- Email: [email protected]
Controlled modification of carbon nanotubes and polyaniline on macroporous graphite felt for high-performance microbial fuel cell anode
Journal of Power Sources 283 (2015) 46e53
Contents lists available at ScienceDirect
Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour
Controlled modification of carbon nanotubes and polyaniline on macroporous graphite felt for high-performance microbial fuel cell anode Hui-Fang Cui a, *, Lin Du a, Peng-Bo Guo a, Bao Zhu a, John H.T. Luong b a
School of Life Sciences, Zhengzhou University, 100# Science Avenue, Zhengzhou 450001, PR China Innovative Chromatography Group, Irish Separation Science Cluster (ISSC), Department of Chemistry and Analytical, Biological Chemistry Research Facility (ABCRF), University College Cork, Cork, Ireland
b
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
Polyaniline (PANI) was electropolymerized on the surface of graphite felt (GF). Carbon nanotubes (CNTs) were electropheritically immobilized on the PANI/GF. CNT modification increased the effective surface area and electrical conductivity. A mediator-free dual-chamber microbial fuel cell was constructed from the anode. The MFC power output increased drastically with the CNT modification.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 5 December 2014 Received in revised form 2 February 2015 Accepted 16 February 2015 Available online 17 February 2015
Polyaniline (PANI) was electropolymerized on the surface of macroporous graphite felt (GF) followed by the electrophoretic deposition of carbon nanotubes (CNTs). The as-prepared macroporous material was characterized by scanning electron microscopy, water contact angle goniometry and electrochemical techniques. Upon the modification of PANI, a rough and nano-cilia containing film is coated on the surface of the graphite fibers, transforming the surface from hydrophobic to hydrophilic. The subsequent modification by CNTs increases the effective surface area and electrical conductivity of the resulting material. The power output of a mediator-free dual-chamber microbial fuel cell (MFC) constructed from the GF anode and an exoelectrogen Shewanella putrefaciens increases drastically with the CNT modification. The CNT/PANI/GF MFC attains an output voltage of 342 mV across an external resistor of 1.96 kU constant load, and a maximum power density of 257 mW m2, increased by 343% and 186%, compared to that of the pristine GF MFC and the PANI/GF MFC, respectively. More bacteria are attached on the CNT/ PANI/GF anode than on the PANI/GF anode during the working of the MFC. This strategy provides an easy scale-up, simple and controllable method for the preparation of high-performance and low-cost MFC anodes. © 2015 Elsevier B.V. All rights reserved.
Keywords: Microbial fuel cell Carbon nanotubes Mediator-free Polyaniline Graphite felt Shewanella putrefaciens
* Corresponding author. E-mail address: [email protected] (H.-F. Cui). http://dx.doi.org/10.1016/j.jpowsour.2015.02.088 0378-7753/© 2015 Elsevier B.V. All rights reserved.
H.-F. Cui et al. / Journal of Power Sources 283 (2015) 46e53
1. Introduction Microbial fuel cells (MFCs) directly convert chemical energy to electrical energy by harnessing the metabolism of exoelectrogens, microorganisms that can mediate extracellular electron transfer (EET) [1e4]. Ideally, the final electron acceptor of an exoelectrogen should be a solid conductor, i.e., the anode. Owing to the enzymatic diversity and various strain availability of exoelectrogens, MFCs are able to recover the energy of various organic molecules in human excrement [5], marine sediment [6e9], wastewater [10,11], etc. Apparently, MFCs have a great potential for a broad range of applications, such as wastewater treatment and bioremediation with concomitant energy production [10,11], electronic power sources for space shuttles, and biosensors [12e15]. However, low power output remains one of the main obstacles for their widespread practical applications. Reduced oxidoreductases of exoelectrogens at the extracellular membrane rely on three principal, but not mutually exclusive EET mechanisms in transferring their electrons to the exogenous final acceptor: direct electron transfer; indirect electron transfer via shuttling of excreted mediators; and through electrically conductive pili [16e18]. Therefore, development of anode materials, which can directly affect the bacterial attachment, electron transfer and substrate oxidation, is a key factor for the MFC performance and cost-effectiveness [19]. Among different electrode materials used as the anode of MFCs, carbon-based materials such as carbon cloth, carbon paper, especially graphite/carbon felt have been the mostly adopted because of their very low costs, excellent electrical conductivity, chemical stability, non-corrosiveness, and good biocompatibility [19e22]. In brief, graphite felt (GF), a three-dimensional (3D) porous carbon material, possesses a high specific surface area to interface with bacteria, resulting in high power density. However, the hydrophobic surface property of graphite compromises its ability for the bacterial attachment and electron transfer via shuttling of bacterial excreted mediators. Modification of the surface morphology and property might enhance the attachment and the viability of the bacteria, facilitating the EET mechanism, therefore enhancing the MFC performance. Carbon nanotubes (CNTs), a carbon based nanomaterial consisting of cylindrical graphene sheets [23], with unique electrical and structural properties, have been incorporated into MFC anodes
47
[24e26]. The incorporation of 20 wt.% CNT into a conductive polymer anode increases the power density of MFCs [24]. The MFC anode of vertically aligned, forest type multiwalled carbon nanotubes (MWCNTs) increases the anode surface-to-volume ratio, therefore improves the microbial ability to couple and transfer electrons to the anode [26]. As a conductive polymer, polyaniline (PANI) has been used for the MFC anode preparation due to its facile processibility, high hydrophilicity, electrical conductivity, biocompatibility and stability [21,24,27e30]. The positively charged PANI in neutral solutions electrostatically interacts with the negatively charged bacterial membrane [31]. Thus, a high bacterial density with high biodiversity has been observed on the surface of a carbon felt anode modified by PANI, corresponding to a 35% increase in the power output [28]. Of notice is also the decoration of PANI on graphene to promote bacterial adhesion and biofilm formation [21,30]. This paper unravels a new strategy for the electrophoretic adsorption of CNTs on PANI modified macroporous GF, serving as an anode in a mediator-free microbial fuel cell using exoelectrogen Shewanella putrefaciens. The surface morphology, surface hydrophilicity, and the electrochemical properties of the modified GF were characterized. The CNT modification was controllable with high reproducibility and the anode conductivity was not compromised by any nonconductive polymer binder. The effect of PANI and CNT modification on the MFC power output was investigated, and the CNT adsorption conditions were optimized. 2. Experimental 2.1. Materials and reagents Multiwalled carbon nanotubes (MWCNTs) (outer diameter < 8 nm, bundle length 10e30 mm, purity > 95%) were purchased from Chengdu Organic Chemicals (Chengdu, P.R. China). Macroporous graphite felt (GF) (pore size 200e300 mm, thickness: 4 mm, carbon content: 98%) was purchased from Beijing Sanye Carbon (China). Nafion 117 (thickness: 183 mm, density: 360 g m2, conductivity: 0.083 s cm1, exchange capacity: 0.89 meq g1) (DuPont, USA) was used as a proton exchange membrane (PEM) in the microbial fuel cells. MWCNTs were treated by refluxing in a mixture of sulfuric and nitric acids (3:1 v/v) at 25 mg mL1 for 4 h to enhance their hydrophilicity and thus aqueous dispersivity. Ultracentrifugation of the MWCNT slurry at 20,000 rpm followed by the pellet washing to neutral pH by deionized water was repeated several times. The resulting neutral MWCNT-water suspension of ~3.3 mg mL1 was centrifuged at 7000 rpm. The collected supernatant was used for modification of GF. All other chemicals were obtained from a local chemical agent and used without further purification. Deionized water (>18.2 MU cm1) obtained from a Millipore water system was used in all the experiments. 2.2. Instrumentations and analysis
Fig. 1. The picture of the dual-chamber MFC with two equal rectangular cells with a Nafion PEM sandwiched between the two chambers (left: cathodic chamber, right: anodic chamber).
Pristine or modified GF before and after MFC running, was probed by scanning electron microscopy (SEM, JEOL JSM 6700F, operated at 5 kV or 15 kV). Based on static water contact angle measurement, the surface property of the pristine or PANI modified GF was probed by an optical goniometer (JC2000C1, Shanghai Zhongchen Digital Technique Apparatus, China) using the sessile drop method in air atmosphere. The CNT loading amount at the modified GF was determined by measuring the absorbance at 249 nm of the CNT suspension before and after the CNT loading, by using a UVeVis spectrophotometer (UV-2450, Shimadzu Scientific Instrument, Japan). All electrochemical measurements were performed at room temperature (25 C) using a CHI-660C
48
H.-F. Cui et al. / Journal of Power Sources 283 (2015) 46e53
Scheme 1. Schematic procedure for controlled modification of carbon nanotubes (CNTs) and polyaniline (PANI) on graphite felt (GF).
