Bioresource Technology 134 (2013) 251–256

Contents lists available at SciVerse ScienceDirect

Bioresource Technology

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

Improving electricity production in tubular microbial fuel cells throughoptimizing the anolyte flow with spiral spacers

0960-8524/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.biortech.2013.02.010

⇑ Corresponding author. Tel.: +1 414 229 5846; fax: +1 414 229 6958.E-mail address: zhenhe@uwm.edu (Z. He).

Fei Zhang a, Zheng Ge a, Julien Grimaud b, Jim Hurst b, Zhen He a,⇑a Department of Civil Engineering and Mechanics, University of Wisconsin–Milwaukee, Milwaukee, WI 53211, United Statesb Veolia Water North America, 101 W Washington Street, Indianapolis, IN 46204, United States

h i g h l i g h t s

" The spiral spacers improve electricity production in tubular microbial fuel cells." The spiral spacers are effective in both vertical and horizontal installations." The spiral spacers (with straight electrode) perform better than spiral electrode." The onsite test in a wastewater plant confirms the benefits of the spiral spacers.

a r t i c l e i n f o

Article history:Received 18 December 2012Received in revised form 30 January 2013Accepted 1 February 2013Available online 9 February 2013

Keywords:Microbial fuel cellsSpiral spacersEnergyWastewater treatment

a b s t r a c t

The use of spiral spacers to create a helical flow for improving electricity generation in microbial fuel cells(MFCs) was investigated in both laboratory and on-site tests. The lab tests found that the MFC with thespiral spacers produced more electricity than the one without the spiral spacers at different recirculationrates or organic loading rates, likely due to the improved transport/distribution of ions and electronmediators instead of the substrates because the organic removal efficiency was not obviously affectedby the presence of the spiral spacers. The energy production in the MFC with the spiral spacers reached0.071 or 0.073 kWh/kg COD in either vertical or horizontal installment. The examination of the MFCsinstalled in an aeration tank of a municipal wastewater treatment plant confirmed the advantage of usingthe spiral spacers. Those results demonstrate that spiral spacers could be an effective approach toimprove energy production in MFCs.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

A microbial fuel cell (MFC) is a promising technology that canbe applied to wastewater treatment for energy-efficient pollutantremoval. In the past decade, researchers have made significant pro-gress towards understanding fundamental issues of microbiology,electrochemistry and reactor architecture in MFCs (Arends andVerstraete, 2012). However, MFC development is still hinderedby challenges such as system scaling up and further improvementof electric energy. The power density >1 kW/m3 has been achievedin some studies with very small-scale MFCs (Fan et al., 2007), butlarge-scale MFCs (>1 L) generally had low power output. Substan-tial efforts have been made to improve the power output in MFCsthrough modifying or pre-treating electrode materials, using high-efficiency separators, optimizing reactor configuration, and select-ing efficient microorganisms (Ci et al., 2012; Logan and Rabaey,2012; Zhang et al., 2009).

Optimizing operating conditions is another important approachto improve MFC performance. Besides factors like temperature andpH, using mixing intensity to improve mass transfer could be aneffective method to improve the performance in continuouslyoperated MFCs. It was reported that a higher mixing intensitythrough applying a higher shear rate optimized biofilm formationand thus improved the activities of the electrochemically-activemicrobes in an MFC (Pham et al., 2008). A study of a continuouslyoperated liter-scale MFC found that the improvement of powerproduction via adjusting the mixing intensity was affected by thesubstrate loading rates and higher recirculation rates might notbe effective to increase electricity generation under some condi-tions (Zhang et al., 2010). Those results demonstrate the impor-tance of the flow pattern of the anolyte to MFC performance andalso reveal the problems of MFCs containing high surface area elec-trodes. The high surface area electrodes, such as carbon-fiber basedbrush electrodes, have been proved effective in improving electric-ity generation in MFCs (Logan et al., 2007). Many studies that wereconducted in small-size MFCs had good mixing of their anolytesand did not have obvious issues with the mass transfer of

