The investigation uses design of experiments (DOE) approach to determine the optimal parameters of photo resist (PR) coating process for photolithography in wafer manufacturing.
Plans of experiments via nine experimental runs are based on the orthogonal arrays to determine the optimum factor condition. In this study, the mean thickness and the uniformity of thickness of the PR are adopted as the quality targets of the PR coating process. This partial factorial design of the DOE method provides an economical and systematic method of determining the optimal process parameter.
A model for the mathematical prediction of the mean thickness and the uniformity of thickness for the PR has been developed in terms of the PR temperature, the chamber humidity, the spinning rate, and the dispensation rate by means of the DOE method. The PR temperature and the chamber humidity are found to be the most significant factors in both the mean thickness and the uniformity of thickness for a PR coating process.
This analysis is valid of dynamic dispensing of the specific type of PR with constant material properties and applying onto a wafer size of eight inches.
A systematic method has been developed to find suitable combinations of process parameters; hence, the traditional approach such as the trial‐and‐error method that is very time‐consuming can be avoided. Furthermore, the efficiency of designs the parameter and the quality of the products are greatly improved.
This approach can be easily applied to design an optimal parameter setting to meet various requirements of different types of products.
