To read the full version of this content please select one of the options below:

NOVEL: A NONLINEAR VISCOELASTIC MODEL FOR THERMAL OXIDATION OF SILICON

J.P. Peng (Theoretical Modeling Department IBM, East Fishkill Laboratory, Z/47C Hopewell Junction, NY 12533)
D. Chidambarrao (Theoretical Modeling Department IBM, East Fishkill Laboratory, Z/47C Hopewell Junction, NY 12533)
G.R. Srinivasan (Theoretical Modeling Department IBM, East Fishkill Laboratory, Z/47C Hopewell Junction, NY 12533)

Abstract

We have developed a computer oxidation modeling program, named NOVEL, which has been integrated into our process simulator FINDPRO. It combines the modified Deal‐Grove growth rate model with a nonlinear viscoclastic deformation model to predict both the oxide shape and stress. Modeling the thermal oxidation of silicon presents several numerical challenges. First, the oxide region expands and deforms extensively during the process which has to be modeled as a moving boundary, large deformation problem. Second, the SiO2 mechanical property changes from clastic to viscoclastic to viscous as the processing temperature is changed from a value below the the glass transition temperature (960°C) to one above it. The viscoclastic deformation model which is adequate over the entire temperature range of interest has an intrinsic numerical singularity when the oxide viscosity (divided by time) becomes relatively lower than the elastic modulus at high temperatures. These must be handled appropriately to ensure that the modeled results are correct. In this paper, we present details of how NOVEL solves the above mentioned problems. We show examples of low temperature/high pressure oxidation of a LOCOS structure, trench isolation structure, and the technique by which the finite element program NOVEL interfaces with the finite difference process simulator FINDPRO.

Citation

Peng, J.P., Chidambarrao, D. and Srinivasan, G.R. (1991), "NOVEL: A NONLINEAR VISCOELASTIC MODEL FOR THERMAL OXIDATION OF SILICON", COMPEL - The international journal for computation and mathematics in electrical and electronic engineering, Vol. 10 No. 4, pp. 341-353. https://doi.org/10.1108/eb051711

Publisher

:

MCB UP Ltd

Copyright © 1991, MCB UP Limited