Bioresource Technology 102 (2011) 10459–10464

Contents lists available at SciVerse ScienceDirect

Bioresource Technology

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

Power generation using spinel manganese–cobalt oxide as a cathode catalystfor microbial fuel cell applications

Mohamed Mahmoud a,⇑, Tarek A. Gad-Allah a, K.M. El-Khatib b, Fatma El-Gohary a

a Water Pollution Research Department, National Research Centre, El-Tahrir St., Dokki, P.O. Box 12311, Cairo, Egyptb Chemical Engineering and Pilot-Plant Department, National Research Centre, El-Tahrir St., Dokki, P.O. Box 12311, Cairo, Egypt

a r t i c l e i n f o a b s t r a c t

Article history:Received 19 July 2011Received in revised form 26 August 2011Accepted 29 August 2011Available online 5 September 2011

Keywords:Microbial fuel cellsOxygen reduction reactionCathode catalystManganese–cobalt oxidePower generation

0960-8524/$ - see front matter � 2011 Elsevier Ltd. Adoi:10.1016/j.biortech.2011.08.123

⇑ Corresponding author. Tel./fax: +20 2 33351573.E-mail addresses: mohamed_mahmoud_84@yaho

com (M. Mahmoud).

This study focused on the use of spinel manganese–cobalt (Mn–Co) oxide, prepared by a solid state reac-tion, as a cathode catalyst to replace platinum in microbial fuel cells (MFCs) applications. Spinel Mn–Cooxides, with an Mn/Co atomic ratios of 0.5, 1, and 2, were prepared and examined in an air cathode MFCswhich was fed with a molasses-laden synthetic wastewater and operated in batch mode. Among thethree Mn–Co oxide cathodes and after 300 h of operation, the Mn–Co oxide catalyst with Mn/Co atomicratio of 2 (MnCo-2) exhibited the highest power generation 113 mW/m2 at cell potential of 279 mV,which were lower than those for the Pt catalyst (148 mW/m2 and 325 mV, respectively). This study indi-cated that using spinel Mn–Co oxide to replace platinum as a cathodic catalyst enhances power genera-tion, increases contaminant removal, and substantially reduces the cost of MFCs.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Increased economic growth and social development are leadingto a large gap between energy demands and the availability of fos-sil fuels. Current methods to produce energy are not sustainable,and concerns about climate change and global warming requiredeveloping new methods of energy production using renewableand carbon-neutral sources. The development of electrogenic reac-tors based on microbial fuel cells (MFCs) represents a new ap-proach for generating electricity from waste and biomass (Loganet al., 2006; Rabaey et al., 2005). Although the concept of electricityproduction from bacteria was conceived nearly a century ago(Cohen, 1931), only recently the technology has been sufficientlyimproved to make it useful as a method for energy generation.One near-term application of MFCs will be to produce electricityfrom wastewater, providing a new way to simultaneously treatwastewater while obtaining a source of clean and renewable en-ergy (Liu et al., 2004; Yu et al., 2007). Further development will al-low MFCs to be used for renewable energy production from othersources of biomass.

A large effort has been invested to enhance energy productionin MFCs. Most attention was on substrate degradation, electrongeneration, and transfer in anode chambers (Lovely, 2006; Loganet al., 2006; Li et al., 2010). Several important factors including

ll rights reserved.

o.com, m_mahmoud@lycos.-

substrate concentration, pH, conductivity, microbial activity, cir-cuit resistance, electrode, and membrane material have beenextensively investigated (Aelterman et al., 2008; Li et al., 2010).However, in recent studies, oxygen reduction and electron accep-tance in the cathode chamber have been found as the limiting stepfor energy production due to the slow reaction kinetics of the oxy-gen reduction rate (ORR) (Jang et al., 2004; Pham et al., 2004).

Platinum (Pt) has been widely used as the cathode catalyst toaccelerate the ORR and electron acceptance in MFCs (Liu andLogan, 2004), but the high cost of platinum, which estimated ataround 50% of the total capital cost of the system, poses an obstaclein the application of MFCs (Rozendal et al., 2008). Moreover, theeffectiveness of Pt as ORR catalyst would be far diminished due tothe lower power density than fuel cells, and contamination frombiological process (Tartakovsky and Guiot, 2006). Alternatively,several Pt-free non-noble transition metals such as Co, Fe and Mnand their supporting structures have been studied in the form ofiron(III) phthalocyanine (FePc) and cobalt tetramethoxyphenyl-porphyrin (CoTMPP) for ORR (Zhao et al., 2005; Cheng et al.,2006; You et al., 2006; Lefebvre et al., 2009; Duteanu et al., 2010).Moreover, stainless steel alloy and nickel alloys have been usedfor alternative low cost catalysts in microbial electrolysis cells (Sel-embo et al., 2009). They stated that the performance of stainlesssteel and nickel alloys was similar or better than platinum sheetmetal when used as cathodes. Stainless steel A286 showed the bestperformance of all the alloys tested at an applied potential of 0.9 V.A nickel oxide layer electrodeposited on the metal surfaces im-proved their catalytic performance by at least a factor of four. In a

