Thermoelectric Materials DOI: 10.1002/anie.201405683 Thermoelectric Tin Selenide: The Beauty of Simplicity** Hao Zhang and Dmitri V. Talapin* bond anharmonicity · materials science · thermal conductivity · thermoelectrics · tin selenide Solid-state thermoelectric (TE) modules can couple heat flow and electric current. This approach offers a viable route for power generation from different sources of waste heat, such as car exhausts and industrial plants. Compared to competing technologies, such as traditional heat engines, thermoelectric devices that contain no moving parts or fluids are noise- and emission-free. They can be of small size and operate at low powers (milli- or microwatts) as an energy source for remote sensors. [1] TE modules are also used for many cooling applications. The efficiency of the TE module cannot exceed the Carnot limit, h max = (T H ÀT C )/T H , where T H and T C are the temperatures of the hot and the cold side of the device, respectively. Because of irreversible heat losses through the Joule heat and heat conduction, real TE devices operate at only a fraction of the Carnot efficiency: h = h max [(1+ ZT) 1/2 À1]/[(1+ZT) 1/2 + T C /T H ]. This expression introduces the thermoelectric figure of merit, ZT , which relates the device efficiency to the material properties: ZT = (S 2 sT)/k, where S is the thermopower, which is also known as the Seebeck coefficient, s is the electrical conductivity, and k is the thermal conductivity. In most materials, S, s, and k are interrelated in a way that makes it challenging to achieve ZT values greater than unity, which are needed for commer- cial deployment beyond a few niche applications. [2] The strategies for improving this value can be roughly divided into two groups. One set of approaches aims at maximizing the power factor S 2 s through electronic doping and band engineering. [3] The other methods target k, which is a sum of the electronic (k e ) and lattice (k L ) thermal conductivity. In fact, the most significant progress in ZT enhancement came from a reduction in k L by the development of clever strategies for phonon scattering. For example, Biswas et al. reported ZT values of approximately 2.2 at 915 K in Na-doped PbTe containing phonon-scattering endotaxial SrTe nanostructures and grain boundaries. [4] The high ZT value was attributed primarily to a low k L value of 0.5 W m À1 K À1 . Alternatively, low k values along with high ZT values have been observed in “phonon glass electron crystal” materials where the low k value is attributed to complex unit cells, such as in CeFe 3 CoSb 12 skutterudite and Yb 14 MnSb 11 Zintl phases. [5] To summarize, the rational engineering of TE materials is based on the use of heavy elements and highly doped semiconductor phases, ideally with complex multi-atomic unit cells. The introduction of nanoscale inclusions and grain boundaries improves phonon scattering to further reduce the k L values. [6] Unfortunately, the heavy elements that are used for TE materials are often either toxic (e.g., Pb) or scarce (e.g., Te). Complex unit cells as well as nanostructuring require a tight control over the synthesis process and the associated manufacturing costs. [5] A surprising and potentially transformative discovery was recently reported by Zhao and co-workers : [7] Single crystals of SnSe gave a ZT value of approximately 2.6 at 923 K along a particular crystallographic direction, the b axis. This finding is a new ZT record for bulk materials. A very high ZT value of about 2.3 was also observed along the c axis, whereas a moderate ZT value of approximately 0.8 was measured along the a axis (Figure 1). Such a TE performance goes against established rules as SnSe is made up of light elements and has a small and simple unit cell. Expectedly, this known semiconductor was ignored by the TE community in the past. The record ZT value was achieved without optimization of the carrier concentration, and as a result, the power factors of the SnSe crystals were only moderate: Values of 2.1, 10.1, and 7.7 mW cm À1 K À2 were obtained at 850 K along the a, b, and c axes, respectively. It is likely that the ZT value of SnSe will be further enhanced through systematic optimization by making use of doping, alloying, band engineering, and other methods developed for PbTe and other established TE materials. For comparison, the power factors of optimized p-type BiSbTe [6c] and PbTe [4] are approximately 40 mWcm À1 K À2 and 27 mWcm À1 K À2 and thus several times larger than that of SnSe. SnSe outperformed other materials because of the ultra- low k value observed for the high-temperature Cmcm phase. Above 800 K, the k L values along all of the crystal axes were smaller than 0.25 W m À1 K À1 , which is significantly lower than the k L value of p-type PbTe containing carefully engineered phonon scattering nanostructures and grain boundaries (ca. 0.5 W m À1 K À1 ). [4] Zhao et al. discussed the underlying physics of the extremely low k L value of the SnSe single crystals, and attributed it to the strong anharmonicity of chemical bonds in this layered compound. Density functional theory (DFT) was [*] H. Zhang, Prof. Dr. D. V. Talapin Department of Chemistry and James Franck Institute University of Chicago Chicago, IL 60637 (USA) E-mail: dvtalapin@uchicago.edu Prof. Dr. D. V. Talapin Center for Nanoscale Materials, Argonne National Lab Argonne, IL 60439 (USA) [**] We acknowledge support from the II-VI Foundation. . Angewandte Highlights 2  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2014, 53,2–4 Ü Ü These are not the final page numbers!