The flat-plate plant-microbial fuel cell: the effect of a new design on internal resistances
© Helder et al.; licensee BioMed Central Ltd. 2012
Received: 12 June 2012
Accepted: 24 July 2012
Published: 21 September 2012
Due to a growing world population and increasing welfare, energy demand worldwide is increasing. To meet the increasing energy demand in a sustainable way, new technologies are needed. The Plant-Microbial Fuel Cell (P-MFC) is a technology that could produce sustainable bio-electricity and help meeting the increasing energy demand. Power output of the P-MFC, however, needs to be increased to make it attractive as a renewable and sustainable energy source. To increase power output of the P-MFC internal resistances need to be reduced. With a flat-plate P-MFC design we tried to minimize internal resistances compared to the previously used tubular P-MFC design. With the flat-plate design current and power density per geometric planting area were increased (from 0.15 A/m2 to 1.6 A/m2 and from 0.22 W/m2 to and 0.44 W/m2)as were current and power output per volume (from 7.5 A/m3 to 122 A/m3 and from 1.3 W/m3 to 5.8 W/m3). Internal resistances times volume were decreased, even though internal resistances times membrane surface area were not. Since the membrane in the flat-plate design is placed vertically, membrane surface area per geometric planting area is increased, which allows for lower internal resistances times volume while not decreasing internal resistances times membrane surface area. Anode was split into three different sections on different depths of the system, allowing to calculate internal resistances on different depths. Most electricity was produced where internal resistances were lowest and where most roots were present; in the top section of the system. By measuring electricity production on different depths in the system, electricity production could be linked to root growth. This link offers opportunities for material-reduction in new designs. Concurrent reduction in material use and increase in power output brings the P-MFC a step closer to usable energy density and economic feasibility.
KeywordsPlant-microbial fuel cell Design Flat-plate Internal resistance Root growth Spartina anglica Sustainable electricity
Plant-Microbial Fuel Cell the technology to produce electricity with living plants
Microbial Fuel Cell bio-electrochemical system to produce electricity from organic matter
Electromotive force driving force for electrons
Cation exchange membrane selective membrane for cations
Open Cell Voltage voltage at open circuit (zero current).
With a growing world population and increasing welfare, energy demand worldwide is increasing . Currently used fossil fuels are unevenly distributed over the world, are being depleted, and are unsustainable [2, 3]. Sustainable alternative energy sources that are available nowadays all have their drawbacks. They are weather dependent (wind, solar power), compete with food/feed production (some biofuels) [3, 4] or involve high investment costs . The Plant-Microbial Fuel Cell (P-MFC) is a technology that can potentially be used weather-independent, at any place in the world where plants can grow, without competition with food or feed production, and with relatively low investment costs [6–11]. The technology is still in its infancy, however, and large improvements should be achieved to make this technology energetically and economically feasible. One of the main challenges with current state of technology is its power output. Even though theoretical power output is estimated at 3.2 W/m2 geometric planting area ,maximum power output has been improved only from 65 mW/m2 in 2008  to 220 mW/m2 in 2010  in systems with plants as sole organic matter source. Current biomass-electricity systems, like anaerobic digestion, produce the same amount of electricity per geometric planting area as the maximum that was achieved in P-MFCs, 220 mW/m2. So even at current power output, the P-MFC could compete with anaerobic digestion based on electricity production per geometric planting area. But maximum power outputs of the P-MFCs have been achieved in short term tests like polarization curves and were not sustained for longer periods of time. Over longer periods of time, average power output is limited to maximally 50 mW/m2 geometric planting area . In some experiments a decrease in power output during runtime of an experiment was observed,due to an increase of membrane resistance and build-up of ion-transport resistances . In one recent publication, however, it is shown that average power output increases with runtime of the experiment. This latter experiment was done with a flat-plate P-MFC .
Equation 1 Power density (P) as function of maximum voltage (Vmax = 1.1 V), current density (i,A/m2) and internal resistance (Ri, Ωm2) in a P-MFC [15, 16]. Derivation of equation can be found in supporting information.
The flat-plate system has been described before, but internal resistances of this design were not characterized. In this paper we describe the different partial internal resistances in the flat-plate P-MFC after 320 days runtime of the experiment. Internal resistances were determined at different rooting depths in the P-MFC. To enable this, three separate anodes were used in the system, which were not electronically connected.
