Optimizing Process Recipe for Critical Dimensions in Dry Etching Process

Reducing variations of fabrication processes is a primary focus for improvement in etching process. One powerful tool employed towards this goal is Advanced Process Control (APC). APC adjusts recipes to compensate for measured variations. Systemic variations that were acceptable previously can become problematic as node sizes get smaller and structures become more complex. Dry etch can be a major source of variations and will be the focus of this research. Different APC methods have been developed to adjust recipes. However it is challenging to determine an optimal recipe to achieve multiple critical dimensions. In this article, an optimization method is proposed to generate optimal recipes for multiple input and multiple output (MIMO) systems. Models are constructed by applying Gaussian Process Regression and the optimal recipe is predicted using the proposed optimization method. Experimental data in dry etch process were collected and processed for model construction and recipe prediction. The results demonstrate the effectiveness of the proposed method in exploring optimal recipes for MIMO systems.