- Email: [email protected]
CNT nanocomposite anode in mediator-less microbial fuel cell
Biosensors and Bioelectronics 58 (2014) 75–80
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Enhanced electrical contact of microbes using Fe3O4/CNT nanocomposite anode in mediator-less microbial fuel cell In Ho Park a, Maria Christy b, Pil Kim a,b,c, Kee Suk Nahm a,b,c,n a
Specialized Graduate School of Hydrogen and Fuel Cell Engineering, Chonbuk National University, Jeonju 561-756, Republic of Korea R&D Education Center for Fuel Cell Materials & Systems, Chonbuk National University, Jeonju 561-756, Republic of Korea c School of Chemical Engineering, Chonbuk National University, Jeonju 561-756, Republic of Korea b
art ic l e i nf o
a b s t r a c t
Article history: Received 21 October 2013 Received in revised form 27 January 2014 Accepted 17 February 2014 Available online 26 February 2014
A novel Fe3O4/CNT nanocomposite was synthesized and employed for the modification of carbon paper anode in a mediator-less microbial fuel cell (MFC) to enhance its performance. The Fe3O4/CNT composite modified anodes with various Fe3O4 contents were investigated to find the optimum ratio of the nanocomposite for the best MFC performance. The Fe3O4/CNT modified anodes produced much higher power densities than unmodified carbon anode and the 30 wt% Fe3O4/CNT modified anode exhibited a maximum power density of 830 mW/m2. In the Fe3O4/CNT composite modified anode, Fe3O4 helps to attach the CNT on anode surface by its magnetic attraction and forms a multi layered network, whereas CNT offers a better nanostructure environment for bacterial growth and helps electron transfer from E.coli to electrode resulting in the increase in the current production with the catalytic activity of bacteria. The electrocatalytic behavior and all possible mechanism for their better performance are discussed in detail with the help of various structural and electrochemical techniques. & 2014 Elsevier B.V. All rights reserved.
Keywords: Microbial fuel cell Electro-catalysis Power production Electron transfer Nanocomposite Modified anode
1. Introduction Microbial fuel cell (MFC) is one of the promising green energy sources to produce electricity from organic wastes with the help of bacteria as catalyst. It converts the chemical energy in the organic wastes into electricity by bio-electrochemical reactions (Logan et al., 2006; Lovely, 2006). The MFC comprises a bio-anode where the microorganisms disintegrate the organic wastes into small molecules while producing electrons and protons. Thus produced electrons transfer from the microbes to anode and further travel through the external circuit to the cathode where they are used to reduce the electron acceptors. Meanwhile, protons migrate through a proton exchange membrane (PEM) from the anode to the cathode and complete the circuit. This is the basic working mechanism of MFC, which results in the generation of electrical power as well as removal of organic waste simultaneously (Delaney et al., 1984; Roller et al., 1984). In spite of the promising merits of MFC, the power generated from the fuel cell is still low to limit its applications in energy generation industries. The anode material is considered to be one of the key factors that influence the energy conversion in MFC along with cathodes and electrolytes. In the process of electricity generation from MFC, because the electron transfer from microbes to the anode is
n
Corresponding author. Tel.: þ 82 63 270 2311. E-mail address: [email protected] (K.S. Nahm).
http://dx.doi.org/10.1016/j.bios.2014.02.044 0956-5663 & 2014 Elsevier B.V. All rights reserved.
relatively slow process, the use of electron mediators is generally required for rapid electron transfer. Mediators such as 2-hydroxyl 1, 4-naphthoquinone (HNQ) or thionine have been generally used as electron shuttle to facilitate the electron transfer. These types of MFC are called mediated MFC where the mediators need to be replenished continuously to avoid toxicity and to reduce operational losses (Logan et al., 2006). Since Kim et al. demonstrated a possibility of mediatorless MFC with Shewanella species (Hyung Joo Kim et al., 2002), various metal reducing microorganisms such as Geobacter and Shewanella are reported to directly transfer electrons to the electrode in the electrocatalytic process of the mediatorless MFC (Chang et al., 2006; Zhang et al., 2006). These microbial organisms are known to effectively transfer electrons through nanowire without any mediators. The mediatorless MFC has provided stable power generation using various biomasses but has a low power output (Caccavo et al., 1994; You et al., 2006). The relatively low power density and poor energy conversion efficiency of the conventional MFC are resulted from sluggish electron transfer between bacteria and electrode. Modification of anode surface can reduce the charge transfer resistance and increase the electron transfer and there by improve the overall performance of MFC. The modification of anode with various nano-engineering techniques has been proposed to be promising and efficient to facilitate the electron transfer between microbes and anode (Scott et al., 2007). Bacterial attachment and the formation of a biofilm or network on the anode surface are essential for the efficient biological transfer of electrons in an MFC. Various modification strategies
76
I.H. Park et al. / Biosensors and Bioelectronics 58 (2014) 75–80
including nanomaterials and fabrication methods have been developed so far to increase the electron-accepting ability of the electrodes and to improve electron transfer and power density. Qiao et al. (2007) investigated carbon nanotube/polyaniline nanostructure composite as anode materials and reported. They observed the highest electrochemical activity and maximum power density at 20 wt% CNT composite anode in E.coli-based MFC. A unique nanostructured polyaniline/mesoporous TiO2 nanocomposite was also explored as an anode in E.coli MFC (Qiao et al., 2008), which reported that the best bio- and electrocatalytic performance was observed from the composite with 30 wt% PANI. Meanwhile, Feng et al. (2010) attained the maximum power density of 1303 mWm 2 from a polypyrrole/ anthraquinone-2,6-disulphonic disodium salt (PPy/AQDS)-modified carbon felt anode, which is 13 times larger power than that obtained from the MFC equipped with unmodified carbon anode in the presence of Shewanella species. Interestingly, Sun et al. (2010) fabricated a CNT/TP modified anode with enhanced electron transfer and power production. Peng et al. (2012) also showed enhanced behavior of anode by rolling the mixture of Fe3O4 and activated carbon (AC) on anode surface with a maximum power density of 80975 mW/m2. So, the anode surface structure and its ability to interact with bacteria are very important regardless of the electron transfer mechanism. Hence the modification of anode with nanomaterials is an efficient method to significantly enhance the generation of electricity in MFC. With this background, in this work, we investigated a novel Fe3O4/CNT hybrid nanocomposite modified carbon anode in mediatorless MFC. Carbon nanotubes (CNTs) have been served as a promising option for electrode surface modification materials due to their unique mechanical and electrical properties. First we coated mono-dispersed Fe3O4 nanoparticles on CNT by using a solvothermal synthetic method. Then the prepared Fe3O4/CNT nanocomposite was introduced to E.coli culture medium to form a network of E.coli and Fe3O4/CNT nanocomposite. The multilayered network was coated on the carbon paper electrode by applying magnetic force and used as MFC anode. The Fe3O4/CNT anode with various Fe3O4 content is also investigated to find the optimum ratio of the nanocomposite. The enhanced electron transfer and increased MFC performance with different Fe3O4 (in Fe3O4/CNT anode) content is explained with the help of polarization curves and other electrochemical characterizations.
