of the microbial fuel cell (MFC) with and without immobilized anode by agarose gel (i-MFC and c-MFC, respectively) fed by different concentrations of sodium acetate. i-MFC 1, i-MFC 5, i-MFC 10 represent i-MFCs fed by 1, 5, and 10 g/L of sodium acetate, respectively. c-MFC 1, c-MFC 5, c-MFC 10 represent c-MFCs fed by 1, 5, and 10 g/L of sodium acetate, respectively.
Fig. 4. Temporal change of COD removal in the microbial fuel cell (MFC) with and without immobilized anode by agarose gel (i-MFC and c-MFC, respectively) fed by 10 g/L of sodium acetate.
gradually to electricity. The COD removal between the i-MFC and c-MFC was not significantly different, while the total Coulombs were higher in the i-MFC than in the c-MFC (283 ± 20 C vs. 229 ± 27 C) (Fig. 4). These results suggested that agar was suitable
Fig. 5. The cyclic voltammetry (CV) curves of the anodes in the microbial fuel cell (MFC) with and without immobilized anode by agarose gel (i-MFC and c-MFC, respectively) fed by (A) 1.0 g/L and (B) 10 g/L of sodium acetate. i-MFC 1 and i-MFC 10 represent i-MFCs fed by 1 and 10 g/L of sodium acetate, respectively. c-MFC 1 and c-MFC 10 represent c-MFCs fed by 1 and 10 g/L of sodium acetate, respectively.
736
H. Luo et al. / Fuel 182 (2016) 732–739
Fig. 6. The linear sweep voltage (LSV) results of the cathodes in the microbial fuel cell (MFC) with and without immobilized anode by agarose gel (i-MFC and c-MFC, respectively) fed by different concentrations of sodium acetate. i-MFC 1 and i-MFC 10 represent i-MFCs fed by 1 and 10 g/L of sodium acetate, respectively. c-MFC 1 and c-MFC 10 represent c-MFCs fed by 1 and 10 g/L of sodium acetate, respectively.
was much less in the i-MFC than in the c-MFC (1.7% vs. 15%). Shannon diversity index was lower in the i-MFC than in the c-MFC (2.45 vs 2.69) (Table 1). Values of Chao 1 (i.e., the index of the species richness) were 187 and 200 in the i-MFC and c-MFC, respectively. The values of Shannon diversity index and Chao1 showed that the i-MFC had lower species diversity and richness than the c-MFC, indicating that the in-situ immobilization of the anode blocked the development of the microbial community on the anode. The biomass in the anode of c-MFC was much higher than that in the anode of i-MFC (5.3 ± 1.0 vs.2.9 ± 0.7 mg/g). The result was explained as follows. The estimated average pore size inside 1% agarose gel was smaller than 0.5 lm [12], thus it was more difficult for bacteria in the anolyte to move into the gel. On the contrary, the biomass in the anode of c-MFC increased with the operation, as the self-growth of the anode biofilm and the attachment of bacteria from the anolyte. Moreover, the SEM images showed that bacteria were not observed apparently to grow on the surface of agarose gel in the i-MFC (Fig. 8), suggesting that free bacteria in the anolyte might prefer to move onto the cathode instead of anode, which needs to be further studied.
4. Discussion activities of both i-MFC and c-MFC were greatly decreased with the increase of acetate concentrations. For example, at 0.2 V the cathode current decreased from 15.4 ± 0.2 to 1.5 ± 0.2 mA with the acetate concentrations from 1.0 to 10 g/L. Our LSV results of the cathodes in the i-MFC fed with 1 g/L acetate were comparable to other reports [22]. The cathodic biomass measurement showed that a thick biofilm layer was developed on the surface of the cathodes of i-MFC and c-MFC fed by 10 g/L acetate. The cathodic biomass in the i-MFC was almost the same as that in the c-MFC (6.1 ± 1.0 vs. 6.0 ± 1.0 mg/g). 3.3. Community analysis on the anodic biofilm The numbers of OTUs at 97% sequence similarity were 178 and 190 for the i-MFC and c-MFC, respectively (Table 1). The bacterial community distributions shown in Table 2 were based on the percentage of total sequences at the phylum level. The anode biofilm of i-MFC was mainly composed of Proteobacteria (68%), Bacteroidetes (16%), and Synergistetes (6.0%), while 49% of Proteobacteria, 15% of Bacteroidetes and 15% of Lentisphaerae were found in the anode biofilm of c-MFC. Geobacter was identified and dominated the microbial communities at the genus level in the MFCs (Fig. 7). The relative abundance of Geobacter in the i-MFC was much higher than that in the c-MFC (62% vs. 40%). Lentisphaerae in the anode
Table 1 Richness and diversity estimates of the anode bacterial communities in the i-MFC and c-MFC based on the OTUs at 97% similarity. Sample
Reads
OTU
Chao1
Shannon diversity index
i-MFC c-MFC
30489 34129
178 190
187 200
2.45 2.69
i-MFC and c-MFC represent MFCs with and without immobilized anode by agarose gel, respectively.
