Computational Approach for Structure Generation of Anisotropic Particles (CASGAP) for Data Driven Methods to Establish Structure-Property Relationships

The presence of anisotropic building blocks in synthetic or biological soft materials can have a direct influence on their physical properties. Characterization of structural anisotropy is critical in understanding their impact on structure-property relationships and for designing materials with target properties. Computational generation of three-dimensional real space structures with desired structural features will aid in the analysis of experimental characterization results like small angle scattering (SAS) profiles and will also serve to model initial configurations for physics-based simulation techniques that relate structure to properties (e.g., structural color [2], mechanical properties). Moreover, generation of structures with anisotropic particles/domains presents unique challenges as compared to their spherical counterpart and in this talk, we will highlight these challenges and as a solution present the Computational Approach for Structure Generation of Anisotropic Particles (CASGAP) method [1]. This method generates structures with anisotropic particles/domains that have a target distribution of shapes, sizes, and orientational order. We will demonstrate the use of structures generated by CASGAP in preparing large datasets that connect a structure to their scattering profile and to their calculated property such as color. Such datasets can be used to train machine learning models linking design of materials, the 3D structure, and properties, enabling forward and inverse design.<br/>[1] Nitant Gupta, Arthi Jayaraman, “Computational Approach for Structure Generation of Anisotropic Particles (CASGAP) with Targeted Distributions of Particle Design and Orientational Order” <i>submitted for peer review in May 2023</i>