Results and discussion
Anodic resistance is the highest partial internal resistance in flat-plate P-MFC
In case of P-MFC 1 this was true; most roots were found in the top part of the system, less in the middle part and least in the bottom part. Anodic resistance of the bottom part of P-MFC 1 (Figure 4a, b) increases with a lower external resistance, while current density does not increase for this level. This is likely due to the fact that hardly any roots were present (<0.1 g) and so substrate was not available to increase current density. If current density cannot be increased, but external resistance allows more current to flow, overpotential will increase. The same occurred in the middle part of P-MFC 1, albeit to a lesser extent, and even in the top part of P-MFC 1, where most roots were found, anode-resistance increased with a lower external resistance, indicating possible substrate-limitation.
P-MFC 2 shows a different pattern (Figure 4c, d). Top and middle part of P-MFC 2 have shown for the largest part of the runtime of the experiment the same voltage and anode and cathode potentials. Likely an electrical connection between top and middle anode of P-MFC 2 existed, so that these two anodes have actually functioned as one. Top and middle level of P-MFC 2 show therefore the same partial and total internal resistance, even though more roots were present in the top part of P-MFC 2 than in the middle part of P-MFC 2. Since electrons could flow freely between the two anodes, substrate limitation at one of them will not directly lead to a higher anode resistance as long as at the other anode substrate is still available. Opposite to the situation in P-MFC 1, the bottom part of P-MFC 2 shows highest current and lowest internal resistance of the three levels in P-MFC 2. Like in P-MFC 1, the bottom part of P-MFC 2 hardly contained any roots. The bottom part of P-MFC 2, however, did contain at least one root-tip. Literature describes that so-called hotspots can occur in the rhizosphere [18, 19]. A hotspot is a place in the root zone where microbial activity and exudation are enhanced compared to the rest of the rhizosphere . Intensity of turnover processes in these hotspots is at least one order of magnitude higher than in the bulk-soil . Even so, hotspots will likely have occurred in the top and middle anode as well. More research should be done to further explore the hotspot hypothesis and exclude artifacts.
Influence of design on internal resistance
It is likely that the design of the system (tubular versus flat-plate) has a large influence on the internal resistance. In our experiment plant-roots were mainly found in the top part of the system, which was consistent with previous experiments [9, 11, 12, 14]. This means that in the tubular system the average distance for protons produced in the anode (or other cations like K+ or Na+) to travel to the cathode is larger than in the flat-plate system (Figure 2). A larger distance will lead to a higher ionic resistance (Equation 8) and with that to a higher membrane resistance. It was shown in previous research by Timmers et al. [11, 14] that membrane resistance was high and increased during polarization curves, which was not the case with the flat-plate system (Figure 5).
Total internal resistances in Ω.m 2 membrane and Ω.m 3 reactor volume of P-MFC 1 and 2 (flat-plate) and P-MFC (tubular) of Timmers et al. 2012 []
Time-step polarization curve (min)
Current density (mA/m2)
Total internal resistance (Ωm2)
Total internal resistance (Ωm3)
P-MFC 1 top
P-MFC 1 top
P-MFC 2 top
P-MFC 2 top
P-MFC 1 middle
P-MFC 1 middle
P-MFC 2 middle
P-MFC 2 middle
P-MFC 1 bottom
P-MFC 1 bottom
P-MFC 2 bottom
P-MFC 2 bottom
Timmers et al. tubular
Timmers et al. tubular
Timmers et al. Tubular
Timmers et al. Tubular
Influence of difference in plant-growth medium on internal resistance
In the tubular system a large transport resistance was found, whereas transport resistance in the flat-plate system is low. It should be noted though, that the plant-growth medium can influence transport resistance as well. Conductivity was 0.15-0.17 S.m-1 in the experiment of Timmers et al. and 1.1-1.3 S.m-1 in our experiment. Higher conductivity in our experiment will lead to a lower ionic resistance compared to Timmers et al. This would mean, however, that according to Equation 5 transport resistance could be expected to be higher than in the case of Timmers et al. The use of buffer in the experiment of Timmers et al. and the lack of it in our experiment can have an influence on the pH-gradient over the membrane. It is shown in Figure 4 that pH-resistance in our case can be positive or negative, meaning the pH-gradient from cathode to anode can be either positive or negative (Equation 6). When cathode-pH is higher than anode-pH, pH-gradient is positive, protons produced in the anode will have to migrate from anode to cathode. Protons migrate from a low pH to a high pH, so transport is driven by pH-difference. A pH-resistance is present, however, due to the change in electromotive force (EMF), as described by the Nernst equation (Equation 7). When the gradient is negative, however, protons will migrate from cathode to anode and the resistance is reversed to a driving force to produce protons. Based on Equation 5, a positive pH-resistance will reduce the transport resistance, but a negative pH-resistance will increase the transport resistance, which happens in the middle and bottom level of P-MFC 1 (Figure 5). Compared to the experiment of Timmers et al. transport resistance is lower in our experiment in all cases except bottom part of P-MFC 1. Thus, even though the difference in plant-growth medium between the experiments of Timmers et al. and ours could influence the internal resistance, it doesn’t seem to explain the reduction of transport resistance in the flat-plate system compared to the tubular system.
