Application of density functional theory calculations to elucidate the structure of lignin linkages and the intermolecular interactions among proxies of lignin, hemicellulose, and cellulose

Open Access
Author:
Watts, Heath Donald
Graduate Program:
Geosciences
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
December 13, 2011
Committee Members:
  • James David Kubicki, Dissertation Advisor
  • Katherine Haines Freeman, Committee Member
  • Jennifer Macalady, Committee Member
  • Ming Tien, Committee Member
Keywords:
  • lignin
  • DFT
  • density functional theory
  • GIAO
  • NMR
  • coniferyl alcohol
  • biofuels
Abstract:
This work involved the application of quantum chemistry techniques to proxies of plant cell wall (PCW) biopolymers including lignin, hemicellulose, and cellulose. The latter are potential sources of biofuel; however, the extraction of lignin from PCWs, which must occur before biofuel production, is costly and difficult. Lignin formation involves abiotic, random free radical additions of phenylpropanoids. Better understanding the structure, formation, and interactions of lignin could aid in PCW degradation during biofuel production. This work used density functional theory (DFT) computational chemistry calculations to study the structure and formation mechanisms that can occur in lignin, and the ability of DFT methods to predict energetically favorable hydrogen bonding and other intermolecular interactions for PCW materials. One study calculated accurate nuclear magnetic resonance (NMR) chemical shifts of common lignin linkages. These results showed that mPW1PW91/6-31G(d) NMR calculations, when used in conjunction with a multi-standard approach (benzene for sp2 C and H and methanol for sp3 C and H) produced results that accurately matched the experimental chemical shifts for common lignin linkages. A second study proposed reaction mechanisms for the formation of non-cyclic α-linkages; the most favorable mechanism matched the one predicted by experimentalists. The third study modeled hydrogen bonding (H-bonding) between water and 1-methylimidazole accurately, based on infrared spectroscopy data and calculated thermodynamic results. The B3LYP method quantitatively out-performed the M05-2X and MP2 methods for this chemical system. The fourth study used seven DFT methods to evaluate the energetic and structural differences for monomer proxy pairs of lignin, hemicellulose, and cellulose obtained from each method. The CAM-B3LYP and LC-ωPBE methods predicted the lowest energy structures. The results from this work show that the application of an appropriate DFT method to models relevant to PCWs can produce accurate results that could be useful for larger-scale calculations and for predicting experimental outcomes.