Date of Award

Summer 2019

Document Type

Open Access Dissertation

Department

Chemical Engineering

First Advisor

Andreas Heyden

Abstract

Conversion of lignocellulosic biomass into transportation fuels or commodity and specialty chemicals will be an important and fast-growing industry within the United States over the coming decades. Its growth will be driven by a variety of factors, including increased energy demands, environmental considerations, national security, and government mandates. To achieve the desired energy efficiency and economic impact, the emerging biorefining industries need novel heterogeneous catalysts with exceptional activity, selectivity, and stability. Catalytic materials developed in the petrochemical industries are generally not suitable for processing highly functionalized feedstocks typical of the biorefinery landscape. Due to the characteristics of this biomass feedstock (aqueous, highly water soluble, very reactive, and thermally unstable), liquid-phase processing technologies are exceedingly sought after to reduce the process cost as well as to increase the targeted product selectivity. Despite making considerable progress in our understanding of the stability and the surface properties of metal-supported nanoparticles in vapor phase environments, the effect of condensed phase is less investigated and not well-understood due to the added complexity of the reaction system containing both a complex heterogeneous catalyst and a condensed phase.

In order to gain fundamental understanding of the solvation phenomena occurring at solid-liquid interfaces, our research is primarily focused on the development, validation, and application of solvation methods for the rational design of novel heterogeneous transition metal catalysts for biomass conversion processes. As prototypical reactions with relevance to biomass catalysis, we investigated the hydrodeoxygenation (HDO) of various model biomolecules such as ethanol, ethylene glycol, and guaiacol under vapor and aqueous phase processing conditions to elucidate the reaction mechanism and the effect of condensed phase on the reaction kinetics. Using first principles calculations, continuum solvation models, and mean-field microkinetic modeling, we characterized the solvent effects on the kinetics of reactions and product distributions. An important outcome of our study is the identification of uncertainty in computed solvent effects due to the uncertainty of the cavity radius of transition metal atoms in implicit solvation schemes. To further elucidate the role of water on the reaction mechanism, we performed solvation calculations with our explicit solvation scheme for metal surfaces (eSMS). We found that implicit solvation models are most appropriate whenever directional hydrogen bonding is not present or does not change significantly along the reaction coordinate. Explicit solvation calculations agree with the implicit solvation models for C-H and C-OH bond cleavages of polyols where they both predict a small (<0.10 eV) solvent effect. In contrast and unlike the implicit solvation models, our explicit solvation model predicts a larger solvent stabilization (>0.35 eV) for the O-H bond cleavage due to its ability to approximately describe hydrogen bonding. Consequently, O-H bond dissociations are significantly favored over C-H and C-OH bond dissociations of polyols under aqueous processing conditions of biomass.

Share

COinS