The optical inspection system MicroProf® DI enables the inspection of structured and unstructured wafers during the entire manufacturing process. FRT's automated defect inspection enables a maximum of high-quality level. As a result, productivity can be increased and costs can be saved by preventing the further processing of defective wafers.
Intelligent defect analysis is based on advanced software that enables semiconductor manufacturers to prevent yield losses by detecting defects at an early stage. It monitors inspection result data, detects and classifies defects such as spot defects, scratches, smears and inclusions in seconds based on user-defined criteria. It meets the defect analysis requirements of semiconductor fabs and provides the user with a number of functions for checking and displaying inspection data generated by the MicroProf® DI. In addition, the data can be managed and analyzed. The software also allows the user to define regions of interest (ROIs) and classify defects located within these regions.
For structured wafers a die-based evaluation method is used. With the "die to die" evaluation method, an ROI is first defined, which is used to create the "golden image". The "Golden Image" is the averaged image of all dies. In the second step, the "Golden Image" is subtracted from the respective input image to detect any deviations from the averaged Die. Possible defects are detected by a threshold approach, marked in the input image and exported into the KLARF format. There is a disadvantage with this method, however. If errors occur repeatedly on the dies, they are not recognized as defects.
To avoid this, the "Golden Die" evaluation method can be used, in which the individual Dies are compared with an ideal Die. Shape deviations, however, can be caused by an incorrect alignment between the "Golden Image" and the respective input image. This can be caused by a slight rotation or shift between the "Golden Image" and the input image, so that a very precise alignment is essential.
For unstructured wafers, a wafer-based evaluation method is used, which is based on the contrast of the pixel gray values using blob analysis. First, defined wafer structures are automatically excluded and then the defect inspection is performed.
The software allows the user to optimize the recipe for collecting data for each inspection step, thus ensuring enhanced process control. The recipe monitors the memory for the software and reads every file with the inspection results that is written to the folder. It then analyzes the error distribution of the results file according to user-defined criteria and outputs a KLARF file in which the error signatures are classified. The analyzed KLARF file can then be loaded into the inspection results database.
Our defect inspection software provides effective visualization, versatile processing and fast generation of wafer maps as well as precise quantification and comprehensible documentation of defects. The software enables the user to display different views of a selected data set, e.g. wafer map and defect list with number, position, size and grey value of the respective defects. The software provides a wafer map showing the location of each defect in the selected data set. The defects are assigned according to user-defined criteria and synchronized with the defect list.
Whether laboratory, development, quality assurance or production - FRT offers the right inspection solution for your wafer application. Our experts will be glad to help you solve your measurement tasks by putting together the best possible system configuration for you.
[Translate to English:] MicroProf® DI