in Luria–Bertani (LB) medium. The E.coli K12 colony in LB agar medium was extracted into a falcon tube containing 5 mL LB medium. Seed culture was carried out for 12 h at 160 rpm in a shaking incubator (37 1C). 1 mL of the seed culture medium was mixed with 20 mL of LB medium in a falcon tube followed by the main culture for 12 h at 160 rpm in a shaking incubator (37 1C) (Kim et al., 2002). 2.3. Assembly of anode To prepare Fe3O4/CNT modified anode, 0.05 g of Fe3O4/CNT nanocomposite was introduced to E.coli culture medium in LB broth, vortexed for uniform dispersion of particles in the solution and then immobilized bacteria on Fe3O4/CNT nanocomposite by a culturing process for 2–3 h while continuously shaking at 160 rpm in a constant temperature incubator. This process allows the formation of a sort of network between E.coli and Fe3O4/CNT nanocomposite in the solution. An electrode of carbon paper (TGPH-090, Toray International Inc.) wrapped around a magnetic slit (10 10 2 mm3) was placed in the same culture medium to form a multilayer of the E.coli and Fe3O4/CNT nanocomposite on the carbon paper electrode layer. In other words, a network of E.coli and Fe3O4/CNT nanocomposite is formed in a multilayered shape on the magnetic carbon electrode resulting in the formation of a conductive biocatalytic electrode. The conductive biocatalytic electrode was prepared with different Fe contents of Fe3O4/CNT nanocomposite (10, 30, 50 and 80 wt%) and fuel cell performance of the electrodes were comparatively investigated. For a clear understanding the anode assembly is clearly shown in Fig. 1 in the supplementary article. 2.4. Fabrication and evaluation of MFC The detailed MFC fabrication and MFC evaluation with its working conditions and details are explained in supplementary article. 2.5. Physical and electrochemical characterizations The physical and electrochemical characterization details and techniques used to study them are also explained in the supplementary article.
2. Materials and methods 3. Results and discussion 2.1. Synthesis of Fe3O4/CNT 3.1. Structural characterizations Fe3O4 was deposited on CNT using a solvothermal synthetic method reported detail in previous reports (Deng et al., 2012; Yoon et al., 2012). Stoichiometric amounts of Iron III acetylacetonate, Fe(C5H8O2)3 precursor were completely dissolved in triethylene glycol (C6H14O4) (sigma Aldrich) with agitation. They were initially mixed with 0.1 g of CNT (JEO Co.) to form CNT suspended solutions. (The amounts of precursor used for preparing different ratios of Fe3O4 in Fe3O4/CNT are given in Table 1 in supplementary article). The prepared suspension was then reduced with triethylene glycol at 280 1C to form Fe3O4/CNT nanocomposite. The suspension was agitated uniformly throughout the reduction reaction using magnetic stirrer. Fe3O4/CNT nanocomposite with four different Fe3O4 compositions (viz. 10, 30, 50, and 80 wt% ) were synthesized and studied in this work. 2.2. Microbe cultivation A gram-negative bacterium Escherichia coli (E.coli) K12 of 2 μm size was used in this experiment and the aerobic culture was done
Crystallographic studies were performed on the synthesized metal/CNT composites with different Fe3O4 contents by using X-ray diffraction (XRD) spectroscopy. XRD spectra for the synthesized samples are shown in Fig. 1(a) as a function of contents. The characteristic (2 2 0), (3 1 1), (4 0 0), (4 2 2), (5 1 1) and (4 4 0) peaks appeared on the spectra are due to Fe3O4 [JCPDS 01-1111], whereas the (0 0 2) peak corresponds to CNT (Sun and Liu, 2005). This confirms that the deposited material on CNT in this work is Fe3O4. With increasing Fe3O4 content in Fe3O4/CNT, the intensity of the Fe3O4 related peaks increases, while that of CNT decreases, as shown in Fig. 1(b), suggesting the formation of well aligned Fe3O4 over CNTs. The surface morphology of the synthesized Fe3O4/CNT nanocomposite was analyzed by TEM and FESEM measurements. The TEM images (Fig. 2(a)) clearly show the deposition of Fe3O4 nanoparticles over raw carbon nanotube surface and the density of the Fe3O4 nanoparticles increases with the increase of the Fe3O4 loading content. The diameter of CNT is seen to be 10–15 nm and
I.H. Park et al. / Biosensors and Bioelectronics 58 (2014) 75–80
77
Fig. 1. (a) XRD spectra of Fe3O4/CNT nanocomposite with Fe content; (i) CNT, (ii) 10 wt%, (iii) 30 wt%, (iv) 50 wt%, and (v) 80 wt%; and (b) the intensity of Fe content in CNT.
Fig. 2. (a) TEM images of Fe3O4/CNT nanocomposite with Fe content; (i) 10 wt%, (ii) 30 wt%, (iii) 50 wt%, and (iv) 80 wt%. (b) SEM images of (i) Pure CNT; (ii) Fe3O4/CNT nanocomposite and (iii) Fe3O4/CNT after experiment which shows E.coli adherence.
that of the deposited Fe3O4 nanoparticles varies between 3–10 nm as shown in Fig. 2(a). Although not shown in this paper, the FESEM images also presented that the number of Fe3O4 particles deposited on CNT surface increases with Fe3O4 loading content.
3.2. MFC performance Current versus time of mediator-less MFC fabricated with Fe3O4/CNT nanocomposite assembled anode with different Fe3O4
78
I.H. Park et al. / Biosensors and Bioelectronics 58 (2014) 75–80
contents are given in Fig. 3. The anode chamber is first filled with glucose and bacteria suspension and purged with nitrogen for 30 min to remove residual oxygen in the solution and then measured for current as a function of time (up to 120 h). It can be observed from all the samples that, with the increase of time, the current produced from the anode by catalytic activity of bacteria initially increases and reach a maximum value, and then decrease gradually. However, it is seen that the Fe3O4/CNT modified carbon anode generates very high current compared to unmodified bare carbon electrodes. Fig. 3 shows that the highest current of 0.32 mA is attained from 30 wt% Fe3O4/CNT composite for longer time duration (about 45 h). Although the concentration of glucose in the anode chamber and the number of bacteria on anode during the bio-reaction may cause the variation of current as a function of time, it is considered that 30 wt% Fe3O4 in Fe3O4/ CNT composite anode offers the best nanostructure environment for bacterial growth and electron transfer from E-coli to electrode, resulting in the production of higher current by the catalytic activity of bacteria. The power density and polarization curves of mediator-less MFC assembled with Fe3O4/CNT nanocomposite anode is given in Fig. 4(a) and (b), respectively. It is seen from Fig. 4(a) that Fe3O4/ CNT modified anodes show much higher power densities than unmodified carbon anode. It has been reported that exogenous mediators have always been necessary in biofuel cell experiments
Fig. 3. Time and current of mediator-less MFC using Fe3O4/CNT electrode with different Fe3O4 content.
