Bias-dependent drift resistance modeling for accurate DC and AC simulation of asymmetric HV-MOSFET

N. Hefyene1, E. Vestiel2,3, B. Bakeroot4, C. Anghel1, S. Frere5, A.M. Ionescu1, R. Gillon2
1Ecole Polytechnique Fédérale de Lausanne, EPFL-STI-IMM-LEG, Lausanne, Switzerland
2Alcatel Microelectronics, Oudenaarde, Belgium
3Ecole Polytechnique Fédérale de Lausanne, EPFL-STI-IMM-LEG, Switzerland
4ELIS/TFCG-IMEC University of Gent, Ghent, Belgium
5Alcatel Microelectronics, Belgium

Tóm tắt

A detailed investigation of the drift resistance evolution with the gate and drain biases in Lateral DMOS architectures is reported. The extractions are performed using the concept of intrinsic drain voltage, V/sub K/, applied to both simulated and measured data. Some new special test structures (MESDRIFT) have been designed and fabricated in order to investigate the DMOS bias-dependent drift resistance and experimentally confirm 2D numerical simulations. Some of the physical origins, associated with drift resistance dependence on gate and drain bias, are discussed. A simple yet efficient DMOS macro-modeling strategy is reported. It consists of combining a low-voltage BSIM model module with a bias-dependent series resistance described by a quasi-empirical mathematical expression. All LDMOS operation regimes (including quasi-saturation) are captured by the proposed expression and data measured on MESDRIFT is used to calibrate the BSIM and drift parameters. The methodology does not dependent on the drift architecture and can be applied to any similar asymmetric HV MOS devices.

Từ khóa

#Voltage #MOSFETs #Electrical resistance measurement #Testing #Analytical models #Numerical simulation #Integrated circuit modeling #CMOS technology #Data mining #Performance evaluation

Tài liệu tham khảo

liu, 1997, IEEE Transactions on Electron Devices, 44, 1117, 10.1109/16.595939 hefyene, 2001, Proc ESSDERC 2001, 403 frere, 2001, Proc ESSDERC 2001, 399 silvaco, 1998, Atlas User s Manual Device Simulation Software, 1 2 anghel, 2001, Proc ESSDERC 2001, 399 10.1109/16.777172