標題: A parallel adaptive finite volume method for nanoscale double-gate MOSFETs simulation
作者: Li, YM
Yu, SM
友訊交大聯合研發中心
D Link NCTU Joint Res Ctr
關鍵字: parallel algorithm;domain decomposition;adaptive computational method;semiconductor device simulation;quantum correction model;nanoscale device;double-gate MOSFETs
公開日期: 1-Mar-2005
摘要: We propose in this paper a quantum correction transport model for nanoscale double-gate metal-oxide-semiconductor field effect transistor (MOSFET) device simulation. Based on adaptive finite volume, parallel domain decomposition, monotone iterative, and a posteriori error estimation methods, the model is solved numerically on a PC-based Linux cluster with MPI libraries. Quantum mechanical effect plays an important role in semiconductor nanoscale device simulation. To model this effect. a physical-based quantum correction equation is derived and solved with the hydrodynamic transport model. Numerical calculation of the quantum correction transport model is implemented with the parallel adaptive finite volume method which has recently been proposed by us in deep-submicron semiconductor device simulation. A 20 nm double-gate MOSFET is simulated with the developed quantum transport model and computational technique. Compared with a classical transport model. it is found that this model can account for the quantum mechanical effects of the nanoscale double-gate MOSFET quantitatively. Various biasing conditions have been verified on the simulated device to demonstrate its accuracy. Furthermore. for the same tested problem, the parallel adaptive computation shows very good computational performance in terms of the mesh refinements, the parallel speedup, the load-balancing, and the efficiency. (C) 2004 Elsevier B.V. All rights reserved.
URI: http://dx.doi.org/10.1016/j.cam.2004.06.001
http://hdl.handle.net/11536/13968
ISSN: 0377-0427
DOI: 10.1016/j.cam.2004.06.001
期刊: JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS
Volume: 175
Issue: 1
起始頁: 87
結束頁: 99
Appears in Collections:Conferences Paper


Files in This Item:

  1. 000225746700008.pdf