252 F. Zhang et al. / Bioresource Technology 134 (2013) 251–256

substrates and ions. However, in a large-scale MFC, substrates/ionscould be unevenly distributed inside the anode compartment,thereby creating dead zones where substrates/ions supply isinsufficient and electrode surface area is not efficiently used forelectricity generation. For example, in a tubular MFC filled withhigh-density electrode materials, the anolyte will likely go fromthe inlet to the outlet through a pathway with less hydraulicresistance, which only occupies the part of the interior space ofthe anode compartment. Other parts of the anode space that donot receive an active supply of substrates/ions will have to relyon slow diffusion and may have microbes under a starving condi-tion or higher electrolyte resistance, resulting in a low efficiency ofmicrobial activity and electrode use. Therefore, it is necessary tooptimize the flow of the anolyte as well as the substrates/ions’ dis-tribution in continuously operated MFCs containing high surfacearea electrodes.

The anolyte flow can be controlled by designing flow channelson the anode electrode (Min and Logan, 2004), but it limited theapplication of high surface area electrodes. The recent develop-ment of spiral anodes in MFCs has aimed to optimize the anolyteflow for higher electricity generation. A spiral anode channel wascreated by using graphite-coated stainless steel mesh and thisMFC achieved a good performance of both waste treatment andelectricity generation from dairy wastewater (Mardanpour et al.,2012). Another study used an ion exchange membrane to createa spiral channel with carbon cloth as electrodes, which signifi-cantly improved power density compared with conventionaltwo-chamber MFCs (Jia et al., 2012). Both of those studies devel-oped round-disk shape MFCs that present great challenges in beingscaled up to a continuously operated system for practical wastewa-ter treatment. A more detailed study of the anolyte flow pathwaywas reported in a tubular MFC containing a helical anode electrodethat created a helical flow channel (Kim et al., 2012). Their resultsrevealed that the flow pattern improved mass transfer, therebyresulting in more power output. However, the potential issue withthis helical anode electrode is that its manufacturing procedurecould be complicated and its surface area is limited by the carbonmaterials that are used.

In this study, we took advantage of the concept of a helical flowpattern and attempted to use simple spiral spacers to improveelectricity generation in tubular MFCs. Instead of creating spiralelectrodes, those spiral spacers were adapted to the well-provencarbon brush electrodes; thus, they maintained the feature of ahigh surface area of the carbon brush while creating a helical flow

Fig. 1. Preparation of electrodes and the MFCs: (A) spiral spacers made of rubber materporous PVC sleeve for the on-site test; and (D) the assembled MFC for the on-site test.

pattern. The superior performance of the spiral spacers was dem-onstrated through comparison of the MFCs with and without spiralspacers in both laboratory tests and onsite investigation. The labexperiments examined the effects of recirculation rates, organicloading rates, and different installation positions (vertical and hor-izontal). In comparison, an MFC with a spiral anode electrode (car-bon brush also was made into a spiral shape) was studied in thehorizontal installation. The onsite test was conducted by installingtwo MFCs (with and without spiral spacers) in an aeration tank of amunicipal wastewater treatment plant for treating primary efflu-ent. The results were expected to provide a simple and feasible ap-proach to produce more energy through optimizing the anolyteflow in tubular MFCs.

2. Methods

2.1. Lab MFCs setup and operation

Two tubular MFCs (MFClab-1 and MFClab-2) were constructed byrolling up a piece of cation exchange membrane (CEM, MembraneInternational Inc., Ringwood, NJ, USA) around a PVC tube with a3.8-cm diameter that had a length of 70 cm and 1.0-cm holesthroughout the tube. The PVC tube functioned as supporting mate-rial for the CEM tube that contained a 1-m long carbon brush as ananode electrode. The liquid volume of the CEM tube (anode com-partment) was about 1.15 L. The carbon brushes were pretreatedas previously before being used (Wang et al., 2009). The spiralspacers were made of round-shape rubber plates with 4.5 cm indiameter and �2 cm in distance between each plate (Fig. 1A).The rubber plates were connected with titanium wires, and the to-tal number of spacers for one anode electrode was 35. The spiralspacers were installed to the anode electrode of the MFClab-1(Fig. 1B), while the MFClab-2 acted as a control without the spiralspacers. The cathode electrode was a piece of carbon cloth(20 � 70 cm, Zoltek Corporation, St. Louis, MO, USA) containing5 mg/cm2 activated carbon powder (Thermo Fisher Scientific,USA) as a catalyst for oxygen reduction. The activated carbon pow-der was coated to the cathode electrode by using a 10% PTFE solu-tion as a binder agent and heat-treated at 375 �C for half hour. Forvertical installation, each MFC was set up in a PCV tube that had adiameter of 7.6 cm and functioned as a cathode compartment witha liquid volume of 1.3 L. The cathode was aerated with the air at100 mL/min. For horizontal installation, both MFCs were laid down

ials; (B) spiral spacers installed onto a straight carbon brush; (C) tubular MFC and

