Chemical-Data-Driven Validation of Physical Theories of Liquid Crystals

Abstract Physicists have established a general approach for explaining the phase transition behavior using order parameters in phenomenological theories. Order parameters are generally accounted for by single molecular properties in mean-field theories. Such theories build on physicists’ hypotheses, which are validated by agreement between the observed and predicted values, such as the theoretical estimation of phase transition temperatures. In this study, we report chemical-data-driven methods for finding and scrutinizing physical assumptions on which physical theories are based. We demonstrate the effectiveness of our methodology by validating the assumption of the widely accepted phenomenological Landau theory for liquid-crystalline (LC) phases, wherein the symmetry of the lower-temperature phase explains phase transition. Furthermore, our method suggests that the importance of molecular flexibility is underestimated in the major mean-field theories for LC phases, compared with the widely incorporated molecular shape and intermolecular electrostatic interactions.