Influence of time-steps used in polarization curve on internal resistance
Timmers et al. have shown that the time-steps used to make a polarization curve have a large influence on the internal resistance . The longer the time-steps, the higher the internal resistance, due to an increasing anode resistance caused by proton build-up and an increasing membrane resistance caused by accumulation of cations in the anode . In order to properly compare our results with those of Timmers et al. and understand the mechanisms, it is therefore important to realize that time-steps in the experiment of Timmers et al. (2011) were 5 or 60 minutes . In Timmers et al. (2010) time-steps of polarization are not reported . In our experiment time-steps of 10 minutes were used. Compared to the 5 minute time-step results of Timmers et al., a higher internal resistance could be expected in the flat-plate system, but compared to the 60 minute time-step results, our internal resistance was expected to be lower. Total internal resistance in the flat-plate P-MFC was higher than in the tubular system when normalized to membrane surface area, which cannot be explained from the difference in time-steps (Table 1). When we look to internal resistance when normalized to MFC-volume, however, the picture is different and the flat-plate P-MFC does indeed show a lower internal resistance than the tubular P-MFC in all cases except bottom anode of P-MFC 2 at low external resistance.
Internal resistance in relation to power output of the flat-plate P-MFC
Expected and measured power densities (P, W/m 2 ), based on calculated internal resistances (R i , Ωm 2 ) and measured current densities (i, A/m 2 )
Expected P (W/m2)
Measured P (W/m2)
Δ P (W/m2)
The flat-plate P-MFC design resulted in a lower transport and membrane resistance than the previously used tubular P-MFC. It did, however, not result in a lower total internal resistance normalized to membrane surface area. Total internal resistance of the flat-plate P-MFC is at best comparable to total internal resistance of the tubular P-MFC, only differently distributed over several partial internal resistances. In the flat-plate P-MFC the anodic resistance is highest, when using a chemical cathode, due to substrate limitation or mass transfer limitation. To overcome the problem of substrate limitation, either the plant should exudate more, exudates should be converted into electricity more efficiently or other rhizodeposits, like dead root material should be used. Possibly, when rhizosphere is fully mature, more dead root material is available, which will probably lead to higher substrate availability in the P-MFC. Higher exudation rates might be achieved by adapting the plant-growth medium to stimulate exudation . For future research it would be interesting to further reduce anode height, since the middle and bottom level anodes generally generated less electricity (Figure 3) than the top one and most roots were found in the top anode.
Materials and methods
Set-up and operation
The cathode consisted of three sections of graphite felt of 5 cm height, (Grade WDF, 2.8 mm, National Specialty Products Carbon and Graphite Felt, Taiwan), each corresponding with an anode section (Figure 6b). The cathode compartment consisted of two flow-through modules with the graphite felt in between. A gold wire was woven through the graphite felt of each section to serve as current collector. From day 70 ferric cyanide (K3Fe(CN)6, 50 mM) was used as catholyte to stabilize cathode-potential and be able to study anode-potential. Ferric cyanide was replenished whenever cathode-potential dropped below +200 mV (against Ag/AgCl reference electrode).
From day 1 through day 116 top, middle and bottom sections of the anode were electronically connected, thus behaving as one anode. During this period external load between anode and cathode was 300 Ω. From day 117 anode-sections were electronically disconnected. The three anode-sections all were connected to their corresponding cathode over an external load of 900 Ω.
