A New Approach for Quantification of Morphology-Property Linkages In Fuel Cell Materials: A Case Study for PEM Fuel Cells
- Type: Conference Presentation
- Conference Type:
AIChE Annual Meeting
- Presentation Date:
October 17, 2011
- Skill Level:
Due to the complex morphology and small length scales of many fuel cell materials, experimental quantification of the key properties of these materials can be expensive and quite difficult to conduct, if not impossible. As a result, modeling efforts become an imperative approach to evaluate the impact of material structure on fuel cell performance. To date, pore-scale modeling approach is widely used to understand the complex materials-performance relationship in these systems. However, the computational complexity of these models often limits simulation to analyze only a small volume of the material of interest. The model domain selected for the pore-scale models is typically chosen randomly from a much larger microstructure dataset. When considering the complex and heterogeneous internal structure of fuel cell materials, it is highly unlikely that this randomly selected volume would accurately reflect the full material structure and related impact on cell performance.
The objective of this work is two-fold. The first goal is to develop advanced microstructure analysis tools for direct quantification of the key structural properties of the complex fuel cell materials. The second goal is to develop a new approach for intelligently selecting small representative volume elements (RVEs) from much larger material microstructure datasets, which can be confidently used in pore-scale modeling efforts to obtain reliable results regarding the structure-performance relationship. The diffusion media (DM) in polymer electrolyte fuel cells is chosen for initial demonstration of the approach. The microstructure of a dual-layer DM sample (i.e., a micro-porous layer coated on a macro-DM substrate) is quantified using X-ray computed tomography and dual-beam focused ion beam scanning electron microscopy. Computationally efficient algorithms are developed to extract the key structural parameters (e.g., porosity, surface area, phase connectivity) from measured microstructure datasets of these materials. In particular, two novel microstructure analysis techniques are introduced for the quantification of tortuosity and pore size distribution. Using in-house microstructure analysis tools based on n-point statistics and principal component analysis decomposition, sets of small RVEs that accurately represent salient features of tested fuel cell DM samples are selected from the much larger, full datasets. A detailed validation study is performed to assess the reliability of the presented approach.