Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
8148609 | Journal of Crystal Growth | 2018 | 28 Pages |
Abstract
In the polysilicon chemical vapor deposition reactor, the operating parameters are complex to affect the polysilicon's output. Therefore, it is very important to address the coupling problem of multiple parameters and solve the optimization in a computationally efficient manner. Here, we adopted Response Surface Methodology (RSM) to analyze the complex coupling effects of different operating parameters on silicon deposition rate (R) and further achieve effective optimization of the silicon CVD system. Based on finite numerical experiments, an accurate RSM regression model is obtained and applied to predict the R with different operating parameters, including temperature (T), pressure (P), inlet velocity (V), and inlet mole fraction of H2 (M). The analysis of variance is conducted to describe the rationality of regression model and examine the statistical significance of each factor. Consequently, the optimum combination of operating parameters for the silicon CVD reactor is: Tâ¯â¯=â¯â¯1400â¯K, Pâ¯â¯=â¯â¯3.82â¯atm, Vâ¯â¯=â¯â¯3.41â¯m/s, Mâ¯â¯=â¯â¯0.91. The validation tests and optimum solution show that the results are in good agreement with those from CFD model and the deviations of the predicted values are less than 4.19%. This work provides a theoretical guidance to operate the polysilicon CVD process.
Related Topics
Physical Sciences and Engineering
Physics and Astronomy
Condensed Matter Physics
Authors
Li-sha An, Chun-jiao Liu, Ying-wen Liu,