Knowledge-based electrical monitor approach using very large array yield structures to delineate defects during process development and production yield improvement

A knowledge-based system has been developed for a set of very large electrical test pattern structures that allows one to understand and debug process problems in sub-micron EPROM/Flash technology. These structures are called 'MONGOS' due to their size ( approximately 300 K Cells) and were designed using the process target EPROM or Flash cells in an array as an exact replica of the product vehicle. Resistance and IV measurements, on pieces of the MONGOs complete with gate and substrate bias control, provide raw data input. The entire 300 K cells can be treated like one giant transistor. A post-processor program creates defect bins from these data for each layer deconfounding the defects according to rules defined in a set of calculation files. The approach is flexible in that defect bins (models) can be defined and implemented as mechanisms are refined; it is also fast since >300 K cells are tested in parallel. Thus, process improvements can be engineered much faster.<>