Theoretical studies of heat generation and diffusion in Si devices generally assume that hot electrons in Si lose their energy mainly to optical phonons. Here, we briefly review the history… Click to show full abstract
Theoretical studies of heat generation and diffusion in Si devices generally assume that hot electrons in Si lose their energy mainly to optical phonons. Here, we briefly review the history of this assumption, and using full-band Monte Carlo simulations—with electron-phonon scattering rates calculated using the rigid-ion approximation and both empirical pseudopotentials and Harris potentials—we show that, instead, electrons lose as much as 2/3 of their energy to acoustic phonons. The scattering rates that we have calculated have been used to study hot-electron effects, such as impact ionization and injection into SiO2, and are in rough agreement with those obtained using density functional theory. Moreover, direct subpicosecond pump-probe experimental results, some of them dating back to 1994, are consistent with the predictions of our model. We conclude that the study of heat generation and dissipation in nanometer-scale Si devices may require a substantial revision of the assumptions that have been considered “common wisdom” so far.Theoretical studies of heat generation and diffusion in Si devices generally assume that hot electrons in Si lose their energy mainly to optical phonons. Here, we briefly review the history of this assumption, and using full-band Monte Carlo simulations—with electron-phonon scattering rates calculated using the rigid-ion approximation and both empirical pseudopotentials and Harris potentials—we show that, instead, electrons lose as much as 2/3 of their energy to acoustic phonons. The scattering rates that we have calculated have been used to study hot-electron effects, such as impact ionization and injection into SiO2, and are in rough agreement with those obtained using density functional theory. Moreover, direct subpicosecond pump-probe experimental results, some of them dating back to 1994, are consistent with the predictions of our model. We conclude that the study of heat generation and dissipation in nanometer-scale Si devices may require a substantial revision of the assumptions that have been cons...
               
Click one of the above tabs to view related content.