Ultra-wide-field imaging Mueller matrix spectroscopic ellipsometry for semiconductor metrology

Article https://doi.org/10.1038/s41467-025-63511-1 Ultra-wide-field imaging Mueller matrix spectroscopic ellipsometry for semiconductor metrology Juntaek Oh1, Jaehyeon Son1, Changhyeong Yoon1, Eunsoo Hwang1, Jinwoo Ahn1, Jaewon Lee1, Jinsoo Lee1, Jiyong Shin1, Donggun Lee1, Seunga Lim1, Jeongho Ahn2, Younghoon Sohn3, Sangjin Hyun4, Myungjun Lee1 & Taeyong Jo1 We propose an ultra-wide-field imaging Mueller matrix spectroscopic ellip- sometry (IMMSE) system for semiconductor metrology. The IMMSE system achieves large-area measurements with a 20 mm × 20 mm field of view (FOV)— the largest FOV reported to date—and a spatial resolution of 6.5 µm. It enables the acquisition of over 10 million Mueller matrix (MM) spectra within the FOV, while a unique signal correction algorithm ensures spectrum consistency across the FOV. Leveraging this numerous MM spectra and machine learning, spatially dense metrology across the entire wafer area is achieved. This approach provides over 1987 times more metrology data and 662 times higher throughput compared to conventional point-based methods, such as scanning electron microscopy. We experimentally demonstrate the potential of the IMMSE for yield enhancement in semiconductor manufacturing by identifying spatial variations of dynamic random access memory (DRAM) structures within individual chips as well as across the wafer. As semiconductor device features have been continuously scaled down, ensuring accurate manufacturing process control has become increasingly challenging1,2. The narrow dimensional tolerances lead to local structural variations within a chip, which impacts electrical per- formance and potentially results in device failure3,4. To effectively monitor these variations across an entire wafer, advanced semi- conductor metrology techniques with higher spatial density and faster throughput are essential5–7. Spectroscopic analysis techniques, such as spectroscopic reflec- tometry (SR) and spectroscopic ellipsometry (SE), provide essential information on thickness and critical dimensions (CD) due to their high sensitivity and non-destructive nature8–14. Mueller matrix spec- troscopic ellipsometry (MMSE) further extends these capabilities by utilizing Mueller matrix (MM) to characterize the complex structures of modern semiconductor devices, including overlay errors15–17. Despite their advantages, SR, SE, and MMSE share the inherent lim- itation of point-by-point measurement, which makes it challenging to achieve high spatial density or faster throughput for wafer-scale analysis18–20. Scanning electron microscopy (SEM), offers an alternative solution for the characterization of complex device structures with its exceptional resolution21–23. However, its small field of view (FOV) limits the identification of structural variations within individual chips as well as comprehensive wafer-scale metrology24–27. Imaging-based techniques, such as imaging reflectometry, ima- ging ellipsometry and imaging MM ellipsometry have been proposed to overcome the limitations of conventional point-based approches20,28–32. However, their performance is constrained by a lim- ited FOV of only a few millimeters and the restricted utilization of MM components33,34. These limitations restrict rapid metrology across the entire wafer and the precise characterization of complex device Received: 9 January 2025 Accepted: 21 August 2025 Check for updates 1Advanced Process Development Lab 4, Semiconductor R&D Center, Samsung Electronics Co., Ltd., 1-1 Samsungjeonja-ro, hwaseong-si, Gyeonggi-do 18848, Republic of Korea. 2DRAM Process Development Team, Process Development Department, Semiconductor R&D Center, Samsung Electronics Co., Ltd., 1-1 Samsungjeonja-ro, hwaseong-si, Gyeonggi-do 18848, Republic of Korea. 3Metrology and Inspection Team, Samsung Electronics Co., Ltd., 1-1 Samsungjeonja- ro, hwaseong-si, Gyeonggi-do 18848, Republic of Korea. 4Advanced Process Development Team, Semiconductor R&D Center, Samsung Electronics Co.