Novel Fractal Points Implementation for Electron Beam Inspection (EBI) and Data Analysis
In this work, novel fractal points have been introduced for EBI and results analysis in DRAM manufacturing that successfully boost signal resolution from within a die to a fractal point. With the implementation of the EBI recipe, the results show a significant improvement in nuisance defects reduction by 98.1%, as well as an increase in defects of interest (DOI) by 255.7%. Moreover, the Cohen’s d value of −1.32 indicates a substantial improvement in data discriminability. On the other hand, defect results with implementing fractal points demonstrate the new capabilities and value for microscopic EBI data analysis. With 10K-fold enhancement for resolution, specific weak region and defective probability for each unit can be identified. They provide new insights into evaluating die quality that indicate explicit direction to shorten the yield learning cycle and accelerate technology development prior to probe readings.
Authors
- K. Chung (ORCID: https://orcid.org/0000-0002-9843-1495)
- Th Ho
- Y Wang
- Tanemasa Asano (ORCID: https://orcid.org/0000-0003-2887-5055)
- Xing Ji (ORCID: https://orcid.org/0000-0003-2640-5914)
- Po Jen Mo
Institutions
- Micropore Technologies (United Kingdom) (GB)
- Shine Micro (United States) (US)
- Taichung City Government (TW)
- National Taiwan Sport University (TW)
Publication Details
- Published
- 2025-03-24
- DOI
- https://doi.org/10.1109/icmts63811.2025.11068899
- Primary Topic
- Industrial Vision Systems and Defect Detection
- Type
- article
- Field-Weighted Citation Impact
- 0.00