As the temperature of a supercooled liquid is lowered, its relaxation time can increase by many orders of magnitude in a relatively narrow temperature interval. When the relaxation time becomes longer than the experimental time, the material "falls out of thermodynamic equilibrium" and we say that it has become a glass.
Being out of equilibrium gives rise to the appearance of "physical aging": the results of experiments depend on the time that has passed since the material has entered the glass regime. The glass transition is usually accompanied by the presence of strong fluctuations ("dynamical heterogeneities") spanning regions of size of the order of a few times the individual molecule size (or a few particle diameters for the case of colloidal glasses). These fluctuations are believed to be responsible for the presence of anomalous transport properties and non-exponential relaxation. They are now being directly observed by experiments probing nanoscale regions, such as atomic force microscopy for glassy polymer films, or confocal microscopy for colloidal glasses.
Our group performs theoretical calculations and numerical simulations to study non-equilibrium structural relaxation in glasses, and in particular the presence of nanometer-scale dynamical heterogeneities in glassy materials. The analytical approaches include the path-integral formulation of the Martin-Siggia-Rose formalism, sometimes combined with Renormalization Group techniques, to study non-equilibrium dynamics in systems under the presence of disorder and thermal fluctuations. The numerical techniques include classical Molecular Dynamics and dynamical Monte Carlo simulations.