Electronic Transport Simulations in Thermoelectric Materials beyond The Constant Relaxation Time Approximation

The large variety of complex electronic structure materials and their alloys offer highly promising directions for improvements in ZT originating from the power factor (PF). The electronic structure of novel materials contains rich features such as many valleys (typically included in machine learning descriptors [1]), warped features, elongated features, anisotropic bands, resonant states, topological states, and more. The transport dynamics also involve complex scattering physics. Many theoretical studies provide very useful direction in identifying and optimizing these materials, most of them, however, based on the constant relaxation time approximation, which brings an arbitrary degree of error, both qualitatively and quantitatively.<br/><br/>Here we describe our advanced electronic Boltzmann transport simulator, <i>ElecTra</i> [2], together with a method we developed to obtain accurate scattering rates from ab initio calculations. We demonstrate its capabilities in a series of investigations: i) regarding the importance of including all relevant scattering mechanisms in electronic transport computation [3]; ii) regarding the possibility to obtain ultra-high PF in low bandgap materials with highly asymmetric transport properties between the conduction and valence bands [4, 5]; iii) regarding the influence of the shape of the electronic structure with its elongated features and high band anisotropy on the PF, where we show that 2-3x PF variations are observed based on the band shape complexities alone; iv) regarding the possibility of reliably computing transport properties of complex band materials with low symmetry and large unit cells in realistic time scales entirely from first principles.<br/><br/><i>References</i><br/>[1] P. Graziosi, C. Kumarasinghe, N. Neophytou, <i>ACS Appl. Energy Mater.</i>, <b>3</b>, 6, 5913-5926 (2020)<br/>[2] P. Graziosi, Z. Li, N. Neophytou, <i>Computer Physics Communications</i>, <b>287</b>, 108670 (2023)<br/>[3] P. Graziosi, C. Kumarasinghe, N. Neophytou, <i>J. Appl. Phys.</i>, <b>126</b>, 155701 (2019)<br/>[4] P. Graziosi, N. Neophytou, <i>J. Phys. Chem. C</i>, <b>124</b>, 34, 18462 (2020)<br/>[5] P. Graziosi, Z. Li, N. Neophytou, <i>Appl. Phys. Lett.</i>, <b>120</b>, 072102 (2022)<br/><br/><i>Acknowledgments</i><br/>This work has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement No 678763) and and from the UK Research and Innovation fund (project reference EP/X02346X/1).