The main objective of the project ANR PRCI « DENZIP », performed in collaboration with Dr. Umut Aydemir from Koç University in Istanbul (Turkey), is to identify and optimize novel n-type Zintl phases for energy conversion applications by thermoelectric effects.
The thermoelectric properties of a material are characterized by the dimensionless thermoelectric figure of merit ZT defined as the ratio of the thermopower (also known as the Seebeck coefficient) squared times the absolute temperature divided by the product of the electrical resistivity by the total thermal conductivity. Zintl phases are known to generally exhibit very low thermal conductivity values (typically below 1 W m-1 K-1 above 300 K) combined with semiconducting electrical properties that can be tuned through doping. Such a combination of favorable electrical and thermal properties lead to high ZT values (on the order of 1.0) between 600 and 1200 K. However, most of the reported Zintl phases so far show p-type electrical conduction, with only a handful of them being n-type and efficient. A key aspect of the DENZIP project is to identify novel n-type Zintl phases exhibiting high thermoelectric performance. To overcome this challenge, our research activities focus on the synthesis of novel Zintl compounds offering several possible strategies to optimize their physical properties (via substitutions notably) and/or their microstructure.
The DENZIP project concerns the synthesis of novel Zintl phases that may be doped to induce a n-type electrical conduction. These syntheses use the planetary ball-milling technique which consists in loading the elements into jars with balls, that will provide the mechanical energy necessary to form the targeted phase. The obtained powders are characterized in detail by powder x-ray diffraction before being consolidated under high pressure by Spark Plasma Sintering. The bulk dense samples are then cut with a diamond wire saw into samples of appropriate size and geometry for transport properties measurements at low (2 - 300 K) and high temperatures (300 - 1000 K). These measurements include electrical resistivity, thermopower, thermal conductivity, complemented by specific heat measurements. Additional analyses based on scanning electron microscopy and transmission electron microscopy are also carried out to investigate in more detail the crystal structure and the chemical homogeneity of the synthesized samples.
So far, our efforts have been mostly devoted to the synthesis of n-type Zintl phases already reported in the literature (Mg3Sb1.5Bi1.5 and KGaSb4) with the aim at further optimizing their thermoelectric performance using various doping elements. We have initially synthesized the quaternary phase Ba2GaBiS5 that belonged to a list of potential candidates based on electronic band structure calculations. In addition to these syntheses performed at the Institut Jean Lamour (IJL, Nancy, French partner), our turkish colleagues at the Koç University (KU, Istanbul) have focused their efforts on the syntheses of Cu3-xTe2, Mg3Sb1.5Bi1.5 (using an innovative synthesis route) and BaGa2Sb2. These compounds have been characterized in detail to better correlate their microstructure to their electrical and thermal properties. The first results obtained for Mg3Sb1.5Bi1.5 in France and Turkey have demonstrated the possibility to further improve its thermoelectric performance by tuning the microstructure of optimized polycrystalline samples. The studies carried out at KU have also shown for the first time the possibility to induce a n-type conduction in BaGa2Sb2 by varying the off-stoichiometry on the chemical composition. Although the performance achieved remains moderate, these results show that controlling the off-stoichiometry and the associated defects (vacancies and/or antisite defects) is a key aspect to design n-type Zintl phases. Several novel doping elements have also been identified by IJL for the phase KGaSb4, opening new avenues for a detailed optimization of its thermoelectric properties. However, our study has also revealed that this phase is thermally unstable, making the study of the impact of these dopants on the transport properties difficult.
In parallel to these studies, at IJL, we are currently focusing our attention to the n-type phases Mg3Sb1.5Bi1.5 et Mg3Sb2 by considering other dopants on the Mg site in order to further improve their thermoelectric properties. We also envisage the syntheses of the two compounds KMnBi et KSnSb, which have recently been mentioned as potential n-type dopable Zintl phases with high thermoelectric performance.
So far (T0+18 months since the beginning of the project), two studies have been published in high-impact, international peer-reviewed journals (“Enhanced thermoelectric performance in Mg3+xSb1.5Bi0.49Te0.01 via engineering microstructure through melt-centrifugation”, Journal of Materials Chemistry A 9, 1733-1742 (2021) and “Phase-transition-enhanced thermoelectric transport in rickardite mineral Cu3-xTe2”, Chemistry of Materials 33, 1832-1841 (2021)). The results obtained have also been presented during a colloquium in Germany and in the International Conference on Thermoelectrics (ICT 2019 in South Korea). The research activities of the PhD student at IJL involved in this project have been presented during the French virtual conference on thermoelectrics, organized in October 2020.
DENZIP project addresses the issue of waste-heat recovery via the optimization of the thermoelectric properties of materials classified as Zintl phases. These compounds are currently being intensively studied due to their excellent thermoelectric performances. Their complex crystal structure and their extremely rich chemistry are two major advantages making Zintl phases a particularly fertile playground for designing novel highly-efficient thermoelectric materials. All these desirable properties are exemplified by the discovery of the Zintl phase Yb14MnSb11 which is the best high temperature p-type thermoelectric compound known to date capable of operating at temperatures well above 1000 K. For this reason, this compound is currently being developed for use in radioisotope thermoelectric generator by NASA’s Jet Propulsion Laboratory. However, no n-type analogues have been identified so far. The design of such n-type phase with similar thermoelectric properties would help to build an all-Zintl thermoelectric generators with improved efficiency compared to those currently used. This project aims to break this technological issue by identifying n-type Zintl phases and by optimizing their thermoelectric properties. The experimental studies of this project will be realized on single and polycrystalline compounds and will be accompanied by theoretical studies to provide a better understanding of the crystal structure/microstructure/transport properties relationships, should it be on doped or undoped samples. Detailed studies of the lattice dynamics of these materials will be conducted in internationally-renowned neutron scattering facilities and will help to pinpoint the origin of the extremely low lattice thermal conductivity of these compounds.
DENZIP project involves two partners, Koç University in Istanbul (Turkey) and the Institute Jean Lamour in Nancy (France), which both possess a solid experience on the studies of thermoelectric compounds. While the synthesis and main structural and chemical characterizations will be performed in Koç University, the Institute Jean Lamour will devote its efforts to the measurements of the transport properties in a very broad temperature range (2 – 1000 K). The excellent complementarity of this consortium will enable to reach the main ambitious goal of this project, that is, to design a highly-efficient n-type Zintl phase for power generation applications. Furthermore, this project will strengthen the scientific collaboration between France and Turkey over the long term in the area of materials for energy.
Monsieur Christophe Candolfi (Institut Jean Lamour (Matériaux - Métallurgie - Nanosciences - Plasmas - Surfaces))
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
IJL Institut Jean Lamour (Matériaux - Métallurgie - Nanosciences - Plasmas - Surfaces)
KU Koç University
Help of the ANR 184,399 euros
Beginning and duration of the scientific project: April 2019 - 36 Months