with E.coli since without the mediators, MFC using E coli as anode biocatalyst generated only negligible amount of electricity (Park and Zeikus, 2000, 2003). In our experiment, however, the Fe3O4/ CNT modified anodes show higher power densities even though no mediator has been employed. Fig. 4(a) shows that the power density curve for the Fe3O4/CNT modified anode with 30 wt% Fe3O4 content exhibits a maximum power density of 830 mW/m2 at a current density of 1900 mA/m2. The power density of Fe3O4/ CNT modified anode initially increases from 10 wt% to 30 wt% Fe3O4 contents and shows the maximum value at 30 wt% Fe3O4. Meanwhile, the power decreases when increasing Fe3O4 contents from 30 wt% to 50 wt% and further to 80 wt%. The power density is calculated to be 440 mW/m2 for 10 wt%, 830 mW/m2 for 30 wt%, 540 mW/m2 for 50 wt% and 300 mW/m2 for 80 wt% of Fe3O4 in Fe3O4/CNT modified anode. In previous literatures (Qiao et al., 2007, 2008; Feng et al., 2010; Sun et al., 2010; Peng et al., 2012), the surface modification of anode has been reported to enhance the generation of electricity from MFC. Sun et al. constructed a three-dimensional MWNTs interwoven network structure on anode by modifying carbon paper (TP) electrode utilizing a layer by layer (LBL) assemble technique and investigated the anode characteristics (Xie et al., 2011). The electron transfer ability of anode was significantly enhanced due to the interfacial resistance reduction resulting from the LBL self-assembled CNTs modification. It was also observed that the existence of the three dimensional MWNT multilayer increased a large accessible surface area for immobilization of microorganisms which decreased the interfacial charge transfer resistance, resulting in the production of a higher power density with 20% enhancement comparing to the unmodified bare carbon anode. The enhanced anode performance could be influenced by several parameters, such as electrode surface, electrochemical characteristics of electrode, and the mechanism of electron transfer, etc. Few other reports (Peng et al., 2011; Xie et al., 2011, 2012) on the surface modification of anode, mostly using CNTs, also identified the increase of electrode surface area which is useful for the immobilization of microbes and the reduction of electron transfer resistance in thus prepared anode, resulting in the increase of power generation from the MFC. Our observation above presents that all the Fe3O4/CNT modified anodes show much higher power densities than unmodified bare carbon anode and the power density produced from 10 wt% Fe3O4/CNT anode is about two times lower than that from 30 wt% Fe3O4/CNT anode. This presents that the power density obtained from the MFC is significantly influenced on the Fe3O4 weight
Fig. 4. (a) Power output of mediator-less MFC using Fe3O4/CNT electrode with Fe3O4 content (resistance of 100 KΩ–300 Ω); and (b) polarization curves of mediator-less MFC using Fe3O4/CNT electrode with Fe3O4 content (resistance of 100 KΩ–300 Ω).
I.H. Park et al. / Biosensors and Bioelectronics 58 (2014) 75–80
percent in the Fe3O4/CNT composite electrode. On contrary, however, the addition of further Fe3O4 more than 30 wt% has considerably decreased the power densities to be 540 mW/m2 and 300 mW/m2 for 50 wt% and 80 wt% Fe3O4 in Fe3O4/CNT modified anode, respectively. BET technique was employed to measure typical surface area for the Fe3O4/CNT nanocomposite, which are also listed in Table 1 in supplementary article. It is seen from the Table that the 10 wt% Fe3O4 in Fe3O4/CNT anode has the highest surface area and the surface area of the Fe3O4/CNT nano-composite is decreased with the increase of Fe3O4 content in Fe3O4/CNT nanocomposites. This could possibly be attributed to the lower specific surface area of Fe3O4 than CNT. The deposition of Fe3O4 on CNT could reduce the specific surface area of the composite. In the previous TEM and SEM studies on the surface morphology of the synthesized Fe3O4/CNT nanocomposite, it was also seen that the increase of the Fe3O4 loading content increased the number of 3–10 nm Fe3O4 nanoparticles deposited over carbon nanotube surface. This indicates that the increase of the Fe3O4 loading content over CNT results in the decrease of the surface area of Fe3O4/CNT nanocomposite. Differently from the trend of the surface area, however, Fig. 4(a) showed that, initially, with increasing Fe3O4 content from 10 wt% to 30 wt%, the power density of Fe3O4/ CNT modified anode increases, though the surface area decreases. And then both the surface area and the power density decrease when increasing Fe3O4 contents higher than 30 wt% to be 50 wt% and to be 80 wt%. Our observation of the highest power generation from 30 wt% can cause a controversy to the results observed in previous report where the power density produced increases proportional to the surface area (Sun et al., 2010). This means that in our work Fe3O4 deposited on the CNT surface does a significant role to suppress and create a diffusion problem of electron donor and to interfere the electron transfer between the donor and acceptor. Qiao et al. (2008) determined an optimum condition to generate the highest power in their anode modification studies. They explored catalytic activity of nanostructured polyaniline (PANI)/ TiO2 nanocomposite as an anode modification material in E.coli MFC and observed the best bio-and electrocatalytic performance with 30 wt% PANI modified TiO2, although specific surface area of the samples decreases with increasing PANI content in the composites. Although the mechanism of direct electron transfer between the nanocomposite and E.coli observed in their experiment was not clearly proposed, they proposed that the greatest electrocatalytic behavior of the E.coli with 30 wt% PANI/TiO2
79
electrode could be ascribed to optimal surface structure for good mass transfer and low IR drops in the inner pores of anode. The decrease in power production with increase in Fe3O4 content might be explained with respect to the electron transfer between the microbes and the anodes. Since 10 wt% Fe3O4 in Fe3O4/CNT has the highest surface area, it is expected to deliver maximum power density. On contraire, however, 30 wt% Fe3O4 in Fe3O4/CNT anode is optimized to be the best ratio for maximum power production. This phenomenon can be viewed in a different way. The main role of Fe3O4 here is to attach the high surface CNT nanoparticles on the carbon anode surface with the magnetic attraction between Fe3O4 and the magnet. Meanwhile, CNT in Fe3O4/CNT composite contributes to help electron transfer from microbes to anode electrode owing to its high electron conductive property (Sun et al., 2010). So that, the microbes attached on the CNT surface transfers electrons through the multi-channel CNT networks onto the anode surface. In case of 10 wt% Fe3O4 in Fe3O4/ CNT, there is large surface area on the composite anode which can accommodate the adsorption of large number of microbes/E.coli on CNT surface. Owing to less Fe3O4 content in the 10% sample, however, only few CNT are attracted to be coated on the carbon anode. For the case of 30 wt% Fe3O4 in Fe3O4/CNT anodes, on the other hand, the increased amount of Fe3O4 content affords the build-up of multi-layered CNT network over the carbon anode surface, while CNTs provides enough surface area necessary for the adsorption of large microbe population. This large population of microbes engages in rapid and high electron transfer through the high surface multi-channeled CNT networks, resulting in high power production. However, for the other two cases of 50 and 80 wt% Fe3O4 in Fe3O4/CNT anodes, the CNT surface is highly populated with Fe3O4, which prevents bacterial adherence: the microbes and CNT interactions are interfered by high Fe3O4 concentration, resulting in low electron transfer. This in turn decreases the power production in the MFC. This indicates that the catalytic performance of the Fe3O4/CNT nanocomposite modified anode in the MFC can be optimized by adjusting the Fe3O4 percentage in the composite, and the composite with 30 wt% Fe3O4 gives the highest bio-and electrocatalytic performance. The polarization curves (Fig. 4(b)) show a typical shape commonly observed in MFC system (Logan et al., 2006; Sun et al., 2010). As the current density increases, the curves show rapid decrease of potential in the initial current density range originated from the activation loss, slightly decrease in the middle current density range due to ohmic loss, and rapidly decrease in
Fig. 5. (a) Cyclic voltammogram for the anode of mediator-less MFC with Fe3O4 content in N2-saturated 0.1 M PBS. (Scan rate 25 mV s 1); and (b) Nyquist plots of impedance spectra for Fe3O4/CNT electrode with Fe content (frequency range of 100 kHz–1 MHz).
