Endpoint optimization for semiconductor processes

Abstract

A camera may capture reflected light from the surface of the wafer during a semiconductor process that adds or removes material from the wafer, such as an etch process. To accurately determine an endpoint for the process, a camera sampling rate and light source intensity may be optimized in the process recipe. Optimizing the light source intensity may include characterizing light intensities that will be reflected from the waiver using an image of the wafer. Pixel intensities may be used to adjust the light source intensity to compensate for more complex wafer patterns. Optimizing the camera sampling rates may include nondestructively rotating a view of the wafer and converting the sampled intensities to the frequency domain. The camera sampling rate may be increased or decreased to remove spatial noise from the image without oversampling unnecessarily. These optimized parameters may then generate a clean, repeatable trace for endpoint determination.