(cycling between 0.2 and þ0.7 V) was used to effect electropolymerization. For comparison, PANI was electropolymerized onto GF by chronocoulometry with the potential stepping from þ0.8 V (20 s) to þ1.2 V (120 s). CNTs were electrophoretically migrated onto the GF or PANI/GF by applying an anode current of 9 mA (about 2.8 mA cm2) for 10, 20, 30 and 40 min from 25, 50, 75, 100, 125 and 150 mg mL1 of the CNT suspension (denoted as CNT-25, CNT-50, CNT-75, CNT-100, CNT-125 and CNT-150, respectively). For comparison, CNTs were also adsorbed onto the PANI/GF by immersing the PANI/GF into the CNT suspensions for 30 min. 2.4. Electrochemical characterization of the modified graphite felt Pristine or modified GF (square sheets, 1.0 cm 1.0 cm with 2 mm in thickness) was subject to electrochemical characterization. The CV experiment was performed in the phosphate buffer PBS (10 mM, pH 7.0) or 5 mM K3Fe(CN)6/1 M KCl as appropriate. 2.5. Bacterial growth
Fig. 2. Cyclic voltammograms (CVs) for the electropolymerization of PANI onto the surface of GF. The CVs are the 1st, 11th, 21st, 31st, 41st, 51st, 61st, 71st, 81st, 91st and 100th cycle.
electrochemical workstation (Shanghai Chenhua Instrument, CHI, China) in a three-electrode setup consisting of a working electrode (GF), a Pt coiled wire counter electrode and an Ag/AgCl/KCl (3 M) reference electrode. 2.3. Modification of graphite felt For the preparation of MFC anodes, the GF was cut to square sheets (1.8 cm 1.8 cm) and treated by soaking in 1 M HCl, 3% H2O2, and then deionized water (twice) at 50 C, each for 30 min. The astreated GF sheets were oven-dried, and ready for modification. PANI was modified on the GF sheets by electropolymerization from an aniline monomer solution (0.1 M in 0.5 M H2SO4). Cyclic voltammetry (CV) with a scan rate of 0.1 V s1 for 120 scan cycles
Shewanella putrefaciens obtained from Shandong University (P.R. China) was grown anaerobically at 30 C for 24 h in a medium containing 1.64 g L1 acetate, 3 g L1 yeast extract, 0.3 g L1 KH2PO4, 2.5 g L1 NaHCO3, 0.1 g L1 KCl, 0.1 g L1 MgCl2, and 0.1 g L1 CaCl2. The resulting bacteria culture (0.5 mL) at the exponential phase was used to inoculate the medium (45 mL) in the anodic chamber of the MFC for electricity harvesting. Nitrogen was bubbled into the suspension for 30 min to remove oxygen before the chamber was sealed. 2.6. MFC construction and operation A dual-chamber MFC with two equal rectangular cells of 63 mL in volume was constructed (as shown in Fig. 1). A Nafion PEM with an area of 2.8 cm 5 cm (14 cm2) was sandwiched between the two chambers. The pristine GF or GF modified with PANI and/or CNTs was employed as the MFC anode. A square piece of 2.5 cm 4.5 cm (11.25 cm2) plain carbon cloth (National Center for Biotechnology Education, UK) was used as the MFC cathode. The cathodic compartment (the left chamber in Fig. 1) was fed with 48 mL of 0.05 M K3[Fe(CN)6 supported by 0.1 M KCl. The anodic
Fig. 3. SEM images of (A) pristine GF, (B) PANI/GF, and (C) CNT-100/PANI/GF. The anodic current applied for the electrophoretic immobilization of CNTs lasted for 30 min.
H.-F. Cui et al. / Journal of Power Sources 283 (2015) 46e53
49
Fig. 4. The photos of the (A) pristine GF and (B) PANI/GF in the contact angle measurement.
compartment (the right chamber in Fig. 1) was fed with the S. putrefaciens inoculated medium. The constructed MFC was placed inside a temperature controlled incubator at 30 C with a continuous magnetic stirring in the anodic compartment. A constant load of 1.96 kU was connected between the anode and cathode to close the electrical circuit. The voltage across the external resistor was recorded using a USB digital multimeter (RBH8251, Beijing RUIBOHUA Control Technology). The polarization and power density curves were measured by varying the load resistance from 100 to 5000 U when the MFC output voltage reached plateau. The current and power outputs were normalized to the lateral geometric area of the GF anode. 2.7. Investigation of power producing mechanisms When the MFC output voltage reached plateau, the spent medium sample was collected and subject to a glassy carbon (GC) electrode as the working electrode for CV measurement. The CV measurement of a fresh medium was performed for comparison. 3. Results and discussion 3.1. Modification of graphite felt PANI was electropolymerized on GF by CV or chronocoulometry (Scheme 1). The former produced a uniform black PANI layer on the entire macroporous gray-black GF surface, whereas only the outer 1e1.5 mm thick layer of the GF was covered with PANI by the latter. Apparently, the steric hindrance impedes the diffusion of aniline to the inner site of the 4 mm thick GF. However, the CV method dynamically changes the applied potential, producing sufficient energy to overcome the steric hindrance. Hence, the CV method was used for all subsequent experiments. During the PANI electropolymerization process, the reduction peak at ~þ0.04 V and the oxidation peak at ~þ0.21 V increased monotonically and drastically (Fig. 2). After the modification of PANI, CNTs in the uniform CNT suspension were modified onto the GF by applying an anodic current, as shown in Scheme 1. Accordingly, the PANI/GF surface was protonized and positively charged, whereas the CNTs carrying carboxylic groups would be electrophoretically adsorbed onto the PANI/GF surface. 3.2. Surface morphology and hydrophilicity of the modified graphite felt The SEM images of the pristine and modified GF are shown in Fig. 3. The pristine graphite fiber surface is mesoporous, containing furrow along the fiber axial direction, produced during the electrospinning of polyacrylonitrile. The electrodeposition of PANI
introduced a rough layer onto the graphite fiber. The high magnification SEM micrograph (Inset in Fig. 3B) shows that the PANI film is rough and loose, containing many nano-cilia. The graphite fibers at the outer part of the GF was deposited with more PANI than those at the inner part, owing to the steric hindrance for aniline molecules from entering the inner site of the 4 mm thick GF. The CV method for PANI electrodeposition can produce energy to significantly overcome the steric hindrance. The electrophoretic adsorption of CNTs onto the surface of PANI/GF was confirmed by the SEM measurement. The CNTs irregularly lay on or are embedded-in the PANI layer of the PANI/GF. The anodic current applied at the PANI/ GF exerts an electrophoretic force to mobilize CNTs from the suspension to the PANI/GF as well as protonates and positively charges the PANI film for the immobilization of the negatively charged CNTs. The pristine GF exhibited a static water contact angle of 113e120 (Fig. 4), indicating its surface hydrophobicity. Upon the modification with PANI, the water drop in the measurement was completely and rapidly adsorbed into the PANI/GF. Such behavior attested the hydrophilic nature of the surface, resulting from the deposition of the PANI film. The capillary force enables the migration of water to both the macropores and mesopores of the PANI/ GF.
3.3. Electrochemical characterization of the modified graphite felt The CVs of the pristine or modified GF in neutral PBS are illustrated in Fig. 5A. The modification of PANI obviously increased the double layer current, suggesting an increase in the surface area. In addition, the PANI film modified on the graphite surface introduced electrochemical active groups, reflected by the appearance of new redox peaks. The redox peaks at the potentials ranging between 0.17 V and þ0.67 V are the characteristic redox peaks from PANI (originated from the redox transition between the leucoemeraldine and the polaronicemeraldine form) [30]. With the coating of CNT-25 on the PANI/GF surface, the double layer current is higher without any faradic current suppression, confirming the successful deposition of CNTs. As highly conductive and high specific surface area materials [32], the CNTs deposited on the PANI/GF appreciably increases the electrode surface area as expected. The CVs with K3Fe(CN)6 as a probe are shown in Fig. 5B. The probe exhibited well-defined quasi-reversible redox peaks at the GF electrodes without and with modifications. The effective surface areas (A) of the electrodes were estimated from the peak currents (Ip) at different scan rates according to the RandleseSevcik equation [33,34]. The Ip value at the GF electrode and the PANI/GF electrode at the scan rate of 0.02 V s1 was 11.6 mA and 11.2 mA; the redox peak potential difference (DEp) 231 mV and 252 mV; and the A value 25.7 cm2 and 24.6 cm2, respectively. PANI is
50
H.-F. Cui et al. / Journal of Power Sources 283 (2015) 46e53
Fig. 5. (A) CVs in 0.01 M PBS at (a) pristine GF, (b) PANI/GF (b), and (c) CNT-25/PANI/GF electrodes. Scan rate: 0.05 V s1. (B) CVs in 5 mM K3Fe(CN)6 supported by 1 M KCl at pristine GF (solid line), PANI/GF (dotted line), and CNT-25/PANI/GF (dashed line) electrodes. Scan rate: 0.02 V s1. The anodic current applied for the electrophoretic immobilization of CNT-25 lasted for 30 min.