80
I.H. Park et al. / Biosensors and Bioelectronics 58 (2014) 75–80
the later stage because of concentration loss. From the slopes of the polarization curves in each step, it is seen that the amounts of the activation losses and ohmic losses are almost equal for all the Fe3O4/CNT modified electrodes. But the concentration losses vary depending on the samples when applying higher current densities. The mass transfer losses increase in the order of 304 504 10 480 wt% Fe3O4. This also indicates that 30 wt% Fe3O4/CNT modified electrode has the smallest mass transfer loss due to the appropriate composition of Fe3O4/CNT nanocomposite, resulting in high power density. Although unmodified carbon paper electrode shows negligible faradaic current generation, as seen in Fig. 5(a), the featureless and rectangular shaped CV curves, which appear from the typical carbon double layer behavior owing to its structural variations, are observed from all the Fe3O4/CNT modified electrodes with their distinct oxidation peak potentials, indicating the production of higher faradaic currents (Zhang et al., 2013). CV spectra show faradaic peak current shift towards right side or positive potential with the increase in Fe3O4 content. This means there is increased electron transfer efficiency of E.coli which forms a multi-layered network with Fe3O4/CNT modification and the oxidation potential increases with Fe3O4 content, resulting in the decrease in MFC performance. Consequently, this CV data qualitatively indicates that E.coli turns highly electro active due to the presence of Fe3O4/ CNT nanocomposite in the anode coating which is supported by impedance measurements. The electrochemical impedance spectra of Fe3O4/CNT modified anode with different Fe3O4 contents are given in Fig. 5(b). From Nyquist plot, the x-intercept gives the electrolyte resistance (Rs) in the high frequency region and the diameter of semicircle represents the charge transfer resistance (Rct). The smaller the Rct the faster the rate of charge transfer i.e., the charge transfer resistance (Rct) at electrode/electrolyte interface is equal to the diameter of the semicircle. Each anode showed significant difference in the observed electrolyte resistance (Rs) and all four composite electrodes represent well defined frequency dependent semicircle impedance curves over high frequencies. The Nyquist plot shows that 30 wt% Fe3O4 in Fe3O4/CNT modified anode has the smallest semicircle exhibiting very low charge transfer resistance. This result supports that 30 wt% Fe3O4 is an optimum ratio in Fe3O4/ CNT nanocomposite to attain high power density from the MFC. As the Fe3O4 content in Fe3O4/CNT increases, there is also significant increase in the resistance. In order to see the state of E.coli adsorbed on the modified electrode surface, SEM images of the anode surfaces are given in Fig. 2(b). (i), (ii) and (iii) in Fig. 2(b) are SEM images of pure CNT, Fe3O4/CNT nanocomposite without E.coli, and Fe3O4/CNT after experiment which includes E.coli adherence, respectively. It can be seen that the Fe3O4/CNTs are completely covered with an E.coli bio-film after cell test. This indicates that the CNTs and E.coli forms a multilayered network on the anode surface to make a conductive biocatalytic electrode with the help of Fe3O4. The Fe3O4/CNT nanocomposites are strongly adsorbed on the electrode surface due to the magnetic force of Fe3O4. Consequently, the interaction of magnetized Fe3O4 particles and the formation of complex network with conductive CNT were found to be important factors for improving electrical contact of E.coli and improving fuel cell performance. 4. Conclusions A novel Fe3O4/CNT hybrid nanocomposite modified carbon anode was investigated in mediatorless MFC. Mono-dispersed Fe3O4 nanoparticles were deposited on CNT by using solvothermal synthetic method and the prepared Fe3O4/CNT nanocomposite was employed to modify carbon paper electrode of MFC anode.
The Fe3O4/CNT anode with various Fe3O4 content was investigated to find the optimum ratio of the nanocomposite. The TEM images showed that 3–10 nm sized Fe3O4 nanoparticles were well deposited over carbon nanotube surface and the density of the Fe3O4 nanoparticles increased with the increase of the Fe3O4 loading. The Fe3O4/CNT modified anode improved the overall performance of MFC without any mediators. Specifically, 30 wt% Fe3O4 in the Fe3O4/CNT composite anode is found to be the optimum ratio and exhibits better catalytic performance with the highest power density of 830 mWh/m2 than other anodes. This optimum Fe3O4 in Fe3O4/CNT greatly improved the CNT coating and surface area of the electrode. Thus the MFC performance is enhanced by increasing the adhesion of bacteria and efficiency of electron transfer between microbes and anode.
Acknowledgments This work was supported by the Human Resources Development program (No. 20114030200060) of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government Ministry of Trade, Industry and Energy. This research was also supported by Basic Science Research Program through National Research Foundation of Korea (NRF) funded by the Ministry of Education (2009–0094031).
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2014.02.044. References Caccavo, F.J., Lonergan, D.J., Lovley, D.R., Davis, M., Stolz, J.F., McInerney, M.J., 1994. Appl. Environ. Microbiol. 60, 3752–3759. Chang, I.N.S., Moon, H., Bretschger, O., Jang, J.K., Park, H.I., Nealson, K.H., Kim, B.H., 2006. J. Microbiol. Biotechnol. 16 (2), 163–177. Delaney, G.M., Bennetto, H.P., Mason, J.R., Roller, H.D., Stirling, J.L., Thurtson, C.F.J., 1984. Chem. Tech. Biotechnol. 34B, 13–27. Deng, J., Wena, X., Wang, Q., 2012. Mater. Res. Bull. 47, 3369–3376. Feng, C.H., Ma, L., Li, F.B., Mai, H.J., Lang, X.M., Fan, S.S., 2010. Biosens. Bioelectron. 25, 1516–1520. Kim, H.J., Park, H.S., Hyun, M.S., Chang, I.S., Kim, M., Kim, B.H., 2002. Enzyme Microb. Technol. 30, 145–152. Logan, B.E., Hamelers, B., Rozendal, R., Shroder, U., Keller, J., Freguia, S., Aelterman, P., Verstraete, W., Rabaey, K., 2006. Environ. Sci. Technol. 40, 5181. Lovely, D.R., 2006. Nat. Rev. Microbiol. 4, 497–508. Park, D.H., Zeikus, J.G., 2000. Appl. Environ. Microbiol. 66 (4), 1292–1297. Park, D.H., Zeikus, J.G., 2003. Biotechnol. Bioeng. 81 (3), 348–355. Peng, X., Wang, H., Xia, X., Huang, X., Mo, Y., Cao, X., Fan, M., 2011. Biosens. Bioelectron. 26, 3000–3004. Peng, X., Yu, H., Wang, X., Zhou, Q., Zhang, S., Geng, L., Sun, J., Cai, Z., 2012. Bioresour. Technol. 121, 450–453. Qiao, Y., Li, C.M., Bao, S.J., Bao, Q.L., 2007. J. Power Sources 170, 79–84. Qiao, Y., Bao, S.J., Li, C.M., Cui, X.Q., Lu, Z.S., 2008. ACS Nano 2, 113–119. Roller, H.D., Bennetto, H.P., Delaney, G.M., Mason, J.R., Stirling, J.L., Thurtson, C.F., 1984. J. Chem. Tech. Biotechnol. 34B, 3–12. Scott, K., Rimbu, G.A., Katuri, K.P., Prasad, K.K., Head, I.M., 2007. Process Safety Environ. Protect. 85 (5), 481–488. Sun, Z., Liu, Z., 2005. J. Mater. Chem. 15, 4497–4501. Sun, J.J., Zhao, H.Z., Yang, Q.Z., Song, J., Xue, A., 2010. Electrochim. Acta 55, 3041–3047. Xie, X., Hu, L., Pasta, M., Wells, G., Kong, D., Criddle, C., Cui, Y., 2011. Nano Lett. 11, 291–296. Xie, X., Ye, M., Hu, L., Liu, N., McDonough, J.R., Chen, W., Alshareef, H.N., Criddle, C.S., Cui, Y., 2012. Energy Environ. Sci. 5, 5265. Yoon, S.M., Choi, W.M., Baik, H., Shin, H.J., Song, I., Kwon, M.S., Bae, J.J., Kim, H., Lee, Y.H., Choi, J.Y., 2012. ACS Nano 6 (8), 6803–6811. You, S.J., Zhao, Q.L., Jiang, J.Q., Zhang, J.N., Zhao, S.Q., 2006. J. Environ. Sci. Health 41, 2721–2734. Zhang, T., Cui, C., Chen, S., Ai, X., Yang, H., Shen, P., Peng, Z., 2006. Chem. Commun., 2257–2259 Zhang, F., Xia, X., Luo, Y., Sun, D., Call, D.F., Logan, B.E., 2013. Bioresour. Technol. 133, 74–81.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Enhanced electrical contact of microbes using Fe3O4/CNT nanocomposite anode in mediator-less microbial fuel cell In Ho Park a, Maria Christy b, Pil Kim a,b,c, Kee Suk Nahm a,b,c,n a
Specialized Graduate School of Hydrogen and Fuel Cell Engineering, Chonbuk National University, Jeonju 561-756, Republic of Korea R&D Education Center for Fuel Cell Materials & Systems, Chonbuk National University, Jeonju 561-756, Republic of Korea c School of Chemical Engineering, Chonbuk National University, Jeonju 561-756, Republic of Korea b
art ic l e i nf o
a b s t r a c t
Article history: Received 21 October 2013 Received in revised form 27 January 2014 Accepted 17 February 2014 Available online 26 February 2014
A novel Fe3O4/CNT nanocomposite was synthesized and employed for the modification of carbon paper anode in a mediator-less microbial fuel cell (MFC) to enhance its performance. The Fe3O4/CNT composite modified anodes with various Fe3O4 contents were investigated to find the optimum ratio of the nanocomposite for the best MFC performance. The Fe3O4/CNT modified anodes produced much higher power densities than unmodified carbon anode and the 30 wt% Fe3O4/CNT modified anode exhibited a maximum power density of 830 mW/m2. In the Fe3O4/CNT composite modified anode, Fe3O4 helps to attach the CNT on anode surface by its magnetic attraction and forms a multi layered network, whereas CNT offers a better nanostructure environment for bacterial growth and helps electron transfer from E.coli to electrode resulting in the increase in the current production with the catalytic activity of bacteria. The electrocatalytic behavior and all possible mechanism for their better performance are discussed in detail with the help of various structural and electrochemical techniques. & 2014 Elsevier B.V. All rights reserved.
