Increased performance of single-chamber microbial fuel cells using an improved cathode structure

https://doi.org/10.1016/j.elecom.2006.01.010Get rights and content

Abstract

Maximum power densities by air-driven microbial fuel cells (MFCs) are considerably influenced by cathode performance. We show here that application of successive polytetrafluoroethylene (PTFE) layers (DLs), on a carbon/PTFE base layer, to the air-side of the cathode in a single chamber MFC significantly improved coulombic efficiencies (CEs), maximum power densities, and reduced water loss (through the cathode). Electrochemical tests using carbon cloth electrodes coated with different numbers of DLs indicated an optimum increase in the cathode potential of 117 mV with four-DLs, compared to a <10 mV increase due to the carbon base layer alone. In MFC tests, four-DLs was also found to be the optimum number of coatings, resulting in a 171% increase in the CE (from 19.1% to 32%), a 42% increase in the maximum power density (from 538 to 766 mW m−2), and measurable water loss was prevented. The increase in CE due is believed to result from the increased power output and the increased operation time (due to a reduction in aerobic degradation of substrate sustained by oxygen diffusion through the cathode).

Introduction

A microbial fuel cell (MFC) uses bacteria to generate electricity from the oxidation of organic matter [1], [2], [3], [4], [5], [6], [7], [8]. Bacteria capable of electricity generation have been enriched from domestic wastewater [6], ocean sediments [9], animal wastes [10], and anaerobic sewage sludge [11], [12]. Electricity generation is supported by a range of biodegradable substrates, including glucose, acetate, lactate, butyrate, ethanol and organic matter in wastewater [5], [13], [14], [15], [16]. Several factors affect MFC performance including the microbial inoculum, chemical substrate (fuel), type of proton exchange material (and the absence of this material), cell internal and external resistance, solution ionic strength, electrode materials, and electrode spacing [3], [4], [5], [13], [14], [15], [16], [17], [18], [19], [20].

The cathode is an important factor in the performance of a MFC due to the poor kinetics of oxygen reduction reaction in a neutral pH medium [16], [21]. Using more effective catholytes than oxygen, such as ferricyanide, can substantially increase power output [12], [17], [19]. However, ferrocyanide must be replaced after ferricyanide is converted to ferrocyanide, while systems using oxygen can be continuously operated and therefore, the reaction is self sustaining. Furthermore, a proton exchange membrane (PEM) must be used to prevent ferricyanide from diffusing into the anode chamber. In air-cathode MFCs, removing the PEM increases the maximum power density [5]. Although Pt is an expensive component for MFC construction, it has recently been shown that Pt loading on the water side of an air cathode can be as low as 0.1 mg cm−2 without affecting power densities, and that Pt can be replaced by less expensive alternatives such as Co-tetramethyl phenylporphyrin (CoTMPP) with only small changes in MFC performance [15], [22].

The coulombic efficiency (CE) in MFCs that use oxygen varies widely, ranging from 9–12% for single chamber systems lacking a PEM, to 40–90% for two-chamber systems with aqueous cathodes and a PEM using glucose as substrate [5], [21], [23]. One of the main reasons for low CEs is loss of substrate due to bacterial oxidation using oxygen as an electron acceptor. Oxygen flux through the air cathode is 3.7 times higher than that through the same cathode containing a Nafion PEM [5]. In addition, there can be evaporative losses of water from the chamber at the cathode surface, which produce a gas phase in the anode chamber. The amount of oxygen needed at the MFC cathode is small compared to that required for hydrogen fuel cells, due to the low current densities (0.03–1.5 mA cm−2) [19] in MFCs. Therefore, we hypothesized that power might not be adversely affected by a cathode coating that could reduce water loss, as long as the coating did not completely hinder oxygen diffusion to the Pt catalyst. Such cathode coatings, called diffusion layers (DLs), have been used to improve cathode performance in other types of fuel cells, but these systems operate under much different conditions of pH (pH 7, compared to traditional hydrogen/oxygen fuel cells which can use highly acid or alkaline conditions [24], [25]), temperature (20–30 °C, compared to 50–1000 °C), current density (usually <2 mA cm−2, compared to 100 mA cm−2), and water concentration at the cathode (water saturation of the cathode, compared to systems that use a solid electrolyte).

