Numerical Assessment of SOFC Anode Polarization Based on Three-Dimensional Model Microstructure Reconstructed from FIB-SEM Images

Abstract

A three-dimensional numerical simulation of the solid oxide fuel cell ͑SOFC͒ anode overpotential is conducted in a microstructure which is reconstructed by dual-beam focused ion beam-scanning electron microscopy ͑FIB-SEM͒. Gaseous, ionic, and electronic transport equations are solved by a lattice Boltzmann method with electrochemical reaction at the three-phase boundary. The predicted anode overpotential agrees with the experimental data at the fuel supply of 1.2% H 2 O-98.8% H 2 , while it is larger than the data at 10% H 2 O-90% H 2 . The dependence of exchange current density on steam partial pressure, gas diffusion modeling, as well as computational domain size must be further investigated in the future. Local three-dimensional distributions of electrochemical potential and current density inside the anode microstructure are obtained. Their nonuniformities are attributed to the scattered three-phase boundaries and complex transport paths through the solid phases.Solid oxide fuel cell ͑SOFC͒ is anticipated to play a major role in future energy utilization because of its superior efficiency and fuel flexibility. 1 However, its cost effectiveness and durability must be further improved before market introduction. The electrode microstructure has significant effects on the cell performance and durability of SOFCs. Thus, a basic understanding of the microscopic features of the electrode is indispensable. Quantitative investigations that relate the electrode microstructural parameters obtained from two-dimensional images and the polarization resistances have been reported. Wilson and Barnett 2 related anode polarization resistances to the three-phase boundary ͑TPB͒ densities of Ni-yttria-stabilized zirconia ͑YSZ͒ active layers by means of stereology. Shikazono et al. 3 also used stereology as well as the concept of contiguity theory to investigate the relationship between polarization characteristics and microstructural parameters such as TPB length and effective conductivities. However, a random mixture is assumed in stereology and the concept of contiguity theory, which has to be validated for further investigation. Furthermore, dead ends of the phases and electrochemically inactive TPBs should be rationally removed to quantitatively discuss the effects of microstructure on polarization characteristics. To overcome these issues, it is necessary to establish a method that can directly predict the polarization resistance in the real three-dimensional ͑3D͒ microstructure.Recently, direct measurements of a 3D SOFC electrode microstructure have been carried out using focused ion beam-scanning electron microscopy ͑FIB-SEM͒ 4-10 and X-ray computed tomography ͑XCT͒. 11 As a result, useful quantitative data such as TPB length and tortuosity factor are obtained from the reconstructed 3D microstructures. However, difficulty remains in removing errors that arise from a discretization process and insufficient sample volume size. 9,10,12 In addition to the above experimental studies, numerical simulations have the possibility ...

Investigation of the electrochemical active thickness of solid oxide fuel cell anode

“…6, the EAT decreases from 8.1 μm to 4.5 μm as xH2,inlet decreases from 96.8% to 70% when output voltage Vcell= 0.8V. Similar results can be found in [6], in which a thinner EAT is obtained with a lower xH2,inlet.…”

“…For the standard cell defined with parameters in Table. 3, the EAT is 8.1 μm (Vcell=0.8 V). Such a small value is in the range of [5,6,26,28,29], but differs a lot from [7,25,30]. The differences are probably caused by the quite different exchange current densities used in simulations.…”

“…Jiang et al [7] implies a more complicated underlying relationship between operating parameters and EAT. Besides, it is reported by Naoki Shikazono et al [6] that a thinner EAT is found with 10% humidified H2 as fuel, compared with the EAT of about 10-15 μm when 1.2% humidified H2 is used as fuel at 1273K. This result also can't be explained by our common knowledge that the EAT is controlled by the electrode ionic conductivity.…”

“…[4] that the EAT in anode is about 20 μm when the operating temperature is 1073K, current density is 0.5 A/m 2 and average particle radius is 0.1 μm. Furthermore, it is revealed in their study that the EAT is increased to about 60 μm when operating temperature is increased to 1273 K. Yoshinori Suzue et al [5] and Naoki Shikazono et al [6] solved the governing equations describing gas, ion and electron transport using Lattice Boltzmann Method (LBM) based on their reconstructed…”

Towards the 3D Modelling of the Effective Conductivity of Solid Oxide Fuel Cell Electrodes – Validation against experimental measurements and prediction of electrochemical performance

“…Suzue et al [15] used the LBM to model the transport phenomena and electrochemical reactions in 3D SOFC anode microstructures derived using stochastic correlation reconstruction, through which 3D distributions of potential and current were obtained. Gas diffusion was assumed to be due to the combination of binary Fickian diffusion of H 2 and H 2 O and Knudsen diffusion, and the Butler-Volmer equation was used to describe the charge-transfer kinetics.…”