Predictive Spectroscopic and Thermodynamic Properties of Complex Solutions and Interfaces

William Thomas Smith

doi:10.7273/000005254

Complex chemical systems, particularly those made of many constituents, often exhibit macroscopic behavior that results in intricate experimental observables that are challenging to interpret. This may derive from the fact that the subsystem components may have overlapping responses. This Dissertation focuses on the use of theoretical methods to study these systems at the atomic level and predict how the fundamental interactions manifest themselves in the thermodynamic properties or spectroscopic response. Aluminum oxyhydroxide mineral/aqueous interfaces, relevant to the production of aluminum oxide (Bayer Process) and the Superfund site at Hanford, WA, encompass the first study. The interface was studied using a combination of classical and DFT methods to sample a wide breadth of time and spatial scales. In the first part of the study, surface proprieties, such as edge structure and acidity, were calculated and used in conjunction with larger-scale surface complexation models to predict pH-dependent aggregation behavior. The subsequent phase of this study explored the adsorption of a series of nitrates on the mineral surface and how it affects solution organization and dynamics. The study concludes with the application of molecular dynamics to predict reactivity of water on the aluminum oxyhydroxide surface and interpretation of inelastic neutron scattering measurements. The outcome of this study was the development of a ground-up approach that connects atomistic simulations to experimentally observed nanoparticle aggregation. The other study of this Dissertation examined a series of molten salts relevant to the new generation of nuclear reactors. The intermolecular interactions involving the f-elements manifest into experimental observables that are difficult to interpret. In an effort to increase worker safety and limit the proliferation of sensitive nuclear materials, in situ monitoring of the reactor via Raman spectroscopy has been proposed. High-temperature systems of LiCl–KCl with variable loading of UCl3 and LaCl3 were modeled exclusively with recently developed DFT methods in order to best investigate the electronic response of these f-elements. An analysis, aided by the use of graph theory, was used to propose how subensemble species change with an increased presence of lanthanides/actinides. An efficient workflow for predicting Raman spectrum from density functional perturbation theory is benchmarked against experimentally available data for the LaCl3 system and the first predicted spectra of a UCl3 molten salt is presented. The results from this investigation indicate, under low concentrations, the studied lanthanides/actinides produce similar Raman spectra and thus this method may not be sufficient as a stand alone methodology for online monitoring.

Predictive Spectroscopic and Thermodynamic Properties of Complex Solutions and Interfaces

Files and links (1)

Abstract

Metrics

Details