10460 M. Mahmoud et al. / Bioresource Technology 102 (2011) 10459–10464

recent study, Li et al. (2010) investigated manganese oxides with acryptomelane-type octahedral molecular sieve (OMS-2) structureas a cathode catalyst in order to replace platinum. They investigatedthree catalysts doped with cobalt (Co-OMS-2), copper (Cu-OMS-2),and cerium (Ce-OMS-2). The results showed that after 400 h ofoperation, the Co-OMS-2 and Cu-OMS-2 exhibited good catalyticperformance in ORR with power density of 180 and 165 mW/m2,respectively. In another study, Liu et al. (2010) demonstrated thata nano-structured manganese oxide (MnOx), prepared by an elec-trochemically deposition method, could be an effective catalystfor oxygen reduction in an MFC to generate electricity with themaximum power density of 772.8 mW/m3.

Thus, the improved catalytic oxygen reduction by cost-effectivecathode materials would significantly enhance the overall perfor-mance of MFC. In the search for an ideal catalyst, spinel Mn–Cooxide has been studied as a promising alternative to platinum inmethanol and borohybride fuel cells (Restovic et al., 2002; Nissinenet al., 2002). Generally, spinels form a large class of compounds ofgeneral formula AB2X4, where A and B are metal cations and X is ananion such as O, S, Se or Te. Depending on the elements, the spinelstructure will be normal or inverse, cubic or tetrahedral, and thecations will occupy octahedral or tetrahedral sites and have posi-tive charges between I and IV (Nissinen, 2004). The high catalyticactivity of spinel Mn–Co oxides toward ORR could be due to themixed semi-conductivity of spinel oxides enables their direct useas electrode materials (Nissinen et al., 2002). The cations canchange their oxidation state, and ‘‘electroconductivity chains’’ arethus formed providing movement of either electrons or holes(Wen and Kang, 1998).

Therefore, the objective of this study was to investigate the fea-sibility of using spinel Mn–Co oxide as cathode catalyst in MFCs toenhance power generation and substrate degradation. The pre-pared Mn–Co oxide with different initial atomic ratio of Mn andCo were examined in terms of electrochemical and biochemicalreactions in MFCs. Moreover, the performance of Mn–Co oxide cat-alysts was compared with that of platinum to explore the possibil-ity of using Mn–Co oxide to replace platinum as cost-effectivecathode catalyst in MFC applications.

2. Methods

2.1. Catalyst preparation and characterization

Spinel Mn–Co oxides, with Mn/Co atomic ratios of 0.5, 1 and 2,have been prepared by a modified scheme of method described byJinénez Mateos et al. (1989). Briefly, different atomic ratios of Mnand Co carbonate precursors were mixed and heat treated at400 �C for 3 h in air atmosphere. The prepared samples were la-beled as MnCo-0.5, MnCo-1 and MnCo-2, for Mn/Co atomic ratiosof 0.5, 1 and 2, respectively, for simplicity. All chemicals used werereagent grade and used without any further purification.

Prepared samples were characterized using X-ray diffraction(XRD) and Brunauer–Emmer–Teller (BET) surface area measure-ments. The XRD patterns were collected on a Bruker diffractometer(Germany) with graphite monochromatized CuKa radiation. Thesurface areas of oxide powders were measured according to theBET equation by N2 physisorption at 77.4 K using a QuantachromeNOVA automated gas sorption system (Gad-Allah et al., 2008).

2.2. Cathode electrodes preparation

Cathode electrodes for MFC studies were prepared using nonwet-proofed carbon paper (E-Tek). On the air facing side, fourpoly-tetrafluoroethylene (PTFE) diffusion layers were appliedaccording to previously reported procedures (Cheng et al., 2006).