F. Zhang et al. / Bioresource Technology 134 (2013) 251–256 253

with about 2� angles respective to the horizontal level and sub-merged in a tank with a liquid volume of 25 L that was aeratedwith the air at 200 mL/min. The anode and the cathode electrodeswere connected to an external circuit across a resistor of 10 O, un-less stated otherwise.

Both MFCs were continuously operated under the same condi-tion at room temperature �20 �C. The anodes were inoculated withthe digested sludge collected from the South Shore Waste Recla-mation Facility (Milwaukee, WI, USA). A synthetic solution wasused as an anolyte containing (per liter of tap water): CH3COONa,1 g; NaCl, 0.5 g; MgSO4, 0.015 g; CaCl2, 0.02 g; KH2PO4, 0.53 g;K2HPO4, 1.07 g; NaHCO3, 1 g; and trace element, 1 mL (He et al.,2006). The anolyte feeding rate ranged from 0.6 to 2.4 mL/min,resulting in a hydraulic retention time (HRT) of 32 to 8 h. Tap waterwas used as a catholyte in both MFCs and was fed at the samespeed as the anolytes. The anolytes were recirculated at 50, 150or 300 mL/min.

2.2. Onsite MFCs setup and operation

Two tubular MFCs (MFConsite-1 and MFConsite-2) were con-structed using CEM tubes with a length of 100 cm and a diameterof 5 cm. No PVC tube was placed inside the CEM tube. The anodecompartment that had a liquid volume of 2.0 L contained a 1-mlong pre-treated carbon brush as the anode electrode. The cathodeelectrode was a piece of carbon cloth that wrapped the CEM tubeand was coated with 5 mg/cm2 activated carbon powder in thesame procedure as the MFCs in the lab. The spiral spacers were in-stalled in the anode of the MFConsite-1, while the MFConsite-2 actedas a control for comparison. The MFCs were placed in a PVC-tubesleeve that had a diameter of 7.6 cm and a length of 100 cm; thePVC tube contained 2.2-cm holes throughout (Fig. 1C and D). Thecompleted MFCs were installed in an aeration tank (submergedin water) at the South Shore Waste Reclamation Facility. The an-odes of both MFCs were not particularly inoculated. The primaryeffluent pumped from a sample site was fed into the MFCs at3 mL/min, resulting in an anolyte HRT of 11.1 h. The anolytes wererecirculated at 200 mL/min by a peristaltic pump. The MFCs tookadvantage of aeration in the aeration tank for the oxygen supplyto their cathode electrodes.

2.3. Measurement and analysis

The MFC voltages were monitored by digital meters (2700,Keithley Instruments, Inc., Cleveland, OH, USA) every 5 min. Theconcentration of chemical oxygen demand (COD) was measuredusing a colorimeter (DR/89, Hach Company, Loveland, CO, USA)according to the manufacturer’s procedure. Polarization tests wereconducted by using a potentiostat (Reference 600, Gamry Instru-ments, Warminster, PA, USA) at a scan rate of 0.1 mV/S. Power den-sity, current density, and COD loading and removal rate werecalculated based on the liquid volume of the anode compartment.Columbic efficiency was calculated according to the followingequation:

CE ¼Qoutput

Q input¼

PI � t

96485� CODtotal � 4

where CE is the coulombic efficiency based on organic substrate,Qoutput is the produced charge, Qinput is the total charge availablein the substrate that has been removed, and t (s) is the time.CODtotal (mol) is the total COD removed by the MFC in the periodof time t.

