No Arabic abstract
A theoretical treatment of deeply supercooled liquids is difficult because their properties emerge from spatial inhomogeneities that are self-induced, transient, and nanoscopic. I use computer simulations to analyse self-induced static and dynamic heterogeneity in equilibrium systems approaching the experimental glass transition. I characterise the broad sample-to-sample fluctuations of salient dynamic and thermodynamic properties in elementary mesoscopic systems. Findings regarding local lifetimes and distributions of dynamic heterogeneity are in excellent agreement with recent single molecule studies. Surprisingly broad thermodynamic fluctuations are also found, which correlate well with dynamics fluctuations, thus providing a local test of the thermodynamic origin of slow dynamics.
Freezing is a fundamental physical phenomenon that has been studied over many decades; yet the role played by surfaces in determining nucleation has remained elusive. Here we report direct computational evidence of surface induced nucleation in supercooled systems with a negative slope of their melting line (dP/dT < 0). This unexpected result is related to the density decrease occurring upon crystallization, and to surface tension facilitating the initial nucleus formation. Our findings support the hypothesis of surface induced crystallization of ice in the atmosphere, and provide insight, at the atomistic level, into nucleation mechanisms of widely used semiconductors.
We numerically study the relaxation dynamics of several glass-forming models to their inherent structures, following quenches from equilibrium configurations sampled across a wide range of temperatures. In a mean-field Mari-Kurchan model, we find that relaxation changes from a power-law to an exponential decay below a well-defined temperature, consistent with recent findings in mean-field $p$-spin models. By contrast, for finite-dimensional systems, the relaxation is always algebraic, with a non-trivial universal exponent at high temperatures crossing over to a harmonic value at low temperatures. We demonstrate that this apparent evolution is controlled by a temperature-dependent population of localised excitations. Our work unifies several recent lines of studies aiming at a detailed characterization of the complex potential energy landscape of glass-formers.
We obtain analytic expressions for the time correlation functions of a liquid of spherical particles, exact in the limit of high dimensions $d$. The derivation is long but straightforward: a dynamic virial expansion for which only the first two terms survive, followed by a change to generalized spherical coordinates in the dynamic variables leading to saddle-point evaluation of integrals for large $d$. The problem is thus mapped onto a one-dimensional diffusion in a perturbed harmonic potential with colored noise. At high density, an ergodicity-breaking glass transition is found. In this regime, our results agree with thermodynamics, consistently with the general Random First Order Transition scenario. The glass transition density is higher than the best known lower bound for hard sphere packings in large $d$. Because our calculation is, if not rigorous, elementary, an improvement in the bound for sphere packings in large dimensions is at hand.
We compare the spatial correlations of bond-breaking events and bond-orientational relaxation in a model two-dimensional liquid undergoing Newtonian dynamics. We find that the relaxation time of the bond-breaking correlation function is much longer than the relaxation time of the bond-orientational correlation function and self-intermediate scattering function. However, the relaxation time of the bond-orientational correlation function increases faster with decreasing temperature than the relaxation time of the bond-breaking correlation function and the self-intermediate scattering function. Moreover, the dynamic correlation length that characterizes the size of correlated bond-orientational relaxation grows faster with decreasing temperature than the dynamic correlation length that characterizes the size of correlated bond-breaking events. We also examine the ensemble-dependent and ensemble-independent dynamic susceptibilities for both bond-breaking correlations and bond-orientational correlations. We find that for both correlations, the ensemble-dependent and ensemble-independent susceptibilities exhibit a maximum at nearly the same time, and this maximum occurs at a time slightly shorter than the peak position of the dynamic correlation length.
We provide a compact derivation of the static and dynamic equations for infinite-dimensional particle systems in the liquid and glass phases. The static derivation is based on the introduction of an auxiliary disorder and the use of the replica method. The dynamic derivation is based on the general analogy between replicas and the supersymmetric formulation of dynamics. We show that static and dynamic results are consistent, and follow the Random First Order Transition scenario of mean field disordered glassy systems.