The catalyst layer was prepared by mixing of Mn–Co oxide catalystwith carbon Vulcan XC-72R and Nafion solution (5% Nafion solu-tion from Aldrich) to form catalyst ink which painted on the sideof the carbon paper exposed to solution. The catalyst ink formula-tion was: 6.25 mg of catalyst, 16.66 mg of carbon Vulcan XC-72R,270 lL of isopropyl alcohol, 217 lL of 5% Nafion solution and135 lL of double distilled water. The mixture was ultrasonicatedfor 30 min. For MFC cathode fabrication, the catalyst loading(Mn–Co oxide) was 0.5 mg/cm2. The first layer was obtained byspreading the ink on carbon paper electrode area, using amaskandabrush. Several successive layers were then deposited on top of eachother to reach the loading of 0.5 mg/cm2. It was ascertained thatthe previous layers were dry before adding another layer. Pt wasused as a control for the cathode catalysts. The platinum cathodewas made following the previously described method, in which0.5 mg Pt/cm2 catalyst was bonded on a carbon paper electrodeusing a Nafion solution. All electrodes were dried at room temper-ature for 24 h before use.

2.3. MFC set-up and operation

All MFC tests were carried out using a single-chamber air cath-ode MFC with effective volume of 38 mL. A cylindrical MFC cham-ber with a diameter of 4.0 cm and a length of 3.0 cm was made ofplexiglass. Non wet-proofed carbon paper (E-Tek) was used as ananode without further treatment. The anode and cathode wereplaced on opposite sides with the oxygen catalyst coating layer fac-ing the anode and the polytetrafluoroethylene (PTFE) coated gasdiffusion layer facing air. Prior to startup, each of the MFCs wasfilled with 32 mL of molasses-laden synthetic wastewater withan initial chemical oxygen demand (COD) of 1650 ± 300 mg/L andprepared in phosphate buffered solution (PBS) containing Na2HPO4

(6.02 g), KH2PO4 (1.024 g), NH4Cl (0.41 g) and mineral media(10 mL). The mineral media had the following composition in1 L: EDTA (0.5 g), CoCl2�6H2O (0.082 g), CaCl2�2H2O (0.114 g),H3BO3 (0.01 g), Na2MoO4�2H2O (0.02 g), Na2SeO3 (0.001 g), Na2-

WO4�2H2O (0.01 g), NiCl2�6H2O (0.02 g), MgCl2 (1.16 g),MnCl2�4H2O (0.59 g), ZnCl2 (0.05 g), CuSO4�5H2O (0.01 g), and Al-K(SO4)2 (0.01 g). The source of microorganisms was a digestedsludge (6 ml) collected from anaerobic reactor treating municipalwastewater (2.5 g VSS/L) (Mahmoud et al., 2011). MFCs were oper-ated in fed-batch mode at room temperature (25 ± 2 �C) and fedsolution was replaced when the cell voltage dropped below30 mV, forming one operating cycle. An external resistance (Rext)of 0.5 kX was used unless otherwise stated. The performance ofMFCs was evaluated with respect to power generation and sub-strate degradation. The experiments were repeated six times.

2.4. Electrochemical cell test

The cathode current was measured by applying different volt-age in potentiometry tests (linear sweep voltammetry (LSV)) usinga three-chambered electrochemical cell containing a working elec-trode (cathode electrode with 0.5 cm2 projected surface area), aplatinum plate counter electrode, and an Ag/AgCl reference elec-trode. The catalyst-coated side of the cathode was placed facingthe solution, with the uncoated side exposed directly to air. LSVstudies were conducted using a potentiostat (VoltaLab 06) byapplying potential from 0.4 to �0.6 V with a scan rate of 10 mV/sand recording the current. A curve of the cathode potentials againstcurrent densities was used to evaluate the performance of cath-odes. For a fixed current density, the higher the potential the betterthe performance of the cathode material. All the measurementswere conducted at room temperature (25 ± 2 �C) in oxygen satu-rated PBS.

M. Mahmoud et al. / Bioresource Technology 102 (2011) 10459–10464 10461

2.5. Analysis and calculations

Cell voltage across external resistance was measured and re-corded every 5 min using a data acquisition system (LabJack U6)connected to a personal computer. Current density (mA/m2),power density (mW/m2) and coulombic efficiency (CE) were calcu-lated as previously described (Logan et al., 2006). The polarizationcurve was plotted to determine the internal resistance (Rin) and thepower generation at different external resistances (Rext). Polariza-tion curves were obtained by varying external resistance from12,000 to 50 X with each resistor used per batch cycle. Rin was cal-culated at the maximum power point on the plot where Rin wasequal to Rext (Logan et al., 2006).

Moreover, chemical oxygen demand (COD), biochemical oxygendemand (BOD5) and total suspended solids were analyzed accord-ing to the standard methods (APHA, 2005).

