CE05 - Une énergie durable, propre, sûre et efficace

Low Cost Thermoelectric Heusler Alloys – LoCoThermH

Low-cost thermoelectric alloy

Bi2Te3 is currently the best thermoelectric material at 300 K. However, the scarcity of tellurium (Te) prevents its use in a mass market. The Fe2VAl alloy, made up of abundant chemical elements, could replace Bi2Te3 if its thermoelectric properties were improved.

Improvement of the thermoelectric properties of the low-cost alloy Fe2VAl

Thermoelectric materials make it possible to directly convert a heat flow into electric current and vice versa. They allow the development of all-solid, compact, and reliable refrigeration or thermoelectric generation applications. Bismuth telluride (Bi2Te3) is currently the reference material for 300 K thermoelectric applications. However, the cost, scarcity and toxicity of tellurium prevent its use in a mass market. Fe2VAl could become a substitute for Bi2Te3: indeed, its elements are abundant and inexpensive and the combination of its Seebeck coefficient (n- or p-type) and its electrical conductivity is better. than that of Bi2Te3. However, its thermal conductivity is ten times greater than that of Bi2Te3 and this leads to ten times lower thermoelectric performance

To decrease the thermal conductivity of Fe2VAl and improve its thermoelectric properties, a multi-scale approach is implemented, where the whole spectrum of phonons that transport heat is strongly scattered at the atomic, nanometric and mesoscopic scale (< 1µm). To enhance phonon scattering on an atomic scale, solid solutions with multiple substitutions as in Fe2V1 yTyAl1-zXz (T, X = other elements of the periodic table) are examined experimentally and theoretically. To intensify the scattering of phonons on a nanometric scale, nano-inclusions (~20 nm) of a secondary phase are introduced. At a mesoscopic scale, the reduction in grain size leads to the multiplication of grain boundaries, which also scatter phonons. By combining these effects in a single alloy, the thermal conductivity is greatly reduced and thus the thermoelectric performance of Fe2VAl is improved.

1 - Antisite defects modify the electronic structure of Fe2VAl: at high concentration, these defects open a band gap.
2 – A thermoelectric power factor optimum has been reached in n-type Fe2V1.03Al0.97; a high but not optimum value was reached in p-type Fe2V0.985Al1.015.
3 – Calculations and experiments show that Fe2V1-xTaxAl1-xSnx has a lower thermal conductivity than Fe2V1-2xTa2xAl.
4- The thermoelectric figure of merit equals the record value ZT = 0.3 at 300 K in Fe2V0.96Ta0.07Al0.97 with mesostructured grains and decreased thermal conductivity.

From a theoretical point of view, it would be useful to identify the chemical descriptors responsible for the gap opening of Fe2VAl and to study other dopants/substituents to further increase the figure of merit. From an experimental point of view, Fe2VAl could be implemented as thin films in planar thermogenerator.

The theoretical refutation of the possibility of an extreme figure of merit in a thin film of Fe2V0.8W0.2Al was first published. Then, the effect of antisite defects on the electronic structure of Fe2VAl and the lower thermal conductivity of Fe2V1-xTaxAl1-xSnx gave rise to two other papers. The high power factors in Fe2V1+xAl1-x have also been published while ZT = 0.3 in Fe2V0.96Ta0.07Al0.97 has so far only been communicated in congresses.

Thermoelectric cooling (TEC) based on materials with a dimensionless thermoelectric figure of merit (ZT) larger than 0.2 at 300 K can compete with absorption refrigeration while TEC based on materials with ZT larger than 1 at 300 K can compete with current air conditioning systems. Such materials can also be the basis of small-scale thermoelectric generator (TEG) to power the wireless sensors required by applications like the “Internet of Things” or the “Factory 4.0”. Bi2Te3 and its derivatives (ZT = 1 for both n & p type doped) already exist and are indeed commercialized in TEC & TEG applications. However, the scarcity, the cost and the toxicity of tellurium prevent a broader use of Bi2Te3 in applications. Cheap and non-toxic thermoelectric materials are thus needed to substitute Bi2Te3. The Heusler alloy Fe2VAl could be considered as a substitute to Bi2Te3 for TEC & TEG applications at 300K. Its chemical elements are indeed non-toxic and inexpensive. Moreover, Fe2VAl can be obtained both as n-type and p-type and the combination of its Seebeck coefficient and electrical conductivity is better than in Bi2Te3. However, its thermal conductivity (28 W m-1 K-1 at 300 K in pristine Fe2VAl) is unfavorable to applications since it is an order of magnitude larger than in Bi2Te3. This leads to a best ZT of 0.2 in Fe2VAl, a value too small to compete with Bi2Te3.
To solve this problem, the research consortium constituted by Institut de Chimie et de Matériaux Paris-Est (ICMPE, Thiais) and Institut Charles Gerhardt Montpellier (ICGM) proposes to reduce the thermal conductivity in Fe2VAl - without degrading the favorable combination of its Seebeck coefficient and electrical conductivity - by following an original multi-scale approach where the entire spectrum of heat carrying phonons will be strongly scattered: at an atomic scale, by substituting an element by another element with a larger mass; at a nanometric scale, by nano-precipitates of a secondary phase, and at a mesoscopic scale, by the numerous grain boundaries introduced upon reduction of the grain sizes of the bulk polycristal. To intensify scattering of the phonons at an atomic scale, solid solutions with multiple substitutions like Fe2-xMxV1 yTyAl1 zXz (M, T, X = other elements of the periodic table) will be experimentally examined. The choice of these substituent(s) will be guided by prior first principles calculations of the thermoelectric properties, which will be carried out in the frame of the Density Functional Theory (DFT). These calculations will also help the search of the best substituents maximizing the combination of its Seebeck coefficient and electrical conductivity for both n & p type conduction. To intensify scattering of the heat carrying phonon at a nanometric scale, nano-inclusions of a secondary phase will be introduced. At a mesoscale (< 500 nm), the reduction of the grain size by grinding and sintering techniques will lead to the multiplication of the grain boundaries which also scatter the phonons. By combining all these multi-scale effects in a single alloy, the lattice thermal conductivity will be strongly reduced and finally, the thermoelectric figure of merit of Fe2VAl will be improved.
From a fundamental point of view, this project will be a milestone in the search of better thermoelectric materials since very few cases of an intentional reduction of the thermal conductivity by a multi-scale approach are reported in the literature. Moreover, the comparison of the calculations of the thermoelectric properties with the experimental results will lead to a detailed and objective assessment of their predictive character. From a technological point of view, the degree of maturity of Fe2VAl will be increased and the existence of a cheap thermoelectric material for both n & p type doping should trigger further researches and developments in partnership with industry.

Project coordinator

Monsieur Eric ALLENO (Institut de Chimie et des Matériaux Paris-Est)

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.


ICMPE Institut de Chimie et des Matériaux Paris-Est
ICGM Institut de chimie moléculaire et des matériaux - Institut Charles Gerhardt Montpellier

Help of the ANR 394,038 euros
Beginning and duration of the scientific project: November 2018 - 42 Months

Useful links

Explorez notre base de projets financés



ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter