Performance of microbial fuel cells with and without ... - ResearchGate

Bioelectrochemistry 79 (2010) 261–264Contents lists available at ScienceDirectBioelectrochemistryjournal homepage: www.elsevier.com/locate/bioelechemShort communicationPerformance of microbial fuel cells with and without Nafion solution as cathodebinding agentYuelong Huang a , Zhen He b , Florian Mansfeld a, ⁎a Corrosion and Environmental Effects Laboratory (CEEL), The Mork Family Department of Chemical Engineering and Materials Science, University of Southern California, Los Angeles,CA 90089-0241, USAb Department of Civil Engineering and Mechanics, University of Wisconsin—Milwaukee, Milwaukee, WI 53211, USAarticleinfoabstractArticle history:Received 30 November 2009Received in revised form 24 March 2010Accepted 25 March 2010Available online 31 March 2010Keywords:Microbial fuel cell (MFC)NafionElectrochemical impedancespectroscopy (EIS)Internal resistanceElectricity productionThe performance of tubular microbial fuel cells (MFC) with and without Nafion solution as binding agent forthe cathode catalyst preparation was investigated using different electrochemical techniques. The currentoutput of both types of MFCs was monitored as a function of time using an external resistor. The current did notchange much with time and was higher for the water cell (WC) than for the Nafion cell (NC). Cell voltage (U c )–current (I) curves were recorded using a potentiodynamic technique. From the U c –I curves powerconcentration (P)–I and P–U c curves were constructed. The water cell (without Nafion) also achieved ahigher maximum power output. The internal resistance that was determined from the cell voltage at which thepower concentration reached its maximum value was higher for the NC than that for the WC, possibly due tothe higher cathodic polarization resistance of the NC cell. The impedance for the cathodes decreased withexposure time for both cells due to increased porosity of the surface layers covering the cathode materials. Nochanges of the impedance were observed for the WC anode. For the NC anode the impedance spectra changedfrom a one-time constant system to a two-time constant system at the longer exposure time.© 2010 Elsevier B.V. All rights reserved.1. IntroductionMicrobial fuel cells (MFC) are bio-electrochemical devices thatconvert organic materials into electricity using microorganisms asbiocatalysts [1–4]. Bacteria have been identified to be capable ofgenerating electricity from the anaerobic oxidation of organic matter[1–5]. Several factors have been identified as limiting the powerproduction of MFCs [6], including (1) fuel oxidation at the anode,(2) electron transfer from microbial cells to the anode, (3) ohmicresistance of the circuit, (4) proton transfer through the membrane,and (5) rate of oxygen reduction at the cathode.The method of assembling membrane electrodes can significantlyaffect the MFC performance. Many efforts have been made to apply Ptas a catalyst to improve the oxygen reduction rate at the cathode.Nafion solution, which is a perfluroinated ion-exchange solution, isoften used as a binding agent to connect the Pt particles, membrane,and carbon electrode [2,7,8]. A widely used method is that Pt/Cpowder is mixed with Nafion solution to form a paste, which is thenapplied by brushing, spraying or dipping onto an ion-exchangemembrane [7,8]. Employing binding agents is costly and may increasethe polarization resistance of the cathode. It would be of great interest⁎ Corresponding author. Tel.: +1 213 740 3016; fax: +1 213 740 7797.E-mail address: mansfeld@usc.edu (F. Mansfeld).to understand the effect of cathode binding agents on the poweroutput of a MFC in more detail.In this study, a Nafion solution was used as a model cathodebinding agent. The electricity production from two MFCs with andwithout Nafion solution as cathode binding agent was investigated.Electrochemical impedance spectroscopy (EIS) was employed tocharacterize the electrochemical properties of the anode and thecathode and provide information concerning the polarization resistancesof the anode and cathode electrodes. Current–time curves wererecorded as a function of exposure time using an external resistor. Cellvoltage–current curves were determined using a potentiodynamictechnique. From these curves power concentration–current concentrationand power concentration–cell voltage curves were calculatedfor the water cell (WC) and the Nafion cell (NC).2. Experimental approach2.1. MFC configuration and operationA schematic and a photograph of the laboratory prototype MFC areshown in Fig. 1. The MFCs were made of a tubular cation exchangemembrane (CMI-7000, Membrane International Inc., NJ, USA), whichhad a diameter of 4.8 cm and a length of 14 cm. The total volume wasabout 250 cm 3 and the liquid volume was 150 cm 3 . The anode that wasplaced inside the membrane consisted of carbon granules. It wasconnected to the external circuit by two titanium wires. The cathode1567-5394/$ – see front matter © 2010 Elsevier B.V. All rights reserved.doi:10.1016/j.bioelechem.2010.03.009

