Regression analysis for predicting the elasticity of liquid crystal elastomers

Abstract

It is highly desirable but difficult to understand how microscopic molecular details influence the macroscopic material properties, especially for soft materials with complex molecular architectures. In this study we focus on liquid crystal elastomers (LCEs) and aim at identifying the design variables of their molecular architectures that govern their macroscopic deformations. We apply the regression analysis using machine learning (ML) to a database containing the results of coarse grained molecular dynamics simulations of LCEs with various molecular architectures. The predictive performance of a surrogate model generated by the regression analysis is also tested. The database contains design variables for LCE molecular architectures, system and simulation conditions, and stress–strain curves for each LCE molecular system. Regression analysis is applied using the stress–strain curves as objective variables and the other factors as explanatory variables. The results reveal several descriptors governing the stress–strain curves. To test the predictive performance of the surrogate model, stress–strain curves are predicted for LCE molecular architectures that were not used in the ML scheme. The predicted curves capture the characteristics of the results obtained from molecular dynamics simulations. Therefore, the ML scheme has great potential to accelerate LCE material exploration by detecting the key design variables in the molecular architecture and predicting the LCE deformations.

Understanding Photomechanical Behavior of Liquid Crystalline‐Based Actuators

“…As an actuator whose mechanical properties can be tuned by changing just its chemical composition, [1,[9][10][11][12][13] these lightresponsive liquid crystalline-based materials are attractive materials, acting either as a standalone actuator or as integrated into a system. [6,14] One significant challenge for developing novel and advanced applications with these materials is the lack of a basic understanding of the chemistry and physics of these materials, as well as their responses to the applied stimuli, and eventually how this information can be used to yield a certain mechanical performance.…”

Structure-Based Multilevel Descriptors for High-throughput Screening of Elastomers

“…It is worth noting that while generic descriptors can be inadequate, there have been attempts to tailor descriptors specifically for elastomers to predict their mechanical properties. To date, accurately predicting the mechanical properties of elastomers requires descriptors derived from molecular structures, simulations, and/or experimental data. − Descriptors based on molecular structures typically originate from the monomers comprising the elastomers, providing insight into their physicochemical features and stoichiometry. − Descriptors based on numerical simulation, such as electronic parameters (e.g., HOMO/LUMO gap, polarizability) from density functional theory, , thermodynamic parameters from thermodynamic models, elongation simulation data from molecular dynamics, and chain architecture from Monte Carlo simulations, provide crucial information about intermolecular interactions. Descriptors based on experiments, such as FT-IR absorbance, can be utilized to gauge the degree of cross-linking. , Although these descriptors, when combined, provide a comprehensive characterization of elastomers, their dependence on simulation and experimental data limits the applicability of HTS in an elastomer system, as acquiring experimental data or conducting complex simulations for all candidates of interest is often not feasible.…”

“…32−36 Descriptors based on molecular structures typically originate from the monomers comprising the elastomers, providing insight into their physicochemical features and stoichiometry. 32−36 Descriptors based on numerical simulation, such as electronic parameters (e.g., HOMO/LUMO gap, polarizability) from density functional theory, 33,35 thermodynamic parameters from thermodynamic models, 35 elongation simulation data from molecular dynamics, 34 and chain architecture from Monte Carlo simulations, 32 provide crucial information about intermolecular interactions. Descriptors based on experiments, such as FT-IR absorbance, can be utilized to gauge the degree of cross-linking.…”