AUTHOR=Xu Chaoyang , Huang Shixuan , Luo Hu , Li Guoneng , Fan Yinwei , Wei Shutian , Xu Cheng , Guo Wenwen 

TITLE=Numerical analysis of thermoelectric power generation coupled with temperature-dependent material properties

JOURNAL=Frontiers in Energy Research

VOLUME=Volume 11 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2023.1315100

DOI=10.3389/fenrg.2023.1315100

ISSN=2296-598X

ABSTRACT=Thermoelectric generator (TEG) with improved performance is a promising technology in power supply and energy harvesting. Existing studies primarily adopt constant material properties to investigate the TEG performance. However, TE material properties are subjected to considerable variations with temperature. Thus, reasonable doubts are raised concerning the influence level of temperature dependent material properties on the TEG performance. To solve this problem, an efficient and comprehensive one-dimensional numerical model is developed to fully consider the three-order polynomial temperature dependent thermal conductivity, Seebeck coefficient, and electrical resistivity. Control volume and finite difference algorithms are compared, and experiments are conducted to verify the developed numerical model. The temperature distribution along the TE leg obviously differs from the parabolic shape, which is a classic temperature distribution under the assumption of constant material properties. Insights found that the local change rate of thermal conductivity and Thomson effect are the essential reasons for the abovementioned phenomenon. It is found that the Thomson heat is released in part of leg near the cold-end, whereas the Thomson heat is absorbed in the remaining part of the leg near the hot-end. The electric power on the basis of constant material properties is confirmed to be accurate enough by the developed numerical model, but the parabolic-shape of TE efficiency can be only obtained when temperature dependent material properties are considered. Furthermore, it is not wise to improve the TE efficiency by structural optimization. The present work provides an efficient and comprehensive one-dimensional numerical model to include temperature dependent material properties. New insights into the temperature and heat flux distribution, Thomson influence, and structural optimization potential, are also presented for in-depth understanding of TE conversion process.