2026-09-19T07:35:53Zhttps://ecommons.cornell.edu/server/oai/request

oai:ecommons.cornell.edu:1813/1171592026-05-15T19:48:14Zcom_1813_35col_1813_47

Luo, Aileen Singer, Andrej Weinberger, Kilian Suntivich, Jin 2025-06-30T22:03:01Z 2024-12 ProQuest Submission ID: 14793 ProQuest Publication ID: 31760991 https://hdl.handle.net/1813/117159 http://doi.org/10.7298/72rj-4719 16921953 101 pages The development of functional thin films is essential for advancing applications in energy storage, catalysis, electronics, and optics. Synchrotron X-ray microscopy offers spatially resolved, elementally sensitive mapping of materials structure at the nanoscale. Nonetheless, traditional data analysis methods struggle to process the high-dimensional datasets efficiently and precisely, often leading to time-intensive, difficult analysis. This dissertation presents several data-driven approaches for synchrotron-based scanning X-ray diffraction microscopy characterization of functional thin films, incorporating advanced data processing techniques such as data science and machine learning to enhance data interpretation and optimize analysis time. By integrating unsupervised clustering, deep learning, and physics-aware automatic differentiation, the proposed methodologies enable rapid analysis of thin film structural morphology, which plays a critical role in fundamental materials properties. We demonstrate the utility of these approaches through case studies on materials relevant to electrocatalysis and microelectronics, showcasing improvements in both the accuracy and speed of feature extraction, as well as explore potential applications toward in situ and operando experiments. This work not only establishes a robust framework for the data-driven analysis of X-ray diffraction microscopy data but also provides insights into the structure-property relationships key to improving the performance of functional thin films. en Attribution-NonCommercial 4.0 International https://creativecommons.org/licenses/by-nc/4.0/ Data-driven Synchrotron X-ray Microscopy Characterization of Functional Thin Films dissertation or thesis https://newcatalog.library.cornell.edu/catalog/16921953 application/pdf Materials Science and Engineering Cornell University Doctor of Philosophy Ph. D., Materials Science and Engineering https://hdl.handle.net/1813/59810.2 Publication Cornell Theses and Dissertations Luo, Aileen 5893a6ea-7af3-41d7-abc6-04bcd26ab5df open.access open.access <Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="0a35b66e-0631-4fa0-b636-581550dae7da"> <Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type> <Language>en</Language> <Title>Data-driven Synchrotron X-ray Microscopy Characterization of Functional Thin Films</Title> <PublishedIn> <Publication> </Publication> </PublishedIn> <PublicationDate>2024-12</PublicationDate> <DOI>http://doi.org/10.7298/72rj-4719</DOI> <Authors> <Author> <DisplayName>Luo, Aileen</DisplayName> <Affiliation> <OrgUnit> </OrgUnit> </Affiliation> </Author> </Authors> <Editors> </Editors> <Publishers> <Publisher> <OrgUnit /> </Publisher> </Publishers> <License>https://creativecommons.org/licenses/by-nc/4.0/</License> <Abstract>The development of functional thin films is essential for advancing applications in energy storage, catalysis, electronics, and optics. Synchrotron X-ray microscopy offers spatially resolved, elementally sensitive mapping of materials structure at the nanoscale. Nonetheless, traditional data analysis methods struggle to process the high-dimensional datasets efficiently and precisely, often leading to time-intensive, difficult analysis. This dissertation presents several data-driven approaches for synchrotron-based scanning X-ray diffraction microscopy characterization of functional thin films, incorporating advanced data processing techniques such as data science and machine learning to enhance data interpretation and optimize analysis time. By integrating unsupervised clustering, deep learning, and physics-aware automatic differentiation, the proposed methodologies enable rapid analysis of thin film structural morphology, which plays a critical role in fundamental materials properties. We demonstrate the utility of these approaches through case studies on materials relevant to electrocatalysis and microelectronics, showcasing improvements in both the accuracy and speed of feature extraction, as well as explore potential applications toward in situ and operando experiments. This work not only establishes a robust framework for the data-driven analysis of X-ray diffraction microscopy data but also provides insights into the structure-property relationships key to improving the performance of functional thin films.</Abstract> <Access xmlns="http://purl.org/coar/access_right" > </Access> </Publication>