Elsevier

Journal of Power Sources

Volume 194, Issue 1, 20 October 2009, Pages 397-407
Journal of Power Sources

A model-based parametric analysis of a direct ethanol polymer electrolyte membrane fuel cell performance

https://doi.org/10.1016/j.jpowsour.2009.04.064Get rights and content

Abstract

In the present work, a model-based parametric analysis of the performance of a direct ethanol polymer electrolyte membrane fuel cell (DE-PEMFC) is conducted with the purpose to investigate the effect of several parameters on the cell's operation. The analysis is based on a previously validated one-dimensional mathematical model that describes the operation of a DE-PEMFC in steady state. More precisely, the effect of several operational and structural parameters on (i) the ethanol crossover rate from the anode to the cathode side of the cell, (ii) the parasitic current generation (mixed potential formation) and (iii) the total cell performance is investigated. According to the model predictions it was found that the increase of the ethanol feed concentration leads to higher ethanol crossover rates, higher parasitic currents and higher mixed potential values resulting in the decrease of the cell's power density. However there is an optimum ethanol feed concentration (approximately 1.0 mol L−1) for which the cell power density reaches its highest value. The platinum (Pt) loading of the anode and the cathode catalytic layers affects strongly the cell performance. Higher values of Pt loading of the catalytic layers increase the specific reaction surface area resulting in higher cell power densities. An increase of the anode catalyst loading compared to an equal one of the cathode catalyst loading has greater impact on the cell's power density. Another interesting finding is that increasing the diffusion layers’ porosity up to a certain extent, improves the cell power density despite the fact that the parasitic current increases. This is explained by the fact that the reactants’ concentrations over the catalysts are increased, leading to lower activation overpotential values, which are the main source of the total cell overpotentials. Moreover, the use of a thicker membrane leads to lower ethanol crossover rate, lower parasitic current and lower mixed potential values in comparison to the use of a thinner one. Finally, according to the model predictions when the cell operates at low current densities the use of a thick membrane is necessary to reduce the negative effect of the ethanol crossover. However, in the case where the cell operates at higher current densities (lower ethanol crossover rates) a thinner membrane reduces the ohmic overpotential leading to higher power density values.

Introduction

Direct Ethanol PEM Fuel Cells (DE-PEMFCs) attract the increasing interest of many researchers, due to the advantages of the feed fuel, which is hydrogen rich, less toxic and has higher energy density compared to the widely used alcohol in these devices, methanol. Moreover, ethanol as a liquid fuel can be stored, handled and distributed more easily than hydrogen and it is considered renewable, since it can be obtained mainly from the fermentation of biomass. However, the use of ethanol in PEM fuel cells is accompanied with a series of challenges that have to be overcome. The main drawbacks of DE-PEMFCs that limit their application as competitive devices are (i) the slow kinetics of the ethanol electro-oxidation reaction over the anode electrocatalyst, (ii) the fact that the electro-oxidation of ethanol below 100 °C does not proceed all the way to carbon dioxide, but rather to acetaldehyde and acetic acid indicating that the problem of the C–C bond cleavage cannot be sufficiently resolved by the up-to-date tested electrocatalysts and (iii) the ethanol crossover from the anode to the cathode side of the cell leading to the parasitic oxidation reaction of ethanol on the cathode electrocatalyst, hindering the oxygen reduction reaction (ORR). The above mentioned problems have been the subject of several experimental works dealing with DE-PEMFCs [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17].

The performance of DE-PEMFCs depends on numerous parameters, such as the ethanol feed concentration, the operating temperature, the specific area of the catalyst where the ethanol electro-oxidation and the ORR take place, the design parameters of the different layers comprising the fuel cell, the resistance of the catalyst layer, the conductivity of the membrane, the rate of the ethanol crossover, the released products of the electro-oxidation along with their influence on the species transport and so on. However, the investigation of the impact of each of the above mentioned parameters experimentally is cost prohibitive. As a consequence, theoretical investigations are also essential for an in-depth understanding and optimization of the operation of a DE-PEMFC [18], [19], [20], [21], [22], [23], and the analysis of the operating parameters that affect the cell performance is required for the further development of these devices.

In the present work, a model-based parametric analysis of the fuel cell operation is performed in order to investigate the effect of (i) the ethanol feed concentration, (ii) the Pt loading of the anode and cathode catalytic layers, (iii) the specific reaction surface area of the catalytic layers, (iv) the thickness of the Nafion membrane and (v) the porosity and thickness of the anode and the cathode gas diffusion layers and catalysts layers on (i) the ethanol crossover rate, (ii) the parasitic current generation (mixed potential formation) and (iii) the total cell performance.

