Home / Research / Energy Conversion / Thermoelectrics / Nanostructured
Nanoporous Silicon
In this work, we perform theoretical studies on the thermoelectric materials, which can conv
ert heat into electricity and vice versa. Because of the ability to interconvert heat and electricity, an efficient thermoelectric material has technological importance in areas ranging from solid-state refrigeration to power generation due to its environmental benignity. However, conventional bulk-phase thermoelectric materials such as Bi2Te3 do not sho
w efficiency, which is characterized by the figure of merit (ZT), high enough to replace commercial equipments in use these days. ZT is around 1 for Bi2Te3, whereas ZT > 3 is needed to replace, for example, a conventional freon-gas refrigerator, and much effort has been made to find (or design) new materials with high ZT. Since ZT is inversely proportional to thermal conductivity () and most contribution to comes from phonons, many studies have been focused on decreasing by increasing the phonon scattering using nanostructured materials.
Here, we propose Si with nanometer-sized cylindrical pores in periodic arrangement (nanoporous Si) as a potential high-efficiency thermoelectric material. Our numerical simulations show, depending on the volume fraction of pores and the pore size, that the thermal conductivity of nanoporous Si is lower than that of bulk phase by a factor of 300 due to increased phonon scattering. This extremely low thermal conductivity makes naoporous Si very attractive for thermoelectric applications. As can be seen in the figure, when we combine these thermal conductivity calculations with electronic structure calculation of the electronic conductivity and Seebeck coefficient, we see that the figure of merit is more than 100 times larger than that of the bulk phase (d_s is the pore spacing and d_p is the pore diameter).
For further information about this work, see J.-H. Lee, J. C. Grossman, J. Reed, and G. Galli, “Lattice thermal conductivity of nano-porous Si: a molecular dynamics study,” Applied Phys. Lett. 91, 223110 (2007). And, J.-H. Lee, J. C. Grossman, and G. Galli, “Enhanced figure of merit in nano-porous Si,” NanoLetters 8, 3750 (2008) [chosen as a Nature Materials “Highlight”, Nat. Mat. 7, 925 (2008)].
Nanoporous Germanium
The thermoelectric properties of np-Ge are computed using the same combination of approaches as for np-Si. We note that a straightforward extrapolation from bulk behavior to the nanoscale is not necessarily reliable, a
nd indeed our results show that the highest ZTmax value for np-Ge is twice that of np-Si, which is far less than the 9-fold increase in the respective bulk phases (ZTmax for np-Si and np-Ge are shown in this figure by the circles). The striking difference in the increased factor of ZTmax between the respective bulk and nanoporous phases is due to the similarity of the thermal conductivity in the nanoporous geometry; while kappa of bulk Ge is three times lower than that of bulk Si, our calculations show that np-Ge and np-Si have very similar kappa . In contrast, the power factor of bulk Ge is 2.5 times larger than that of bulk Si, and np-Ge shows a similar increase the power factor over np-Si, which suggests that ZTmax of different np systems is almost entirely determined by the differences in electronic properties. Further, our calculations also show that np-Ge may have advantages over np-Si since the carrier concentration for ZTmax is two orders of magnitude lower, less than 1018 cm−3 because of the high dispersiveness in the band structure in np-Ge.
Nanostructured
Thermoelectrics
Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge MA 02139-4307
