Novel ternary sulfide thermoelectric materials from high throughput transport and defect calculations

Abstract

Predicting a novel thermoelectric material requires the simultaneous optimization of several different factors. In this work we use two material design strategies to identify a new n-doped ternary sulfide thermoelectric using a high throughput framework. We have investigated a total of 234 ternary sulfides. Out of these, we identify five candidates that are thermodynamically stable and have bandstructures leading to favourable transport properties. We then predict the likelihood of n-doping, based on a high throughput defect analysis that characterizes the stability of the most favorable compensating intrinsic defects. This proved to be a crucial step which filters only one candidate without any stable electron killer defects. We also explore the possibility of improving the thermoelectric properties with thermal expansion, resulting in carrier pocket alignment. We demonstrate that in CoSbS several positive factors are simultaneously effective, i.e. good thermoelectric properties, correct bandgap, no intrinsic doping limit and an alignment of carrier pockets with volume increase.