semiconductive in neutral pH due to partially deprotonation, therefore its deposition results in an increase of the DEp value. The coating of a rough and loose PANI layer, as observed in the SEM micrographs, did not increase the Ip value and the surface area of the GF. Again, the result could be attributed to the semiconductive nature of PANI in neutral pH as well as the intrinsic roughness of the pristine graphite fiber surface. In contrast, the modification of CNT-25 on the surface of the PANI/GF increased the Ip value from 11.2 mA to 13.3 mA, corresponding to an increase in the effective surface area from 24.6 cm2 to 29.6 cm2 (increased 20%); whereas the DEp value decreased from 252 mV to 240 mV. The comparison experiments with the PANI/GF immersed in CNT suspension for
Fig. 6. (A) The time courses of the voltage output and (B) the power density curves of the MFCs constructed from the (a) pristine GF, (b) PANI/GF, (c) CNT-125/GF, and (d) CNT-125/PANI/GF anodes. Insets: The bar charts showing the values (mean ± S.E.) of (A) stable voltage and (B) maximum power density of the MFCs (n ¼ 4e7). *** represents that the p value is less than 0.001 in the t-test.
30 min showed that the effective surface area of the PANI/GF after immersion in 25 mg mL1 increased only 6% (from 20.6 cm2 to 21.9 cm2). The anodic current applied at the PANI/GF electrode during CNT modification not only drives the negatively charged CNTs to the electrode, but also promotes the protonation of the PANI film to attract the negatively charged CNTs. CNTs possess remarkable electrical and mechanical properties, high chemical stability, and high specific surface [32]. The Ip and A values increased during the course of the CNT electrophoretic adsorption and reached a plateau at 20e30 min of the adsorption for all the CNT concentrations used in this study (data not shown). Therefore,
H.-F. Cui et al. / Journal of Power Sources 283 (2015) 46e53
51
Fig. 7. SEM images of (A, B) the PANI/GF and (C, D) the CNT-125/PANI/GF anodes collected at the stage of stable MFC voltage output.
a period of 30 min for CNT electrophoretic adsorption was applied for all the subsequent MFC anodes. The experiment results also indicated that more CNTs were immobilized onto the PANI/GF electrode using higher CNT concentration, in the range below 125 mg mL1 (Fig. S-2, Supplementary materials). The surface area of the CNT-25/PANI/GF increased by ~20% (from 24.6 cm2 to 29.6 cm2, Fig. 4B), while that of a CNT-100/PANI/GF electrode increased by ~50% (data not shown), compared to the pristine GF. The CNT loading amount of CNT-25/PANI/GF, CNT-50/PANI/GF, CNT50/PANI/GF, CNT-100/PANI/GF, CNT-125/PANI/GF, and CNT-150/ PANI/GF was determined to be 57, 168, 258, 330, 412, and 430 mg cm3 (mean value, based on the geometric volume of the graphite felt), respectively. The loading amount measurement was very reproducible with the CNT concentration equal and less than 125 mg mL1, but showed a big variation for 150 mg mL1 CNT suspension (430 ± 70 mg cm3, mean ± S.E.; n ¼ 3). The CNTs may be likely to aggregate when the CNT concentration reaches a critical value, leading to saturation and variation in CNT loading amount. The anodic current applied at the PANI/GF electrode during CNT modification drives the negatively charged CNTs nearby the electrode migrating to the electrode. More CNTs locates nearby the electrode, more migrating to the electrode. The comparison experiment with the PANI/GF immersed in 125 mg mL1 CNTs for 30 min only loaded CNTs with the amount of 37 mg cm3, much less than its electrodeposition counterpart, confirming the significant role of the applied anodic current in CNT immobilization. 3.4. The MFC power output The performances of the GF with controllable deposition of PANI and CNTs as an MFC anode were evaluated. The PANI/GF felt anode (Fig. 6A and B, curve b) produced an MFC stable voltage of ~211 mV, and a maximum power density of ~80 mW m2, slightly higher than those from a pristine GF anode (curve a, ~188 mV and
49 mW m2, respectively). Similarly, with the PANI modification, the MFC stable voltage and maximum power density from the CNTs modified anodes (curve d) are slightly higher than those without PANI (curve c). Modification of the semiconductive PANI film on the GF surface increases hydrophilicity, and introduces positively charges to the electrode surface, favoring electrostatic interaction with the negatively charged bacterial membrane [31]. As mentioned previously, the coating of PANI offered no noticeable increase in the surface area although this polymer is hydrophilic and caries a positive charge. A series of experiments were conducted to examine the effect of CNT modification on the MFC anode performance, and optimize the CNT concentration required for surface modification. The MFC output voltage from the CNT/PANI/GF anodes increased monotonically with the CNT concentration and reached a plateau of ~342 mV at the CNT concentration of 125 mg mL1 (data not shown). The CNT immobilization is saturated at this CNT concentration, resulting in the saturation in the MFC power output. Fig. 6A illustrates the typical voltageetime curves of the MFCs constructed from the CNT-125/GF anode (without PANI) (curve c) and the CNT125/PANI/GF anode (curve d). The voltage output is stable for more than 80 h, suggesting that the biofilm and the MFC system are stable. The statistic results (n ¼ 4e7) for the stable voltages are shown in the Inset in Fig. 6A. The MFC output voltages from the CNT-125/PANI/GF anode (~348 mV in curve d, 342 mV in mean) and the CNT-125/GF anode (~336 mV in curve c, 340 mV in mean) are significantly higher than those from the PANI/GF anode and the pristine GF anode. Similarly, the MFC maximum power density from the CNT-125/PANI/GF anode (~308 mW m2 in curve d, 257 mW m2 in mean) and the CNT-125/GF anode (~227 mW m2 in curve c, 248 mW m2 in mean) significantly increased compared to those from the PANI/GF anode (~80 mW m2 in curve b, 90 mW m2 in mean) and the pristine GF anode (~49 mW m2 in curve a, 58 mW m2 in mean) (Fig. 6B). The MFC maximum power
52
H.-F. Cui et al. / Journal of Power Sources 283 (2015) 46e53
density from the CNT-125/PANI/GF anode increased by 343% and 186% (mean value), compared to that from the pristine GF anode and the PANI/GF anode, respectively. The power density value from the CNT-125/PANI/GF anode MFC (257 mW m2 in mean) is ~ 5 times higher than that from a 2-hydrox-1,4-naphthoquinone mediated MFC equipped with a CNT/PANI paste anode (42 mW m2) [24]. It should be noted that the MFC reported here is a mediator-free system, therefore, the set-up and operation are more cost-effective, simple and biocompatible compared to a mediator counterpart. Furthermore, the CNT coating greatly improved the attachment of the bacterial film and the microbial growth on the PANI/GF anode, attributed to high electrical conductivity, high effective surface area, and one-dimensional nanoscale morphology of the CNTs. Fig. 7 illustrates the SEM images of the MFC anode PANI/GC and CNT-125/PANI/GF collected at the stage of a stable MFC voltage output. The microbes were attached on the anode surfaces with both spherical and rod shapes. Their stem tails (positioned with the arrow in Fig. 7A) may be formed from extracellular materials and/or microbial flagella, and are attached on the anode surfaces. High magnification images show that the microbes were also attached to the anode surface through the numerous cilia (positioned with the arrow in Fig. 7B) under their spherical body. S. putrefaciens belongs to the a-Proteobacteria, a facultative anaerobe with the ability to reduce iron and manganese [35]. More importantly, this is an electricity-generating microorganism that can use electrodes as electron acceptors [36]. No obvious redox waves were observed in CV analysis of the MFC culture media (data not shown), excluding any indirect electron transfer via the shuttling of excreted mediators as the main EET mechanism. The incorporation of CNTs onto the surface of PANI/GF enhances both electrical conductivity and effective surface area, and introduces highly conductive CNT nano-cilia onto the electrode surface. These combined factors may facilitate the exoelectrogens to metabolize via- and attach onto-the MFC anode, favoring the EET of S. putrefaciens by direct electron transfer and/or through electrically conductive pili. 4. Conclusions PANI and MWCNTs were controllably, stably and binder-freely modified on the surface of macroporous GF by electrochemical techniques. With the modification of a rough, nano-cilia containing PANI film, the hydrophobic surface of GF becomes hydrophilic. Consequently, the surface of PANI/GF modified with CNTs possesses a high effective surface area and electrical conductivity, and its application as an anode in MFCs results in a significantly high power output. With an ease of scale-up, preparation simplicity and high performance, both CNTs modified GF and PANI/GF could well serve as the low-cost MFC anodes for large scale MFC systems. Acknowledgments This work was financially supported by the National Natural Science Foundation of China (NSFC 21345007), the National Recruitment Program of High-End Foreign Experts (China, GDW20124100167), the Henan Key Science and Technology Program (132102310042), and the Henan Open-up and Collaboration Program of Science and Technology (132106000070). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jpowsour.2015.02.088.