Concentrations of macronutrients, micronutrients, and added salts in demineralized water as composition of the plant-growth medium
H2MoO4 (85 % MoO3)
Anode and cathode potential were measured with Ag/AgCl-reference electrodes (3 M KCl, + 205 mV versus SHE, ProSenseQis). Data was logged every 60 seconds with Fieldpoint (Module S, National Instruments) and collected with Labview (National Instruments Software). To test maximum power density polarization curves were made simultaneously for each section. These were made by disconnecting anode and cathode for 30 minutes to reach open cell voltage (OCV) and subsequently connecting anode and cathode over an external load of 900, 500, 250, 100, 10, 100, 250, 500 and 900 Ω, for 10 minutes each. At day 320 anode was sampled at 9 sample-points, which were evenly distributed over the three anode-sections, so three sample-points in each anode-section. Samples of 1 ml each were measured for conductivity (ProLine Plus Qis, ProSense BV, Oosterhout, The Netherlands) and pH (691 pH-meter, Metrohm, Herisau, Switzerland).
The set-ups were placed in a climate chamber (Microclima 1750 Snijders) with an average light intensity of 596 ± 161 μmole m-2 s-1, measured at the top of the set-ups witha light intensity meter (Photodyne 44XLA). The climate chamber was controlled at 25°C with 75% humidityand a day-night regime of 14:10 hours was used.
After 703 days the set-ups were taken apart and root-weight was measured for the three anode-sections per P-MFC.
Equation 2 Internal resistance in a P-MFC (Ω.m2), in which R an = anodic resistance (Ω.m2), R cat = cathodic resistance (Ω.m2), R T = transport resistance (Ω.m2).
Equation 3 Anodic resistance in a P-MFC (Ω.m2), in which = theoretical anode potential (V), E an = measured anode potential at a certain external resistance (V), i = current density (A/m2).
Since the theoretical anode potential in the P-MFC is based on a mixed potential  and concentrations of different exudates are unknown, it is assumed that theoretical anode potential equals open cell potential (Ean,OCP).
Equation 4 Cathodic resistance in a P-MFC (Ω.m2), in which = theoretical cathode potential at 50 mM ferric cyanide solution (V), E cat = measured cathode potential at a certain external resistance (V), i = current density (A/m2).
Equation 5 Transport loss in a P-MFC (V), in which; Emem = measured potential over the membrane (V), EΔpH = pH gradient loss (V), Eionic = ionic loss (V).
Equation 6 pH resistance in an MFC, in which pHcath = cathode pH, pHan = anode pH, and i = current density per membrane area.
Equation 7 Nernst-equation, in which is the standard potential (V), R is the universal gas constant (8.314 J mol-1 K-1), T is the temperature (K), n is the number of electrons involved in the reaction (−), F is Faraday’s constant (96485 C mol-1), is the acetate activity (mol L-1), is the proton activity (mol L-1), and is the bicarbonate activity (mol L-1).
This equation shows that one pH-unit difference will lead to a change in anode potential of 59 mV.
Equation 8 Ionic resistance in the anode of an MFC, in which dan = average distance between point of proton production and membrane, and σan = conductivity of anolyte.
Power densities are expressed per geometric planting area because it gives insight in the possibility to use this technology for large scale electricity production.
We would like to thank Ruud Timmers for his input on the flat-plate design and help with the practical work. We would like to thank the Plantpower consortium for thinking along with us on the design of the flat-plate system and the interpretation of the obtained results. The research leading to these results has received funding from the European Community's Seventh Framework Programme FP7/2007-2013 under grant agreement n° 226532.