The theoretical power requirement for the pumping system wasestimated as (Kim et al., 2011):

Ppumping ¼QcE1000

where P is the power requirement (kW), Q is the flow rate (m3/s), cis 9800 N/m3, and E is the hydraulic pressure head (m). In thisstudy, we estimated hydraulic pressure heads of 0.03 m and0.05 m for the anolyte feeding and recirculation pumps. The energyconsumption by aeration was estimated according to a previouspublication (Larson et al., 2007).

3. Results and discussion

3.1. Vertical installment

Vertical installation is commonly used in bioelectrochemicalsystems with a upflow configuration (Deng et al., 2010; He et al.,2006; Jacobson et al., 2011; Jana et al., 2010; Sukkasem et al.,2011; Zhang et al., 2010). When the two MFCs were set up verti-cally, we examined the effects of the anolyte recirculation ratesand the organic loading rates (or HRTs) on their performance ofelectricity generation and organic removal.

Three recirculation rates, including 50, 150 and 300 mL/min,were tested at a fixed HRT of 15 h. The corresponding upflowspeeds at those recirculation rates are 1.9, 5.7 and 11.3 m/h. Afterthe MFCs achieved stable electricity generation at an externalresistance of 10 O, polarization curves were constructed to evalu-ate the overall power production (Fig. 2). At 50 mL/min, the maxi-mum power density and the maximum current density of theMFClab-1 were 4.9 W/m3 and 43.1 A/m3, respectively, higher than3.1 W/m3 and 24.7 A/m3 of the MFClab-2, demonstrating that thespiral spacers improved electricity generation in an MFC(Fig. 2A). The advantage of the MFClab-1 became greater with an in-creased recirculation rate, and at 300 mL/min, the maximumpower and the maximum current density of the MFClab-1 reached7.1 W/m3 and 62.6 A/m3; at the same recirculation rate, theMFClab-2 produced 4.5 W/m3 and 29.2 A/m3 (Fig. 2C). The COD re-moval efficiency was not obviously different between the twoMFCs at the same recirculation rate but a higher recirculation rateimproved COD removal in both MFCs. For example, at 50 mL/min,the MFClab-1 removed 78.4 ± 0.8% and the MFClab-2 removed79.9 ± 2.7% of the total COD; when the recirculation rate increasedto 300 mL/min, the two MFCs removed 87.8 ± 1.7% and 85.7 ± 1.1%,respectively. Those COD results suggest that the spiral spacersmight not improve the substrate supply to microorganisms, differ-ent from what we expected, although there is a chance that thehelical flow promoted the substrate distribution to electrochemi-cally-active bacteria but further evidence is needed. The improvedelectricity generation with the spiral spacers indicated that elec-tricity production might not be directly limited by microbial activ-ity, and the modified anolyte flow might have accelerated thetransport of ions and chemicals that acted as electron mediators,both of which are key factors to electricity generation.

The effect of organic loading rates was examined through vary-ing the influent flow rate from 0.6 to 2.4 mL/min, resulting in threeHRTs of 32, 15 and 8 h, and the corresponding loading rates rang-ing from 0.57 to 2.30 kg COD/m3/day. A fixed recirculation rate of300 mL/min was applied for the organic loading rate tests. Underthe operation at an external resistor of 10 O, the MFClab-1 produced14.9 ± 0.8 mA and the MFClab-2 generated 9.8 ± 3.7 mA at 0.57 kgCOD/m3/day (or HRT 32 h); both MFCs achieved almost 100% re-moval of the COD. When the organic loading rate increased to2.30 kg COD/m3/day (or HRT 8 h), the MFClab-1 produced22.8 ± 1.1 mA, much higher than 15.4 ± 1.1 mA in the MFClab-2,while the COD removal was similar between the two MFCs (variedbetween 64% and 66%). The overall electricity generation in theMFCs was shown in the polarization curves (Fig. 3). Clearly, the

Fig. 2. The voltage and power curves of the MFClab-1 (blue solid line) and theMFClab-2 (red dash line) at different anolyte recirculation rates: (A) 50 mL/min; (B)150 mL/min; and (C) 300 mL/min. (For interpretation of the references to color inthis figure legend, the reader is referred to the web version of this article.)