Keywords: Microbial fuel cell Electro-catalysis Power production Electron transfer Nanocomposite Modified anode
1. Introduction Microbial fuel cell (MFC) is one of the promising green energy sources to produce electricity from organic wastes with the help of bacteria as catalyst. It converts the chemical energy in the organic wastes into electricity by bio-electrochemical reactions (Logan et al., 2006; Lovely, 2006). The MFC comprises a bio-anode where the microorganisms disintegrate the organic wastes into small molecules while producing electrons and protons. Thus produced electrons transfer from the microbes to anode and further travel through the external circuit to the cathode where they are used to reduce the electron acceptors. Meanwhile, protons migrate through a proton exchange membrane (PEM) from the anode to the cathode and complete the circuit. This is the basic working mechanism of MFC, which results in the generation of electrical power as well as removal of organic waste simultaneously (Delaney et al., 1984; Roller et al., 1984). In spite of the promising merits of MFC, the power generated from the fuel cell is still low to limit its applications in energy generation industries. The anode material is considered to be one of the key factors that influence the energy conversion in MFC along with cathodes and electrolytes. In the process of electricity generation from MFC, because the electron transfer from microbes to the anode is
n
Corresponding author. Tel.: þ 82 63 270 2311. E-mail address: [email protected] (K.S. Nahm).
http://dx.doi.org/10.1016/j.bios.2014.02.044 0956-5663 & 2014 Elsevier B.V. All rights reserved.
relatively slow process, the use of electron mediators is generally required for rapid electron transfer. Mediators such as 2-hydroxyl 1, 4-naphthoquinone (HNQ) or thionine have been generally used as electron shuttle to facilitate the electron transfer. These types of MFC are called mediated MFC where the mediators need to be replenished continuously to avoid toxicity and to reduce operational losses (Logan et al., 2006). Since Kim et al. demonstrated a possibility of mediatorless MFC with Shewanella species (Hyung Joo Kim et al., 2002), various metal reducing microorganisms such as Geobacter and Shewanella are reported to directly transfer electrons to the electrode in the electrocatalytic process of the mediatorless MFC (Chang et al., 2006; Zhang et al., 2006). These microbial organisms are known to effectively transfer electrons through nanowire without any mediators. The mediatorless MFC has provided stable power generation using various biomasses but has a low power output (Caccavo et al., 1994; You et al., 2006). The relatively low power density and poor energy conversion efficiency of the conventional MFC are resulted from sluggish electron transfer between bacteria and electrode. Modification of anode surface can reduce the charge transfer resistance and increase the electron transfer and there by improve the overall performance of MFC. The modification of anode with various nano-engineering techniques has been proposed to be promising and efficient to facilitate the electron transfer between microbes and anode (Scott et al., 2007). Bacterial attachment and the formation of a biofilm or network on the anode surface are essential for the efficient biological transfer of electrons in an MFC. Various modification strategies
76
I.H. Park et al. / Biosensors and Bioelectronics 58 (2014) 75–80
including nanomaterials and fabrication methods have been developed so far to increase the electron-accepting ability of the electrodes and to improve electron transfer and power density. Qiao et al. (2007) investigated carbon nanotube/polyaniline nanostructure composite as anode materials and reported. They observed the highest electrochemical activity and maximum power density at 20 wt% CNT composite anode in E.coli-based MFC. A unique nanostructured polyaniline/mesoporous TiO2 nanocomposite was also explored as an anode in E.coli MFC (Qiao et al., 2008), which reported that the best bio- and electrocatalytic performance was observed from the composite with 30 wt% PANI. Meanwhile, Feng et al. (2010) attained the maximum power density of 1303 mWm 2 from a polypyrrole/ anthraquinone-2,6-disulphonic disodium salt (PPy/AQDS)-modified carbon felt anode, which is 13 times larger power than that obtained from the MFC equipped with unmodified carbon anode in the presence of Shewanella species. Interestingly, Sun et al. (2010) fabricated a CNT/TP modified anode with enhanced electron transfer and power production. Peng et al. (2012) also showed enhanced behavior of anode by rolling the mixture of Fe3O4 and activated carbon (AC) on anode surface with a maximum power density of 80975 mW/m2. So, the anode surface structure and its ability to interact with bacteria are very important regardless of the electron transfer mechanism. Hence the modification of anode with nanomaterials is an efficient method to significantly enhance the generation of electricity in MFC. With this background, in this work, we investigated a novel Fe3O4/CNT hybrid nanocomposite modified carbon anode in mediatorless MFC. Carbon nanotubes (CNTs) have been served as a promising option for electrode surface modification materials due to their unique mechanical and electrical properties. First we coated mono-dispersed Fe3O4 nanoparticles on CNT by using a solvothermal synthetic method. Then the prepared Fe3O4/CNT nanocomposite was introduced to E.coli culture medium to form a network of E.coli and Fe3O4/CNT nanocomposite. The multilayered network was coated on the carbon paper electrode by applying magnetic force and used as MFC anode. The Fe3O4/CNT anode with various Fe3O4 content is also investigated to find the optimum ratio of the nanocomposite. The enhanced electron transfer and increased MFC performance with different Fe3O4 (in Fe3O4/CNT anode) content is explained with the help of polarization curves and other electrochemical characterizations.