Section snippets

Theory

The mathematical model development is based on our previous work [19]; as a consequence, only a brief description of the theoretical part is given here. During the mathematical model development the following assumptions were made: (a) the equations are defined in one direction (through-plane direction—cf. Fig. 1), (b) the cell operates under steady-state, isothermal conditions, (c) the model considers neither a two-phase flow regime nor a phase change taking place during operation, (d) the

Model validation

The mathematical model development is based on our previous work [19] and it has been validated against the experimental data presented in the literature [31]. The base case values of the parameters used in the model development are shown in Table 1.

Effect of ethanol feed concentration on cell performance and operation

The effect of the ethanol feed concentration on the cell performance and the parasitic current formation when the cell operates at 75 °C is depicted in Fig. 2. The Pt loading for the anode and cathode catalyst layers used in the model calculations is

Conclusions

In the present work a parametric analysis regarding the performance of a DE-PEMFC by the aid of a validated one-dimensional mathematical model was undertaken. The model predicts the fuel cell polarization performance in terms of the VI, PI curves, the ethanol crossover rate and the parasitic current formation for different operational and structural parameters. It was found that there is an optimum ethanol feed concentration of ∼1.0 mol L−1 for which the cell power density obtains its highest

Acknowledgements

This work is part of the 03ED897 research project, implemented within the framework of the “Reinforcement Programme of Human Research Manpower” (PENED) and co-financed by National and Community Funds (25% from the Greek Ministry of Development-General Secretariat of Research and Technology and 75% from EU-European Social Fund).

References (54)

  • T. Lopes et al.

    J. Power Sources

    (2007)
  • C. Lamy et al.

    Electrochim. Acta

    (2004)
  • E. Antolini

    J. Power Sources

    (2007)
  • S. Song et al.

    Appl. Catal. B: Environ.

    (2006)
  • E. Antolini et al.

    Electrochem. Commun.

    (2007)
  • F. Colmati et al.

    Appl. Catal. B: Environ.

    (2007)
  • F. Colmati et al.

    J. Power Sources

    (2006)
  • S. Kontou et al.

    J. Power Sources

    (2007)
  • J. Liu et al.

    Electrochem. Commun.

    (2007)
  • J. Ribeiro et al.

    Electrochim. Acta

    (2007)
  • S. Rousseau et al.

    J. Power Sources

    (2006)
  • S. Song et al.

    Appl. Catal. B: Environ.

    (2008)
  • S. Song et al.

    J. Power Sources

    (2005)
  • P.E. Tsiakaras

    J. Power Sources

    (2007)
  • Q. Wang et al.

    J. Power Sources

    (2008)
  • T. Lopes et al.

    Int. J. Hydrogen Energy

    (2008)
  • G. Andreadis et al.

    Chem. Eng. Sci.

    (2006)
  • G.M. Andreadis et al.

    J. Power Sources

    (2008)
  • R. Sousa et al.

    J. Power Sources

    (2008)
  • I. Sarris et al.

    Solid State Ionics

    (2006)
  • H. Hitmi et al.

    Electrochim. Acta

    (1994)
  • C. Lamy et al.

    J. Power Sources

    (2002)
  • K. Yin

    J. Power Sources

    (2007)
  • K.M. Yin

    J. Power Sources

    (2008)
  • C. Marr et al.

    J. Power Sources

    (1999)
  • S. Eccarius et al.

    J. Power Sources

    (2008)
  • K. Jeng et al.

    J. Power Sources

    (2002)
  • Cited by (29)

    • Oxygen reduction reaction in ethanol fuel cells

      2022, Oxygen Reduction Reaction: Fundamentals, Materials, and Applications
    • Direct ethanol fuel cells (DEFCs)

      2020, Direct Liquid Fuel Cells: Fundamentals, Advances and Future
    • A mathematical model for direct ethanol fuel cells based on detailed ethanol electro-oxidation kinetics

      2019, Applied Energy
      Citation Excerpt :

      Maximum power density is reached with 0.5 M ethanol feed concentration, but 1 M reaches almost the same peak power density with a much larger limiting current density, which guarantees a more stable cell operation. This value is similar to other optimum feed concentrations reported in the literature [50,55,70,75,81], what makes us gain confidence in the results of the model. It is also seen that reduced ethanol feed concentrations lead to high fuel utilizations, ca. 100%, due to the reduced crossover rates.

    • A genetically optimized kinetic model for ethanol electro-oxidation on Pt-based binary catalysts used in direct ethanol fuel cells

      2017, Journal of Power Sources
      Citation Excerpt :

      Despite the good agreement in terms of polarization curves, the composition of the products predicted by this model is far from satisfactory, as recently shown by the authors [52]. Due to the importance of crossover in DEFC performance, most models have also included this effect [18,39–45,51]. Since the molecular structures of ethanol and methanol are very similar, all crossover models for ethanol are based on those previously developed for methanol [53], with the crossover flux driven by molecular diffusion and electro-osmotic drag.

    View all citing articles on Scopus
    View full text