262 Y. Huang et al. / Bioelectrochemistry 79 (2010) 261–264with the cathode acting as counter electrode (CE) followed by recordingof the spectrum for the cathode with the anode serving as CE.3. Results and discussionFig. 1. A schematic diagram (a) and photograph of MFC (b).was a layer of a carbon/Pt mixture that was applied on the outside ofthe membrane. The carbon/Pt mixture was prepared by two differentmethods: 1) carbon powder containing 10% of Pt (Vulcan XC 72, BASFFuel Cell, Inc., NJ, USA) was mixed with the Nafion solution (Dupont,Nafion® perfluorinated ion-exchange resin, 1 µL for every 1 mg of Pt/Cpowder) and 2) the Nafion solution was replaced by tap water. Afterthe mixture was applied to the membrane, Ni-coated carbon fibers(TenaX®-J, Toho Tenax Co., Ltd., grade HTS 40, Irvine, CA, USA) servingas current collectors and part of the cathode were wrapped around themembrane and air-dried at room temperature overnight. An Ag/AgClreference electrode was placed in the anode chamber of each MFC.Both MFCs were fed continuously with a solution containingNaC 2 H 3 O 2·3H 2 O (3 g/L); yeast extract (0.2 g/L); NH 4 Cl (1 g/L); MgSO 4(0.25 g/L); NaCl (0.5 g/L); CaCl 2 (15 mg/L); trace solution (1 mL/L) [3]and phosphate buffered nutrient medium (100 mL/L) containingNaH 2 PO 4 (50 g/L) and Na 2 HPO 4 (107 g/L). The hydraulic retentiontime was 8 h. Twenty ml of a mixture of anaerobic and aerobic sludge(50/50), which were collected from a wastewater treatment plant(Joint Water Pollution Plant, Los Angeles, CA), was injected into theanode chamber as inoculum. All tests were conducted at roomtemperature (23 °C–25 °C).Fig. 2 shows an example of the current–time curves that wererecorded over extended time periods. The current did not changemuch with time and the WC produced higher current values than theNC. The performance of the NC and the WC was investigated by therecording U c –I curves.Since the area of the anode was not known, the experimentalcurrent and power values were normalized by the liquid volume ofthe anode compartment which was 150 cm 3 . Fig. 3 gives a comparisonof the current concentration and the power concentration producedby the WC and the NC. Fig. 3a shows the current concentration as afunction of the cell voltage for the two cells. The maximum currentconcentration obtained from the WC (26 µA/cm 3 ) was more thantwice that from the NC (12 µA/cm 3 )(Fig. 3a). Fig. 3b shows the powerconcentration as a function of the current concentration. Themaximum power concentration produced by the WC was about1.8 µW/cm 3 compared to 1.2 µW/cm 3 for the NC. The maximumpower concentration was obtained at a cell voltage of about 200 mV inboth cases as can be seen from the P–U c curves in Fig. 3c.At the cell voltage U max where maximum power output P maxoccurs R int =R ext , where R int is the internal resistance of the MFC andR ext is the external resistor that is placed between the anode and thecathode to obtain U max [9]. R int has been defined as [9]:R int = R a p + R c p + R Ω ;where R p a and R p c are the polarization resistance of the anode and thecathode, respectively and R Ω contains the ohmic contributions such asthe solution resistance between the anode and the cathode, themembrane resistance and the electrical resistance of the electrodes,leads, etc. R int of the WC was calculated as 117 Ω, while for the NC avalue of 195 Ω was obtained. As discussed previously, R int depends oncell voltage [9].The impedance spectra obtained for the NC and the WC at theopen-circuit potential (OCP) for the cathode and the anode during theexposure times of 120 and 140 days are shown in Fig. 4 as Bode-plotsin which the logarithm of the impedance modulus |Z| is plotted vs. thelogarithm of the applied frequency f. The impedance spectra of the NCand the WC cell cathodes are in general agreement with the coatingmodel proposed by Mansfeld et al. [10,11] for the analysis ofimpedance data for polymer coated metals. The decrease of theimpedance of the cathode for both MFCs with the exposure time isconsidered to be due to a decrease of the thickness of the polymerð1Þ2.2. Data collectionThe voltage across a R ext =500Ω external resistor was recordedevery 30 s using a computer based data acquisition system (Keithley2700 multimeter). These measurements were only interrupted whenimpedance spectra or cell voltage–current curves were recorded. Cellvoltage (U c )–current (I) curves were collected by applying a potentiodynamicscan at a scan rate of 0.1 mV/s from the open-circuit cellvoltage U oc to zero cell voltage U o using a Gamry reference 600potentiostat. The power concentration P (µW/cm 3 ) based on the liquidvolume of the MFC was calculated from the U c –I curves and plotted as afunction of current concentration (j) or cell voltage. The impedancespectra were collected at the open-circuit potential (OCP) of the anodeand the cathode using a Gamry reference 600 potentiostat (Garmryinstruments, Warminster, PA, USA). An ac voltage signal of 10 mV wasapplied in a frequency range from 100 kHz to 5 mHz. EIS measurementswere conducted by first recording the impedance spectrum of the anodeFig. 2. Current–time curves for the WC and the NC (R ext =500 Ω) for an exposure timeof 34 days.

