In metallic ruthenium oxides, transport coefficients such as the Seebeck effect S can be described considering a coexistence of metallic carriers together with localized magnetic moment contributions enhancing S[1]. The aim of this work was to investigate the See- beck coefficient in several ruthenium oxides presenting different electronic and magnetic backgrounds to better understand this coexistence effect. Different structures have been investigated: rutile, hollandites[2], R-type ferrites[3] and perovskites. These results have been compared to well known SrRuO3 and Sr2RuO4 and provide evidence of high T Seebeck coefficient protected from band structure considerations in ruthenium oxides. Optimization of chemical composition or process for substituted-GeTe using active learning assisted by machine learning and Bayesian optimization[4, 5]. Random Forest algorithm based on elementary descriptors only has demonstrated relatively good performances in modeling the thermoelectric properties for ternary and quaternary substituted-GeTe and also for Cu2ZnSnS4 Kesterite. Prediction for proposed chemical compositions or process parameters targeting optimized thermoelectric properties.
[1] I. Terasaki, Y. Sasago and K. Uchinokura, Phys. Rev. B 56, R12685 (1997).
[2] F. Pawula, S. H ́ebert, D. Pelloquin, and A. Maignan, J. Mater. Chem. C 7, 235106 (2021).
[3] F. Pawula, R. Daou, S. H ́ebert, D. Pelloquin, J. Juraszek, and A. Maignan, Phys. Rev. B 103, 235106 (2021).
[4] A. S. Nugraha, G. Lambard, J. Na, M. S. A. Hossain, T. Asahi, W. Chaikittisilp, and Y. Yamauchi, J. Mater. Chem. A 8, 13532 (2020).
[5] G. Lambard, T. T. Sasaki, K. Sodeyama, T. Ohkubo, and K. Hono, Scr. Mater. 209 114341 (2022).
