Session W4: Glassy Dynamics and Jamming

Growing length-scales at the glass and jamming transition

Growing timescales are usually associated to growing length-scales. The glass transition is no exception. Recent experimental, numerical and theoretical results unveiled that the slowing down of the dynamics is accompanied by growing dynamic and static correlation lengths.\\ The aim of this talk is to present an overview of what has been recently understood about growing length-scales at the glass transition and their role in glassy dynamics. I will also discuss the main open questions and the predictions of different theoretical approaches.   

Structural rearrangements that govern flow in colloidal glasses

We use colloidal glasses to obtain insight into the flow of amorphous materials. In three dimensions and real time, we track the individual colloidal particles in a flowing glass, and we visualize structural rearrangements that occur during flow. The individual particle trajectories are used to identify regions of non-affine deformation, in which shear concentrates. Under slow shear, we observe thermally activated `shear transformation zones' embedded in an otherwise elastic amorphous material. Connections between these zones result in flow, which is homogeneous on macroscopic length scales. We calculate correlation functions for the fluctuations of non-affine displacements, and find a remarkable scaling, indicating that the flow of glasses is highly correlated in space. By reconstructing the entire three-dimensional strain distribution, we demonstrate that these system-spanning correlations arise from the elastic interactions between shear transformations.   

Order in amorphous solids

A solid is a system which has density modulations that are not erased by thermal motion, even on long timescales. The typical example is a crystal. Particles with soft, inter-penetrable cores may form such a solid. The remarkable fact that one can build a hard building with soft bricks can only be explained by the cooperation between an infinite number of particles. If soft particles may form a true, glassy solid phase, then we are forced to accept that an infinite coherence length must also exist for them. And yet, when we look at glass configurations, they appear definitely disordered, liquid-like. This is the mystery of glasses, which motivates the quest for a hidden order. I shall describe a coherence length that is accessible experimentally, and should diverge in an ideal glass state.   

Low frequency vibrational modes and particle rearrangements in colloidal glasses

We investigate the correlation between low frequency vibrational modes and fragile regions in two dimensional binary colloidal glasses, consisting of thermosensitive microgel particles. The sample packing fraction is tuned by small changes in temperature. The particles remain in their equilibrium positions in jammed states, and rearrangements are observed during temperature changes when packing fraction is changing. Using the particle displacement covariance matrix, we extract the intrinsic vibrational modes of the ``shadow'' colloidal network (i.e., with same geometric configuration and interactions but absent damping). Spatial correlations are observed between low frequency quasi-localized modes and rearranging clusters. The low frequency modes are found to contribute much more to particle rearrangement than high frequency modes. The number of rearranging clusters, as well as the size of particle rearrangements, increases as the system approaches jamming transition.   

Space-time phase transitions in models of glasses

Glass forming systems have a much richer dynamical phase structure than their thermodynamics would suggest. I will show how to explore this by means of large-deviation methods. In particular, I will demonstrate the existence of space-time phase transitions in kinetically constrained models of glasses. Similar space-time transitions seem to be present in atomistic models of supercooled liquids. In contrast to equilibrium phase transitions, which occur in configuration space, these transitions occurs in trajectory space, and are controlled by variables that drive the system out of equilibrium. Glass formers appear to live at, or close to, first-order coexistence between two distinct dynamical phases: an active and equilibrium phase, and an inactive and non-equilibrium one. This space-time coexistence helps explain observed fluctuation effects such as dynamic heterogeneity and transport decoupling. The connection of the glass transition to a true order-disorder dynamical transition offers the possibility of a unified picture of glassy phenomena.