Statistical description of the thermodynamics of glass-forming liquids

The design of new glass products centers around connecting the structural features of a network to their macroscopic properties. Since glass-forming liquids lack long-range order, their local structure is described through statistical mechanical means. Though it is possible to determine the distribution of probable microstates through stochastic sampling, such techniques may be computationally inefficient. In this paper, we introduce a statistical mechanical approach to calculate the probability and degeneracy of each possible microstate for a given composition and temperature. These probability distributions are used to calculate the corresponding thermodynamic properties, including configurational entropy, configurational enthalpy, and configurational Gibbs free energy, for an arbitrary glass-forming liquid.