Implementing FDC in the Wafer Dicing Process to Improve Product Quality

In the wafer dicing process, quality is strongly impacted by die chipping caused by the dicer saw during cutting. A tool-based FDC (Fault Detection and Classification) system can collect extensive tool sensor data from a dicer saw, generate meaningful statistical data, and store them in a database. This FDC data can be correlated by lot and wafer to measured metrology wafer metrology data. Then, advanced statistical techniques can identify which tool signals most influence die chipping then monitor those signals with FDC models, completing the circle to improve product quality.

Dicer saw process and equipment information can be collected real-time from multiple sensors on the tool by the Equipment Sentinel (ES) FDC system. User-defined models collect lot and wafer trace data, generate statistics from sensor data, and create alarms and tool interdictions. A Rudolph Technologies NSX wafer metrology tool collects chipping defect data, which is correlated with the FDC data into a single database for analysis. The Genesis advanced analysis package can be used to find relationships between chipping and FDC tool data. These critical tool parameters can then be more closely monitored.