Fig. 3. The voltage and power curves of the MFClab-1 (blue solid line) and theMFClab-2 (red dash line) at different organic loading rates (or HRTs): (A) 0.57 kgCOD/m3/day (33 h); (B) 1.14 kg COD/m3/day (15 h); and (C) 2.30 kg COD/m3/day(8 h). (For interpretation of the references to color in this figure legend, the reader isreferred to the web version of this article.)

254 F. Zhang et al. / Bioresource Technology 134 (2013) 251–256

MFClab-1 had outcompeted the MFClab-2, confirming that the spiralspacers were beneficial to electricity generation. The power outputincreased with the increased organic loading rates (or decreasedHRTs) because of more substrate supply. For instance, the maxi-mum power density of the MFClab-1 increased from 4.0 W/m3 at0.57 kg COD/m3/day (or HRT 32 h) to 8.2 W/m3 at 2.30 kg COD/m3/day (or HRT 8 h).

An effective approach to evaluate the electricity generation inan MFC is to establish an energy balance. Energy analysis has beenmissing in the MFC studies for a long time but is clearly important(He, 2013). It was only recently that energy balances have been re-ported in MFCs (Ge et al., 2013a,b; Xiao et al., 2012). In this study,we have built energy balances for both MFCs under a few condi-tions. In the vertical installation, the energy balance was analyzedat a recirculation rate of 300 mL/min and an organic loading rate of2.30 kg COD/m3/day (or HRT 8 h). The MFClab-1 produced an en-ergy intensity of 0.071 kWh/kg COD (or 0.036 kWh/m3), whilethe MFClab-2 produced only 0.033 kWh/kg COD (or 0.016 kWh/m3) (Table 1). The overall energy balances were negative for bothMFCs, but the MFClab-1 had a less negative balance because ofmore energy production. The aeration accounted for 70% of the

energy consumption; without aeration, the energy balances basedon the pumping system would be positive for the MFClab-1 but stillnegative in the MFClab-2. Therefore, to achieve an energy-neutral(or surplus) treatment process using the MFC technology, aerationmust be eliminated or maintained at a minimum. The non-aerationcathode can be accomplished through a passive air supply that wasdemonstrated in a previous tubular MFC (Zhang et al., 2010).

3.2. Horizontal installment

Some tubular MFCs were operated in a horizontal position(Zhuang et al., 2012a,b). Horizontal installation could be moreadvantageous over vertical installation when multiple MFCs areconnected in a series and the produced biogas needs to be drivenout of the tubular reactor. Therefore, we also compared the perfor-mance of the MFClab-1 and the MFClab-2 when they were horizon-tally installed in a water tank containing tap water as thecatholyte. A fixed recirculation rate of 300 mL/min and an organicloading rate of 2.30 kg COD/m3/day (or HRT 8 h) were employedfor the test. At an external resistance of 10 O, the MFClab-1

Table 1Analysis of energy production and consumption in the MFCs under a certain conditions. The unit of energy is kWh/kg COD.

Production Consumption Energy balance

Pumps Aeration Total Pumps only Total

Vertical installmentMFClab-1 0.071 0.034 0.081 0.115 0.037 �0.044MFClab-2 0.033 0.035 0.082 0.117 �0.002 �0.084MFConsite-1 0.205 0.141 – 0.141 0.064 0.064MFConsite-2 0.053 0.130 – 0.130 �0.077 �0.077

Horizontal installmentMFClab-1 0.073 0.035 0.085 0.120 0.038 �0.047MFClab-2 0.028 0.036 0.087 0.123 �0.008 �0.095New MFClab-2 0.043 0.036 0.088 0.124 0.007 �0.081

Fig. 4. The voltage and power curves of the MFClab-1 (green solid line) and theMFClab-2 (red dotted line) in the horizontal installation. The new MFClab-2 (bluedash line) contained a spiral anode electrode, as shown in the inset figures. (Forinterpretation of the references to colour in this figure legend, the reader is referredto the web version of this article.)

