Automated In-situ Large-area De-processing of ICs with High Throughput

The objective of this proposal is to demonstrate the feasibility of producing an automated delayering and imaging system with end point detection, material density detection with built in neural network error correction. This process, coined fast Automated Delayering-Image Capture System (ADICS) leverages off of the existing Pix2Net which is a proven automated imaging 3D microchip reconstruction software suite.  This will provide a fast track to a system that automates both the delayering and imaging in situ with initial feasibility studies done on commercially available dual beam Focused Ion Beam (FIB) / Scanning Electron Microscope (SEM) system, and more specialized wide spot FIB systems.  MSI’s proven reverse engineering workflow and existing Pix2Net™ software tool suite has enabled rapid extraction of IC feature sets to form a comprehensive ‘signature’ for IP analysis verification, complete GDS II and design extractions. Despite this success, automated delayering of the microchip has been elusive. A significant opportunity exists leveraging off of the existing Pix2Net™ software suite, were the in situ removal of layers, end point detection, and image capture can be automated and incorporated into a new API leveraging off of the Pix2Net™ workflow.