Optimization of fuel cell performance using computational fluid dynamics

Tabbi Wilberforce*, Oluwatosin Ijaodola, Ogungbemi Emmanuel, James Thompson, Abdul Ghani Olabi, Mohammad Ali Abdelkareem, Enas Taha Sayed, Khaled Elsaid, Hussein M. Maghrabie

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

13 Citations (Scopus)
41 Downloads (Pure)


A low cost bipolar plate materials with a high fuel cell performance is important for the establishment of Proton Exchange Membrane (PEM ) fuel cells into the competitive world market. In this research, the effect of different bipolar plates material such as Aluminum (Al), Copper (Cu), and Stainless Steel (SS) of a single stack of proton exchange membrane (PEM) fuel cells was investigated both numerically and experimentally. Firstly, a three dimensional (3D) PEM fuel cell model was developed, and simulations were conducted using commercial computational fluid dynamics (CFD) ANSYS FLUENT to examine the effect of each bipolar plate materials on cell performance. Along with cell performance, significant parameters distributions like temperature, pressure, a mass fraction of hydrogen, oxygen, and water is presented. Then, an experimental study of a single cell of Al, Cu, and SS bipolar plate material was used in the verification of the numerical investigation. Finally, polarization curves of numerical and experimental results was compared for validation, and the result shows that Al serpentine bipolar plate material performed better than Cu and SS materials. The outcome of the investigation was in tandem to the fact that due to adsorption on metal surfaces, hydrogen molecules is more stable on Al surface than Cu and SS surfaces.
Original languageEnglish
Article number146
Pages (from-to)1-21
Number of pages21
Issue number2
Publication statusPublished - 20 Feb 2021


  • PEM fuel cell
  • serpentine bipolar plate
  • polarization curve
  • stainless steel
  • copper


Dive into the research topics of 'Optimization of fuel cell performance using computational fluid dynamics'. Together they form a unique fingerprint.

Cite this