F. Zhang et al. / Bioresource Technology 134 (2013) 251–256 255

produced 22.6 ± 0.9 mA, higher than 13.7 ± 0.7 mA in the MFClab-2,demonstrating that the spiral spacers were also effective to im-prove electricity generation in the horizontal installation. TheCOD removal was similar between the two MFCs, varying between62% and 64%. The maximum power density of the MFClab-1 was8.8 W/m3, about 1.87 times the one of the MFClab-2 (4.7 W/m3)(Fig. 4). The energy production in the MFClab-1 was 0.073 kWh/kg COD, 160% higher than 0.028 kWh/kg COD in the MFClab-2. Sim-ilar to the vertical installation, the MFClab-1 achieved a less nega-tive energy balance than the MFClab-2 (Table 1).

Because the previous study showed the improved electricityproduction with spiral anode electrodes (Kim et al., 2012), it couldbe of interest to investigate whether having a carbon brush in aspiral arrangement along the spiral spacers would further improvethe MFC performance. To do this, we twisted a carbon brush thatwas similar to the anode electrode of the MFClab-1 and modifiedthe MFClab-2 with this spiral anode electrode (and the spiral spac-ers) (the inset pictures of Fig. 4). The new MFClab-2 produced moreelectricity than the previous one (Fig. 4), indicating that the spiralarrangement inside an anode compartment indeed helped to im-prove electricity production. However, the spiral anode electrodedid not exhibit superior performance to that of the spiral spacersonly. The maximum power density of the new MFClab-2 was6.4 W/m3, lower than that of the MFClab-1. Likewise, the energyproduction in the new MFClab-2 was 0.043 kWh/kg COD, also lowerthan that in the MFClab-1 (Table 1). The lower performance of thenew MFClab-2, compared with the MFClab-1, was possibly due toseveral reasons. First, for a fair comparison between the newMFClab-2 and the MFClab-1, we used carbon brushes with the samedimension; the twisted carbon brush in the MFClab-2 became

shorter than the CEM tube and thus a portion of the CEM in theMFClab-2 was not well used for electricity generation. Second, thetwisted carbon brush increased the density of the carbon fiber be-tween the spiral spacers and could hinder the water flow, therebyreducing the effect of the helical flow. The detailed reasons requirefurther investigation; however, from the perspective of electrodefabrication and our experiences in the lab, we feel that adding spir-al spacers to a straight carbon brush will be easier and simpler thantwisting a carbon brush with the spiral spacers. Therefore, wechose the model of the MFClab-1 for the on-site test.

3.3. Onsite test

To further demonstrate the technical viability and the advanta-ges of the spiral spacers, we conducted an on-site test by installingtwo MFCs in an aeration tank of a municipal wastewater reclama-tion facility. Both MFCs were used to treat the primary effluent andtook advantage of aeration for the oxygen supply to their cathode.Such a concept has been studied in the lab but not in an actualwastewater treatment process (Cha et al., 2010; Liu et al., 2011).At an HRT of 11.1 h and an average organic loading rate of0.23 kg COD/m3/day, the MFConsite-1 containing the spiral spacersproduced more electricity than the MFConsite-2, although the cur-rent generation fluctuated strongly due to the varied organic con-centration in the primary effluent and the strong motion of theMFCs disturbed by aeration. The average current of the MFConsite-1 in the 60-d operation was 15.5 mA, almost twice the current ofthe MFConsite-2 (8.2 mA) (Fig. 5A). The CEs were 36.3% and 20.0%for the MFConsite-1 and the MFConsite-2, respectively. On average,the operating power density of the MFConsite-1 was 1.20 W/m3,which was 3.5 times that obtained from the MFConsite-2 (0.34 W/m3). The MFConsite-1 achieved slightly higher COD removal effi-ciency than the MFConsite-2 (Fig. 5B). At day 55, the MFConsite-2had a negative TCOD removal efficiency, which was related to avery low COD concentration (<20 mg/L) in the primary effluentafter a major storm. Unlike the lab test, the spiral spacers led toa positive net energy in the MFConsite-1, while the MFConsite-2 stillhad a negative energy balance (Table 1). However, it should benoted that we did not include the aeration energy into our energyanalysis because the estimate of aeration energy for the MFCs in anactual aeration tank would be very difficult. The MFConsite-1 gener-ated higher energy intensity (0.205 kWh/kg COD) than those of thelab MFCs, mainly due to a lower organic loading rate in the waste-water treatment plant. Those results from the on-site tests furtherconfirmed our findings from the lab tests that spiral spacers con-tributed to improved electricity production in MFCs.