References [1] B.E. Logan, Exoelectrogenic bacteria that power microbial fuel cells, Nat. Rev. Microbiol. 7 (2009) 375e381. [2] D.R. Lovley, Bug juice: harvesting electricity with microorganisms, Nat. Rev. Microbiol. 4 (2006) 497e508. [3] B.E. Rittmann, R. Krajmalnik-Brown, R.U. Halden, Pre-genomic, genomic and post-genomic study of microbial communities involved in bioenergy, Nat. Rev. Microbiol. 6 (2008) 604e612. [4] X. Xie, G. Yu, N. Liu, Z. Bao, C.S. Criddle, Y. Cui, Grapheneesponges as highperformance low-cost anodes for microbial fuel cells, Energy Environ. Sci. 5 (2012) 6862e6866. [5] J.J. Konikoff, L.W. Reynolds, E.S. Harris, Electrical energy from biological systems, Aerosp. Med. 34 (1963) 1129e1133. [6] D.R. Bond, D.E. Holmes, L.M. Tender, D.R. Lovley, Electrode-reducing microorganisms that harvest energy from marine sediments, Science 295 (2002) 483e485. [7] M.E. Nielsen, C.E. Reimers, H.A. Stecher, Enhanced power from chambered benthic microbial fuel cells, Environ. Sci. Technol. 41 (2007) 7895e7900. [8] M.E. Nielsen, C.E. Reimers, H.K. White, S. Sharma, P.R. Girguis, Sustainable energy from deep ocean cold seeps, Energy Environ. Sci. 1 (2008) 584e593. [9] L.M. Tender, C.E. Reimers, H.A. Stecher, D.E. Holmes, D.R. Bond, D.A. Lowy, K. Pilobello, S.J. Fertig, D.R. Lovley, Harnessing microbially generated power on the seafloor, Nat. Biotechnol. 20 (2002) 821e825. [10] H. Liu, R. Ramnarayanan, B.E. Logan, Production of electricity during wastewater treatment using a single chamber microbial fuel cell, Environ. Sci. Technol. 38 (2004) 2281e2285. [11] B.E. Logan, Simultaneous wastewater treatment and biological electricity generation, Water Sci. Technol. 52 (2005) 31e37. [12] C. Donovan, A. Dewan, D. Heo, H. Beyenal, Batteryless, wireless sensor powered by a sediment microbial fuel cell, Environ. Sci. Technol. 42 (2008) 8591e8596. [13] L.M. Tender, S.A. Gray, E. Groveman, D.A. Lowy, P. Kauffman, J. Melhado, R.C. Tyce, D. Flynn, R. Petrecca, J. Dobarro, The first demonstration of a microbial fuel cell as a viable power supply: powering a meteorological buoy, J. Power Sources 179 (2008) 571e575. €tze, J.B. Hughes, A.M. Puzrin, Microbial fuel cell [14] J.M. Tront, J.D. Fortner, M. Plo biosensor for in situ assessment of microbial activity, Biosens. Bioelectron. 24 (2008) 586e590. [15] A. Shantaram, H. Beyenal, R.R.A. Veluchamy, Z. Lewandowski, Wireless sensors powered by microbial fuel cells, Environ. Sci. Technol. 39 (2005) 5037e5042. [16] B.H. Kim, H.J. Kim, M.S. Hyun, D.H. Park, Direct electrode reaction of Fe(III)reducing bacterium, Shewanella putrefaciens, J. Microbiol. Biotechnol. 9 (1999) 127e131. [17] D.R. Lovley, The microbe electric: conversion of organic matter to electricity, Curr. Opin. Biotechnol. 19 (2008) 564e571. €fte, W. Verstraete, Microbial phenazine production [18] K. Rabaey, N. Boon, M. Ho enhances electron transfer in biofuel cells, Environ. Sci. Technol. 39 (2005) 3401e3408. [19] M. Zhou, M. Chi, J. Luo, H. He, T. Jin, An overview of electrode materials in microbial fuel cells, J. Power Sources 196 (2011) 4427e4435. [20] Y.Y. Yu, H.L. Chen, Y.C. Yong, D.H. Kim, H. Song, Conductive artificial biofilm dramatically enhances bioelectricity production in Shewanella-inoculated microbial fuel cells, Chem. Commun. 47 (2011) 12825e12827. [21] J. Hou, Z. Liu, P. Zhang, A new method for fabrication of graphene/polyaniline nanocomplex modified microbial fuel cell anodes, J. Power Sources 224 (2013) 139e144. [22] K. Watanabe, Recent developments in microbial fuel cell technologies for sustainable bioenergy, J. Biosci. Bioeng. 106 (2008) 528e536. [23] S. Iijima, Helical microtubules of graphitic carbon, Nature 354 (1991) 56e58. [24] Y. Qiao, C.M. Li, S.J. Bao, Q.L. Bao, Carbon nanotube/polyaniline composite as anode material for microbial fuel cells, J. Power Sources 170 (2007) 79e84. [25] X. Xie, L. Hu, M. Pasta, G.F. Wells, D. Kong, C.S. Criddle, Y. Cui, Three-dimensional carbon nanotube-textile anode for high-performance microbial fuel cells, Nano Lett. 11 (2011) 291e296. [26] J.E. Mink, J.P. Rojas, B.E. Logan, M.M. Hussain, Vertically grown multiwalled carbon nanotube anode and nickel silicide integrated high performance microsized (1.25 mL) microbial fuel cell, Nano Lett. 12 (2012) 791e795. [27] K. Scott, G.A. Rimbu, K.P. Katuri, K.K. Prasad, I.M. Head, Application of modified carbon anodes in microbial fuel cells, Process Saf. Environ. Prot. 85 (2007) 481e488. [28] C. Li, L. Zhang, L. Ding, H. Ren, H. Cui, Effect of conductive polymers coated anode on the performance of microbial fuel cells (MFCs) and its biodiversity analysis, Biosens. Bioelectron. 26 (2011) 4169e4176. € der, M. Rosenbaum, F. Scholz, Fluorinated polyanilines as [29] J. Niessen, U. Schro superior materials for electrocatalytic anodes in bacterial fuel cells, Electrochem. Commun. 6 (2004) 571e575. [30] Y.-C. Yong, X.-C. Dong, M.B. Chan-Park, H. Song, P. Chen, Macroporous and monolithic anode based on polyaniline hybridized three-dimensional graphene for high-performance microbial fuel cells, ACS Nano 6 (2012) 2394e2400. [31] B. Lai, X. Tang, H. Li, Z. Du, X. Liu, Q. Zhang, Power production enhancement with a polyaniline modified anode in microbial fuel cells, Biosens. Bioelectron.
H.-F. Cui et al. / Journal of Power Sources 283 (2015) 46e53 28 (2011) 373e377. [32] H. Dai, E.W. Wong, C.M. Lieber, Probing electrical transport in nanomaterials: conductivity of individual carbon nanotubes, Science 272 (1996) 523e526. [33] A.J. Bard, L.R. Faulkner, Electrochemical Methods: Fundamentals and Applications, second ed., John Wiley & Sons, New York, 2001. [34] J.E.B. Randles, Kinetics of rapid electrode reactions, part 2drate constants and activation energies of electrode reactions, Trans. Faraday Soc. 48 (1952)
53
828e832. [35] D.A. Saffarini, T.J. DiChristina, D. Bermudes, K.H. Nealson, Anaerobic respiration of Shewanella putrefaciens requires both chromosomal and plasmidborne genes, FEMS Microbiol. Lett. 119 (1994) 271e277. [36] A.P. Borole, G. Reguera, B. Ringeisen, Z.W. Wang, Y. Feng, B.H. Kim, Electroactive biofilms: current status and future research needs, Energy Environ. Sci. 4 (2011) 4813e4834.
