Multiphysics Modeling for Solid Oxide Electrolyzer Cell with Heterogenous Synthetic Microstructure
Xinfang Jin,Puvikkarasan Jayapragasam,Yasser Shoukry
DOI: https://doi.org/10.1149/ma2023-0154163mtgabs
2023-08-28
ECS Meeting Abstracts
Abstract:Solid oxide fuel cell (SOFC) and electrolyzer cell (SOEC) are the main focus for researchers in recent times to diminish greenhouse gas emissions and meet cleaner energy demand, as these devices have highest energy conversion rates and efficiencies [1]. The major hinderance in commercializing SOEC is due to its faster cell degradation under higher current density, compared against its SOFC counterpart. The delamination of oxygen electrode (OE) from the electrolyte (EL) is one of the major causes for the cell degradation [2]. We employed a multi-physics numerical model to understand the electrochemical performance of a SOEC half-cell more precisely at a microscale. In this study, widely used mixed ionic and electronic conductor (MIEC) La 1-x Sr x Co 1-y Fe y O 3-d (LSCF) and Gadolinium-doped Ceria (GDC) composite is used as the oxygen electrode and, pure ionic conductor 8 mol% Y 2 O 3 doped ZrO 2 (YSZ) is used as the electrolyte. A GDC buffer layer is tucked in between to prevent undesirable side reactions between OE and EL during long term polarization operation. Until now, a homogenous OE structure has commonly been used with effective physicochemical properties to predict the performance of SOFC/SOEC [3,4]. Only a few researchers have worked on heterogenous microstructure model [5] and simulated the electrochemical performance of the cell. Synthetic microstructure of LSCF-GDC porous electrode as shown in Figure 1 are constructed with an opensource software Dream 3D with particle size ranging from 0.45um - 1.5um. We developed in-house Matlab code for labelling domains, interfaces (2PB and 3PB) and digitally removing the unconnected phases to reduce numerical instabilities while solving the mathematical model in COMSOL Multiphysics 6.0. Mesh in these domains is constructed using Iso2Mesh toolbox [6] and translated as mphtxt file which is later imported into COMSOL. This practice reduces the computational time and memory. We employed Nernst-Planck theory for solving the transport of species like vacancy, hole and electron within LSCF inside OE. Multi-step charge transfer at two phase boundary (2PB) and triple phase boundary (3PB) interfaces for oxygen evolution reaction are formulated by Butler-Volmer expressions [4]. A physics-based impedance model under DC bias is developed to separate the electrochemical contributions from 2PB and 3PB transport pathways. The voltage-current curve during long-term polarization test is coupled with the model for validation and approximation of process parameters. Furthermore, we will simulate the crack propagation at OE-buffer layer interface in various directions and orientations to understand the failure mechanism on the synthetically constructed microstructure. References: A. Laguna-Bercero, “Recent advances in high temperature electrolysis using solid oxide fuel cells: a review”, J. Power Sources , vol. 203, no. 4, 2012 N. Rashkeev and M. V. Glazoff, “Atomic-scale mechanisms of oxygen electrode delamination in solid oxide electrolyzer cells”, Int. J. Hydrogen Energy , vol. 37, no. 2, pp. 1280–1291, 2012 Pakalapati, K. Gerdes, H. Finklea, M. Gong, X. Liu and I. Celik, “Micro scale dynamic modeling of LSM/YSZ composite cathodes”, Solid State Ionics, vol. 258, pp. 45-60, 2014 Cook, J. Wrubel, Z. Ma, K. Huang, X. Jin, “Modeling Electrokinetics of Oxygen Electrodes in Solid Oxide Electrolyzer Cells”, Journal of The Electrochemical Society, vol. 168, no. 11, pp. 114510, 2021 Su, Z. Zhong and Z. Jiao, “A novel multi-physics coupled heterogenous single-cell numerical model for solid oxide fuel cell based on 3D microstructure reconstructions”, Energy & Environmental Science, vol. 15 pp. 2410 – 2424, 2022 Fang and D. Boas, “Tetrahedral mesh generation from volumetric binary and grayscale images”, Proceedings of IEEE International Symposium on Biomedical Imaging , 2009, pp. 1142–1145 Figure 1