2.6. Scanning electron microscopy (SEM) observation of the cathodes

The morphology of cathodes was observed using SEM (JEOL,JXA-840A) according to the method described by Mahmoud et al.(2010). Briefly, the cathodes were fixed for 4 h with 2% (w/v) glu-taraldehyde in phosphate buffer at room temperature, and dehy-drated through a graded series of water–ethanol mixtures of 30%,50%, 60%, 70%, 80%, 90%, and 100%, for 20 min per step. The sam-ples were kept in each mixture for 10 min, and finally dried bythe vacuum freeze-dried method (Uyanik et al., 2002). After beingsputter coated with gold particles, SEM was performed with ascanning electron microscope (JEOL, JXA-840A) at 30 kV.

3. Results and discussion

3.1. Characterization of the Mn–Co oxide cathodes

The crystalline structures of the synthesized Mn–Co oxideswere determined by XRD as shown in Fig. S1. The obtained pat-terns showed that the samples had seven main planes of (111),(220), (311), (440), (422), (511), and (440), attributable toMnCo2O4 (pdf# 84-0482). The above results indicated that the pureindividual phase of Mn–Co oxides was successfully formed via thesolid state preparation method. The BET surface areas were foundto increase with the increase of Mn content. The BET surface areaswere determined to be 46.6, 92.33 and 149.7 m2/g for 0.5, 1 and 2Mn/Co ratio, respectively.

The morphology and compositions of the cathode catalysts ana-lyzed by SEM showed that the application of Mn–Co oxide pro-duced structures having filamentous shape (Fig. S2a) which aredifferent from the Pt-coated cathode which have surface with a flatshape (figure not shown). For the Mn–Co oxide catalysts, the fila-mentous structures increase the surface area and are easier forthe organic substrates to be adsorbed on the cathodes. The highsurface areas of Mn–Co oxide catalysts could enhance the oxygenabsorption and electron acceptance on the catalysts/carbon sur-face. With the filamentous surface properties and the existenceof oxygen vacancies, the Mn–Co oxide is expected to substantiallyincrease the oxygen reaction rate and electron acceptance capabil-ity. On the other hand, the specific pore and tunnel structures ofthe Mn–Co oxide catalysts facilitated bacterial growth and adhe-sion. On the other hand, biofilm was clearly observed on the cath-ode after 300 h of operation (Fig. S2b).

3.2. Electrochemical study of oxygen reduction

The open circuit potential (OCP) was measured to evaluate theelectrochemical or biochemical reaction rates on the cathode. The

higher OCP values are related to a higher reaction rate (Logan et al.,2006). The results presented in Table 1 revealed that the OCPs ofthe air cathode catalysts were increased by increasing the atomicratio of Mn/Co, with a slightly lower value measured for Pt catalyst.The higher OCP with Mn/Co ratio of 2 could be caused by increasedamounts of oxygen adsorbed on the catalyst/carbon surface as a re-sult of higher surface area.

The oxygen reduction activities of the Mn–Co oxides and Pt cat-alyst were compared using a LSV by scanning potential from 0.4 to�0.6 V and the resulting currents are shown in Fig. 1. The onsetpotentials of MnCo-2 and MnCo-1 in LSV reduction curves weresignificantly shifted to more positive value as compared with thoseof MnCo-0.5. The MnCo-2 has a more positive onset potential thanthat of MnCo-1 (0.10 vs. �0.05 V), therefore representing higherlimiting current density due to a metal catalyst. The available re-sults showed that the order of the reduction current produced bythe Mn–Co oxide catalysts was MnCo-2 > MnCo-1 > MnCo-0.5,which was consistent with the OCP results (Table 1). The combinedresults of the LSV and OCP indicated that the MnCo-2 has the high-est catalytic activity. Moreover, among the three samples, MnCo-2had the highest current at the peak potential, likely related to itshighest BET surface area, which would favor more adsorption ofO2 on the surface of MnCo-2. Thus, it can be expected that, amongthe three samples, MnCo-2 is the most effective cathode catalyticmaterial for MFC, which is determined by the interactive effect ofBET surface area.

On the other hand, previous studies reported good oxygenreduction activities from Mn–Co oxides in alkaline media, in whichthe electrochemical reduction of O2 occurred through the oxidationof Mn (III), cyclically produced by reduction of Mn (IV) (Lima et al.,2006; Zhang et al., 2007). Consistent with previous mechanism inthe catalyzed electrodes, Roche and Scott (2009) investigated man-ganese oxides which were chemically deposited onto high specificsurface area carbon as a cathode catalyst. They stated that MnO2

was initially reduced to MnOOH, which was consequently reoxidi-zed to Mn (IV) by oxygen. Due to the resistive effects induced bythe low concentration of hydroxide ions in the solution, the ORRmechanism was difficult to determine at neutral pH. However,the mechanism for oxygen reduction at neutral pH tended to bea two-electron path because more peroxide was formed per mole-cule of reduced oxygen in the ORR potential range.

