Screening of ternary intermetallic compounds for thermoelectric applications – ScrIn

In order to screen a large number of possible configurations, a dual approach, which combines a theoretical part with an experimental part is used in our project. Density Functional Theory (DFT) is applied to calculate the enthalpy of formation at 0 K and the electronic band structure of each configurations (a configuration corresponds to a defined T-M-X combination crystallising in a defined structure-type). Several criteria are then applied to screen the configurations and identify the stable and non-metallic ones. In some cases, additional calculations are possible as phonon or Seebeck calculations.
In the second part, experiments allow to validate the theoretical predictions. First, in order to verify their stability, the screened compounds are synthetized. For the stable and monophasic ones, electronic properties (Seebeck coefficient, electrical resistivity, Hall effect) as well as the thermal conductivity are measured. For the most promising compounds, to increase their figure of merit, doping and substitution will be implemented to optimize their thermoelectric properties.

Main results of the project deal with highlighting new possible thermoelectric compounds. Furthermore, the improvement of the theoretical part allowed to validate the robustness of our method which can now be applied to other materials or/and applications. Our methodology and our theoretical results lead to international collaboration. The massive amount of calculated data in this project also generates the establishment of a database which is exploited in a PhD thesis in our team. This thesis is part of a new study mainly based on the effects of using artificial intelligence for massive screening.

Both aspects of our project lead to the development of two main axes of perspective. First one based on the experimental study of screened compounds: some of them have not yet been synthetized whereas other, already synthetized, can be optimized through doping for example. At the same time, our methodology can be applied to more complex material family as well as other goals.

C. Barreteau, J.-C. Crivello, J.-M. Joubert and E. Alleno, Optimization of criteria for an efficient screening of new thermoelectric compounds: the TiNiSi structure-type as a case-study, (2020), ACS Combinatorial Science, 22 (12), 813-820

Moll, A., Hamidou, A., Crivello, J.-C. Joubert, J.-M., Alleno, E. and Barreteau, C., SrCuP and SrCuSb Zintl phases as potential thermoelectric materials, (2022) ACS Applied Energy Material (under review)

Thermoelectric devices can be used as refrigerators or as electric power generators. Although these devices exhibit several benefits compared to others (compactness, reliability), their small conversion efficiency as well as their cost, restrict them to niche markets. In order to overcome these restrictions, two main way of research are possible: the optimization of existing materials or the search for new promising compounds. This project « Jeunes Chercheuses Jeunes Chercheurs » belongs the second category. Indeed, it aims at discovering new thermoelectric materials through a dual approach which combines first principles calculations and experimental methods.

In the first part of the project, a screening will be performed on all possible combinations among ternary intermetallic T-M-X with M an element from the first line of the transition metals, X a sp element and T a transition metal from the Ti, V, Cr columns, Sr, Ba and La. The choice of the T elements is justified by their reduced cost and their high atomic mass, required for good thermoelectric performance. Previously to calculations, more than 450 prototypes have been identified from crystallographic databases, according to our pre-define subset of elements. First principles calculations based on the Density Functional Theory, implemented on the most interesting prototypes, aim to exclude instable compositions as well as metals in order to keep stable semiconducting compounds, which can be promising thermoelectric materials.

The second stage of our project, involves studying experimentally compounds from the theoretical screening. Synthesis, structural characterization and measurement of the thermoelectric properties will be undertaken on these selected compounds in order to validate the theoretical results and find new promising compounds. For those compounds, to increase their figure of merit, doping and substitution will be implemented to optimize their thermoelectric properties.

This project, thanks to its dual approach which combines first principles calculations and experiments investigations, allows the exploration of thousands of potentials compounds. Moreover, the originality of our project lies in the fact, that both the theoretical and experimental stages are gathered under the responsibility of one single scientist, offering a constant and immediate consistency between the two stages.