DLs have not been previously investigated for air cathode MFCs. A DL that is hydrophobic can also potentially improve performance by decreasing the water flooding of the catalyst. In this paper, we demonstrate that a cathode containing a hydrophobic layer (to reduce water losses) on the air-side of the cathode can significantly increase the coulombic efficiency and power density of an MFC, and reduce water loss through the cathode. Power output and coulombic efficiency of MFCs containing a single carbon/polytetrafluoroethylene (PTFE) base layer was compared to reactors containing additional PTFE DLs.

Section snippets

MFC electrodes

Anodes were made of non-wet proofed carbon cloth (type A, E-TEK), while cathodes were wet-proofed (30%) carbon cloth (type B, E-TEK). Cathodes were either used as supplied by the manufacturer (No DL), or they were coated on the air-facing side of the cathode with a carbon/PTFE layer (T0) or one or more additional DLs consisting of PTFE (Fig. 1). The carbon base layer was prepared by applying a mixture of carbon powder (Vulcan XC-72) and 30 wt% PTFE solution (20 μl/mg of carbon power) onto one

Cathode performance in electrochemical cell tests

Cathodes were initially evaluated on the basis of final potentials in electrochemical tests (no bacteria). As expected, application DLs had no effect on the open circuit potential (OCP) of the cathodes, with all values equal to 302 ± 4 mV (vs Ag/AgCl). The application of the initial carbon base layer to the cathode (T0) increased the potential by less than 10 mV, compared to a plain cathode (No DL) (Fig. 2). Application of successive PTFE DLs increased the cathode potentials for up to four-PTFE

Discussion

The addition of DLs increased the maximum power and the CE. Four DLs added to a carbon/PTFE base was found to be optimal for increasing power generation by 42% (to 766 mW m−2) as further DLs decreased the maximum power output. DLs consistently increased CE, with 8 DLs increasing CE by 32%. This increase in CE was 171% higher than that obtained using the same cathode lacking a DL, and 200% higher than that obtained using a commercially available cathode with pre-loaded catalyst.

The application of

Conclusions

Application of a DL, consisting of mixture of PTFE and carbon, can be used to increase the performance of a MFC. It was found here that the application of four DLs (equivalent to 20 mg cm−2 of PTFE and 2.5 mg cm−2 of carbon) produced the best performance of the MFC by increasing the maximum power density by 42% and the CE by 200%, as compared to a commercially available cathode.

Acknowledgements

This research was supported by National Science Foundation Grant BES-0401885, a seed grant from The Huck Institutes of the Life Sciences at Penn State, and the Stan and Flora Kappe Endowment.

References (31)

  • J. Niessen et al.

    Electrochem. Commun.

    (2004)
  • L.T. Angenent et al.

    Trend Biotech.

    (2004)
  • J.R. Kim et al.

    Appl. Microbiol. Biotechnol.

    (2005)
  • G.C. Gil et al.

    Biosens. Bioelectron.

    (2003)
  • F. Zhao et al.

    Electrochem. Commun.

    (2005)
  • I.S. Chang et al.

    Biosens. Bioelectron.

    (2004)
  • E. Han et al.

    Int. J. Hydrogen Energy

    (2000)
  • S. Srinivasan et al.

    J. Power Source

    (1991)
  • B. Carnes et al.

    J. Power Source

    (2005)
  • D. Bevers et al.

    J. Power Source

    (1996)
  • S. Suzuki et al.

    Biotechnol. Bioeng. Symp.

    (1978)
  • R.M. Allen et al.

    Appl. Biochem. Biotechnol.

    (1993)
  • D.R. Bond et al.

    Appl. Environ. Microbiol.

    (2003)
  • H. Liu et al.

    Environ. Sci. Technol.

    (2004)
  • H. Liu et al.

    Environ. Sci. Technol.

    (2004)
  • Cited by (1009)

    View all citing articles on Scopus
    View full text