3.3. The performance of MFC with different cathodic catalysts

The preliminary results from above tests demonstrated that theMn–Co oxide had relatively high electrochemical characteristics ascompared to a Pt-coated cathode, indicating that they could beused as the cathodic catalysts in MFCs. In the next step, the Mn–Co oxide catalysts were examined in MFC and compared with Ptin terms of power density and substrate removal. The MFCs wereoperated in a batch mode for about 300 h with the syntheticmolasses wastewater as the substrate (six consecutive batches).A refill was done within each cycle before the voltage dropped be-low 30 mV.

Among the three Mn–Co oxide cathode catalysts, the MnCo-2catalyst exhibited the maximum power density generation113 mW/m2 at cell potential 279 mV (at 0.5 kX external resis-tance), which were lower than those for the Pt catalyst(148 mW/m2 at 325 mV) as shown in Table 1. The MnCo-1 andMnCo-0.5 catalysts showed lower cell potential, with the voltagesof 229 and 198 mV, respectively. The corresponding power densi-ties were 77 and 57 mW/m2, respectively. The higher cell potentialof MnCo-2 as compared to the other Mn–Co oxide cathode cata-lysts could be explained by its lower Rin (Table 1). It has been pro-ven that MFCs performance is restricted by internal resistances,also known as cathodic ohmic overpotentials (Zhao et al., 2006;

Table 1Performance of MFC based on different cathodic catalysts.

Catalysts Cathodic opencircuit potential vs.Ag/AgCl (mV)

Internalresistance(X)

Maximumpotential(mV)

Maximumpower density(mW/m2)

MnCo-0.5

116 ± 12 154 ± 25 198 ± 51 57 ± 14

MnCo-1 149 ± 14 132 ± 19 229 ± 59 77 ± 24MnCo-2 159 ± 19 91 ± 29 279 ± 41 113 ± 25Pt 167 ± 10 81 ± 24 325 ± 70 148 ± 21

Fig. 1. LSV obtained in 50 mM PBS saturated by O2 modified by different cathodiccatalysts (catalyst loading = 0.5 mg/cm2, T = 25 ± 4 �C, scan rate: 10 mV/s).

10462 M. Mahmoud et al. / Bioresource Technology 102 (2011) 10459–10464

Morris et al., 2007; Rismani-Yazdi et al., 2008). The Rin consists oftwo parts: the electrolyte ohmic loss caused by the movement ofelectrons through the electrolyte and the electrode ohmic losscaused by the movement of electrons through the electrode andwires (Li et al., 2010). Because the MFCs used in this study hadthe same configuration (e.g., electrode distance, dimension, andanode materials), the differences of Rin between MFCs may have re-sulted from the electrical characteristics of different cathodes,especially the conductivity. Regarding the Mn–Co oxide catalysts,the conductivity included two types: (i) electrical conductivity thatwas performed under normal (room) temperature; (ii) ionic con-ductivity that was performed under high temperatures (Chenet al., 2004; Li et al., 2010). The results obtained at room tempera-ture showed that the Pt, MnCo-2, MnCo-1, and MnCo-0.5 catalystshad good electrical conductivity that led to the low Rin.

The available results showed that the Mn–Co oxide cathodeshad the same reaction rate as the Pt cathode. The cycle durationof the MFC with Mn–Co oxide cathodes and Pt-coated cathodewas in range of 40–45 h. Moreover, the Mn–Co oxide catalystshad slightly higher COD removal efficiencies (86–89%) than thePt-coated cathode (87%). Similar results were observed for BOD5

and TSS (Table 2). From the presented results, we can conclude thatthe ORR of the Mn–Co oxide catalysts was comparable with thatfor the Pt-coated cathode.

On the other hand, the CE values were used to evaluate the con-version of the organic substrates to electricity. Compared to thecase with the Pt catalyst, the Mn–Co oxide catalysts had lowerpower generation but slightly higher COD removal, meaning that

Table 2Performance summary of MFC different cathodic catalysts.

Catalysts COD BOD

Influent Effluent R% Influent

MnCo-0.5 1610 ± 300 220 ± 105 86 ± 6 945 ± 751MnCo-1 1689 ± 341 189 ± 98 88 ± 7 978 ± 464MnCo-2 1675 ± 350 177 ± 130 89 ± 8 984 ± 468Pt 1620 ± 210 197 ± 102 87 ± 9 926 ± 449

the Mn–Co cathode catalysts are effective for COD removal andelectricity generation, but have lower efficiency than Pt catalystcathode at converting organic substrates to electricity. For in-stance, the CE value of the Pt-cathode was 14.8%, but the MnCo-2catalyst with the highest voltage generation among the threeMn–Co oxide catalysts only had a CE value of 12.1%. The MnCo-0.5 catalyst with the highest Rin had the lowest CE values of 6.9%.However, the MnCo-1 had CE of 10.2%. The CE values of all thecathodes tested were lower than 13%; the results were consistentwith a previous report that high COD loadings led to low CE values(Liu et al., 2005; Li et al., 2010), which was probably caused by theconsumption of organic substrates by bacterial growth rather thanby electricity generation. The biofilms were observed on the cath-odes (Fig. S2b), which indicated that the bacterial cells grew fast athigh COD concentration.

