@article{Zhang2026, 
author = {Xuemei Zhang and Jingyu Li and Shuping Guo and Lulu Huang and Mi Qin and Jianbo Zhu and Xiaoqiang Ma and Zhixin Hui and Yongsheng Zhang},
title = {Charge defect design within PbTe grain boundaries to influence the mechanical properties},
year = {2026},
journal = {Journal of Materiomics},
volume = {12},
number = {2},
keywords = {Thermoelectric, Point defects, PbTe grain boundaries, Shear modulus, Machine learning},
url = {https://www.sciopen.com/article/10.1016/j.jmat.2025.101142},
doi = {10.1016/j.jmat.2025.101142},
abstract = {Defect engineering is a key strategy for optimizing the thermoelectric (TE) properties of PbTe-based materials, and investigating charged defects in PbTe grain boundaries (GBs) is crucial for understanding its thermoelectric properties. In this study, focusing the PbTe(111)<112>/PbTe(111)<112> GBs, we perform a high-throughput investigation of the formation energies with various charged point (intrinsic and extrinsic) defects and their effects on the mechanical properties, the shear modulus. The GBs can facilitate the formation of the charged point defects (such as  VPb2−, SbPb1+), indicating the accumulations of the defects within the GBs region. Such defect accumulation can strongly increase the phonon scatterings. Furthermore, charge defects within Te—PbTe GBs lower the shear modulus to <33.1 GPa, due to the weakening interactions between Pb—Te bonds. The soft bonds around GBs will induce the stronger anharmonicity and further suppress the lattice thermal conductivity. Employing the machine learning method, we establish the relationship between the shear modulus and physical descriptors, which can efficiently screen or design the various purposes of PbTe compounds. Our work bridges the gap in understanding charged defects at grain boundaries in PbTe-based thermoelectric materials and giving rise to the design methodology to achieve high promising thermoelectric performance through charged defect influenced mechanical properties.}
}