Por Gabriel Constantinescu
Since it was first discovered in 1821, the Seebeck effect has been used for measuring temperatures (thermocouples and thermopiles), in high-frequency power sensors, in powering Space probes (radioactive TE generators, RTGs), in increasing fuel efficiencies in automobiles (automotive TE generators) and in solar thermoelectric (TE) generators (STEGs), using from selected dissimilar metals and metal alloys, to specially designed TE materials, modules and generators [1-4].
TE materials can convert a temperature gradient directly into electricity due to the Seebeck effect and are considered one of the most promising means to produce ‘green’ electricity from the various available waste heat sources . Since about two thirds of the energy produced and used globally each year is lost to the environment as waste heat, the use of TE materials for direct heat-to-electricity conversion (in solid-state, with no moving parts and no maintenance) is considered to be an attractive solution for solving multiple problems addressed in current societal challenges (e.g., reducing the greenhouse gasses emissions, easy access to clean and cheap energy, contributing to the constantly increasing global energy demands).
The efficiencies of TE materials are mainly given by the dimensionless figure of merit, ZT (=(S2σT)/κ, with S, Seebeck coefficient, σ, electrical conductivity, T, absolute temperature and κ, total thermal conductivity; the electrical part (S2σ) is called power factor, PF), which depends entirely on the material’s intrinsic properties, and the temperature difference, ΔT, between hot and cold junctions. The maximum ΔT depends on the thermal and chemical stabilities of the constituent materials and on the operating temperatures of the heat and cold sources. For a given temperature range, the figure of merit Z can be maximized by increasing the numerator, S2σ, and/or reducing the denominator, κ. A good TE material should therefore possess high PF and low κ at operating temperature T. Unfortunately, S, σ and κ are not independent of each other in a given material. For example, a large σ usually implies a small S and vice versa. A maximum PF is observed for a charge carrier concentration between around 1019 and 1021 carriers/cm3, which corresponds to narrow band gap semiconductors or semimetals. On the other hand, κ=κe+κph (κe and κph are the electron and phonon contributions to κ, respectively). As κe is related to σ via the Wiedemann-Franz law (κe/σ=LT; L=2.44x10-8 WΩK-2), it follows that the maximization of ZT requires the minimization of κph. Therefore, in this perspective, a good TE material must be a good electrical conductor and a bad thermal conductor, simultaneously.
A TE generator consists of TE modules. A TE module is composed of at least one p-type and one n-type TE material, connected electrically in series and thermally in parallel (see Figure 1). If a current flows from the n-type material to the p-type one, the dominant carriers in both materials move away from the junction and carry the heat away. The junction thus becomes cold because the electrical current pumps heat away from the junction.
In order to achieve a maximum efficiency η in a TE generator, the figure of merit Z should have the highest possible value over the widest possible temperature range (highest thermal gradient), or, equivalently, ZT should be as high as possible. Therefore, the major focus of research in the area of thermoelectricity applications is the development of TE materials with high ZT values which can operate in a wide temperature range, especially at high temperatures.
Figure 1. TE generator, TE modules and TE materials. Typical design.
The traditional state-of-the-art TE materials (Bi2Te3, CoSb3, PbTe, β-FeSi2 etc.) are usually heavily doped and narrow band-gap semiconductors with ZT values around one . They have relatively low efficiencies (<10%), are far from being cost effective, cannot work at high temperatures (>300°C) without degradation and/or evaporation, and in several cases contain toxic, heavy and scarce elements (see Figure 2). Consequently, they are used mainly in niche applications, where their advantages outweigh their disadvantages.