Power densities of the Mn–Co oxide catalysts and the Pt catalystwere compared using a polarization curve measurement. The Ptcathode had a higher maximum power density (330 mW/m2) thanthe Mn–Co oxide cathodes (Fig. 2a). The MnCo-2 had maximumpower density of 284 mW/m2, while MnCo-1 and MnCo-0.5 hadpower densities of 230 and 200 mW/m2, respectively. The powerdensities of the Mn–Co oxide cathodes followed the trends of theOCP values and the voltage generation (Table 1). That is, the higherOCP values and higher voltage generation were related to the high-er power density.

3.4. Mn–Co oxide as alternative cathodic catalyst

From the application point of view, costs associated with theMFC construction should be lowered when the overall MFC perfor-mance is required to be kept sustainable. Lowering the activationoverpotential for oxygen reduction with non-precious metal cata-lysts is important for the scale-up and application of MFCs (Zhaoet al., 2005; Cheng et al., 2006). It is helpful to reduce the activationenergy of oxygen reduction through developing excellent nano-structured non-precious metal catalysts (Bashyam and Zelenay,2006; Gong et al., 2009; Liu et al., 2010). In this study, an approachwas proposed in order to reduce the oxygen reduction overpoten-tial based on the active Mn–Co oxide catalysts prepared throughsolid state reaction. The use of Mn–Co oxide as a cathodic catalystenables effective oxygen reduction in the single-chamber MFC.Moreover the performance of the catalysts developed in this studywas as good as a Pt catalyzed MFC. Therefore, the Mn–Co oxide cat-alysts tested in this study exhibited a great potential to replaceplatinum as a cathode catalyst for MFC applications in large-scalewastewater treatment plants for efficient contaminant removaland high power generation.

In this work, the preliminary cost estimates for Mn–Co oxidewas 0.0058 $/cm2 as compared with 0.0447 $/cm2 of the commer-cialized Pt/C based cathode catalyst (Morris et al., 2007). The opti-mization of non-transition metal catalyst loading and heattreatment were not carried out in this paper. It is expected that fur-ther optimizing the catalyst structure and stability while consider-ing the MFC condition might lead to an improvement in catalystperformance and lower the overall cost.

TSS

Effluent R% Influent Effluent R%

88 ± 54 91 ± 6 710 ± 210 209 ± 109 71 ± 1276 ± 40 92 ± 5 725 ± 279 205 ± 104 72 ± 1076 ± 39 92 ± 5 730 ± 277 197 ± 103 73 ± 1189 ± 51 90 ± 8 726 ± 265 201 ± 100 72 ± 19

Fig. 2. Power output (a) and polarization curve (b) of the MFCs with different cathodes.

M. Mahmoud et al. / Bioresource Technology 102 (2011) 10459–10464 10463

In general, the non-noble transient metal catalysts are still be-low Pt-based catalyst in catalytic activity and stability as wasshown in this study and in the previous reports. In this work,the MnCo-2 in MFC test produces 23.6% lower power density thanPt/C (based on cathode electrode with external resistance of0.5 kX). In addition, transition alloy catalysts might be able to re-duce Pt loading and/or replace expensive noble catalysts in PEMfuel cells; thus, this technical advancement can be applied toMFCs.

4. Conclusion

In this study, Mn–Co oxides with different atomic ratio of Mnand Co were examined in order to replace platinum as the cathodiccatalyst in MFCs application. From the experimental results, therewere three major conclusions drawn:

First, increasing the Mn content enhanced power generationdue to the increase in BET surface area.

Second, compared with a platinum catalyst, the Mn–Co oxidecatalyst catalysts showed good activity toward oxygen reductionin neutral media.

Third, since the cost of Mn–Co oxide is expected to be just afraction (�1/10) of that of Pt, the Mn–Co oxide catalysts tested inthis study exhibited a great potential to replace platinum as a cath-ode catalyst for MFC applications.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.biortech.2011.08.123.

References

Aelterman, P., Versichele, M., Marzorati, M., Boon, N., Verstraete, W., 2008. Loadingrate and external resistance control the electricity generation of microbial fuelcells with different three-dimensional anodes. Bioresour. Technol. 99, 8895–8902.

