Accurate Mask Model Approaches for Wafer Hot Spot Prepdition and Verificaition

摘要

Hotspots often lead to unexpected critical dimension (CD) behavior, degradation of process window and ultimately impact wafer yield. The most important aspects of the patterning process in device manufacturing are to calibrate the OPC model, correct for process window (PW) limiting hotspots, and correct defect issues. Despite advanced techniques for low kl lithography such as optical proximity correction, source mask optimization, multiple patterning, and EUV lithography, PW is still narrowing and it is hard to estimate the impacts of hot spots as device features shrink at an accelerated rate from node to node. In this paper, we will demonstrate the limitation of traditional hotspot detection technology and introduce a practical lithography hotspot identification method using a mask process model. Mask model-based hotspot verification is a better approach to precisely identify lithography hotspots and will provide the information needed to improve hotspots lithographic performance. Hot spot detection and correction are under development with the goal of minimizing process window impact of weak points. As standard operating procedure (SOP) of our advanced mask characterization and optimization (AMCO), contour-based prediction is the key part of hot spot modeling in order to match wafer process results. AMCO is a series of techniques designed to enhance wafer performance by harmonizing the mask and wafer processes. Our contour based approaches show good feasibility to predict hot spots in lithography.