4. Conclusions

This study has presented a simple approach to use spiralspacers to optimize the anolyte flow for improving electricity

Fig. 5. Current generation (A) and the removal of total COD (B) in the MFCsinstalled in an aeration tank of a municipal wastewater reclamation facility.

256 F. Zhang et al. / Bioresource Technology 134 (2013) 251–256

production in MFCs. This method is effective in both vertical andhorizontal installations of MFC reactors. The advantage of the spir-al spacers becomes greater at a higher recirculation rate or a higherorganic loading rate. Although some issues, such as optimal spacergaps, selection of spacer materials and better manufacturing meth-od of spiral spacers, need to be further explored, the results fromboth lab tests and on-site examination have clearly demonstratedthat using spiral spacers benefits electricity generation in MFCs.

Acknowledgements

This study was financially supported by a Grant from VeoliaWater North America. The authors thank Milwaukee MetropolitanSewage District (MMSD) for providing a testing site at South ShoreWater Reclamation Facility. The authors also thank Yann Moreau(Veolia), Caroline Dale (Veolia), Khristopher Radke (Veolia), Chris-topher Magruder (MMSD), Scott Royer (Veolia), Mark Swayne(Veolia), and Kyle Jacobson (UW-Milwaukee) for their help withthe project, and Dr. Marjorie P. Piechowski (UW-Milwaukee) forproofreading the manuscript.

References

Arends, J.B.A., Verstraete, W., 2012. 100 years of microbial electricity production:three concepts for the future. Microb. Biotechnol. 5, 333–346.

Cha, J., Choi, S., Yu, H., Kim, H., Kim, C., 2010. Directly applicable microbial fuel cellsin aeration tank for wastewater treatment. Bioelectrochemistry 78, 72–79.

Ci, S., Wen, Z., Chen, J., He, Z., 2012. Decorating anode with bamboo-like nitrogen-doped carbon nanotubes for microbial fuel cells. Electrochem. Commun. 14, 71–74.

Deng, Q., Li, X., Zuo, J., Ling, A., Logan, B.E., 2010. Power generation using anactivated carbon fiber felt cathode in an upflow microbial fuel cell. J. PowerSources 195, 1130–1135.

Fan, Y.Z., Hu, H.Q., Liu, H., 2007. Enhanced coulombic efficiency and power densityof air-cathode microbial fuel cells with an improved cell configuration. J. PowerSources 171, 348–354.

Ge, Z., Ping, Q., He, Z., 2013a. Hollow-fiber membrane bioelectrochemical reactor fordomestic wastewater treatment. J. Chem. Technol. Biotechnol.. http://dx.doi.org/10.1002/jctb.4009.

Ge, Z., Ping, Q., Xiao, L., He, Z., 2013b. Reducing effluent discharge and recoveringbioenergy in an osmotic microbial fuel cell treating domestic wastewater.Desalination 312, 52–59.

He, Z., 2013. Microbial fuel cells: now let us talk about energy. Environ. Sci. Technol.47, 332–333.

He, Z., Wagner, N., Minteer, S.D., Angenent, L.T., 2006. An upflow microbial fuel cellwith an interior cathode: assessment of the internal resistance by impedancespectroscopy. Environ. Sci. Technol. 40, 5212–5217.

Jacobson, K.S., Drew, D., He, Z., 2011. Use of a liter-scale microbial desalination cellas a platform to study bioelectrochemical desalination with salt solution orartificial seawater. Environ. Sci. Technol. 45, 4652–4657.

Jana, P.S., Behera, M., Ghangrekar, M.M., 2010. Performance comparison of up-flowmicrobial fuel cells fabricated using proton exchange membrane and earthencylinder. Int. J. Hydrogen Energy 35, 5681–5686.

Jia, B., Hu, D., Xie, B., Dong, K., Liu, H., 2012. Increased power density from a spiralwound microbial fuel cell. Biosens Bioelectron 41, 894–897.

