edfas.org 11 ELECTRONIC DEVICE FAILURE ANALYSIS | VOLUME 28 NO. 1 ADVANCED SEMICONDUCTOR FAILURE ANALYSIS USING IN-SITU AFM IN FIB-SEM Radek Dao, Ondřej Novotný, Veronika Hegrová, Rosalinda Ring, and Jan Neuman NenoVision sro, Brno, Czech Republic radek.dao@nenovision.com EDFAAO (2026) 1:11-13 1537-0755/$19.00 ©ASM International® INTRODUCTION As the semiconductor industry moves toward advanced nodes and complex 3D architectures, failure analysis (FA) must respond with improved spatial resolu- tion, analytical sensitivity, and localized data correlation. Traditional workflows often fall short due to timeconsuming inter-instrument transfers, environmental contamination risks, and poor contextual continuity. To address these limitations, the authors introduce a correlative, in-situ methodology that integrates atomic force microscopy (AFM) directly into the vacuum chamber of a scanning electron microscope (SEM) or focused ion beam (FIB) platform (Fig. 1). This enables site-specific analysis under constant vacuum conditions, ensuring that topographical and electrical measurements can be carried out on the same regions of interest (ROI) without delay or repositioning. Using the AFM module LiteScope, which is integrated into the FIB/SEM, it is possible to reveal the structures beneath the sample surface and measure various properties at the same exact location while preserving the same in-situ conditions and preventing sample environmental changes such as contamination or differential pressure. Sample preparation is a cornerstone for the in-situ electrical failure analysis of semiconductors. This work is particularly focused on the combination of plasma FIB (PFIB) delayering with subsequent in-situ conductive AFM (C-AFM) analysis. Automated navigation workflows streamline probe positioning to the exact region of interest (ROI), reducing imaging artifacts and user dependency. This dual-beam configuration allows for real-time switch between milling and probing modes, thus eliminating contamination risk and enabling reproducible current and I/V spectroscopy measurements across device lay- ers in the depth. AFM-IN-SEM ANALYSIS WORKFLOW ON 3D NAND Figure 2 demonstrates an AFM-in-SEM analysis workflow on 3D NAND memory cells, where sub-50 nm features were probed without compromising the surface integrity. First, a larger area of the 3D NAND sample was sputtered with 22 nm of tungsten layer using a gas injection system (GIS) and then delayered using the PFIB, while the AFM tip was safely hidden to avoid material redeposition. The AFM tip was then readily moved to the ROI using a SEM overview with a 30° tilt. Because the AFM tip is clearly visible, the exact location of the ROI cannot be missed. Afterward, both SEM and C-AFM images were taken at the first layer. The whole process was repeated several times, opening the structure in depth, and each time acquiring SEM and C-AFM properties. This allows for the Fig. 1 Example of AFM-in-SEM integration.
RkJQdWJsaXNoZXIy MTYyMzk3NQ==