Steadily periodical electricity was generated in the i-MFC for more than 1100 h (5 cycles), indicating that the performance of i-MFC was kept stable with the agarose gel. Compared with other methods, such as polyvinyl alcohol with powered active carbon gel immobilization [11] and alginate with graphite granules immobilization [26], our immobilization strategy resulted in better performance of MFC (Table 3). With 5 g/L acetate, the maximum power density and voltage in the i-MFC reached 610 mW/m2 and 560 mV, which were higher than those reported in the literature. The in-situ agarose gel immobilization method also made the startup time much shorter. The i-MFC in our test could generate electricity within one hour after refreshing substrate. Compared to other immobilization methods, such as adsorption and binding, our in-situ immobilization had the shortest startup time as far as we known. With those immobilization methods, several to hundreds hours were needed for the exoelectrogens to locate in the anode surface and to form a stable biofilm. For example, the startup time of MFC with the carbon nanoparticles immobilized anode was about 40 h, which was used for extracting bacteria from the matured biofilm (1 year operation), acclimating (12 h), and entrapping bacteria [26]. A startup time of 120 h was taken in the MFC with layer-by-layer assembled gold nanoparticles modified anode [27]. The anode immobilization through adsorption of the electron shuttles, such as methylene blue and neutral red, required more than 6 days to startup [28]. The performance improvement in the i-MFC were explained as follows. The exoelectrogens had high biocompatibility with the natural polymer of agarose. The high relative abundance of Geobacter in the anode biofilm indicated that the main mechanism in the MFC should be a direct electron transfer from Geobacter to the anode with ‘‘nanowires” [29]. The direct electron transfer between Geobacter and the anode reduced the start-up time compared to other immobilization methods. In addition, the appropriate pore size of agarose gel can minimize
Table 2 Microbial community distributions in percentage of total sequences at the phylum level for the anodes of the microbial fuel cells (MFCs).
i-MFC c-MFC
Proteobacteria
Bacteroidetes
Synergistetes
Firmicutes
Lentisphaerae
Actinobacteria
Verrucomicrobia
Acidobacteria
Others
67.8 49.2
16.0 14.5
6.0 6.5
3.4 3.5
1.7 15.2
1.6 4.3
1.0 4.7
0.2 0.3
2.3 1.8
i-MFC and c-MFC represent MFCs with and without immobilized anode by agarose gel, respectively.
H. Luo et al. / Fuel 182 (2016) 732–739
A
737
Geobacter Aminiphilus Victivallis Spirochaeta
26.4%
Christensenellaceae Gordonia Proteiniphilum Azospirillum
1.0% 1.0%
Stappia
Synergistaceae
1.2% 1.4% 1.6% 1.7% 3.9%
Aminivibrio
61.8%
Comamonadaceae Azoarcus Others
B 31.2% 39.8%
1.6% 1.2% 1.1% 3.4% 1.4%
1.0%
15.1%
4.3%
Fig. 7. The relative abundance of major bacterial orders based 16S rDNA sequences in (A). the microbial fuel cell (MFC) with immobilized anode by agarose gel (i-MFC); (B) the control MFC without immobilized anode by agarose gel (c-MFC) fed by 10 g/ L of sodium acetate.
