Machine Learning‐Accelerated Reconstruction of Periodic Nanostructures with X‐ray Fluorescence Spectroscopy Methods

Abstract With advancements in the semiconductor industry, the complexity of three‐dimensional (3D) nanostructures becomes higher with continuously decreasing feature sizes. In order to monitor the processing steps, it is crucial to accurately determine the critical dimensions and composition of thes...

Full description

Saved in:
Bibliographic Details
Main Authors: Vinh‐Binh Truong, Analía Fernández Herrero, Kas Andrle, Victor Soltwisch, Philipp Hönicke
Format: Article
Language:English
Published: Wiley-VCH 2025-05-01
Series:Advanced Materials Interfaces
Subjects:
Online Access:https://doi.org/10.1002/admi.202400898
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850151301741019136
author Vinh‐Binh Truong
Analía Fernández Herrero
Kas Andrle
Victor Soltwisch
Philipp Hönicke
author_facet Vinh‐Binh Truong
Analía Fernández Herrero
Kas Andrle
Victor Soltwisch
Philipp Hönicke
author_sort Vinh‐Binh Truong
collection DOAJ
description Abstract With advancements in the semiconductor industry, the complexity of three‐dimensional (3D) nanostructures becomes higher with continuously decreasing feature sizes. In order to monitor the processing steps, it is crucial to accurately determine the critical dimensions and composition of these nanostructures. Early detection of production malfunctions is crucial for reasonable production yields, as even minor imperfections can heavily impede device performance. Grazing X‐ray fluorescence spectroscopy methods are non‐destructive and element‐specific methods with high sensitivity for nanostructured surfaces. They enable the quantification and localization of elements within the sample, with information depths sufficient also for buried features. Such measurements can be utilized to reconstruct the shape and size of nanostructure features, with simulated data based on finite element method (FEM) Maxwell calculations. However, the computation time of FEM calculations poses a challenge (especially in the X‐ray energy regime), prolonging the reconstruction process and making it impractical for the characterization of multiple samples. To address this issue, a neural network‐based approach is adopted to replace the time‐consuming FEM‐based forward calculations. In this study, the feasibility of utilizing neural networks for nanostructure reconstruction is demonstrated on various nanostructures. The discrimination limit for different model parameters is assessed and compared against conventional FEM calculation results.
format Article
id doaj-art-5652ede0a13746f8919ed5f291d35c2f
institution OA Journals
issn 2196-7350
language English
publishDate 2025-05-01
publisher Wiley-VCH
record_format Article
series Advanced Materials Interfaces
spelling doaj-art-5652ede0a13746f8919ed5f291d35c2f2025-08-20T02:26:19ZengWiley-VCHAdvanced Materials Interfaces2196-73502025-05-011210n/an/a10.1002/admi.202400898Machine Learning‐Accelerated Reconstruction of Periodic Nanostructures with X‐ray Fluorescence Spectroscopy MethodsVinh‐Binh Truong0Analía Fernández Herrero1Kas Andrle2Victor Soltwisch3Philipp Hönicke4Physikalisch‐Technische Bundesanstalt ‐ National Metrology Institute Abbestr. 2‐12 10587 Berlin GermanyPhysikalisch‐Technische Bundesanstalt ‐ National Metrology Institute Abbestr. 2‐12 10587 Berlin GermanyLawrence Berkeley National Laboratory 1 Cyclotron Road Berkeley CA 94720 USAPhysikalisch‐Technische Bundesanstalt ‐ National Metrology Institute Abbestr. 2‐12 10587 Berlin GermanyPhysikalisch‐Technische Bundesanstalt ‐ National Metrology Institute Abbestr. 2‐12 10587 Berlin GermanyAbstract With advancements in the semiconductor industry, the complexity of three‐dimensional (3D) nanostructures becomes higher with continuously decreasing feature sizes. In order to monitor the processing steps, it is crucial to accurately determine the critical dimensions and composition of these nanostructures. Early detection of production malfunctions is crucial for reasonable production yields, as even minor imperfections can heavily impede device performance. Grazing X‐ray fluorescence spectroscopy methods are non‐destructive and element‐specific methods with high sensitivity for nanostructured surfaces. They enable the quantification and localization of elements within the sample, with information depths sufficient also for buried features. Such measurements can be utilized to reconstruct the shape and size of nanostructure features, with simulated data based on finite element method (FEM) Maxwell calculations. However, the computation time of FEM calculations poses a challenge (especially in the X‐ray energy regime), prolonging the reconstruction process and making it impractical for the characterization of multiple samples. To address this issue, a neural network‐based approach is adopted to replace the time‐consuming FEM‐based forward calculations. In this study, the feasibility of utilizing neural networks for nanostructure reconstruction is demonstrated on various nanostructures. The discrimination limit for different model parameters is assessed and compared against conventional FEM calculation results.https://doi.org/10.1002/admi.202400898finite element methodmachine learningnanostructure characterizationneural networkX‐ray fluorescence
spellingShingle Vinh‐Binh Truong
Analía Fernández Herrero
Kas Andrle
Victor Soltwisch
Philipp Hönicke
Machine Learning‐Accelerated Reconstruction of Periodic Nanostructures with X‐ray Fluorescence Spectroscopy Methods
Advanced Materials Interfaces
finite element method
machine learning
nanostructure characterization
neural network
X‐ray fluorescence
title Machine Learning‐Accelerated Reconstruction of Periodic Nanostructures with X‐ray Fluorescence Spectroscopy Methods
title_full Machine Learning‐Accelerated Reconstruction of Periodic Nanostructures with X‐ray Fluorescence Spectroscopy Methods
title_fullStr Machine Learning‐Accelerated Reconstruction of Periodic Nanostructures with X‐ray Fluorescence Spectroscopy Methods
title_full_unstemmed Machine Learning‐Accelerated Reconstruction of Periodic Nanostructures with X‐ray Fluorescence Spectroscopy Methods
title_short Machine Learning‐Accelerated Reconstruction of Periodic Nanostructures with X‐ray Fluorescence Spectroscopy Methods
title_sort machine learning accelerated reconstruction of periodic nanostructures with x ray fluorescence spectroscopy methods
topic finite element method
machine learning
nanostructure characterization
neural network
X‐ray fluorescence
url https://doi.org/10.1002/admi.202400898
work_keys_str_mv AT vinhbinhtruong machinelearningacceleratedreconstructionofperiodicnanostructureswithxrayfluorescencespectroscopymethods
AT analiafernandezherrero machinelearningacceleratedreconstructionofperiodicnanostructureswithxrayfluorescencespectroscopymethods
AT kasandrle machinelearningacceleratedreconstructionofperiodicnanostructureswithxrayfluorescencespectroscopymethods
AT victorsoltwisch machinelearningacceleratedreconstructionofperiodicnanostructureswithxrayfluorescencespectroscopymethods
AT philipphonicke machinelearningacceleratedreconstructionofperiodicnanostructureswithxrayfluorescencespectroscopymethods