in Luria–Bertani (LB) medium. The E.coli K12 colony in LB agar medium was extracted into a falcon tube containing 5 mL LB medium. Seed culture was carried out for 12 h at 160 rpm in a shaking incubator (37 1C). 1 mL of the seed culture medium was mixed with 20 mL of LB medium in a falcon tube followed by the main culture for 12 h at 160 rpm in a shaking incubator (37 1C) (Kim et al., 2002). 2.3. Assembly of anode To prepare Fe3O4/CNT modified anode, 0.05 g of Fe3O4/CNT nanocomposite was introduced to E.coli culture medium in LB broth, vortexed for uniform dispersion of particles in the solution and then immobilized bacteria on Fe3O4/CNT nanocomposite by a culturing process for 2–3 h while continuously shaking at 160 rpm in a constant temperature incubator. This process allows the formation of a sort of network between E.coli and Fe3O4/CNT nanocomposite in the solution. An electrode of carbon paper (TGPH-090, Toray International Inc.) wrapped around a magnetic slit (10 10 2 mm3) was placed in the same culture medium to form a multilayer of the E.coli and Fe3O4/CNT nanocomposite on the carbon paper electrode layer. In other words, a network of E.coli and Fe3O4/CNT nanocomposite is formed in a multilayered shape on the magnetic carbon electrode resulting in the formation of a conductive biocatalytic electrode. The conductive biocatalytic electrode was prepared with different Fe contents of Fe3O4/CNT nanocomposite (10, 30, 50 and 80 wt%) and fuel cell performance of the electrodes were comparatively investigated. For a clear understanding the anode assembly is clearly shown in Fig. 1 in the supplementary article. 2.4. Fabrication and evaluation of MFC The detailed MFC fabrication and MFC evaluation with its working conditions and details are explained in supplementary article. 2.5. Physical and electrochemical characterizations The physical and electrochemical characterization details and techniques used to study them are also explained in the supplementary article.
2. Materials and methods 3. Results and discussion 2.1. Synthesis of Fe3O4/CNT 3.1. Structural characterizations Fe3O4 was deposited on CNT using a solvothermal synthetic method reported detail in previous reports (Deng et al., 2012; Yoon et al., 2012). Stoichiometric amounts of Iron III acetylacetonate, Fe(C5H8O2)3 precursor were completely dissolved in triethylene glycol (C6H14O4) (sigma Aldrich) with agitation. They were initially mixed with 0.1 g of CNT (JEO Co.) to form CNT suspended solutions. (The amounts of precursor used for preparing different ratios of Fe3O4 in Fe3O4/CNT are given in Table 1 in supplementary article). The prepared suspension was then reduced with triethylene glycol at 280 1C to form Fe3O4/CNT nanocomposite. The suspension was agitated uniformly throughout the reduction reaction using magnetic stirrer. Fe3O4/CNT nanocomposite with four different Fe3O4 compositions (viz. 10, 30, 50, and 80 wt% ) were synthesized and studied in this work. 2.2. Microbe cultivation A gram-negative bacterium Escherichia coli (E.coli) K12 of 2 μm size was used in this experiment and the aerobic culture was done
Crystallographic studies were performed on the synthesized metal/CNT composites with different Fe3O4 contents by using X-ray diffraction (XRD) spectroscopy. XRD spectra for the synthesized samples are shown in Fig. 1(a) as a function of contents. The characteristic (2 2 0), (3 1 1), (4 0 0), (4 2 2), (5 1 1) and (4 4 0) peaks appeared on the spectra are due to Fe3O4 [JCPDS 01-1111], whereas the (0 0 2) peak corresponds to CNT (Sun and Liu, 2005). This confirms that the deposited material on CNT in this work is Fe3O4. With increasing Fe3O4 content in Fe3O4/CNT, the intensity of the Fe3O4 related peaks increases, while that of CNT decreases, as shown in Fig. 1(b), suggesting the formation of well aligned Fe3O4 over CNTs. The surface morphology of the synthesized Fe3O4/CNT nanocomposite was analyzed by TEM and FESEM measurements. The TEM images (Fig. 2(a)) clearly show the deposition of Fe3O4 nanoparticles over raw carbon nanotube surface and the density of the Fe3O4 nanoparticles increases with the increase of the Fe3O4 loading content. The diameter of CNT is seen to be 10–15 nm and
I.H. Park et al. / Biosensors and Bioelectronics 58 (2014) 75–80
77
Fig. 1. (a) XRD spectra of Fe3O4/CNT nanocomposite with Fe content; (i) CNT, (ii) 10 wt%, (iii) 30 wt%, (iv) 50 wt%, and (v) 80 wt%; and (b) the intensity of Fe content in CNT.
Fig. 2. (a) TEM images of Fe3O4/CNT nanocomposite with Fe content; (i) 10 wt%, (ii) 30 wt%, (iii) 50 wt%, and (iv) 80 wt%. (b) SEM images of (i) Pure CNT; (ii) Fe3O4/CNT nanocomposite and (iii) Fe3O4/CNT after experiment which shows E.coli adherence.
that of the deposited Fe3O4 nanoparticles varies between 3–10 nm as shown in Fig. 2(a). Although not shown in this paper, the FESEM images also presented that the number of Fe3O4 particles deposited on CNT surface increases with Fe3O4 loading content.
3.2. MFC performance Current versus time of mediator-less MFC fabricated with Fe3O4/CNT nanocomposite assembled anode with different Fe3O4
78
I.H. Park et al. / Biosensors and Bioelectronics 58 (2014) 75–80
contents are given in Fig. 3. The anode chamber is first filled with glucose and bacteria suspension and purged with nitrogen for 30 min to remove residual oxygen in the solution and then measured for current as a function of time (up to 120 h). It can be observed from all the samples that, with the increase of time, the current produced from the anode by catalytic activity of bacteria initially increases and reach a maximum value, and then decrease gradually. However, it is seen that the Fe3O4/CNT modified carbon anode generates very high current compared to unmodified bare carbon electrodes. Fig. 3 shows that the highest current of 0.32 mA is attained from 30 wt% Fe3O4/CNT composite for longer time duration (about 45 h). Although the concentration of glucose in the anode chamber and the number of bacteria on anode during the bio-reaction may cause the variation of current as a function of time, it is considered that 30 wt% Fe3O4 in Fe3O4/ CNT composite anode offers the best nanostructure environment for bacterial growth and electron transfer from E-coli to electrode, resulting in the production of higher current by the catalytic activity of bacteria. The power density and polarization curves of mediator-less MFC assembled with Fe3O4/CNT nanocomposite anode is given in Fig. 4(a) and (b), respectively. It is seen from Fig. 4(a) that Fe3O4/ CNT modified anodes show much higher power densities than unmodified carbon anode. It has been reported that exogenous mediators have always been necessary in biofuel cell experiments
Fig. 3. Time and current of mediator-less MFC using Fe3O4/CNT electrode with different Fe3O4 content.
with E.coli since without the mediators, MFC using E coli as anode biocatalyst generated only negligible amount of electricity (Park and Zeikus, 2000, 2003). In our experiment, however, the Fe3O4/ CNT modified anodes show higher power densities even though no mediator has been employed. Fig. 4(a) shows that the power density curve for the Fe3O4/CNT modified anode with 30 wt% Fe3O4 content exhibits a maximum power density of 830 mW/m2 at a current density of 1900 mA/m2. The power density of Fe3O4/ CNT modified anode initially increases from 10 wt% to 30 wt% Fe3O4 contents and shows the maximum value at 30 wt% Fe3O4. Meanwhile, the power decreases when increasing Fe3O4 contents from 30 wt% to 50 wt% and further to 80 wt%. The power density is calculated to be 440 mW/m2 for 10 wt%, 830 mW/m2 for 30 wt%, 540 mW/m2 for 50 wt% and 300 mW/m2 for 80 wt% of Fe3O4 in Fe3O4/CNT modified anode. In previous literatures (Qiao et al., 2007, 2008; Feng et al., 2010; Sun et al., 2010; Peng et al., 2012), the surface modification of anode has been reported to enhance the generation of electricity from MFC. Sun et al. constructed a three-dimensional MWNTs interwoven network structure on anode by modifying carbon paper (TP) electrode utilizing a layer by layer (LBL) assemble technique and investigated the anode characteristics (Xie et al., 2011). The electron transfer ability of anode was significantly enhanced due to the interfacial resistance reduction resulting from the LBL self-assembled CNTs modification. It was also observed that the existence of the three dimensional MWNT multilayer increased a large accessible surface area for immobilization of microorganisms which decreased the interfacial charge transfer resistance, resulting in the production of a higher power density with 20% enhancement comparing to the unmodified bare carbon anode. The enhanced anode performance could be influenced by several parameters, such as electrode surface, electrochemical characteristics of electrode, and the mechanism of electron transfer, etc. Few other reports (Peng et al., 2011; Xie et al., 2011, 2012) on the surface modification of anode, mostly using CNTs, also identified the increase of electrode surface area which is useful for the immobilization of microbes and the reduction of electron transfer resistance in thus prepared anode, resulting in the increase of power generation from the MFC. Our observation above presents that all the Fe3O4/CNT modified anodes show much higher power densities than unmodified bare carbon anode and the power density produced from 10 wt% Fe3O4/CNT anode is about two times lower than that from 30 wt% Fe3O4/CNT anode. This presents that the power density obtained from the MFC is significantly influenced on the Fe3O4 weight
Fig. 4. (a) Power output of mediator-less MFC using Fe3O4/CNT electrode with Fe3O4 content (resistance of 100 KΩ–300 Ω); and (b) polarization curves of mediator-less MFC using Fe3O4/CNT electrode with Fe3O4 content (resistance of 100 KΩ–300 Ω).