the concentration gradient of substances exchanging between bacteria and solutions. Similar to Blanchet et al. [25], the bacterial community of i-MFC demonstrated that more than 90% of the microbial community distributions was composed by Proteobacteria, Bacteroidetes, and Synergistetes. Geobacter as one of typical exoelectrogens has been widely reported in the acetate-fed MFC and the MFC performance can be related to the dominance of Geobacter species [25]. Therefore, higher percentage of Geobacter in the anode of i-MFC (62%) should result in better performance of the i-MFC than that of the c-MFC. Lentisphaerae has been reported in the MFC fed with glucose and pentachlorophenol [30]. The difference of the microbial communities between the i-MFC and c-MFC further confirmed that high acetate concentration negatively influenced the anode biofilm, resulting low performance of MFCs. Under high acetate concentrations, more electrons were consumed for bacterial growth instead of electricity generation, and some enzymes may bind with the outer membrane of cells, which hinders the electron transfer from exoelectrogens to the anode [4,6]. With the anode immobilization, exoelectrogens were surrounded with agar, which retarded the diffusion of high acetate concentration, thus improved the MFC performance. On the other hand, the growth of biofilm on the cathode could suppress the catalyst activity and increase the internal resistance [31]. Therefore, the deterioration of MFC performance with increase of substrate concentrations was attributable to the increase of cathodic biofilm, which reduced the catalytic activity in the cathode. The cathode deterioration with bacterial growth may be eliminated by incorporating antimicrobial, such as silver into the catalyst layer [31]. The advantages of the entrapment methods in the MFC have not been fully recognized so far. Compared with the c-MFC, inhibition
Fig. 8. The SEM images of the anode surfaces in the microbial fuel cell (MFC) with and without immobilized anode by agarose gel (i-MFC and c-MFC, respectively). (A) i-MFC 230; (B) i-MFC 3500; (C) c-MFC 10000.
of electron transfer and acetate adsorption were not observed in the i-MFC (Figs. 3 and 4). Yuan et al. [26] demonstrated that the maximum power density of 1947 mW/m2 in the MFC could be achieved via the anode entrapped exoelectrogens into carbon nanoparticles matrix. The results indicated that a suitable entrapment could enhance the performance of MFC instead of retarding electron transfer, waste removal, and substrate absorption. The
738
H. Luo et al. / Fuel 182 (2016) 732–739
Table 3 Comparison of different MFC with anode immobilization on power density and maximum voltage. Type of substrates
Concentration
Type of MFC
Immobilization method
Power density (mW/m2)
Maximum voltage (mV)
Reference
Lactate Acetate Acetate Simultaneous distillery wastewater Lactate Acetate
2.24 g/L (20 mM) 1.64 g/L (20 mM) 1.39 g/L (17 mM) 800–3600 mg/L (COD)
Dual-chamber Dual-chamber Single-chamber Single-chamber
129 35 – 36.8–72.9
386a 395a 520 544–598a
[10] [10] [32] [11]
2.02 g/L (18 mM) 1 g/L
Dual-chamber Single-chamber
34 1947
259 mV 500 mV
[9] [27]
Acetate
1 g/L 5 g/L 10 g/L 1 g/L 5 g/L 10 g/L
Single-chamber Single-chamber Single-chamber Single-chamber Single-chamber Single-chamber
Silica coatings Silica coatings Latex coatings Polyvinyl alcohol + powered active carbon gel Alginate/graphite granules Carbon nanoparticle and Teflon entrapment In-situ agarose gel immobilization
680 ± 55 610 ± 50 370 ± 40 610 ± 50 343 ± 30 240 ± 20
600 560 500 600 460 300
This study
Acetate
a
Unmodified anode
This study
Open circuit voltage.
lower CEs in the i-MFC were attributable to low exoelectrogen acclimation in the anode and the cathode deterioration. However, the CE values were still comparable to those reported previously [17]. The maximum power density and CE of the i-MFC can be further improved via other novel anode materials and optimal potential for the exoelectrogen acclimation in the anode [33,34]. Moreover, the long-term voltage output in the i-MFC fed with 10 g/L acetate was comparable with that reported previously [16] (500 vs. 480 mV). Nevertheless, the effect of immobilization process on the exoelectrogens should be further investigated using pure culture, such as Geobacter species. Some bacteria identified in different MFCs, such as Azoarcus sp. [30], Gordonia sp. [35], and Stappia sp. [36], have not been considered as exoelectrogens in the literature. Victivallis as fermentative bacterium has been identified in the anode of MFCs [37,38], while with 2 g/L acetate, the relative abundance of Victivallis was less than 5% [37]. The high abundance of Victivallis in the anode of c-MFC (15%) indicated that 10 g/L acetate was still suitable for the fermentative bacterium growth, resulting in competition with the exoelectrogens and decrease of electricity generation in the MFC. Further research is needed to study the relationship between these nonexolectrogens and Geobacter. In addition, agarose degradation and corresponding changes in anode biofilm during long-term operation for real wastewater treatment should be further investigated. 5. Conclusion In this study, an easily performed in-situ immobilization method was developed for anode of the MFC using agarose gel. The MFC with immobilized anode showed better performance than the control MFC. With the high concentration of acetate (10 g/L), the maximum voltage of the i-MFC was much higher than that of the c-MFC (500 vs. 300 mV). With the acetate concentrations of 5 and 10 g/L, the maximum power densities in the i-MFC were 78% and 54% higher than those in the c-MFC, respectively. The performance improvement was mainly attributable the much higher relative abundance of Geobacter in the i-MFC than in the c-MFC (62% vs 40%). The immobilization of anode biofilm using agarose gel should be a promising method to improve the performance of MFC to treat wastewater containing high concentration organics. Acknowledgments This work was partly supported by Grants from the National Natural Science Foundation of China (Nos. 51278500, 41471181, and 51308557), the Natural Science Foundation of Guangdong Province (S2013010012984, 2015A030313102), and the Program
of Guangzhou Science & Technology Department (Nos. 2015100 10125, 201604010043).