- Sheffield J: World population growth and the role of annual energy use per capita. Technol Forecast Soc Change 1998, 59:55–87.View Article
- Asif M, Muneer T: Energy supply, its demand and security issues for developed and emerging economies. Renew Sustain Energy Rev 2007, 11:1388–1413.View Article
- Hill J, Nelson E, Tilman D, Polasky S, Tiffany D: Environmental, economic, and energetic costs and benefits of biodiesel and ethanol biofuels. Proc Natl Acad Sci USA 2006, 103:11206–11210.View Article
- Fargione J, Hill J, Tilman D, Polasky S, Hawthorne P: Land clearing and the biofuel carbon debt. Science 2008, 319:1235–1238.View Article
- Huesemann MH: Can advances in science and technology prevent global warming? A critical review of limitations and challenges. Mitig Adapt Strat Glob Chang 2006, 11:539–577.View Article
- Kaku N, Yonezawa N, Kodama Y, Watanabe K: Plant/microbe cooperation for electricity generation in a rice paddy field. Appl Microbiol Biotechnol 2008, 79:43–49.View Article
- De Schamphelaire L, Van Den Bossche L, Hai SD, Hofte M, Boon N, Rabaey K, Verstraete W: Microbial fuel cells generating electricity from rhizodeposits of rice plants. Environ Sci Technol 2008, 42:3053–3058.View Article
- Strik DPBTB, Hamelers HVM, Snel JFH, Buisman CJN: Green electricity production with living plants and bacteria in a fuel cell. Int J Energ Res 2008, 32:870–876.View Article
- Helder M, Strik DPBTB, Hamelers HVM, Kuhn AJ, Blok C, Buisman CJN: Concurrent bio-electricity and biomass production in three Plant-Microbial Fuel Cells usingSpartina anglica, Arundinella anomala and Arundo donax. Bioresour Technol 2010, 101:3541–3547.View Article
- Strik DPBTB, Timmers RA, Helder M, Steinbusch KJJ, Hamelers HVM, Buisman CJN: Microbial solar cells: applying photosynthetic and electrochemically active organisms. Trends Biotechnol 2011, 29:41–49.View Article
- Timmers RA, Strik DPBTB, Hamelers HVM, Buisman CJN: Long-term performance of a plant microbial fuel cell with Spartina anglica. Appl Microbiol Biotechnol 2010, 86:973–981.View Article
- Strik DPBTB, Hamelers HVM, Snel JFH, Buisman CJN: Green electricity production with living plants and bacteria in a fuel cell. Int J Energ Res 2008., 10.1002:
- Helder M, Strik DPBTB, Hamelers HVM, Kuijken RCP, Buisman CJN: New plant-growth medium for increased power output of the Plant-Microbial Fuel Cell. Bioresource Technology 2012, 104:417–423.View Article
- Timmers RA, Strik DPBTB, Hamelers HVM, Buisman CJN: Characterization of the internal resistance of a plant microbial fuel cell. Electrochimica Acta 2012. article in press
- Logan BE, Hamelers B, Rozendal R, Schröder U, Keller J, Freguia S, Aelterman P, Verstraete W, Rabaey K: Microbial fuel cells: methodology and technology. Environ Sci Technol 2006, 40:5181–5192.View Article
- Fan Y, Sharbrough E, Liu H: Quantification of the internal resistance distribution of microbial fuel cells. Environ Sci Technol 2008, 42:8101–8107.View Article
- Hamelers HVM, Ter Heijne A, Sleutels THJA, Jeremiasse AW, Strik DPBTB, Buisman CJN: New applications and performance of bioelectrochemical systems. Appl Microbiol Biotechnol 2010, 85:1673–1685.View Article
- Kuzyakov Y: Priming effects: interactions between living and dead organic matter. Soil Biol Biochem 2010, 42:1363–1371.View Article
- Pausch J, Kuzyakov Y: Photoassimilate allocation and dynamics of hotspots in roots visualized by 14 C phosphor imaging. J Plant Nutr Soil Sci 2011, 174:12–19.View Article
- Rozendal RA, Hamelers HVM, Rabaey K, Keller J, Buisman CJN: Towards practical implementation of bioelectrochemical wastewater treatment. Trends Biotechnol 2008, 26:450–459.View Article
- Harnisch F, Schröder U: From MFC to MXC: Chemical and biological cathodes and their potential for microbial bioelectrochemical systems. Chem Soc Rev 2010, 39:4433–4448.View Article
- Blossfeld S, Perriguey J, Sterckeman T, Morel JL, Lösch R: Rhizosphere pH dynamics in trace-metal-contaminated soils, monitored with planar pH optodes. Plant Soil 2010, 330:173–184.View Article
- Sleutels THJA, Hamelers HVM, Rozendal RA, Buisman CJN: Ion transport resistance in microbial electrolysis cells with anion and cation exchange membranes. Int J Hydrogen Energ 2009, 34:3612–3620.View Article
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.