2012-03-14
10:15 at HCI J 574From the theory of critical phenomena we learn that divergences in the space domain and in the time domain go together: whenever a relaxation time explodes, a correlation length does too. Yet, one of the more frustrating aspects in the phenomenology of glass-forming liquids is the the fact that the spectacular increase of the relaxation time is apparently unaccompanied by the growth of any obvious correlation length. Bluntly put, the static structure of a deeply supercooled liquid in its highly viscous phase, is virtually impossible to distinguish from that of a high temperature, very fluid liquid. In fact, this is true as long as we use standard structural correlation functions. Recent developments in the theory of the glass transition, though, have shown that this dilemma may be solved if we introduce some nonstandard means to detect the growth of structural order in glass-formers. What is intriguing in this line of research is that we have to unveil growing "order" in an intrinsically disordered material (glass), so to make any interpretation of the sentence "growing order" exotic, to say the least. In this talk, I will examine one of the new methods (there are several), based on the (actually quite old) idea that in the presence of large correlations, boundary conditions rule entirely the thermodynamic state of the system. Looking for a growing correlation length in glass-forming liquids
Andrea Cavagna
Institute for Complex system, Roma, Italy
© Dec 2024 mk 719 entries