1. control of the structural phase transitions that occurs in n-type AgBiSe2 and preclude their use in conversion modules due to instability during thermal cycling, either by using alternative synthesis routes or chemical pressure induced by proper substitutions.
Between the application and the beginning of the project, preliminary results had been obtained in the host laboratory showing that the crystal structure of AgBiSe2-based materials could be modified by applying pressure. Therefore, instead of using hydrothermal synthesis to control the dimensionality of the particles (nanoparticles with controlled shape), we have decided to use chemical pressure, which is easier to control. We have shown that by finely tuning the chemical pressure through controlled substitutions, the structural phase transitions can be suppressed, leading to materials that are stable upon thermal cycling. Then, we have been able to dope the materials in order to tune their thermoelectric properties, while keeping them transition-free. The obtain powders have been densified, and at the end of the project the characterization of their performances for thermoelectric energy conversion is about to be performed.
2. improvement of the performances of p-type BiCuSeO by producing platelet grains by alternative synthesis routes (2a) or by making composites (2b).
2a. The methodology proposed to reach this objective was based on a study published in 2008 by Stampler et al., which showed that BiCuSeO powder could be produced using hydrothermal synthesis. This synthesis method is known to be very versatile and can be used to produce a large variety of microstructures by playing with the nature of the surfactants added to the synthesis medium. Therefore, we were rather confident that platelet grains could be obtained, being known that BiCuSeO crystallizes in a layered structure.
However, although many different synthesis conditions have been tested, we have never been able to reproduce the results of Stampler et al. After a thorough analysis of the influence of the various possible parameters, we have come to the conclusion that the synthesis of BiCuSeO-based materials by hydrothermal method is not possible based on this previous study, and that the observed result of Stampler et al. was actually linked to a self-combustion reaction initiated by a reaction between the reducing agent used in the synthesis and the filtering paper during drying of the powders.
However, we have been able to suggest a completely different methodology to reach the objective. By playing with the microstructure of the precursors used during the synthesis and the conditions of densification of the powders into pellets, we have been able to obtain textured pellets constituted by stacking of platelet grains.
2b. In parallel from (2a), we have produced and characterized composites between BiCuSeO-based materials and graphene. Ba-doped BiCuSeO powder has been synthesized in large amounts by using mechanical alloying, and mixed with various proportions of graphene powder, of different specific surface. The stability of the pellets in real application conditions has been tested, and surprisingly a reaction has been observed between BiCuSeO and graphene, leading to a degradation of the materials, and limiting the possible applications to near room temperature. The performances for thermoelectric energy conversion have been studied, but no improvement has been observed as compared to graphene-free materials, which shows that the formation of BiCuSeO-graphene composites does not constitute a promising strategy to improve the thermoelectric performances.