Kim, J., Kim, K., Ye, H., Lee, E., Shin, C., McCarty, P.L., Bae, J., 2011. Anaerobic fluidizedbed membrane bioreactor for wastewater treatment. Environ. Sci. Technol. 45,576–581.

Kim, J.R., Boghani, H.C., Amini, N., Aguey-Zinsou, K.-F., Michie, I., Dinsdale, R.M.,Guwy, A.J., Guo, Z.X., Premier, G.C., 2012. Porous anodes with helical flowpathways in bioelectrochemical systems: the effects of fluid dynamics andoperating regimes. J. Power Sources 213, 382–390.

Larson, L., Rosso, D., Leu, S.-Y.B., Stenstrom, M.K., 2007. Energy-Conservation in FinePore Diffuser Installations in Activated Sludge Processes. University ofCalifornia, Los Angeles.

Liu, X.W., Wang, Y.P., Huang, Y.X., Sun, X.F., Sheng, G.P., Zeng, R.J., Li, F., Dong, F.,Wang, S.G., Tong, Z.H., Yu, H.Q., 2011. Integration of a microbial fuel cell withactivated sludge process for energy-saving wastewater treatment: taking asequencing batch reactor as an example. Biotechnol. Bioeng. 108, 1260–1267.

Logan, B., Cheng, S., Watson, V., Estadt, G., 2007. Graphite fiber brush anodes forincreased power production in air-cathode microbial fuel cells. Environ. Sci.Technol. 41, 3341–3346.

Logan, B., Rabaey, K., 2012. Conversion of wastes into bioelectricity and chemicalsusing microbial electrochemical technologies. Science 337, 686–690.

Mardanpour, M.M., Nasr Esfahany, M., Behzad, T., Sedaqatvand, R., 2012. Singlechamber microbial fuel cell with spiral anode for dairy wastewater treatment.Biosens. Bioelectron. 38, 264–269.

Min, B., Logan, B.E., 2004. Continuous electricity generation from domesticwastewater and organic substrates in a flat plate microbial fuel cell. Environ.Sci. Technol. 38, 5809–5814.

Pham, H.T., Boon, N., Aelterman, P., Clauwaert, P., De Schamphelaire, L., vanOostveldt, P., Verbeken, K., Rabaey, K., Verstraete, W., 2008. High shearenrichment improves the performance of the anodophilic microbialconsortium in a microbial fuel cell. Microb. Biotechnol. 1, 487–496.

Sukkasem, C., Laehlah, S., Hniman, A., O’thong, S., Boonsawang, P., Rarngnarong, A.,Nisoa, M., Kirdtongmee, P., 2011. Upflow bio-filter circuit (UBFC): biocatalystmicrobial fuel cell (MFC) configuration and application to biodiesel wastewatertreatment. Bioresour. Technol. 102, 10363–10370.

Wang, X., Cheng, S., Feng, Y., Merrill, M.D., Saito, T., Logan, B.E., 2009. Use of carbonmesh anodes and the effect of different pretreatment methods on powerproduction in microbial fuel cells. Environ. Sci. Technol. 43, 6870–6874.

Xiao, L., Young, E.B., Berges, J.A., He, Z., 2012. Integrated photo-bioelectrochemicalsystem for contaminants removal and bioenergy production. Environ. Sci.Technol. 46, 11459–11466.

Zhang, F., Jacobson, K.S., Torres, P., He, Z., 2010. Effects of anolyte recirculation ratesand catholytes on electricity generation in a liter-scale upflow microbial fuelcell. Energy Environ. Sci. 3, 1347–1352.

Zhang, X., Cheng, S., Wang, X., Huang, X., Logan, B.E., 2009. Separator characteristicsfor increasing performance of microbial fuel cells. Environ. Sci. Technol. 43,8456–8461.

Zhuang, L., Yuan, Y., Wang, Y., Zhou, S., 2012a. Long-term evaluation of a 10-literserpentine-type microbial fuel cell stack treating brewery wastewater.Bioresour. Technol. 123, 406–412.

Zhuang, L., Zheng, Y., Zhou, S., Yuan, Y., Yuan, H., Chen, Y., 2012b. Scalable microbialfuel cell (MFC) stack for continuous real wastewater treatment. Bioresour.Technol. 106, 82–88.