Wafer defect extraction algorithm and analysis for TDI-based line scanning dark-field scattering

In semiconductor manufacturing, wafer defect inspection is critical, both in front-end and back-end processes. Despite the presence of conventional techniques such as scanning electron microscopy (SEM), atomic force microscopy (AFM), and others for inspecting wafer surface defects, these techniques are not sufficient to meet the urgent demands of in-line inspection for wafer fabs. This paper introduces a new method to fast inspect wafer defects utilizing dark-field scattering in conjunction with time delay integration (TDI). In addition, a wafer surface defect extraction algorithm is designed for this method. The comparison of the postprocessing results with images captured by digital microscopy shows that the algorithm can effectively reduce the influence of column fixed pattern noise (CFPN) on the subsequent defect enhancement and can extract the defect signal from the background noise. This method provides a more targeted reference for wafer surface defect inspection and has important application potential in semiconductor manufacturing.