<title>Automated inspection of tellurium inclusions in cadmium zinc telluride (CdZnTe)</title>

As the demand for high quality, low cost X-ray, (gamma) -ray and imaging detector devices increases, there is a need to improve the quality and production yield of semiconductor materials used in these devices. One effective strategy for improving semiconductor device yield is through the use of better device characterization tools that can rapidly and nondestructively identify defects at early stages in the fabrication process. Early screening helps to elucidate the underlying causes of defects and to reduce downstream costs associated with processing defect laden materials that are ultimately scrapped. We report here a method for characterizing tellurium inclusion defects in cadmium zinc telluride semiconductor materials based on near infrared imaging. With this approach, large area wafers are inspected rapidly and non-destructively in two and three spatial dimensions by collecting NIR image frames at multiple regions of interest throughout the wafer using an automated NIR imaging system. The NIR image frames are subjected to image processing algorithms including background correction and image binarization. Particle analysis is performed on the binarized images to reveal tellurium inclusion statistics, sufficient to pass or fail wafers. In addition, data visualization software is used to view the tellurium inclusions in two and three spatial dimensions.