Silicon tin (Si1−x Sn x ) binary alloys are attractive materials for next-generation Si-based photonics and thermoelectronics. The energy band calculations predicted that the conduction band at the Γ point is more rapidly decreased than the others by the Sn substitution, and direct band gap semiconductors can be realized at more than sufficient Sn content [1], although they varied from 25 to 90% by the calculation methods. The corresponding wavelength range will be matched to the optical communication band, suggesting Si1−x Sn x can be applied to near-infrared light source and detector. Another interest is the dramatic reduction of the thermal conductivity by the Sn substitution owing to the mass-difference phonon scattering. Theoretically [2], the thermal conductivity of bulk Si (145 Wm−1K−1 [3]) can be decreased to ~5 Wm−1K−1 by the 10% Sn substitution; the lowest thermal conductivity (3 Wm−1K−1) is obtained at 50% Sn, which is the lowest value among bulk group-IV alloys. The low thermal conductivity achieves high thermoelectric performance and can help to ensure temperature differences within small thermoelectric generators (TEGs) integrated on Si chips, as is the case of Si nanowire TEGs [4].In contrast to the theoretical studies mentioned above, experimental studies are very limited compared with other group-IV alloys such as silicon germanium and germanium tin. Difficulties of the Si1−x Sn x synthesizing are the extremely low solid solubility of Sn in Si (0.1%), a tenth of that in Ge, and the large (∼20%) mismatch between the Si and α-Sn lattices. Nevertheless, some research groups, including us, showed some possibilities to achieve a high Sn content Si1−x Sn x thin films (x>10%) experimentally; reported the optical properties such as photoluminescence and photo-absorption. However, the technology for freely controlling Sn content is still in its infancy; the Si1−x Sn x ’s thermoelectric properties have not been clarified yet. We will discuss the recent progress on the growth techniques of Si1−x Sn x thin films using solid phase epitaxy/crystallization [5-7], molecular beam epitaxy [8,9], sputtering [10], and the potential in thin-film TEGs application. Acknowledgments This work was partly supported by JSPS KAKENHI (Nos. 19K21971, 20H05188, and 21H01366), PRESTO (No. JPMJPR15R2), and CREST (No. JPMJCR19Q5) from JST, the TEPCO Memorial Foundation, and the Naito Research Grant.
Read full abstract