Degree
Doctor of Philosophy (PhD)
Department
Chemical Engineering
Document Type
Dissertation
Abstract
Metal thin films are widely used in technologies such as photodetectors, electro-optic modulators, and plasmonic devices, where understanding their optical and electronic properties are critical to improve their performance. Unlike bulk materials, thin films can be influenced by factors such as strain, quantum confinement, and defects such, grain boundaries, which modify their electronic structure and optical response. Alloying provides an approach for tuning properties such as carrier generation and mobility, which are important for improving the performance and efficiency of these technologies. However, the electronic structure and optical properties of alloys do not necessarily vary linearly with composition, as effects such as band hybridization and chemical disorder can produce complex composition-dependent behavior. Experimental techniques such as spectroscopic ellipsometry and ultrafast pump-probe spectroscopy provide valuable information for understanding these systems but exploring large compositional spaces experimentally and through first-principles calculations can be time- and resource-intensive.
In this context, combining physical models with data-driven approaches can provide a pathway to efficiently understand and predict the properties of alloy thin films using limited experimental data. This dissertation develops a hybrid methodology that combines experimental characterization, physical models, and machine learning to investigate and predict the optical properties of metal alloy thin films across compositional spaces. The methodology integrates optical properties measurements with physically motivated descriptors, dimensionality reduction, Gaussian Process Regression, and adaptive sampling to establish relationships between alloy composition and optical response while identifying compositions that provide the most informative experimental data. This framework provides an approach for reducing the experimental effort required to explore complex alloy systems and for improving the understanding and prediction of their composition-dependent optical properties.
Date
8-20-2026
Recommended Citation
Rojas Murillo, Lina F., "OPTICAL PROPERTIES OF COMPLEX METAL ALLOY THIN FILMS: DEPOSITION, CHARACTERIZATION, AND PHYSICS-INFORMED MACHINE LEARNING" (2026). LSU Doctoral Dissertations. 7200.
https://repository.lsu.edu/gradschool_dissertations/7200
Committee Chair
Kevin M. McPeak
LSU Acknowledgement
1
LSU Accessibility Acknowledgment
1