Investigation of Non-Covalent Interactions in the Solid-State and in Solution Phase

Abstract

Non-covalent interactions are critical components in the synthetic toolbox to modulate self-assembly processes and molecular recognition. One of the major goals in supramolecular chemistry is to understand how the properties of non-covalently bound structures correlate to interaction propensity in different states (i.e. gas phase, solution, and solid-state). Various computational methods involving electrostatic properties (i.e. electrostatic potentials) and theoretical binding energies (ΔEbind) have been explored in order to establish prediction criteria to aid the development of new compounds within the field of active pharmaceutical ingredients (APIs) and energetic materials. Methodologies that support the formation of strong and directional hydrogen bonding and halogen bonding interactions, a directional non-covalent interaction between a halogen atom and an electron rich donor, most commonly feature the concept of cocrystallization which combines two or more neutral compounds in a defined ratio to produce crystalline complexes. A few ways to modify the properties of existing compounds are through changes in crystallization conditions which result in various forms (polymorph, hydrate, solvate, or cocrystal). The work presented here in this thesis is geared toward the expansion of current understanding of how complementary synthon selection parameters can be translated into powerful, predictive tools toward the development of novel supramolecular assemblies. The utility of prediction methods toward the pursuit of new, strongly bound, energetic and non-energetic hydrogen and halogen bonded complexes are highlighted. Moreover, emphasis on how a structural library-based approach can be used to understand the role structural modifications (i.e. changes in intramolecular bonding patterns and substituents) have on substrate scope could lead to the development of more efficient synthon screening techniques. In Chapter 2, a method is implemented to identify hydrogen peroxide solvate (peroxosolvate) formation within N-heterocyclic compounds having known hydrates to improve physical and chemical properties relative to the anhydrous form. This approach relies on the rich structural information within the Cambridge Structural Database (CSD) and understanding the reactivity of pyridyl-groups in the presence of concentrated hydrogen peroxide. In Chapter 3, a discrepancy between predictions based on electrostatic potential energy difference (ΔVs) and computed ΔEbind is presented then assessed through experimental determination of the relative strength of halogen bonding interactions in solution by the direct observation of halogen bonded complexes (iodine interacting with pyridyl nitrogen: C-I···N) by Raman spectroscopy. Experimental observations are rationalized using absolutely-localized molecular orbital energy decomposition analysis (ALMO-EDA) which points to the dominant contribution charge transfer (CT) has in the ranking of binding strengths. Chapter 4 details a proof of concept toward capitalizing on the unique nature of halogen bonds to form predictable and strong interactions in the solid-state. Previous studies on 1,3,5-trihalo-2,4,6-trinitrobenzenes (TXTNB, where X=Cl, Br, and I) and solution phase halogen bonding are utilized as springboards toward the synthesis, cocrystallization, and halogen bonding interaction strength determination of strongly bound poly-nitrated benzene derivatives. The utilization of cocrystallization strategies that translate prediction models between different phases display limitations that could be addressed when considering the affects stability and donor/acceptor pair compatibility have on halogen bonding interactions. To close, Chapter 5 is a conclusion to the work presented in this thesis that provides groundwork for the development and stabilization of new supramolecular assemblies.