Modeling Hierarchical Non-Precious Metal Catalyst Cathodes for PEFCs Using Multi-Scale X-ray CT Imaging

Polymer electrolyte fuel cells (PEFCs) are a promising energy conversion technology for a wide variety of transportation, portable, and stationary applications. State-of-the-art PEFCs rely on platinum and platinum-alloy catalyst nano-particles supported on high surface area carbon black. Unfortunately, the current high Pt loadings needed for the oxygen reduction reaction (ORR) incur significant costs due to the Pt raw material cost. Thus, there is significant interest in developing non-precious metal catalysts (NPMCs) for the ORR in low temperature, acidic fuel cells. Such catalysts are historically challenged by a combination of low activity and instability, but there has been notable progress made by several groups in recent years (1,2). Although the ORR activity has improved, thick electrodes are needed to achieve adequate current densities. Unfortunately, increasing electrode thickness yields increased transport losses for both oxygen and protons. Thus, there is an important need for optimized electrode architectures for thick, NPMC electrodes.

                This work focuses on the optimization of NPMC electrode architectures using a generalized model with morphological parameter inputs from nano-scale resolution X-ray computed tomography (nano-CT) (3). The model is a complete cell, continuum model that includes an agglomerate model (4,5) representation of the cathode. The particular electrodes being studied feature two-distinct length scales of interest to this model. First, there is a dense active region that consists of the primary carbon support particles and the interspersed mesopores and smaller macropores (<500 nm). Second, at a larger length-scale there are distinct volumes of dense regions surrounded by large macropores (500 nm – 10 µm). As Figure 1 shows, this NPMC electrode structure has been characterized by nano-CT at two resolution levels. A large field of view (FOV = 65 µm, 150 nm resolution) mode image provides the macropore morphology and the outer structure of the large, dense, agglomerate-like, regions. A high resolution scan (50 nm resolution, 16 nm voxels) provides the morphology within the dense regions, including the mesopore and smaller macropore morphology.

                As the Figure illustrates, the cathode is modeled using an agglomerate model approach where the high resolution imaging of the dense regions is used to define the internal morphology and transport properties of agglomerates. Likewise, the large FOV imaging is used to define the outer morphology of the dense region, including effective agglomerate and pore diameter distributions (5). The effective transport properties for the representative elementary volume models are extracted from pore-scale finite-element simulations. The complete-cell model is compared to experimental characterization of PEFCs with the same electrodes to evaluate its predictive capabilities. Finally, parametric studies of the morphological properties are performed to identify improved architectures for NPMC cathodes.

REFERENCES

1.     G. Wu, K. L. More, C. M. Johnston and P. Zelenay, Science, 332, 443 (2011).

2.     M. Lefevre, E. Proietti, F. Jaouen and J. P. Dodelet, Science, 324, 71 (2009).

3.     W. K. Epting, J. Gelb and S. Litster, Advanced Functional Materials, 22, 555 (2012).

4.     W. Sun, B. A. Peppley and K. Karan, Electrochimica Acta, 50, 3359 (2005).

5.     W. K. Epting and S. Litster, International Journal of Hydrogen Energy, 37, 8505 (2012).