Some comments on the dynamics of super-cooled liquids near the glass transition

Apparently contradictory evidence for and against cooperativity, as well as for and against spatial homogeneity, has been cited in dynamics of super-cooled liquids near the glass transition. It is possible to unify both sets of claims. Cooperativity should be understood in terms of 'sequential correlations' of individual molecules, or atoms. Then, the timescale of the relaxation is defined by a single rate (selected from a distribution according to geometrically based criteria), while the shape of the relaxation is affected by the correlations. Second, the relevance of inhomogeneity is a function of the length scale investigated. Over small length scales (and, therefore, short time scales of relaxation), super-cooled liquids look inhomogeneous; over larger length scales, they appear homogeneous. A percolation-based theoretical approach consistent with all these basic conditions has also yielded results in accord with a large body of experimental data.