Author ORCID Identifier
https://orcid.org/0000-0003-2671-8216
Date Available
12-12-2025
Year of Publication
2025
Document Type
Master's Thesis
Degree Name
Master of Science (MS)
College
Engineering
Department/School/Program
Chemical and Materials Engineering
Faculty
Qing Shao
Faculty
Zach Hilt
Abstract
Understanding ionic hydration remains a central challenge in physical chemistry and materials science, as the interactions between ions and water molecules govern diverse phenomena ranging from electrolyte transport to selective ion separation. While experimental techniques have provided invaluable insights into solvation energetics and coordination numbers, they often lack atomistic resolution, particularly under nanoscale confinement where direct measurement becomes infeasible. Molecular dynamics (MD) simulations can bridge this gap; however, conventional classical force fields are limited by their simplified, fixed functional forms and empirical parameterization, whereas ab initio molecular dynamics (AIMD) achieves higher accuracy at the expense of severe computational cost and restricted time and length scales. Recent advances in machine learning force fields (MLFFs) offer a promising compromise, combining near-quantum accuracy with the scalability of classical simulations. By learning interatomic interactions directly from electronic structure data, MLFFs such as MACE (MultiAtomic Cluster Expansion) enable detailed and transferable modeling of complex solvation environments.
This work investigates ion hydration using pretrained MACE MLFF models in both bulk water and carbon nanotube (CNT) confinement, providing a comprehensive benchmark for the accuracy and predictive capabilities of MLFFs. Chapter 2 examines the hydration structure and dynamics of alkali (Li⁺, Na⁺, K⁺), alkaline earth (Mg²⁺, Ca²⁺), and halide (F⁻, Cl⁻, Br⁻) ions in bulk water. Analysis of radial distribution functions (RDFs), coordination numbers, and orientational distributions demonstrates close agreement with both classical and ab initio MD literature. Small and highly charged ions exhibit sharp first-shell RDF peaks and strong orientational ordering of surrounding water molecules, while larger ions produce broader, more diffuse hydration shells. Although the MACE models reproduce structural trends with high fidelity, the predicted self-diffusion coefficients and residence times show larger variance and reduced quantitative accuracy, consistent with known limitations of MLFFs in capturing long-timescale dynamical processes.
Chapter 3 extends this framework to ion hydration within (7,7), (8,8), and (9,9) carbon nanotubes, where spatial confinement and interfacial interactions strongly perturb solvation behavior. Radial density profiles reveal distinct ion positioning within the CNT interior: small, high-charge-density cations such as Li⁺ and Mg²⁺ localize near the tube center, while larger ions such as Na⁺ and K⁺ shift toward the water–CNT interface. Residence time and orientational analyses indicate that confinement enhances hydration rigidity for strongly solvated ions and disrupts coordination for weakly solvated species, producing hydration trends that qualitatively mirror bulk behavior but with confinement-dependent modulation. These results demonstrate the ability of MACE MLFFs to capture ion-specific and environment-dependent hydration phenomena across diverse spatial regimes.
Finally, Chapter 4 summarizes these findings, emphasizing the balance between accuracy and efficiency achieved by MLFF-based simulations. The strong agreement with experimental and high-level computational data validates MACE as a valuable tool for probing solvation phenomena, particularly for systems and materials that remain experimentally inaccessible. By combining transferable machine-learned potentials with molecular simulation, this work establishes a framework for studying ion hydration and transport in both bulk and nanoconfined environments, offering predictive insight for the rational design of selective membranes and next-generation separation technologies.
Digital Object Identifier (DOI)
https://doi.org/10.13023/etd.2025.575
Funding Information
This study was supported by the National Science Foundation's Grant No. 2154996 in 2023-2025.
Recommended Citation
Baker, Zachary D., "ION HYDRATION IN BULK AND NANOCONFINED WATER: INSIGHTS FROM MACHINE LEARNING FORCE FIELDS" (2025). Theses and Dissertations--Chemical and Materials Engineering. 179.
https://uknowledge.uky.edu/cme_etds/179