Contents lists available at ScienceDirect
Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour
Controlled modification of carbon nanotubes and polyaniline on macroporous graphite felt for high-performance microbial fuel cell anode Hui-Fang Cui a, *, Lin Du a, Peng-Bo Guo a, Bao Zhu a, John H.T. Luong b a
School of Life Sciences, Zhengzhou University, 100# Science Avenue, Zhengzhou 450001, PR China Innovative Chromatography Group, Irish Separation Science Cluster (ISSC), Department of Chemistry and Analytical, Biological Chemistry Research Facility (ABCRF), University College Cork, Cork, Ireland
b
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
Polyaniline (PANI) was electropolymerized on the surface of graphite felt (GF). Carbon nanotubes (CNTs) were electropheritically immobilized on the PANI/GF. CNT modification increased the effective surface area and electrical conductivity. A mediator-free dual-chamber microbial fuel cell was constructed from the anode. The MFC power output increased drastically with the CNT modification.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 5 December 2014 Received in revised form 2 February 2015 Accepted 16 February 2015 Available online 17 February 2015
Polyaniline (PANI) was electropolymerized on the surface of macroporous graphite felt (GF) followed by the electrophoretic deposition of carbon nanotubes (CNTs). The as-prepared macroporous material was characterized by scanning electron microscopy, water contact angle goniometry and electrochemical techniques. Upon the modification of PANI, a rough and nano-cilia containing film is coated on the surface of the graphite fibers, transforming the surface from hydrophobic to hydrophilic. The subsequent modification by CNTs increases the effective surface area and electrical conductivity of the resulting material. The power output of a mediator-free dual-chamber microbial fuel cell (MFC) constructed from the GF anode and an exoelectrogen Shewanella putrefaciens increases drastically with the CNT modification. The CNT/PANI/GF MFC attains an output voltage of 342 mV across an external resistor of 1.96 kU constant load, and a maximum power density of 257 mW m2, increased by 343% and 186%, compared to that of the pristine GF MFC and the PANI/GF MFC, respectively. More bacteria are attached on the CNT/ PANI/GF anode than on the PANI/GF anode during the working of the MFC. This strategy provides an easy scale-up, simple and controllable method for the preparation of high-performance and low-cost MFC anodes. © 2015 Elsevier B.V. All rights reserved.
Keywords: Microbial fuel cell Carbon nanotubes Mediator-free Polyaniline Graphite felt Shewanella putrefaciens
* Corresponding author. E-mail address: [email protected] (H.-F. Cui). http://dx.doi.org/10.1016/j.jpowsour.2015.02.088 0378-7753/© 2015 Elsevier B.V. All rights reserved.
H.-F. Cui et al. / Journal of Power Sources 283 (2015) 46e53
1. Introduction Microbial fuel cells (MFCs) directly convert chemical energy to electrical energy by harnessing the metabolism of exoelectrogens, microorganisms that can mediate extracellular electron transfer (EET) [1e4]. Ideally, the final electron acceptor of an exoelectrogen should be a solid conductor, i.e., the anode. Owing to the enzymatic diversity and various strain availability of exoelectrogens, MFCs are able to recover the energy of various organic molecules in human excrement [5], marine sediment [6e9], wastewater [10,11], etc. Apparently, MFCs have a great potential for a broad range of applications, such as wastewater treatment and bioremediation with concomitant energy production [10,11], electronic power sources for space shuttles, and biosensors [12e15]. However, low power output remains one of the main obstacles for their widespread practical applications. Reduced oxidoreductases of exoelectrogens at the extracellular membrane rely on three principal, but not mutually exclusive EET mechanisms in transferring their electrons to the exogenous final acceptor: direct electron transfer; indirect electron transfer via shuttling of excreted mediators; and through electrically conductive pili [16e18]. Therefore, development of anode materials, which can directly affect the bacterial attachment, electron transfer and substrate oxidation, is a key factor for the MFC performance and cost-effectiveness [19]. Among different electrode materials used as the anode of MFCs, carbon-based materials such as carbon cloth, carbon paper, especially graphite/carbon felt have been the mostly adopted because of their very low costs, excellent electrical conductivity, chemical stability, non-corrosiveness, and good biocompatibility [19e22]. In brief, graphite felt (GF), a three-dimensional (3D) porous carbon material, possesses a high specific surface area to interface with bacteria, resulting in high power density. However, the hydrophobic surface property of graphite compromises its ability for the bacterial attachment and electron transfer via shuttling of bacterial excreted mediators. Modification of the surface morphology and property might enhance the attachment and the viability of the bacteria, facilitating the EET mechanism, therefore enhancing the MFC performance. Carbon nanotubes (CNTs), a carbon based nanomaterial consisting of cylindrical graphene sheets [23], with unique electrical and structural properties, have been incorporated into MFC anodes
47
[24e26]. The incorporation of 20 wt.% CNT into a conductive polymer anode increases the power density of MFCs [24]. The MFC anode of vertically aligned, forest type multiwalled carbon nanotubes (MWCNTs) increases the anode surface-to-volume ratio, therefore improves the microbial ability to couple and transfer electrons to the anode [26]. As a conductive polymer, polyaniline (PANI) has been used for the MFC anode preparation due to its facile processibility, high hydrophilicity, electrical conductivity, biocompatibility and stability [21,24,27e30]. The positively charged PANI in neutral solutions electrostatically interacts with the negatively charged bacterial membrane [31]. Thus, a high bacterial density with high biodiversity has been observed on the surface of a carbon felt anode modified by PANI, corresponding to a 35% increase in the power output [28]. Of notice is also the decoration of PANI on graphene to promote bacterial adhesion and biofilm formation [21,30]. This paper unravels a new strategy for the electrophoretic adsorption of CNTs on PANI modified macroporous GF, serving as an anode in a mediator-free microbial fuel cell using exoelectrogen Shewanella putrefaciens. The surface morphology, surface hydrophilicity, and the electrochemical properties of the modified GF were characterized. The CNT modification was controllable with high reproducibility and the anode conductivity was not compromised by any nonconductive polymer binder. The effect of PANI and CNT modification on the MFC power output was investigated, and the CNT adsorption conditions were optimized. 2. Experimental 2.1. Materials and reagents Multiwalled carbon nanotubes (MWCNTs) (outer diameter < 8 nm, bundle length 10e30 mm, purity > 95%) were purchased from Chengdu Organic Chemicals (Chengdu, P.R. China). Macroporous graphite felt (GF) (pore size 200e300 mm, thickness: 4 mm, carbon content: 98%) was purchased from Beijing Sanye Carbon (China). Nafion 117 (thickness: 183 mm, density: 360 g m2, conductivity: 0.083 s cm1, exchange capacity: 0.89 meq g1) (DuPont, USA) was used as a proton exchange membrane (PEM) in the microbial fuel cells. MWCNTs were treated by refluxing in a mixture of sulfuric and nitric acids (3:1 v/v) at 25 mg mL1 for 4 h to enhance their hydrophilicity and thus aqueous dispersivity. Ultracentrifugation of the MWCNT slurry at 20,000 rpm followed by the pellet washing to neutral pH by deionized water was repeated several times. The resulting neutral MWCNT-water suspension of ~3.3 mg mL1 was centrifuged at 7000 rpm. The collected supernatant was used for modification of GF. All other chemicals were obtained from a local chemical agent and used without further purification. Deionized water (>18.2 MU cm1) obtained from a Millipore water system was used in all the experiments. 2.2. Instrumentations and analysis
Fig. 1. The picture of the dual-chamber MFC with two equal rectangular cells with a Nafion PEM sandwiched between the two chambers (left: cathodic chamber, right: anodic chamber).
Pristine or modified GF before and after MFC running, was probed by scanning electron microscopy (SEM, JEOL JSM 6700F, operated at 5 kV or 15 kV). Based on static water contact angle measurement, the surface property of the pristine or PANI modified GF was probed by an optical goniometer (JC2000C1, Shanghai Zhongchen Digital Technique Apparatus, China) using the sessile drop method in air atmosphere. The CNT loading amount at the modified GF was determined by measuring the absorbance at 249 nm of the CNT suspension before and after the CNT loading, by using a UVeVis spectrophotometer (UV-2450, Shimadzu Scientific Instrument, Japan). All electrochemical measurements were performed at room temperature (25 C) using a CHI-660C
48
H.-F. Cui et al. / Journal of Power Sources 283 (2015) 46e53
Scheme 1. Schematic procedure for controlled modification of carbon nanotubes (CNTs) and polyaniline (PANI) on graphite felt (GF).
