Machine learning and edge placement error improve electron beam inspection capture rate

This paper presents a holistic solution for advanced memory manufacturing, addressing challenges in electron beam inspection (EBI) capture rate and patterning defectivity with an integrated approach. Our solution links these aspects to establish a process flow with critical monitoring and detection steps, optimizing process performance and accelerating the yield learning curve. This integration utilizes machine learning to optimize the inspection sampling, enhances the utility of EBI as an early yield indicator and improves inspection time utilization, paving the way towards autonomous defect inspection. We demonstrate a substantial improvement in defect capture rate by using computational guided inspection compared to process of record (POR) EBI sampling, achieving strong correlation with final yield (electrical test). Additionally, the introduction of a dual-layer EPE metric enables the detection of yield-impacting patterning defects several weeks earlier than POR EBI step. This comprehensive approach not only improves defect detection but also supports faster and more efficient yield optimization in the fab.

Authors

Institutions

Publication Details

Published
2025-05-01
DOI
https://doi.org/10.1117/12.3051452
Primary Topic
Electron and X-Ray Spectroscopy Techniques
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Machine learning and edge placement error improve electron beam inspection capture rate

Gratiela Isai, V. Rodin, Sangyoon Shim, Ian Yi‐Feng Chang et al.
Electron and X-Ray Spectroscopy Techniques
article

Machine learning and edge placement error improve electron beam inspection capture rate

Gratiela Isai, V. Rodin, Sangyoon Shim, Ian Yi‐Feng Chang, Noor Hasnah Moin, Joy Guo, Hyosung Lee, Bas van den Broek, Souk Kim, Bob Boo, Susan Chen, W.H. Cha, Hyeon-Bo Shim, Ryan Chen, Abhishek Vikram, G. Kim, Cheol-Ho Jeong, J.H. Kim, Samuel Chun‐Lap Lo, Zhihuan Wang, M. Song, Tim Y. Tang, Cheolho Jeon, H.I. Kwon, Jin Sung Kim, Guanchen He, Fuming Wang, Younghoon Sohn, L. Dworkin, Chih-Hung Hsieh, Achim Woessner, Kateryna Lyakhova
article en

Abstract

This paper presents a holistic solution for advanced memory manufacturing, addressing challenges in electron beam inspection (EBI) capture rate and patterning defectivity with an integrated approach. Our solution links these aspects to establish a process flow with critical monitoring and detection steps, optimizing process performance and accelerating the yield learning curve. This integration utilizes machine learning to optimize the inspection sampling, enhances the utility of EBI as an early yield indicator and improves inspection time utilization, paving the way towards autonomous defect inspection. We demonstrate a substantial improvement in defect capture rate by using computational guided inspection compared to process of record (POR) EBI sampling, achieving strong correlation with final yield (electrical test). Additionally, the introduction of a dual-layer EPE metric enables the detection of yield-impacting patterning defects several weeks earlier than POR EBI step. This comprehensive approach not only improves defect detection but also supports faster and more efficient yield optimization in the fab.

Samsung (South Korea) (KR), ASML (United States) (US), ASML (Netherlands) (NL)
Affordable and clean energy
Openalex Percentile: Top 91%
Electron and X-Ray Spectroscopy Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.