Files
Abstract
Computational chemistry is a cornerstone for understanding a wide variety of chemical phenomena and different computational methods are specifically tailored to effectively analyze chemical problems. By choosing the proper chemical methods, I have provided new insights to systems with applications to biofuels and pharmaceutical development. My first study employs high-accuracy ab initio methods to obtain energetics for the combustion pathways connecting n-butanol to the toxic byproducts formaldehyde and acetaldehyde. Results from “gold standard” coupled cluster methods and robust focal-point analysis indicate that the favored pathway is one long assumed to be unimportant.
Heterocycles abound in pharmaceutical compounds, and many heterocycles can exist as more than one tautomer. Identifying the most favorable among possible tautomers is vital for structure-based drug design, as different tautomers can engage in different non-covalent interactions in drug binding sites. At the same time, these same non-covalent interactions can impact the favored tautomer, in some cases leading to preferential formation of a tautomer that is disfavored for the isolated molecule. While the role of hydrogen-bonding on tautomerism is well-appreciated, stacking interactions (roughly parallel interactions of planar π-systems) have the potential to tune the preferred tautomeric state of drug-like heterocycles. To enhance our understanding of these effects, I first benchmarked density functional theory (DFT) methods against robust CCSD(T)/CBS computations, showing that M06-2X/def2-TZVP provides accurate tautomerization energies. Applying this level of theory to a large set of annular tautomeric systems (including both known heterocycles and those that should be synthetically viable but have not yet been synthesized), I classified and identified structural and energetic trends. Subsequently, I applied these findings to assess the impact of stacking interactions with the aromatic amino acids Phe, Tyr, and Trp on tautomeric equilibria, identifying a number of cases for which stacking interactions are predicted to change the preferred tautomer compared to that present in solution.
Heterocycles abound in pharmaceutical compounds, and many heterocycles can exist as more than one tautomer. Identifying the most favorable among possible tautomers is vital for structure-based drug design, as different tautomers can engage in different non-covalent interactions in drug binding sites. At the same time, these same non-covalent interactions can impact the favored tautomer, in some cases leading to preferential formation of a tautomer that is disfavored for the isolated molecule. While the role of hydrogen-bonding on tautomerism is well-appreciated, stacking interactions (roughly parallel interactions of planar π-systems) have the potential to tune the preferred tautomeric state of drug-like heterocycles. To enhance our understanding of these effects, I first benchmarked density functional theory (DFT) methods against robust CCSD(T)/CBS computations, showing that M06-2X/def2-TZVP provides accurate tautomerization energies. Applying this level of theory to a large set of annular tautomeric systems (including both known heterocycles and those that should be synthetically viable but have not yet been synthesized), I classified and identified structural and energetic trends. Subsequently, I applied these findings to assess the impact of stacking interactions with the aromatic amino acids Phe, Tyr, and Trp on tautomeric equilibria, identifying a number of cases for which stacking interactions are predicted to change the preferred tautomer compared to that present in solution.