Degradation Analysis of SOFC Performance for Long-Term Operation with High Fuel Utilization

Abstract

A national project in Japan for SOFC development has been initiated from FY2020 to FY2024 in NEDO’s project in order to establish the method for degradation diagnosis of SOFC stacks with durability and reliability for 130,000 hours or the method for evaluation of expanding application. Central Research Institute of Electric Power Industry (CRIEPI) has operated three types of SOFC stacks, which are developed by KYOCERA, MORIMUA SOFC TECHNOLOGY, and DENSO and investigated their long-term durability in this project. We have developed the electrode polarization model for SOFC performance which is able to express the SOFC performance such as Nernst loss, cathode overvoltage, anode overvoltage, and internal resistance (IR) loss under various conditions with high prediction accuracy. Each SOFC stack has been operated with high fuel utilization (Uf = 80% and 85%) as a severe condition for high efficiency towards 20,000-hours operation, and degradation factors for SOFC durability has been analyzed by using our electrode polarization model. Analysis results until now showed that dominant degradation factors for both Uf = 80% and 85% were increase of cathode overvoltage or IR loss as same as that of the SOFC stacks in the previous term [1]. Although anode overvoltage was a minor degradation factor, increase rate of anode overvoltage under Uf = 85% was relatively larger than that under Uf = 80%. In addition, thermal property testing will be conducted for each SOFC stack in order to include temperature term into our electrode polarization model. This new model can predict SOFC performance more accurately under various operating conditions. This work has been carried out under the project of the New Energy and Industrial Technology Development Organization (NEDO), Japan as a contract research. We would like to express our gratitude to NEDO, KYOCERA, MORIMUA SOFC TECHNOLOGY, Denso as well as to university partners and National Institute of Advanced Industrial Science and Technology (AIST).Reference[1] A. Ido et al., ECS Transactions, 91(1), 801-808 (2019).