Nonequilibrium Simulations to Discern Property, Performance and Sensitivity of Faceted Molecular Crystals

Molecular solids often exist as highly facetted crystals. Interfacial behavior differs from bulk configuration for materials and this difference is especially pronounced in the case of energetic molecular crystals where interfacial atomic configuration can affect properties like sensitivity and stability. The nature and intended use of these materials necessitate an in-depth understanding of how interfacial interactions contribute to overall crystal nucleation and structure, bonding, shock sensitivity, and material stability. Material packing and affinity for shear dislocation often complement material stability and low impact sensitivity—making understanding crystal packing and morphology equally important to understanding shock sensitivity. Additionally, interfaces play a unique and substantial role in modifying properties and response in materials. The property, performance and sensitivity of faceted molecular crystals is hard to determine. Experimental investigations of molecular crystals often reveal only a limited subset of properties that are hard to ascribe to their observed behavior. A rigorous thermodynamic and kinetic analysis that is length-scale and time-scale dependent is needed for a comprehensive understanding. In this thesis, we extend first-principles informed molecular simulations to quantify non-equilibrium properties and response of energetic molecular crystals and their oriented surfaces. We explore surface free energy (SFE) in place of surface energy (SE) as a better descriptor for determining their stability and response. Furthermore, molecular crystals are often stabilized using polymeric binders. The crystals are often embedded in a polymeric matrix. These polymeric binders modify the response of energetic composites. We explore the orientation-dependent work of adhesion between molecular crystals and polymeric binders. We then examine the orientation-dependent onset of chemical reactions using reactive molecular dynamics (RMD) and explore the thermal and mechanical response. We conclude by discussing the potential for using machine learning for comprehensive chemical kinetics models.