A study on the construction of nano-oxidation process parameter prediction model by combining a processing database and multiple regression equation

作者：Huang Jen Ching^*; Weng Yung Jin

来源：Microsystem Technologies, 2014, 20(3): 367-378.

DOI：10.1007/s00542-013-1987-5

This study conducted nano-oxidation experiments in an exploration of applied voltage, time of voltage application, and relative humidity, in order to establish a nano-oxidation processing database. Multiple regression analysis (MRA) was then used to construct the impact of nano-oxidation process parameters on the oxide weight and height relationship. Finally, this study conducted a confirmation experiment and explored the impact of additional parameter amplification on the predictive power of the relationship equation. It was found that the relationship equation of the oxide height and width generated by combining the nano-oxidation processing database and the process parameters obtained using MRA was quite reasonable, and that the addition of database parameters that are in line with physical meaning could improve the predictive power of the relationship equation.

• 出版日期2014-3