MODELING COLD START OF POLYMER ELECTROLYTE FUEL CELLS

Open Access
Author:
Mao, Leng
Graduate Program:
Mechanical Engineering
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
July 18, 2006
Committee Members:
  • Stefan Thynell, Committee Member
  • Long Qing Chen, Committee Member
  • Chao Yang Wang, Committee Chair
  • Fan Bill B Cheung, Committee Member
Keywords:
  • fuel cell
  • model
  • subzero
  • cold start
Abstract:
While substantial research efforts are presently directed toward performance and durability improvement of polymer electrolyte fuel cells (PEFC), PEFC cold start capability and survivability at subzero temperatures remains a challenge and has been scarcely researched. In order to achieve a fundamental understanding of PEFC cold start, a mechanistic model is developed in this thesis to describe the complex physics involved, along with necessary experimental validation. First, a theoretical analysis focusing on the catalyst layer is performed to study the importance of coupled water and heat balance, as well as to elucidate the main effects of ice formation on fuel cell performance. Second, to fully describe the multi-scale physics involved in PEFC cold start and the coupled nature of water and heat transport, a comprehensive multi-dimensional, multi-component, multiphase, transient model is developed. The model accounts for ice/frost precipitation and growth in the cathode catalyst layer and gas diffusion layer and for water transport characteristics at very low temperatures. Governing equations of species, heat and charge transport are developed using a single-domain approach and solved with a finite-volume based computational fluid dynamics (CFD) technique. Validated by extensive experimental data, this computational model is used to identify key parameters that define PEFC cold start capability. Finally, effects of operating conditions, such as initial membrane water content, applied current load and start up temperature, are investigated using this model. PEFC cold start performance with different cell configurations or designs are also studied using this model.