Condensed Matter Physics Seminar: Ruben Zakine

Abstract: First-order nonequilibrium phase transitions observed in active matter, fluid dynamics, biology, climate science, and other systems with irreversible dynamics are challenging to analyze because they cannot be inferred from a simple free energy minimization principle. Rather, the mechanism of these transitions depends crucially on the system’s dynamics, which requires us to work in trajectory space rather than in state space. Here we consider situations where the path of the transitions between competing metastable states can be characterized as the minimizer of an action, whose minimum value appears in a nonequilibrium generalization of Arrhenius' law. We introduce a new numerical tool for the minimization of the action when the fluctuations in the microscopic system are non-Gaussian and the dynamics is not governed by the standard Langevin equation with additive noise. As an illustration of the method, I will pinpoint the first-order phase transition of two spatially-extended nonequilibrium systems: the one of a reaction-diffusion network based on the Schlögl model, and the one of the Active Model B, the natural nonequilibrium extension of the Cahn-Hilliard dynamics of phase demixion. I will also explain how one can use a neural network approach to identify the paths of the transitions and the critical nuclei when the spatial dimension is too large for classical methods.