(cycling between 0.2 and þ0.7 V) was used to effect electropolymerization. For comparison, PANI was electropolymerized onto GF by chronocoulometry with the potential stepping from þ0.8 V (20 s) to þ1.2 V (120 s). CNTs were electrophoretically migrated onto the GF or PANI/GF by applying an anode current of 9 mA (about 2.8 mA cm2) for 10, 20, 30 and 40 min from 25, 50, 75, 100, 125 and 150 mg mL1 of the CNT suspension (denoted as CNT-25, CNT-50, CNT-75, CNT-100, CNT-125 and CNT-150, respectively). For comparison, CNTs were also adsorbed onto the PANI/GF by immersing the PANI/GF into the CNT suspensions for 30 min. 2.4. Electrochemical characterization of the modified graphite felt Pristine or modified GF (square sheets, 1.0 cm 1.0 cm with 2 mm in thickness) was subject to electrochemical characterization. The CV experiment was performed in the phosphate buffer PBS (10 mM, pH 7.0) or 5 mM K3Fe(CN)6/1 M KCl as appropriate. 2.5. Bacterial growth
Fig. 2. Cyclic voltammograms (CVs) for the electropolymerization of PANI onto the surface of GF. The CVs are the 1st, 11th, 21st, 31st, 41st, 51st, 61st, 71st, 81st, 91st and 100th cycle.
electrochemical workstation (Shanghai Chenhua Instrument, CHI, China) in a three-electrode setup consisting of a working electrode (GF), a Pt coiled wire counter electrode and an Ag/AgCl/KCl (3 M) reference electrode. 2.3. Modification of graphite felt For the preparation of MFC anodes, the GF was cut to square sheets (1.8 cm 1.8 cm) and treated by soaking in 1 M HCl, 3% H2O2, and then deionized water (twice) at 50 C, each for 30 min. The astreated GF sheets were oven-dried, and ready for modification. PANI was modified on the GF sheets by electropolymerization from an aniline monomer solution (0.1 M in 0.5 M H2SO4). Cyclic voltammetry (CV) with a scan rate of 0.1 V s1 for 120 scan cycles
Shewanella putrefaciens obtained from Shandong University (P.R. China) was grown anaerobically at 30 C for 24 h in a medium containing 1.64 g L1 acetate, 3 g L1 yeast extract, 0.3 g L1 KH2PO4, 2.5 g L1 NaHCO3, 0.1 g L1 KCl, 0.1 g L1 MgCl2, and 0.1 g L1 CaCl2. The resulting bacteria culture (0.5 mL) at the exponential phase was used to inoculate the medium (45 mL) in the anodic chamber of the MFC for electricity harvesting. Nitrogen was bubbled into the suspension for 30 min to remove oxygen before the chamber was sealed. 2.6. MFC construction and operation A dual-chamber MFC with two equal rectangular cells of 63 mL in volume was constructed (as shown in Fig. 1). A Nafion PEM with an area of 2.8 cm 5 cm (14 cm2) was sandwiched between the two chambers. The pristine GF or GF modified with PANI and/or CNTs was employed as the MFC anode. A square piece of 2.5 cm 4.5 cm (11.25 cm2) plain carbon cloth (National Center for Biotechnology Education, UK) was used as the MFC cathode. The cathodic compartment (the left chamber in Fig. 1) was fed with 48 mL of 0.05 M K3[Fe(CN)6 supported by 0.1 M KCl. The anodic
Fig. 3. SEM images of (A) pristine GF, (B) PANI/GF, and (C) CNT-100/PANI/GF. The anodic current applied for the electrophoretic immobilization of CNTs lasted for 30 min.
H.-F. Cui et al. / Journal of Power Sources 283 (2015) 46e53
49
Fig. 4. The photos of the (A) pristine GF and (B) PANI/GF in the contact angle measurement.
compartment (the right chamber in Fig. 1) was fed with the S. putrefaciens inoculated medium. The constructed MFC was placed inside a temperature controlled incubator at 30 C with a continuous magnetic stirring in the anodic compartment. A constant load of 1.96 kU was connected between the anode and cathode to close the electrical circuit. The voltage across the external resistor was recorded using a USB digital multimeter (RBH8251, Beijing RUIBOHUA Control Technology). The polarization and power density curves were measured by varying the load resistance from 100 to 5000 U when the MFC output voltage reached plateau. The current and power outputs were normalized to the lateral geometric area of the GF anode. 2.7. Investigation of power producing mechanisms When the MFC output voltage reached plateau, the spent medium sample was collected and subject to a glassy carbon (GC) electrode as the working electrode for CV measurement. The CV measurement of a fresh medium was performed for comparison. 3. Results and discussion 3.1. Modification of graphite felt PANI was electropolymerized on GF by CV or chronocoulometry (Scheme 1). The former produced a uniform black PANI layer on the entire macroporous gray-black GF surface, whereas only the outer 1e1.5 mm thick layer of the GF was covered with PANI by the latter. Apparently, the steric hindrance impedes the diffusion of aniline to the inner site of the 4 mm thick GF. However, the CV method dynamically changes the applied potential, producing sufficient energy to overcome the steric hindrance. Hence, the CV method was used for all subsequent experiments. During the PANI electropolymerization process, the reduction peak at ~þ0.04 V and the oxidation peak at ~þ0.21 V increased monotonically and drastically (Fig. 2). After the modification of PANI, CNTs in the uniform CNT suspension were modified onto the GF by applying an anodic current, as shown in Scheme 1. Accordingly, the PANI/GF surface was protonized and positively charged, whereas the CNTs carrying carboxylic groups would be electrophoretically adsorbed onto the PANI/GF surface. 3.2. Surface morphology and hydrophilicity of the modified graphite felt The SEM images of the pristine and modified GF are shown in Fig. 3. The pristine graphite fiber surface is mesoporous, containing furrow along the fiber axial direction, produced during the electrospinning of polyacrylonitrile. The electrodeposition of PANI
introduced a rough layer onto the graphite fiber. The high magnification SEM micrograph (Inset in Fig. 3B) shows that the PANI film is rough and loose, containing many nano-cilia. The graphite fibers at the outer part of the GF was deposited with more PANI than those at the inner part, owing to the steric hindrance for aniline molecules from entering the inner site of the 4 mm thick GF. The CV method for PANI electrodeposition can produce energy to significantly overcome the steric hindrance. The electrophoretic adsorption of CNTs onto the surface of PANI/GF was confirmed by the SEM measurement. The CNTs irregularly lay on or are embedded-in the PANI layer of the PANI/GF. The anodic current applied at the PANI/ GF exerts an electrophoretic force to mobilize CNTs from the suspension to the PANI/GF as well as protonates and positively charges the PANI film for the immobilization of the negatively charged CNTs. The pristine GF exhibited a static water contact angle of 113e120 (Fig. 4), indicating its surface hydrophobicity. Upon the modification with PANI, the water drop in the measurement was completely and rapidly adsorbed into the PANI/GF. Such behavior attested the hydrophilic nature of the surface, resulting from the deposition of the PANI film. The capillary force enables the migration of water to both the macropores and mesopores of the PANI/ GF.
3.3. Electrochemical characterization of the modified graphite felt The CVs of the pristine or modified GF in neutral PBS are illustrated in Fig. 5A. The modification of PANI obviously increased the double layer current, suggesting an increase in the surface area. In addition, the PANI film modified on the graphite surface introduced electrochemical active groups, reflected by the appearance of new redox peaks. The redox peaks at the potentials ranging between 0.17 V and þ0.67 V are the characteristic redox peaks from PANI (originated from the redox transition between the leucoemeraldine and the polaronicemeraldine form) [30]. With the coating of CNT-25 on the PANI/GF surface, the double layer current is higher without any faradic current suppression, confirming the successful deposition of CNTs. As highly conductive and high specific surface area materials [32], the CNTs deposited on the PANI/GF appreciably increases the electrode surface area as expected. The CVs with K3Fe(CN)6 as a probe are shown in Fig. 5B. The probe exhibited well-defined quasi-reversible redox peaks at the GF electrodes without and with modifications. The effective surface areas (A) of the electrodes were estimated from the peak currents (Ip) at different scan rates according to the RandleseSevcik equation [33,34]. The Ip value at the GF electrode and the PANI/GF electrode at the scan rate of 0.02 V s1 was 11.6 mA and 11.2 mA; the redox peak potential difference (DEp) 231 mV and 252 mV; and the A value 25.7 cm2 and 24.6 cm2, respectively. PANI is
50
H.-F. Cui et al. / Journal of Power Sources 283 (2015) 46e53
Fig. 5. (A) CVs in 0.01 M PBS at (a) pristine GF, (b) PANI/GF (b), and (c) CNT-25/PANI/GF electrodes. Scan rate: 0.05 V s1. (B) CVs in 5 mM K3Fe(CN)6 supported by 1 M KCl at pristine GF (solid line), PANI/GF (dotted line), and CNT-25/PANI/GF (dashed line) electrodes. Scan rate: 0.02 V s1. The anodic current applied for the electrophoretic immobilization of CNT-25 lasted for 30 min.
