Effect of cathode pore volume on PEM fuel cell cold start

Ashis Nandy, Fangming Jiang, Shanhai Ge, Chao Yang Wang, Ken S. Chen

Research output: Contribution to journalArticlepeer-review

36 Scopus citations


Start-up of a proton exchange membrane (PEM) fuel cell from subzero temperatures, commonly referred to as cold start, remains a major challenge for automotive applications. In this work, we theoretically and experimentally study the effect of catalyst layer (CL) pore volume (or, more directly, CL thickness) on the cold-start performance of a PEM fuel cell for both isothermal and nonisothermal operations. Special attention is directed to determining the limits of a cold-start performance with an ultrathin CL (1 μm). The cold-start product water or the operational time approaches a minimum nonzero asymptote as the CL is gradually made infinitesimally thin. For a PEM fuel cell with standard cell thermal mass, e.g., 0.4 J/ cm2 K, and with moderately low initial membrane water content (λ0 =7), successful start-up from -20°C at 100 mA/ cm2 can be achieved for CL thicknesses of 10 μm and above, whereas a CL thickness of 20 μm is required for successful self-start-up from -30°C. However, successful start-up can be achieved even with a 1 μm thick CL, given certain adjustments to cell design and material properties. In particular, we study the effects of cell thermal mass and membrane water diffusivity and present a design map for self-start-up of a 1 μm CL PEM fuel cell from various subfreezing temperatures.

Original languageEnglish (US)
Pages (from-to)B726-B736
JournalJournal of the Electrochemical Society
Issue number5
StatePublished - 2010

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Materials Chemistry
  • Surfaces, Coatings and Films
  • Electrochemistry
  • Renewable Energy, Sustainability and the Environment


Dive into the research topics of 'Effect of cathode pore volume on PEM fuel cell cold start'. Together they form a unique fingerprint.

Cite this