Automated systematic discovery for development and production

摘要

In many semiconductor Fabs a combination of in-line photo inspection (PLY) and targeted physical failure analysis (PFA) is used to identify and monitor random defectivity, process excursions, and systematic defects. This analysis is fed back to process and design teams to create actions and fixes which drive yield ramps for integration and development. At the 45nm design node and below the number of process steps involved has greatly increased the time and cost from wafer start to testable product. To meet compressed time to market schedules, semiconductor companies must be more reliant on in-line wafer inspection and defect classification for yield learning, excursion flagging, and process split analysis.