semiconductive in neutral pH due to partially deprotonation, therefore its deposition results in an increase of the DEp value. The coating of a rough and loose PANI layer, as observed in the SEM micrographs, did not increase the Ip value and the surface area of the GF. Again, the result could be attributed to the semiconductive nature of PANI in neutral pH as well as the intrinsic roughness of the pristine graphite fiber surface. In contrast, the modification of CNT-25 on the surface of the PANI/GF increased the Ip value from 11.2 mA to 13.3 mA, corresponding to an increase in the effective surface area from 24.6 cm2 to 29.6 cm2 (increased 20%); whereas the DEp value decreased from 252 mV to 240 mV. The comparison experiments with the PANI/GF immersed in CNT suspension for
Fig. 6. (A) The time courses of the voltage output and (B) the power density curves of the MFCs constructed from the (a) pristine GF, (b) PANI/GF, (c) CNT-125/GF, and (d) CNT-125/PANI/GF anodes. Insets: The bar charts showing the values (mean ± S.E.) of (A) stable voltage and (B) maximum power density of the MFCs (n ¼ 4e7). *** represents that the p value is less than 0.001 in the t-test.
30 min showed that the effective surface area of the PANI/GF after immersion in 25 mg mL1 increased only 6% (from 20.6 cm2 to 21.9 cm2). The anodic current applied at the PANI/GF electrode during CNT modification not only drives the negatively charged CNTs to the electrode, but also promotes the protonation of the PANI film to attract the negatively charged CNTs. CNTs possess remarkable electrical and mechanical properties, high chemical stability, and high specific surface [32]. The Ip and A values increased during the course of the CNT electrophoretic adsorption and reached a plateau at 20e30 min of the adsorption for all the CNT concentrations used in this study (data not shown). Therefore,
H.-F. Cui et al. / Journal of Power Sources 283 (2015) 46e53
51
Fig. 7. SEM images of (A, B) the PANI/GF and (C, D) the CNT-125/PANI/GF anodes collected at the stage of stable MFC voltage output.
a period of 30 min for CNT electrophoretic adsorption was applied for all the subsequent MFC anodes. The experiment results also indicated that more CNTs were immobilized onto the PANI/GF electrode using higher CNT concentration, in the range below 125 mg mL1 (Fig. S-2, Supplementary materials). The surface area of the CNT-25/PANI/GF increased by ~20% (from 24.6 cm2 to 29.6 cm2, Fig. 4B), while that of a CNT-100/PANI/GF electrode increased by ~50% (data not shown), compared to the pristine GF. The CNT loading amount of CNT-25/PANI/GF, CNT-50/PANI/GF, CNT50/PANI/GF, CNT-100/PANI/GF, CNT-125/PANI/GF, and CNT-150/ PANI/GF was determined to be 57, 168, 258, 330, 412, and 430 mg cm3 (mean value, based on the geometric volume of the graphite felt), respectively. The loading amount measurement was very reproducible with the CNT concentration equal and less than 125 mg mL1, but showed a big variation for 150 mg mL1 CNT suspension (430 ± 70 mg cm3, mean ± S.E.; n ¼ 3). The CNTs may be likely to aggregate when the CNT concentration reaches a critical value, leading to saturation and variation in CNT loading amount. The anodic current applied at the PANI/GF electrode during CNT modification drives the negatively charged CNTs nearby the electrode migrating to the electrode. More CNTs locates nearby the electrode, more migrating to the electrode. The comparison experiment with the PANI/GF immersed in 125 mg mL1 CNTs for 30 min only loaded CNTs with the amount of 37 mg cm3, much less than its electrodeposition counterpart, confirming the significant role of the applied anodic current in CNT immobilization. 3.4. The MFC power output The performances of the GF with controllable deposition of PANI and CNTs as an MFC anode were evaluated. The PANI/GF felt anode (Fig. 6A and B, curve b) produced an MFC stable voltage of ~211 mV, and a maximum power density of ~80 mW m2, slightly higher than those from a pristine GF anode (curve a, ~188 mV and
49 mW m2, respectively). Similarly, with the PANI modification, the MFC stable voltage and maximum power density from the CNTs modified anodes (curve d) are slightly higher than those without PANI (curve c). Modification of the semiconductive PANI film on the GF surface increases hydrophilicity, and introduces positively charges to the electrode surface, favoring electrostatic interaction with the negatively charged bacterial membrane [31]. As mentioned previously, the coating of PANI offered no noticeable increase in the surface area although this polymer is hydrophilic and caries a positive charge. A series of experiments were conducted to examine the effect of CNT modification on the MFC anode performance, and optimize the CNT concentration required for surface modification. The MFC output voltage from the CNT/PANI/GF anodes increased monotonically with the CNT concentration and reached a plateau of ~342 mV at the CNT concentration of 125 mg mL1 (data not shown). The CNT immobilization is saturated at this CNT concentration, resulting in the saturation in the MFC power output. Fig. 6A illustrates the typical voltageetime curves of the MFCs constructed from the CNT-125/GF anode (without PANI) (curve c) and the CNT125/PANI/GF anode (curve d). The voltage output is stable for more than 80 h, suggesting that the biofilm and the MFC system are stable. The statistic results (n ¼ 4e7) for the stable voltages are shown in the Inset in Fig. 6A. The MFC output voltages from the CNT-125/PANI/GF anode (~348 mV in curve d, 342 mV in mean) and the CNT-125/GF anode (~336 mV in curve c, 340 mV in mean) are significantly higher than those from the PANI/GF anode and the pristine GF anode. Similarly, the MFC maximum power density from the CNT-125/PANI/GF anode (~308 mW m2 in curve d, 257 mW m2 in mean) and the CNT-125/GF anode (~227 mW m2 in curve c, 248 mW m2 in mean) significantly increased compared to those from the PANI/GF anode (~80 mW m2 in curve b, 90 mW m2 in mean) and the pristine GF anode (~49 mW m2 in curve a, 58 mW m2 in mean) (Fig. 6B). The MFC maximum power
52
H.-F. Cui et al. / Journal of Power Sources 283 (2015) 46e53
density from the CNT-125/PANI/GF anode increased by 343% and 186% (mean value), compared to that from the pristine GF anode and the PANI/GF anode, respectively. The power density value from the CNT-125/PANI/GF anode MFC (257 mW m2 in mean) is ~ 5 times higher than that from a 2-hydrox-1,4-naphthoquinone mediated MFC equipped with a CNT/PANI paste anode (42 mW m2) [24]. It should be noted that the MFC reported here is a mediator-free system, therefore, the set-up and operation are more cost-effective, simple and biocompatible compared to a mediator counterpart. Furthermore, the CNT coating greatly improved the attachment of the bacterial film and the microbial growth on the PANI/GF anode, attributed to high electrical conductivity, high effective surface area, and one-dimensional nanoscale morphology of the CNTs. Fig. 7 illustrates the SEM images of the MFC anode PANI/GC and CNT-125/PANI/GF collected at the stage of a stable MFC voltage output. The microbes were attached on the anode surfaces with both spherical and rod shapes. Their stem tails (positioned with the arrow in Fig. 7A) may be formed from extracellular materials and/or microbial flagella, and are attached on the anode surfaces. High magnification images show that the microbes were also attached to the anode surface through the numerous cilia (positioned with the arrow in Fig. 7B) under their spherical body. S. putrefaciens belongs to the a-Proteobacteria, a facultative anaerobe with the ability to reduce iron and manganese [35]. More importantly, this is an electricity-generating microorganism that can use electrodes as electron acceptors [36]. No obvious redox waves were observed in CV analysis of the MFC culture media (data not shown), excluding any indirect electron transfer via the shuttling of excreted mediators as the main EET mechanism. The incorporation of CNTs onto the surface of PANI/GF enhances both electrical conductivity and effective surface area, and introduces highly conductive CNT nano-cilia onto the electrode surface. These combined factors may facilitate the exoelectrogens to metabolize via- and attach onto-the MFC anode, favoring the EET of S. putrefaciens by direct electron transfer and/or through electrically conductive pili. 4. Conclusions PANI and MWCNTs were controllably, stably and binder-freely modified on the surface of macroporous GF by electrochemical techniques. With the modification of a rough, nano-cilia containing PANI film, the hydrophobic surface of GF becomes hydrophilic. Consequently, the surface of PANI/GF modified with CNTs possesses a high effective surface area and electrical conductivity, and its application as an anode in MFCs results in a significantly high power output. With an ease of scale-up, preparation simplicity and high performance, both CNTs modified GF and PANI/GF could well serve as the low-cost MFC anodes for large scale MFC systems. Acknowledgments This work was financially supported by the National Natural Science Foundation of China (NSFC 21345007), the National Recruitment Program of High-End Foreign Experts (China, GDW20124100167), the Henan Key Science and Technology Program (132102310042), and the Henan Open-up and Collaboration Program of Science and Technology (132106000070). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jpowsour.2015.02.088.
