Techno-environmental analysis of material substitution in thermoelectric modules: non-oxide (bismuth telluride alloys) vs. oxide-based (lanthanum-doped strontium titanate and calcium cobaltite) materials

Due to high toxicity, thermal instability at high temperature, low availability, and the high cost of raw metallic alloys such as Bi2Te3 for thermoelectric (TE) applications, there has been a drive to develop earth-abundant and eco-benign TE materials suitable for high-temperature applications. Oxide-based TEs have lately been touted to satisfy these criteria, but a lifecycle assessment (LCA) and energy payback period (EPBP) assessment of both classes of materials have not been conducted. This paper presents a comparative LCA of two laboratory-based TE modules namely, non-oxide n-type selenium-doped Bi2Te3 and p-type antimony-doped Bi2Te3 (Module A) versus oxide-based n-type lanthanum-doped SrTiO3 and p-type layered Ca3Co4O9 (Module B). Electrical energy consumption (EEC) during fabrication constitutes the largest impact for both modules, even under a decarbonised grid scenario, although Module B has an overall lower EEC. Nonetheless, for Module A, the use of tellurium and antimony exhibited noticeable environmental toxicity impacts, but smaller compared to EEC. The rare earth elements contained in the n-type component of Module B, showed negligible environmental toxicity impact, but those from its p-type component is noticeably high due to the presence of cobalt oxide. Computations of performance characteristics based on the material configurations of both modules showed that Module A yielded a higher power output compared to Module B, and as the power output increases, the EPBP becomes almost identical for both modules, underscoring its integral role to EEC offsetting. Key challenges, therefore, once EEC is diminished for large-scale applications are raw materials availability and cost, alongside performance.