I.H. Park et al. / Biosensors and Bioelectronics 58 (2014) 75–80
percent in the Fe3O4/CNT composite electrode. On contrary, however, the addition of further Fe3O4 more than 30 wt% has considerably decreased the power densities to be 540 mW/m2 and 300 mW/m2 for 50 wt% and 80 wt% Fe3O4 in Fe3O4/CNT modified anode, respectively. BET technique was employed to measure typical surface area for the Fe3O4/CNT nanocomposite, which are also listed in Table 1 in supplementary article. It is seen from the Table that the 10 wt% Fe3O4 in Fe3O4/CNT anode has the highest surface area and the surface area of the Fe3O4/CNT nano-composite is decreased with the increase of Fe3O4 content in Fe3O4/CNT nanocomposites. This could possibly be attributed to the lower specific surface area of Fe3O4 than CNT. The deposition of Fe3O4 on CNT could reduce the specific surface area of the composite. In the previous TEM and SEM studies on the surface morphology of the synthesized Fe3O4/CNT nanocomposite, it was also seen that the increase of the Fe3O4 loading content increased the number of 3–10 nm Fe3O4 nanoparticles deposited over carbon nanotube surface. This indicates that the increase of the Fe3O4 loading content over CNT results in the decrease of the surface area of Fe3O4/CNT nanocomposite. Differently from the trend of the surface area, however, Fig. 4(a) showed that, initially, with increasing Fe3O4 content from 10 wt% to 30 wt%, the power density of Fe3O4/ CNT modified anode increases, though the surface area decreases. And then both the surface area and the power density decrease when increasing Fe3O4 contents higher than 30 wt% to be 50 wt% and to be 80 wt%. Our observation of the highest power generation from 30 wt% can cause a controversy to the results observed in previous report where the power density produced increases proportional to the surface area (Sun et al., 2010). This means that in our work Fe3O4 deposited on the CNT surface does a significant role to suppress and create a diffusion problem of electron donor and to interfere the electron transfer between the donor and acceptor. Qiao et al. (2008) determined an optimum condition to generate the highest power in their anode modification studies. They explored catalytic activity of nanostructured polyaniline (PANI)/ TiO2 nanocomposite as an anode modification material in E.coli MFC and observed the best bio-and electrocatalytic performance with 30 wt% PANI modified TiO2, although specific surface area of the samples decreases with increasing PANI content in the composites. Although the mechanism of direct electron transfer between the nanocomposite and E.coli observed in their experiment was not clearly proposed, they proposed that the greatest electrocatalytic behavior of the E.coli with 30 wt% PANI/TiO2
79
electrode could be ascribed to optimal surface structure for good mass transfer and low IR drops in the inner pores of anode. The decrease in power production with increase in Fe3O4 content might be explained with respect to the electron transfer between the microbes and the anodes. Since 10 wt% Fe3O4 in Fe3O4/CNT has the highest surface area, it is expected to deliver maximum power density. On contraire, however, 30 wt% Fe3O4 in Fe3O4/CNT anode is optimized to be the best ratio for maximum power production. This phenomenon can be viewed in a different way. The main role of Fe3O4 here is to attach the high surface CNT nanoparticles on the carbon anode surface with the magnetic attraction between Fe3O4 and the magnet. Meanwhile, CNT in Fe3O4/CNT composite contributes to help electron transfer from microbes to anode electrode owing to its high electron conductive property (Sun et al., 2010). So that, the microbes attached on the CNT surface transfers electrons through the multi-channel CNT networks onto the anode surface. In case of 10 wt% Fe3O4 in Fe3O4/ CNT, there is large surface area on the composite anode which can accommodate the adsorption of large number of microbes/E.coli on CNT surface. Owing to less Fe3O4 content in the 10% sample, however, only few CNT are attracted to be coated on the carbon anode. For the case of 30 wt% Fe3O4 in Fe3O4/CNT anodes, on the other hand, the increased amount of Fe3O4 content affords the build-up of multi-layered CNT network over the carbon anode surface, while CNTs provides enough surface area necessary for the adsorption of large microbe population. This large population of microbes engages in rapid and high electron transfer through the high surface multi-channeled CNT networks, resulting in high power production. However, for the other two cases of 50 and 80 wt% Fe3O4 in Fe3O4/CNT anodes, the CNT surface is highly populated with Fe3O4, which prevents bacterial adherence: the microbes and CNT interactions are interfered by high Fe3O4 concentration, resulting in low electron transfer. This in turn decreases the power production in the MFC. This indicates that the catalytic performance of the Fe3O4/CNT nanocomposite modified anode in the MFC can be optimized by adjusting the Fe3O4 percentage in the composite, and the composite with 30 wt% Fe3O4 gives the highest bio-and electrocatalytic performance. The polarization curves (Fig. 4(b)) show a typical shape commonly observed in MFC system (Logan et al., 2006; Sun et al., 2010). As the current density increases, the curves show rapid decrease of potential in the initial current density range originated from the activation loss, slightly decrease in the middle current density range due to ohmic loss, and rapidly decrease in
Fig. 5. (a) Cyclic voltammogram for the anode of mediator-less MFC with Fe3O4 content in N2-saturated 0.1 M PBS. (Scan rate 25 mV s 1); and (b) Nyquist plots of impedance spectra for Fe3O4/CNT electrode with Fe content (frequency range of 100 kHz–1 MHz).
