Publication Date

5-2025

Date of Final Oral Examination (Defense)

3-12-2025

Type of Culminating Activity

Dissertation

Degree Title

Doctor of Philosophy in Materials Science and Engineering

Department

Materials Science and Engineering

Supervisory Committee Chair

Oliviero Andreussi, Ph.D.

Supervisory Committee Member

Eric Jankowski, Ph.D.

Supervisory Committee Member

Jenée Cyran, Ph.D.

Supervisory Committee Member

Aurora Clark, Ph.D.

Abstract

Understanding and modeling complex systems is one of the core challenges in modern science. Whether it be the intricate interactions within molecular crystals, the evolving mechanisms of bacterial resistance, or the dynamics of catalytic surfaces, accurately representing these systems is essential for scientific progress. This dissertation addresses these challenges by focusing on advances in computational modeling, demonstrating how various methods such as Density Functional Theory (DFT), machine learning, and custom-developed methods can simplify and help predict complex behaviors across multiple domains. By applying and understanding the scope of these methods to each field, this work underscores how computational modeling bridges the gap between intricate physical phenomena and practical scientific applications.

The first section employs DFT and terahertz (THz) spectroscopy to study the vibrational properties of molecular crystals, emphasizing the adaptation of frequency scaling techniques to address temperature-induced volumetric changes in DFT calculations. This approach enhances our understanding of polymorphic behaviors and the use of computational techniques for facilitating the interpretation of THz spectra in molecular crystals.

The second section addresses the urgent need for new antibiotics due to rising antibiotic resistance. It demonstrates how machine learning and in silico methods expedite the drug discovery process, utilizing large datasets to predict drug efficacy and model bacterial resistance mechanisms efficiently. This segment also considers the potential integration of quantum computing to further accelerate antibiotic development by exploring complex chemical spaces.

The final section presents a novel approach to addressing the complexities of catalytic surface interactions. In heterogeneous catalysis, the identification of active sites and the understanding of surface interactions are critical for optimizing reactions. By automating the generation of local symmetry-invariant descriptors and identification of unique sites, this dissertation shows how computational models can reduce reliance on intuition and arbitrary decision-making, thereby facilitating systematic automation of high throughput screening for the discovery of more efficient catalysts across a wide range of materials.

Overall, this dissertation underscores the critical role of advanced computational methods in understanding and modeling complex systems in materials science, biology, and chemistry. By bridging the gap between theoretical and practical scientific applications, it highlights the pivotal contributions of computational modeling to future scientific discoveries.

DOI

10.18122/td.2367.boisestate

Download

Files over 30MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS