Implementation of an efficient defect classification method in photomask mass production

摘要

In photomask production environments, increasing productivity of defect inspection and improving fidelity of defect classification are important for mask makers to improve capacity of defect inspection tools and to enhance quality of production. In particular, defect classification time corresponds directly to the cost and the cycle time of mask manufacturing and new product development. KLA-Tencor has introduced an automatic defect grouping tool "ReviewSmart" which automatically bins defects with high fidelity. ReviewSmart has been reported in engineering R&D and evaluation [1], [2]. In this paper, we focus on implementation of ReviewSmart in photomask production. 592 plates were processed during the evaluation period. Those plates are for products of logic, memory and flash. Technology nodes are from 65nm to 180nm. With optimized production setting, the automatic defect grouping tool - ReviewSmart improves productivity of defect inspection by 7% with 100% fidelity. In addition to improve productivity, ReviewSmart is helpful to classify aggressive OPC caused nuisance, troubleshoot process issues and expedite product development and improve usable inspection sensitivity as well.