References [1] Cusick RD, Bryan B, Parker DS, Merrill MD, Mehanna M, Kiely PD, et al. Performance of a pilot-scale continuous flow microbial electrolysis cell fed winery wastewater. Appl Microbiol Biot 2011;89:2053–63. [2] Bakheet B, Yuan S, Li ZX, Wang HJ, Zuo JN, Komarneni S, et al. Electro-peroxone treatment of Orange II dye wastewater. Water Res 2013;47:6234–43. [3] Das R, Abd Hamid SB, Ali ME, Ismail AF, Annuar MSM, Ramakrishna S. Multifunctional carbon nanotubes in water treatment: the present, past and future. Desalination 2014;354:160–79. [4] Kim MS, Cha J, Kim DH. Enhancing factors of electricity generation in a microbial fuel cell using Geobacter sulfurreducens. J Microbiol Biotechn 2012;22:1395–400. [5] Park DH, Zeikus JG. Impact of electrode composition on electricity generation in a single-compartment fuel cell using Shewanella putrefaciens. Appl Microbiol Biot 2002;59:58–61. [6] Sharma Y, Li BK. The variation of power generation with organic substrates in single-chamber microbial fuel cells (SCMFCs). Bioresour Technol 2010;101:1844–50. [7] Yang XY, Tian G, Jiang N, Su BL. Immobilization technology: a sustainable solution for biofuel cell design. Energ Environ Sci 2012;5:5540–63. [8] Bai X, Gu HX, Chen W, Shi HC, Yang B, Huang X, et al. Immobilized Laccase on Activated Poly(Vinyl Alcohol) Microspheres for enzyme thermistor application. Appl Biochem Biotech 2014;173:1097–107. [9] Yong YC, Liao ZH, Sun JZ, Zheng T, Jiang RR, Song H. Enhancement of coulombic efficiency and salt tolerance in microbial fuel cells by graphite/alginate granules immobilization of Shewanella oneidensis MR-1. Process Biochem 2013;48:1947–51. [10] Luckarift HR, Sizemore SR, Roy J, Lau C, Gupta G, Atanassov P, et al. Standardized microbial fuel cell anodes of silica-immobilized Shewanella oneidensis. Chem Commun 2010;46:6048–50. [11] Lin CW, Wu CH, Huang WT, Tsai SL. Evaluation of different cell-immobilization strategies for simultaneous distillery wastewater treatment and electricity generation in microbial fuel cells. Fuel 2015;144:1–8. [12] Maaloum M, Pernodet N, Tinland B. Pore size of agarose gels by atomic force microscopy. Electrophoresis 1997;18:55–8. [13] Nussinovitch A, Nussinovitch M, Shapira R, Gershon Z. Influence of immobilization of bacteria, yeasts and fungal spores on the mechanical properties of agar and alginate gels. Food Hydrocolloid 1994;3:361–72. [14] Cassidy MB, Lee H, Trevors JT. Environmental applications of immobilized microbial cells: a review. J Ind Microbiol 1996;2:79–101. [15] Jiang S, Shi T, Zhan X, Long H, Xi S, Hu H, et al. High-performance all-solid-state flexible supercapacitors based on two-step activated carbon cloth. J Power Sources 2014;272:16–23. [16] Kiely PD, Rader G, Regan JM, Logan BE. Long-term cathode performance and the microbial communities that develop in microbial fuel cells fed different fermentation endproducts. Bioresour Technol 2011;102:361–6. [17] Liu H, Cheng SA, Logan BE. Production of electricity from acetate or butyrate using a single-chamber microbial fuel cell. Environ Sci Technol 2005;39:658–62. [18] Clesceri LS, Greenberg AE, Eaton AD. Standard methods for the examination of water and wastewater. Washington, DC: APHA; 1998. [19] Zhang XY, Cheng SA, Wang X, Huang X, Logan BE. Separator characteristics for increasing performance of microbial fuel cells. Environ Sci Technol 2009;43:8456–61. [20] Zhang CP, Li MC, Liu GL, Luo HP, Zhang RD. Pyridine degradation in the microbial fuel cells. J Hazard Mater 2009;172:465–71.
