In this presentation, we present a set of methods for evaluating detectability of EUV resist defects using a SEM simulation framework based on the Nebula SEM simulator developed at Delft University. The full framework allows pattern generation and perturbation, SEM simulation, defect detection, and dispositioning and analysis to be integrated in an automated flow.
We use this flow to examine the limits of defect detectability in EUV-relevant pattern geometries. We present ways in which the framework can be used to evaluate different defect detection and inspection strategies. As examples, we will present comparisons between typical contour-based methods, die-to-die and die-to-database methods, and machine learning approaches. This framework provides insights into expected on-wafer SEM defect detection capabilities by providing explicit simulation-based linkages between defect geometries and SEM responses.
|