Electron Mobility Model for Strained-Si Devices

Current transport models for nanoscale semiconductor devices

“…Several sets of parameters may be found in the literature [59,78,58] fitting equally well to the available low-field experimental data. The choice of the parameter values is not unique and, as it was pointed out in [87], depends on the model used to describe lattice dynamics and electron-phonon coupling.…”

“…One advances the model (27) representing the mobility tensor as a product of a scalar mobility m L and the scaled inverse mass tensor and making the inverse mass tensor stress dependent [58]: …”

“…The after-scattering state can be also located in a non-equivalent valley ( f-type scattering) [59]. The scalar electron mobility in Si is modeled by Matthiessen's rule including the effect of g-type and f-type phonon scattering, and the dependence on the doping concentration N I via the corresponding relaxation times t equiv ; t neq ðDE ðiÞ C Þ, and t I ðN I Þ as [58]:…”

“…The electron mobility for the ith valley in strained Si can be written as [58] m ðiÞ n;str ðN I ; DE…”

“…wherem À1 ðiÞ denotes the scaled effective mass tensor for the ith valley given by (30), and h is a known function of T and DE ðiÞ C [58]. The parameter b ¼ f Á m t =m c , where f is the mobility enhancement factor, defined as the ratio of the saturation value of the transversal mobility in the valley i at high values of strain to the unstrained mobility m L .…”