Production use of an integrated automatic defect classification (ADC) system operating in a laser confocal/white light imaging defect review station

This paper reports results obtained using a fully automatic defect classification package integrated in the system processor of a Laser Confocal/White Light Defect Review Tool. This system classifies defects utilizing algorithms which automatically load a cassette of wafers previously scanned by an Automatic Defect Inspection tool (ADI), aligns the wafers, and image processes to re-detect the defect. The defect region image is then analyzed for its characterization content, and the results matched with a data base of previously stored examples. System calibration is performed by presenting the system with a series of examples of each type of defect, and is edited with on-line tools provided for that purpose. We report results using the system in a production environment on a series of process levels. This effort is primarily with wafers containing logic circuitry (nonrepeating pattern). Accuracy obtained with white light only imaging is reported, as are improvements when laser confocal imaging (3-D, enhanced resolution) is added. We report also studies of optical and mechanical components of the microscope system which are critical to automatic guidelines for implementation of ADC in the production environment