80
I.H. Park et al. / Biosensors and Bioelectronics 58 (2014) 75–80
the later stage because of concentration loss. From the slopes of the polarization curves in each step, it is seen that the amounts of the activation losses and ohmic losses are almost equal for all the Fe3O4/CNT modified electrodes. But the concentration losses vary depending on the samples when applying higher current densities. The mass transfer losses increase in the order of 304 504 10 480 wt% Fe3O4. This also indicates that 30 wt% Fe3O4/CNT modified electrode has the smallest mass transfer loss due to the appropriate composition of Fe3O4/CNT nanocomposite, resulting in high power density. Although unmodified carbon paper electrode shows negligible faradaic current generation, as seen in Fig. 5(a), the featureless and rectangular shaped CV curves, which appear from the typical carbon double layer behavior owing to its structural variations, are observed from all the Fe3O4/CNT modified electrodes with their distinct oxidation peak potentials, indicating the production of higher faradaic currents (Zhang et al., 2013). CV spectra show faradaic peak current shift towards right side or positive potential with the increase in Fe3O4 content. This means there is increased electron transfer efficiency of E.coli which forms a multi-layered network with Fe3O4/CNT modification and the oxidation potential increases with Fe3O4 content, resulting in the decrease in MFC performance. Consequently, this CV data qualitatively indicates that E.coli turns highly electro active due to the presence of Fe3O4/ CNT nanocomposite in the anode coating which is supported by impedance measurements. The electrochemical impedance spectra of Fe3O4/CNT modified anode with different Fe3O4 contents are given in Fig. 5(b). From Nyquist plot, the x-intercept gives the electrolyte resistance (Rs) in the high frequency region and the diameter of semicircle represents the charge transfer resistance (Rct). The smaller the Rct the faster the rate of charge transfer i.e., the charge transfer resistance (Rct) at electrode/electrolyte interface is equal to the diameter of the semicircle. Each anode showed significant difference in the observed electrolyte resistance (Rs) and all four composite electrodes represent well defined frequency dependent semicircle impedance curves over high frequencies. The Nyquist plot shows that 30 wt% Fe3O4 in Fe3O4/CNT modified anode has the smallest semicircle exhibiting very low charge transfer resistance. This result supports that 30 wt% Fe3O4 is an optimum ratio in Fe3O4/ CNT nanocomposite to attain high power density from the MFC. As the Fe3O4 content in Fe3O4/CNT increases, there is also significant increase in the resistance. In order to see the state of E.coli adsorbed on the modified electrode surface, SEM images of the anode surfaces are given in Fig. 2(b). (i), (ii) and (iii) in Fig. 2(b) are SEM images of pure CNT, Fe3O4/CNT nanocomposite without E.coli, and Fe3O4/CNT after experiment which includes E.coli adherence, respectively. It can be seen that the Fe3O4/CNTs are completely covered with an E.coli bio-film after cell test. This indicates that the CNTs and E.coli forms a multilayered network on the anode surface to make a conductive biocatalytic electrode with the help of Fe3O4. The Fe3O4/CNT nanocomposites are strongly adsorbed on the electrode surface due to the magnetic force of Fe3O4. Consequently, the interaction of magnetized Fe3O4 particles and the formation of complex network with conductive CNT were found to be important factors for improving electrical contact of E.coli and improving fuel cell performance. 4. Conclusions A novel Fe3O4/CNT hybrid nanocomposite modified carbon anode was investigated in mediatorless MFC. Mono-dispersed Fe3O4 nanoparticles were deposited on CNT by using solvothermal synthetic method and the prepared Fe3O4/CNT nanocomposite was employed to modify carbon paper electrode of MFC anode.
The Fe3O4/CNT anode with various Fe3O4 content was investigated to find the optimum ratio of the nanocomposite. The TEM images showed that 3–10 nm sized Fe3O4 nanoparticles were well deposited over carbon nanotube surface and the density of the Fe3O4 nanoparticles increased with the increase of the Fe3O4 loading. The Fe3O4/CNT modified anode improved the overall performance of MFC without any mediators. Specifically, 30 wt% Fe3O4 in the Fe3O4/CNT composite anode is found to be the optimum ratio and exhibits better catalytic performance with the highest power density of 830 mWh/m2 than other anodes. This optimum Fe3O4 in Fe3O4/CNT greatly improved the CNT coating and surface area of the electrode. Thus the MFC performance is enhanced by increasing the adhesion of bacteria and efficiency of electron transfer between microbes and anode.
Acknowledgments This work was supported by the Human Resources Development program (No. 20114030200060) of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government Ministry of Trade, Industry and Energy. This research was also supported by Basic Science Research Program through National Research Foundation of Korea (NRF) funded by the Ministry of Education (2009–0094031).
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2014.02.044. References Caccavo, F.J., Lonergan, D.J., Lovley, D.R., Davis, M., Stolz, J.F., McInerney, M.J., 1994. Appl. Environ. Microbiol. 60, 3752–3759. Chang, I.N.S., Moon, H., Bretschger, O., Jang, J.K., Park, H.I., Nealson, K.H., Kim, B.H., 2006. J. Microbiol. Biotechnol. 16 (2), 163–177. Delaney, G.M., Bennetto, H.P., Mason, J.R., Roller, H.D., Stirling, J.L., Thurtson, C.F.J., 1984. Chem. Tech. Biotechnol. 34B, 13–27. Deng, J., Wena, X., Wang, Q., 2012. Mater. Res. Bull. 47, 3369–3376. Feng, C.H., Ma, L., Li, F.B., Mai, H.J., Lang, X.M., Fan, S.S., 2010. Biosens. Bioelectron. 25, 1516–1520. Kim, H.J., Park, H.S., Hyun, M.S., Chang, I.S., Kim, M., Kim, B.H., 2002. Enzyme Microb. Technol. 30, 145–152. Logan, B.E., Hamelers, B., Rozendal, R., Shroder, U., Keller, J., Freguia, S., Aelterman, P., Verstraete, W., Rabaey, K., 2006. Environ. Sci. Technol. 40, 5181. Lovely, D.R., 2006. Nat. Rev. Microbiol. 4, 497–508. Park, D.H., Zeikus, J.G., 2000. Appl. Environ. Microbiol. 66 (4), 1292–1297. Park, D.H., Zeikus, J.G., 2003. Biotechnol. Bioeng. 81 (3), 348–355. Peng, X., Wang, H., Xia, X., Huang, X., Mo, Y., Cao, X., Fan, M., 2011. Biosens. Bioelectron. 26, 3000–3004. Peng, X., Yu, H., Wang, X., Zhou, Q., Zhang, S., Geng, L., Sun, J., Cai, Z., 2012. Bioresour. Technol. 121, 450–453. Qiao, Y., Li, C.M., Bao, S.J., Bao, Q.L., 2007. J. Power Sources 170, 79–84. Qiao, Y., Bao, S.J., Li, C.M., Cui, X.Q., Lu, Z.S., 2008. ACS Nano 2, 113–119. Roller, H.D., Bennetto, H.P., Delaney, G.M., Mason, J.R., Stirling, J.L., Thurtson, C.F., 1984. J. Chem. Tech. Biotechnol. 34B, 3–12. Scott, K., Rimbu, G.A., Katuri, K.P., Prasad, K.K., Head, I.M., 2007. Process Safety Environ. Protect. 85 (5), 481–488. Sun, Z., Liu, Z., 2005. J. Mater. Chem. 15, 4497–4501. Sun, J.J., Zhao, H.Z., Yang, Q.Z., Song, J., Xue, A., 2010. Electrochim. Acta 55, 3041–3047. Xie, X., Hu, L., Pasta, M., Wells, G., Kong, D., Criddle, C., Cui, Y., 2011. Nano Lett. 11, 291–296. Xie, X., Ye, M., Hu, L., Liu, N., McDonough, J.R., Chen, W., Alshareef, H.N., Criddle, C.S., Cui, Y., 2012. Energy Environ. Sci. 5, 5265. Yoon, S.M., Choi, W.M., Baik, H., Shin, H.J., Song, I., Kwon, M.S., Bae, J.J., Kim, H., Lee, Y.H., Choi, J.Y., 2012. ACS Nano 6 (8), 6803–6811. You, S.J., Zhao, Q.L., Jiang, J.Q., Zhang, J.N., Zhao, S.Q., 2006. J. Environ. Sci. Health 41, 2721–2734. Zhang, T., Cui, C., Chen, S., Ai, X., Yang, H., Shen, P., Peng, Z., 2006. Chem. Commun., 2257–2259 Zhang, F., Xia, X., Luo, Y., Sun, D., Call, D.F., Logan, B.E., 2013. Bioresour. Technol. 133, 74–81.