Accurate Prediction of Mechanical Property of Organic Crystals Using Molecular Dynamics-Based Nanoindentation Simulations.

Abstract

Developing reliable methods to predict the mechanical response of organic crystals is essential due to their growing recognition as soft, compliant, lightweight, and ultrafast smart dynamic materials that have applications ranging from pharmaceuticals to biomaterials. Traditional experimental methods, such as nanoindentation, come with practical convenience such as simplicity and efficacy; however, they are restrictive with regard to the surface quality and are subject to other experimental constraints such as the availability of suitable samples and accessibility of as-grown crystal faces. Computational approaches, including quantum mechanics and classical molecular dynamics (MD), offer an alternative approach that is independent of the sample, yet they are known to vary greatly in their predictive accuracy. To account for this shortfall of predictive tools, in this study, we systematically benchmark three major computational methods, including density functional theory (DFT), MD-based approaches based on deformation, and direct nanoindentation simulations, against experimental Young's moduli. Our results show that the MD-simulated process of indentation conforms most favorably with the experiments and has the lowest mean absolute error. The intermolecular interactions, crystal size, indenter spacing, and unloading rates are identified as the key factors that determine the stiffness, as given by the modulus. Our findings underscore the importance of incorporating both thermal effects and experimental conditions into computational models for improved predictive accuracy and highlight the potential of nanoindentation simulations as a reliable tool for the assessment of the mechanical properties and, possibly, also for the design of organic crystals with predetermined mechanical properties and dynamic behavior.