Feature-driven three-stage hybrid model for thin film parameter estimation from XRR curves
Source abstract
Abstract
Neural network-based models offer a promising approach for X-ray reflectivity (XRR) prediction, enabling millisecond-scale estimation of thin film parameters. However, their accuracy is often limited by the non-uniqueness of solutions. To address this challenge, we propose a three-stage hybrid network model comprising an XRR curve feature extraction module, a neural network prediction module, and an interior-point method optimization module. Experimental results on Mo/Si bilayer thin films show that, compared to a traditional single-stage network model trained on raw XRR curves, the proposed three-stage model significantly reduces ε R (defined as the mean squared error between predicted and measured XRR curves). Specifically, when evaluated on experimentally measured data, the model lowers ε R to approximately 37.8% at the second stage and further reduces it to around 12.9% at the third stage.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Block Copolymer Self-Assembly · Generative Adversarial Networks and Image Synthesis · Machine Learning in Materials Science
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
University of Chinese Academy of Sciences
Funding
Strategic Priority Research Program of the Chinese Academy of Sciences · CAS Project for Young Scientists in Basic Research
References (21)
Show source-provided references
1. Saadeh · Opt. Mater. Express · 2023 · 13 · 7810.1364/OME.474887 ↗
2. Chkhalo · Opt. Lett. · 2020 · 45 · 466610.1364/OL.400526 ↗
3. Broekhuijsen · Opt. Mater. Express · 2023 · 13 · 114010.1364/OME.481049 ↗
4. Rui · Multibody Syst. Dyn. · 2005 · 14 · 31710.1007/s11044-005-5006-1 ↗
5. Penzkofer · Opt. Commun. · 1998 · 158 · 22110.1016/S0030-4018(98)00535-5 ↗
6. Nelder · Comput. J. · 1965 · 7 · 30810.1093/comjnl/7.4.308 ↗
7. Martin · Appl. Optics · 1995 · 34 · 224710.1364/AO.34.002247 ↗
8. Zorkal’tsev · Comput. Math. Math. Phys. · 2019 · 59 · 159710.1134/S0965542519100178 ↗
9. Hinderhofer · J. Appl. Crystallogr · 2023 · 56 · 310.1107/S1600576722011566 ↗
10. Pithan · J Synchrotron Rad. · 2023 · 30 · 106410.1107/S160057752300749X ↗
11. Munteanu · J. Appl. Crystallogr · 2024 · 57 · 45610.1107/S1600576724002115 ↗
12. Kim · J. Appl. Crystallogr · 2021 · 54 · 157210.1107/S1600576721009043 ↗
13. Andrejevic · Appl. Phys. Rev. · 2022 · 9 · 01142110.1063/5.0078814 ↗
14. Yao · Vacuum · 2025 · 238 · 11428710.1016/j.vacuum.2025.114287 ↗
15. Liu · Chin. J. Lasers · 2024 · 51 · 070101410.3788/CJL231540 ↗
16. Aquila · Opt. Express · 2006 · 14 · 1007310.1364/OE.14.010073 ↗
17. Burcklen · Opt. Express · 2019 · 27 · 729110.1364/OE.27.007291 ↗
18. Henke · Atom. Data Nucl. Data Tables · 1993 · 54 · 18110.1006/adnd.1993.1013 ↗
19. Yakshin · Physica B · 2000 · 283 · 14310.1016/S0921-4526(99)01909-2 ↗
20. Spearman · Am. J. Psychol. · 1904 · 15 · 7210.2307/1412159 ↗
21. Schumi-Mareček · J. Appl. Crystallogr · 2024 · 57 · 31410.1107/S1600576724001171 ↗