Y. Huang et al. / Bioelectrochemistry 79 (2010) 261–264263Fig. 4. Bode-plots for the cathode (a) and the anode (b) of the WC and the NC for anexposure time of 65 days.explained at present, however they point to a complex interactionbetween the anode and the cathode for the NC.4. Summary and conclusionsFig. 3. Cell voltage–current concentration (a), power concentration–current concentration(b) and power concentration–cell voltage curves (c) for the WC and the NC foran exposure time of 55 days.layers covering the graphite fibers (increase of high-frequencycapacitance) and increased porosity of these layers (decrease ofpore resistance at intermediate frequencies) (Fig. 4b). The impedanceof the cathode for the NC is higher than that for the WC for bothexposure times due to the polymer layers formed during treatmentwith the Nafion solution.The impedance spectra for the WC anode did not changesignificantly with time indicating stable behavior (Fig. 4a). Theimpedance for fb10 4 Hz is due to the solution resistance and is verylow. The impedance for fb1 Hz is due to the capacitance of the anode.The impedance spectra for the NC anode at 120 days were differentfrom those of the WC at the same exposure time with a significantlylower capacitance value. The spectra changed to a two-time constantsystem for an exposure time of 140 days. These changes cannot beThe performance of tubular MFCs (Fig. 1) with and without Nafionsolution as a binding agent was investigated using different electrochemicaltechniques. The current–time curves (Fig. 2) andthepowerconcentration–current concentration or power concentration–cellvoltage curves that were obtained from the experimental cell voltage–current curves (Fig. 3) have demonstrated that the power production bythe water cell is higher than that of the Nafion cell. The internalresistance determined from the power density–cell voltage curves(Fig. 3c) was higher for the Nafion cell. The impedance spectra for thecathodes of the two MFCs suggest that this higher value is due to thehigher polarization resistance of the Nafion cell cathode. The ohmiccontribution to the internal resistance (Eq. (1)) was very small (Fig. 4).The Nafion polymer layers covering the cathode material (Pt particlesand Ni-coated carbon fibers) apparently block access of oxygen andthereby reduce the rate of oxygen reduction.AcknowledgmentThe authors acknowledge financial support from the ChevronCorporation.

264 Y. Huang et al. / Bioelectrochemistry 79 (2010) 261–264References[1] B. Logan, B. Hamelers, R. Rozendal, U. Schroder, J. Keller, S. Freguia, P. Aelterman,W. Verstraete, K. Rabaey, Microbial fuel cells: methodology and technology,Environ. Sci. Technol. 40 (17) (2006) 5181–5192.[2] K. Rabaey, J. Rodríguez, L.L. Blackall, J. Keller, P. Gross, D. Batstone, W. Verstraete,K.H. Nealson, Microbial ecology meets electrochemistry: electricity-driven anddriving communities, ISME J. 1 (2007) 9–18.[3] Z. He, N. Wagner, S.D. Minteer, L.T. Angenent, An upflow microbial fuel cell with aninterior cathode: assessment of the internal resistance by impedance spectroscopy,Environ. Sci. Technol. 40 (2006) 5212–5217.[4] A.K. Shukla, P. Suresh, S. Berchmans, A. Rajendran, Biological fuel cells and theirapplications, Curr. Sci. 87 (4) (2004) 455–468.[5] R.M. Allen, H.P. Bennetto, Microbial fuel-cells: electricity production fromcarbohydrates, Appl. Biochem. Biotechnol. 39/40 (1993) 27–40.[6] G.C. Gil, I.S. Chang, B.H. Kim, M. Kim, J.K. Jang, H.S. Park, H.J. Kim, Operationalparameters affecting the performance of a mediator-less microbial fuel cell,Biosens. Bioelectron. 18 (2003) 327–334.[7] J.R. Kim, G.C. Premier, F.R. Hawkes, R.M. Dinsdals, A.J. Guwy, Development of atubular microbial fuel cell (MFC) employing a membrane electrode assemblycathode, J. Power Sources 187 (2009) 393–399.[8] S. Cheng, H. Liu, B.E. Logan, Increased performance of single-chamber microbial fuelcells using an improved cathode structure, Electrochem. Commun. 8 (2006) 489–494.[9] A.K. Manohar, O. Bretschger, K.H. Nealson, F. Mansfeld, The use of electrochemicalimpedance spectroscopy (EIS) in the evaluation of the electrochemical propertiesof a microbial fuel cell, Bioelectrochemistry 72 (2008) 149–154.[10] F. Mansfeld, Use of electrochemical impedance spectroscopy for the study ofcorrosion protection by polymer coatings, J Appl. Electrochem. 25 (1995) 187–202.[11] F. Mansfeld, L.T. Han, C.C. Lee, G. Zhang, Evaluation of corrosion protection bypolymer coatings with electrochemical impedance spectroscopy and noiseanalysis, Electrochim. Acta 43 (1998) 2933–2945.