A Calculation Method of Deposition Profiles in CVD Reactors Using Genetic Algorithms

In this study, we introduced a novel and generalized calculation method to reproduce the deposition profiles in various types of chemical vapor deposition reactors. Robust and accurate calculations along with reduced computing cost were achieved by this method. Both boundary value problems and initial value problems for estimating the mass balance equations of the deposition species by iterations of numerical integrations were changed into problems of finding the linear combinations consisting of a few “basis functions,” which are inherent in the reactors and deposition species. The coefficients of the linear combinations were optimized using genetic algorithms. We could demonstrate the validity of the proposed method using various examples of the reaction mechanisms and conditions.