References [1] B.E. Logan, Exoelectrogenic bacteria that power microbial fuel cells, Nat. Rev. Microbiol. 7 (2009) 375e381. [2] D.R. Lovley, Bug juice: harvesting electricity with microorganisms, Nat. Rev. Microbiol. 4 (2006) 497e508. [3] B.E. Rittmann, R. Krajmalnik-Brown, R.U. Halden, Pre-genomic, genomic and post-genomic study of microbial communities involved in bioenergy, Nat. Rev. Microbiol. 6 (2008) 604e612. [4] X. Xie, G. Yu, N. Liu, Z. Bao, C.S. Criddle, Y. Cui, Grapheneesponges as highperformance low-cost anodes for microbial fuel cells, Energy Environ. Sci. 5 (2012) 6862e6866. [5] J.J. Konikoff, L.W. Reynolds, E.S. Harris, Electrical energy from biological systems, Aerosp. Med. 34 (1963) 1129e1133. [6] D.R. Bond, D.E. Holmes, L.M. Tender, D.R. Lovley, Electrode-reducing microorganisms that harvest energy from marine sediments, Science 295 (2002) 483e485. [7] M.E. Nielsen, C.E. Reimers, H.A. Stecher, Enhanced power from chambered benthic microbial fuel cells, Environ. Sci. Technol. 41 (2007) 7895e7900. [8] M.E. Nielsen, C.E. Reimers, H.K. White, S. Sharma, P.R. Girguis, Sustainable energy from deep ocean cold seeps, Energy Environ. Sci. 1 (2008) 584e593. [9] L.M. Tender, C.E. Reimers, H.A. Stecher, D.E. Holmes, D.R. Bond, D.A. Lowy, K. Pilobello, S.J. Fertig, D.R. Lovley, Harnessing microbially generated power on the seafloor, Nat. Biotechnol. 20 (2002) 821e825. [10] H. Liu, R. Ramnarayanan, B.E. Logan, Production of electricity during wastewater treatment using a single chamber microbial fuel cell, Environ. Sci. Technol. 38 (2004) 2281e2285. [11] B.E. Logan, Simultaneous wastewater treatment and biological electricity generation, Water Sci. Technol. 52 (2005) 31e37. [12] C. Donovan, A. Dewan, D. Heo, H. Beyenal, Batteryless, wireless sensor powered by a sediment microbial fuel cell, Environ. Sci. Technol. 42 (2008) 8591e8596. [13] L.M. Tender, S.A. Gray, E. Groveman, D.A. Lowy, P. Kauffman, J. Melhado, R.C. Tyce, D. Flynn, R. Petrecca, J. Dobarro, The first demonstration of a microbial fuel cell as a viable power supply: powering a meteorological buoy, J. Power Sources 179 (2008) 571e575. €tze, J.B. Hughes, A.M. Puzrin, Microbial fuel cell [14] J.M. Tront, J.D. Fortner, M. Plo biosensor for in situ assessment of microbial activity, Biosens. Bioelectron. 24 (2008) 586e590. [15] A. Shantaram, H. Beyenal, R.R.A. Veluchamy, Z. Lewandowski, Wireless sensors powered by microbial fuel cells, Environ. Sci. Technol. 39 (2005) 5037e5042. [16] B.H. Kim, H.J. Kim, M.S. Hyun, D.H. Park, Direct electrode reaction of Fe(III)reducing bacterium, Shewanella putrefaciens, J. Microbiol. Biotechnol. 9 (1999) 127e131. [17] D.R. Lovley, The microbe electric: conversion of organic matter to electricity, Curr. Opin. Biotechnol. 19 (2008) 564e571. €fte, W. Verstraete, Microbial phenazine production [18] K. Rabaey, N. Boon, M. Ho enhances electron transfer in biofuel cells, Environ. Sci. Technol. 39 (2005) 3401e3408. [19] M. Zhou, M. Chi, J. Luo, H. He, T. Jin, An overview of electrode materials in microbial fuel cells, J. Power Sources 196 (2011) 4427e4435. [20] Y.Y. Yu, H.L. Chen, Y.C. Yong, D.H. Kim, H. Song, Conductive artificial biofilm dramatically enhances bioelectricity production in Shewanella-inoculated microbial fuel cells, Chem. Commun. 47 (2011) 12825e12827. [21] J. Hou, Z. Liu, P. Zhang, A new method for fabrication of graphene/polyaniline nanocomplex modified microbial fuel cell anodes, J. Power Sources 224 (2013) 139e144. [22] K. Watanabe, Recent developments in microbial fuel cell technologies for sustainable bioenergy, J. Biosci. Bioeng. 106 (2008) 528e536. [23] S. Iijima, Helical microtubules of graphitic carbon, Nature 354 (1991) 56e58. [24] Y. Qiao, C.M. Li, S.J. Bao, Q.L. Bao, Carbon nanotube/polyaniline composite as anode material for microbial fuel cells, J. Power Sources 170 (2007) 79e84. [25] X. Xie, L. Hu, M. Pasta, G.F. Wells, D. Kong, C.S. Criddle, Y. Cui, Three-dimensional carbon nanotube-textile anode for high-performance microbial fuel cells, Nano Lett. 11 (2011) 291e296. [26] J.E. Mink, J.P. Rojas, B.E. Logan, M.M. Hussain, Vertically grown multiwalled carbon nanotube anode and nickel silicide integrated high performance microsized (1.25 mL) microbial fuel cell, Nano Lett. 12 (2012) 791e795. [27] K. Scott, G.A. Rimbu, K.P. Katuri, K.K. Prasad, I.M. Head, Application of modified carbon anodes in microbial fuel cells, Process Saf. Environ. Prot. 85 (2007) 481e488. [28] C. Li, L. Zhang, L. Ding, H. Ren, H. Cui, Effect of conductive polymers coated anode on the performance of microbial fuel cells (MFCs) and its biodiversity analysis, Biosens. Bioelectron. 26 (2011) 4169e4176. € der, M. Rosenbaum, F. Scholz, Fluorinated polyanilines as [29] J. Niessen, U. Schro superior materials for electrocatalytic anodes in bacterial fuel cells, Electrochem. Commun. 6 (2004) 571e575. [30] Y.-C. Yong, X.-C. Dong, M.B. Chan-Park, H. Song, P. Chen, Macroporous and monolithic anode based on polyaniline hybridized three-dimensional graphene for high-performance microbial fuel cells, ACS Nano 6 (2012) 2394e2400. [31] B. Lai, X. Tang, H. Li, Z. Du, X. Liu, Q. Zhang, Power production enhancement with a polyaniline modified anode in microbial fuel cells, Biosens. Bioelectron.
H.-F. Cui et al. / Journal of Power Sources 283 (2015) 46e53 28 (2011) 373e377. [32] H. Dai, E.W. Wong, C.M. Lieber, Probing electrical transport in nanomaterials: conductivity of individual carbon nanotubes, Science 272 (1996) 523e526. [33] A.J. Bard, L.R. Faulkner, Electrochemical Methods: Fundamentals and Applications, second ed., John Wiley & Sons, New York, 2001. [34] J.E.B. Randles, Kinetics of rapid electrode reactions, part 2drate constants and activation energies of electrode reactions, Trans. Faraday Soc. 48 (1952)
53
828e832. [35] D.A. Saffarini, T.J. DiChristina, D. Bermudes, K.H. Nealson, Anaerobic respiration of Shewanella putrefaciens requires both chromosomal and plasmidborne genes, FEMS Microbiol. Lett. 119 (1994) 271e277. [36] A.P. Borole, G. Reguera, B. Ringeisen, Z.W. Wang, Y. Feng, B.H. Kim, Electroactive biofilms: current status and future research needs, Energy Environ. Sci. 4 (2011) 4813e4834.