Author ORCID Identifier

0000-0002-2434-5201

Document Type

Dissertation

Date of Award

5-31-2026

Degree Name

Doctor of Philosophy in Mechanical Engineering - (Ph.D.)

Department

Mechanical and Industrial Engineering

First Advisor

Dibakar Datta

Second Advisor

Eon Soo Lee

Third Advisor

Bo Shen

Fourth Advisor

Samaneh Farokhirad

Fifth Advisor

Mengqiang Zhao

Abstract

The growing demand for sustainable, high-energy-density electrochemical storage has motivated the exploration of multivalent-ion batteries based on earth-abundant elements such as aluminum, calcium, magnesium, and zinc. While multivalent charge carriers offer higher theoretical energy density than lithium, their practical deployment is hindered by sluggish ion transport, strong ion-host interactions, and structural degradation of electrode materials. Identifying host materials that can reversibly accommodate multivalent ions while maintaining structural integrity remains a fundamental challenge. The dissertation develops a scalable, end-to-end computational framework that integrates density functional theory (DFT), machine learning (ML), and generative artificial intelligence (GenAI) to accelerate the discovery of next-generation multivalent-ion battery materials. The work focuses on open-tunnel transition metal oxides, whose interconnected nanoporous channels provide natural diffusion pathways for large, highly charged ions. First-principles DFT calculations are used to systematically investigate multivalent-ion insertion in representative tunnel oxides, revealing the critical role of tunnel geometry and topology in governing ion accommodation and stability. Guided by these physical insights, a generative materials discovery framework is developed by combining a Crystal Diffusion Variational Autoencoder (CDVAE) with a fine-tuned Large Language Model (LLM) trained on over 44,000 known inorganic crystal structures. This framework generates more than 20,000 candidate transition metal oxide structures, which are subsequently screened using graph based machine learning models for thermodynamic stability and electronic properties. Promising candidates are rigorously validated through full DFT structural relaxations and phonon calculations, leading to the identification of several AI-generated materials that can be implemented for industry-scale multivalent-ion-based battery applications. Collectively, this work demonstrates a generalizable and scalable paradigm for AI-driven materials discovery that bridges physics-based modeling, data-driven generation, and first-principles validation. To further connect computational discovery with experimental realization, the dissertation also introduces a retrieval-augmented synthesis planning framework, enabling data-driven guidance for the synthesis of AI-designed materials. The proposed methodology provides a foundation for industry-relevant, large-scale discovery of multivalent-ion battery materials and can be extended to other functional material classes.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.