Microscopic Modeling of Surface Roughness Scattering in Inversion Layers of MOSFETs Based on Ando's Linear Model
TLDR
This paper proposes a new microscopic model for surface roughness scattering in MOSFET inversion layers, improving accuracy over conventional methods.
Key contributions
- Proposes a new microscopic model for surface roughness scattering in MOSFET inversion layers.
- Introduces a probability density for roughness position, aligning parameters with experimental data.
- Derives a nonlocal (nondiagonal) SR scattering rate using the Green's function scheme.
- Demonstrates the model predicts higher SR-limited mobility than conventional Fermi's golden rule.
Why it matters
Accurate modeling of surface roughness scattering is crucial for predicting and optimizing MOSFET performance. This new model offers a more precise understanding, especially in strong fields and low electron energies, correcting previous underestimations of mobility.
Original Abstract
A microscopic model of surface roughness (SR) scattering in inversion layers of bulk-MOSFETs based on Ando's linear model is proposed. Taking into account the stochastic nature of roughness position induced by discontinuity of the spatial derivatives of electrostatic potential and wave-function at the semiconductor/dielectric interface, a probability density of roughness position is introduced at each atomic site. The roughness parameters in the proposed model are consistent with those from experiments, and thus, there is no discrepancy between theory and experiment. The SR scattering rate is then derived by using the Green's function scheme, and we find that the scattering rate is intrinsically nonlocal (nondiagonal) with respect to subband indices and position. In addition, the self-consistent scattering rate greatly deviates from those obtained by Fermi's golden rule in the regimes of strong effective fields and low electron energies. As a result, the conventional model tends to predict smaller SR-limited mobility.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.