H. Luo et al. / Fuel 182 (2016) 732–739 [21] Zhang F, Liu J, Ivanov I, Hatzell MC, Yang WL, Ahn Y, et al. Reference and counter electrode positions affect electrochemical characterization of bioanodes in different bioelectrochemical systems. Biotechnol Bioeng 2014;111:1931–9. [22] Cheng SA, Wu JC. Air-cathode preparation with activated carbon as catalyst, PTFE as binder and nickel foam as current collector for microbial fuel cells. Bioelectrochemistry 2013;92:22–6. [23] Gao CY, Wang AJ, Wu WM, Yin YL, Zhao YG. Enrichment of anodic biofilm inoculated with anaerobic or aerobic sludge in single chambered air-cathode microbial fuel cells. Bioresour Technol 2014;167:124–32. [24] Xing YP, Liu C, Zhou XH, Shi HC. Label-free detection of kanamycin based on a G-quadruplex DNA aptamer-based fluorescent intercalator displacement assay. Sci Rep-UK 2015;5:8125. [25] Blanchet E, Desmond E, Erable B, Bridier A, Bouchez T, Bergel A. Comparison of synthetic medium and wastewater used as dilution medium to design scalable microbial anodes: application to food waste treatment. Bioresour Technol 2015;185:106–15. [26] Yuan Y, Zhou SG, Xu N, Zhuang L. Microorganism-immobilized carbon nanoparticle anode for microbial fuel cells based on direct electron transfer. Appl Microbiol Biot 2011;89:1629–35. [27] Guo W, Pi YQ, Song H, Tang W, Sun JH. Layer-by-layer assembled gold nanoparticles modified anode and its application in microbial fuel cells. Colloid Surface A 2012;415:105–11. [28] Popov AL, Kim JR, Dinsdale RM, Esteves SR, Guwy AJ, Premier GC. The effect of physico-chemically immobilized methylene blue and neutral red on the anode of microbial fuel cell. Biotechnol Bioproc E 2012;17:361–70. [29] Logan BE. Exoelectrogenic bacteria that power microbial fuel cells. Nat Rev Microbiol 2009;5:375–81.
739
[30] Kim JR, Jung SH, Regan JM, Logan BE. Electricity generation and microbial community analysis of alcohol powered microbial fuel cells. Bioresour Technol 2007;98:2568–77. [31] Liu WF, Cheng SA, Sun D, Huang HB, Chen J, Cen KF. Inhibition of microbial growth on air cathodes of single chamber microbial fuel cells by incorporating enrofloxacin into the catalyst layer. Biosens Bioelectron 2015;72:44–50. [32] Wagner R, Porter-Gill S, Logan B. Immobilization of anode-attached microbes in a microbial fuel cell. AMB Expr 2012;2:1–6. [33] Guo K, Soeriyadi AH, Feng HJ, Prevoteau A, Patil SA, Gooding JJ, et al. Heattreated stainless steel felt as scalable anode material for bioelectrochemical systems. Bioresour Technol 2015;195:46–50. [34] Torres CI, Krajmalnik-Brown R, Parameswaran P, Marcus AK, Wanger G, Gorby YA, et al. Selecting anode-respiring bacteria based on anode potential: phylogenetic, electrochemical, and microscopic characterization. Environ Sci Technol 2009;43:9519–24. [35] Huang LP, Chai XL, Quan X, Logan BE, Chen GH. Reductive dechlorination and mineralization of pentachlorophenol in biocathode microbial fuel cells. Bioresour Technol 2012;111:167–74. [36] Cristiani P, Franzetti A, Gandolfi I, Guerrini E, Bestetti G. Bacterial DGGE fingerprints of biofilms on electrodes of membraneless microbial fuel cells. Int Biodeter Biodegr 2013;84:211–9. [37] Liu LH, Tsyganova O, Lee DJ, Su A, Chang JS, Wang AJ, et al. Anodic biofilm in single-chamber microbial fuel cells cultivated under different temperatures. Int J Hydrogen Energ 2012;37:15792–800. [38] Wang SS, Huang LP, Gan LL, Quan X, Li N, Chen GH, et al. Combined effects of enrichment procedure and non-fermentable or fermentable co-substrate on performance and bacterial community for pentachlorophenol degradation in microbial fuel cells. Bioresour Technol 2012;120:120–6.