Transition metal oxides (TMOs) [7-11] are a promising solution for waste heat recovery applications due to reasonable TE properties, high-temperature stability, low toxicity, natural abundance, tunable properties and established synthesis routes [12,13]. The best candidates for TE modules operating at mid-to-high temperatures [14,15] include p-type NaCo2O4 , the first TMO discovered to possess attractive TE properties, Bi2Sr2Co2Oy, Ca3Co4O9 [7,14], and n-type SrTiO3 , CaMnO3  and ZnO . Most TMOs exhibit low σ in bulk form due to low mobility of charge carriers . In TMOs, various nanostructuring and doping techniques can be employed to alter the carrier concentrations and mobility by adjusting the band-gaps and surface energies [12,20]. Many TMOs, such as TiO2, ZnO, CoO and MnO2, exhibit large S. The high S values usually arise from either high charge carrier effective masses due to electronic correlations  or from electron-electron interactions . An enhancement in the S of TMOs can be achieved by altering the electronic density of states (DOS) through nanostructuring techniques and/or quantum confinement . The incorporation of dopants and modifications of stoichiometry are widely employed for adjusting the TE parameters of these TMOs . The main drawback of the known TE TMOs, however, is their relatively low TE performances (ZT<0.4–0.5). Even so, oxide-based TE modules are being increasingly employed for waste heat recovery in vehicles by all major auto-makers to comply with stricter emission requirements . TE generators convert the waste heat from the exhaust into useful power that can be used to operate the on-board electrical systems, thereby increasing fuel efficiency . Additionally, the application of TMO-based TE modules in the industrial waste heat recovery sector is also gaining increasing attention due to the tremendous economic and environmental benefits that can be realized, especially in the aluminum smelting, glass manufacturing, and cement production industries .
Figure 2. Typical applications, operating temperature ranges and constituent’s abundance of various TE materials.
With all the current advancements in nanotechnology and its rapid evolution, a remarkable increase in the ZT of selected TE TMOs is inevitable!
 D. Champier, Energy Convers. Manag. 140, 167 (2017);
 T. M. Tritt, M. A. Subramanian, MRS Bull. 31, 188 (2006);
 K. M. Saqr, M. N. Musa, Therm. Sci. 13, 165 (2009);
 S. B. Riffat, X. Ma, Appl. Therm. Eng. 23, 913 (2003);
 D. M. Rowe, Int. J. Innov. Energy Sys. 1, 13 (2006);
 A. I. Hochbaum et al., Nature 451, 163 (2008);
 R. Funahashi et al., Appl. Surf. Sci. 223, 44 (2004);
 R. Robert et al., Acta Mater. 55, 4965 (2007);
 H.C. Wang et al., Curr. Appl. Phys. 10, 866 (2010);
 H. Ohta, Materials Today 10, 44 (2007);
 G. Constantinescu et al., Scripta Mater. 68, 75 (2013);
 S. Walia et al., Prog. Mater. Sci. 58, 1443 (2013);
 K. Koumoto et al., MRS Bull. 31, 206 (2006);
 I. Terasaki et al., Dalton Transactions 39, 1005 (2010);
 M. Ohtaki, J. Ceram. Soc. Jpn. 119, 770 (2011);
 I. Terasaki et al., Phys. Rev. B 56, R12685 (1997);
 M. Schrade et al., J. Appl. Phys. 115, 103705 (2014);
 T. Tsubota et al., J. Mater. Chem. 7, 85 (1997);
 A. Shakouri, Annu. Rev. Mater. Res. 41, 399 (2011);
 N. Tsuda, K. Nasu, A. Fujimori, K. Siratori, “Electronic conduction in oxides”, Springer, (2000);
 P. Limelette et al., Phys. Rev. Lett. 97, 46601 (2006);
 Y. Wang et al., Nature 423, 425 (2003);
 J. R. Sootsman et al., Angew. Chem. Int. Ed. 48, 8616 (2009);
 J. W. Fergus, J. Eur. Ceram. Soc. 32, 525 (2012);
 T. M. Tritt in: D. R. Clarke, P. Fratzl, Annu. Rev. Mater. Res. 412011, 433 (2011);
Jeronimo de Ayanz Building
Public University of Navarre
Campus de Arrosadia 31006 - Pamplona
Tel. +34 948 169512
Contacto por email