APHA, 2005. Standard Methods for the Examination of Water and Wastewater, 21sted. American Public Health Association, Washington, DC.

Bashyam, R., Zelenay, P., 2006. A class of non-precious metal composite catalysts forfuel cells. Nature 443, 63–66.

Chen, L., Chen, C.L., Huang, D.X., Lin, Y., Chen, X., Jacobson, A.J., 2004. Hightemperature electrical conductivity of epitaxial Gd-doped CeO2 thin films. SolidState Ionics 175, 103–106.

10464 M. Mahmoud et al. / Bioresource Technology 102 (2011) 10459–10464

Cheng, S., Liu, H., Logan, B.E., 2006. Power densities using different cathode catalysts(Pt and CoTMPP) and polymer binders (nafion and PTFE) in single chambermicrobial fuel cells. Environ. Sci. Technol. 40, 364–369.

Cohen, B., 1931. The bacterial culture as an electrical half-cell. J. Bacteriol. 21, 18–19.Duteanu, N., Erable, B., Kumar, S.M.S., Ghangrekar, M.M., Scott, K., 2010. Effect of

chemically modified Vulcan XC-72R on the performance of air-breathingcathode in a single-chamber microbial fuel cell. Bioresour. Technol. 101,5250–5255.

Gad-Allah, T.A., Fujimura, K., Kato, S., Satokawa, S., Kojima, T., 2008. Preparation andcharacterization of magnetically separable photocatalyst (TiO2/SiO2/Fe3O4):effect of carbon coating and calcination temperature. J. Hazard. Mater. 154,572–577.

Gong, K.P., Du, F., Xia, Z.H., Durstock, M., Dai, L.M., 2009. Nitrogen-doped carbonnanotube arrays with high electrocatalytic activity for oxygen reduction.Science 323, 760–764.

Jang, J.K., Pham, T.H., Chang, I.S., Kang, K.H., Moon, H., Cho, K.S., Kim, B.H., 2004.Construction and operation of a novel mediator- and membrane-less microbialfuel cell. Process Biochem. 39, 1007–1012.

Jinénez Mateos, J.M., Morales, J., Tirado, J.L., 1989. Cation deficient Mn, Co spineloxides obtained by thermal decomposition of carbonate precursors. J. SolidState Chem. 82, 87–94.

Lefebvre, O., Ooi, W.K., Tang, Z., Abdullah-Al-Mamun, M., Chua, D.H.C., Ng, H.Y.,2009. Optimization of a Pt-free cathode suitable for practical applications ofmicrobial fuel cells. Bioresour. Technol. 100, 4907–4910.

Li, X., Hu, B., Suibb, S., Lei, Y., Li, B., 2010. Manganese dioxide as a new cathodecatalyst in microbial fuel cells. J. Power Sources 195, 2586–2591.

Lima, F.H.B., Calegaro, M.L., Ticianelli, E.A., 2006. Investigations of the catalyticproperties of manganese oxides for the oxygen reduction reaction in alkalinemedia. J. Electroanal. Chem. 590, 152–160.

Liu, H., Logan, B.E., 2004. Electricity generation using an air-cathode single chambermicrobial fuel cell in the presence and absence of a proton exchange membrane.Environ. Sci. Technol. 38, 4040–4046.

Liu, H., Cheng, S.A., Logan, B.E., 2005. Production of electricity from acetate orbutyrate using a single-chamber microbial fuel cell. Environ. Sci. Technol. 39,658–662.

Liu, H., Ramnarayanan, R., Logan, B.E., 2004. Production of electricity duringwastewater treatment using a single chamber microbial fuel cell. Environ. Sci.Technol. 38, 2281–2285.

Liu, X.W., Sun, X.F., Huang, Y.X., Sheng, G.P., Zhou, K., Zeng, R.J., Fang, D., Wang, S.G.,Xu, A.W., Tong, Z.H., Yu, H.Q., 2010. Nano-structured manganese oxide as acathodic catalyst for enhanced oxygen reduction in a microbial fuel cell fed witha synthetic wastewater. Water Res. 44, 5298–5305.

Logan, B.E., Aelterman, P., Hamelers, B., Rozendal, R., Schröeder, U., Keller, J., Freguia,S., Verstraete, W., Rabaey, K., 2006. Microbial fuel cells: methodology andtechnology. Environ. Sci. Technol. 40, 5181–5192.

Lovely, D.R., 2006. Bug juice. Harvesting electricity with microorganisms. Nat. Rev.Microbiol. 4, 497–508.

Mahmoud, M., Tawfik, A., El-Gohary, F., 2011. Use of down-flow hanging sponge(DHS) reactor as a promising post-treatment system for municipal wastewater.Chem. Eng. J. 168, 535–543.

Mahmoud, M., Tawfik, A., El-Gohary, F., 2010. Simultaneous organic and nutrientremoval in a naturally ventilated bio-tower treating pre-settled municipalwastewater. J. Environ. Eng. ASCE 136, 301–307.

Morris, J.M., Jin, S., Wang, J.Q., Zhu, C.Z., Urynowicz, M.A., 2007. Lead dioxide as analternative catalyst to platinum in microbial fuel cells. Electrochem. Commun.9, 1730–1734.

Nissinen, T., 2004. Preparation and characterisation of MnCo2O4 and Raney nickelcatalysts for alkaline fuel cells. Ph.d. thesis, University of Helsinki, Finland.

Nissinen, T., Valo, T., Gasik, M., Rantanen, J., Lampinen, M., 2002. Microwavesynthesis of catalyst spinel MnCo2O4 for alkaline fuel cell. J. Power Sources 106,109–115.

Pham, T.H., Jang, J.K., Chang, I.S., Kim, B.H., 2004. Improvement of cathodereaction of a mediatorless microbial fuel cell. J. Microbiol. Biotechnol. 14,324–329.

Rabaey, K., Clauwaert, P., Aelterman, P., Verstraete, W., 2005. Tubular microbialfuel cells for efficient electricity generation. Environ. Sci. Technol. 39, 8077–8082.

Restovic, A., Rı́os, E., Barbato, S., Ortiz, J., Gautier, J.L., 2002. Oxygen reduction inalkaline medium at thin MnxCo3�xO4 (0 6 x 6 1) spinel films prepared by spraypyrolysis. Effect of oxide cation composition on the reaction kinetics. J.Electroanal. Chem. 522, 141–151.

Rismani-Yazdi, H., Carver, S.M., Christy, A.D., Tuovinen, O., 2008. Cathodiclimitations in microbial fuel cells: an overview. J. Power Sources 180, 683–694.

Roche, I., Scott, K., 2009. Carbon-supported manganese oxide nanoparticles aselectrocatalysts for oxygen reduction reaction (orr) in neutral solution. J. Appl.Electrochem. 39, 197–204.

Rozendal, R.A., Hamelers, H.V.M., Rabaey, K., Keller, J., Buisman, C.J.N., 2008.Towards practical implementation of bioelectrochemical wastewatertreatment. Trends Biotechnol. 26, 450–459.

Selembo, P.A., Merrill, M.D., Logan, B.E., 2009. The use of stainless steel and nickelalloys as low-cost cathodes in microbial electrolysis cells. J. Power Sources 190,271–278.

Tartakovsky, B., Guiot, S.R., 2006. A comparison of air and hydrogen peroxideoxygenated microbial fuel cell reactors. Biotechnol. Prog. 22, 241–246.

Uyanik, S., Sallis, P.J., Anderson, G.K., 2002. The effect of polymer addition ongranulation in an anaerobic baffled reactor (ABR) Part II: compartmentalizationof bacterial populations. Water Res. 36, 944–955.

You, S., Zhao, Q., Zhang, J., Jiang, J., Zhao, S., 2006. A microbial fuel cell usingpermanganate as the cathodic electron acceptor. J. Power Sources 162, 1409–1415.

Yu, E.H., Cheng, S., Scott, K., Logan, B., 2007. Microbial fuel cell performance withnon-Pt cathode catalysts. J. Power Sources 171, 275–281.

Wen, T.C., Kang, H.M., 1998. Co–Ni–Cu ternary spinel oxide-coated electrodes foroxygen evolution in alkaline solution. Electrochim. Acta 43, 1729–1745.

Zhang, D., Chi, D., Okajima, T., Ohsaka, T., 2007. Catalytic activity of dual catalystssystem based on nano-manganese oxide and cobalt octacyanophthalocyaninetoward four-electron reduction of oxygen in alkaline media. Electrochim. Acta52, 5400–5406.

Zhao, F., Harnisch, F., Schroder, U., Scholz, F., Bogdanoff, P., Herrmann, I., 2005.Application of pyrolysed iron(II) phthalocyanine and CoTMPP based oxygenreduction catalysts as cathode materials in microbial fuel cells. Electrochem.Commun. 7, 1405–1410.

Zhao, F., Harnisch, F., Schröder, U., Scholz, F., Bogdanoff, P., Herrmann, I., 2006.Challenges and constraints of using oxygen cathodes in microbial fuel cells.Environ. Sci. Technol. 40, 5193–5199.