Simple aging in molecular glasses

The glass transition takes place when the structural (alpha) relaxation freezes in and the liquid enters a non-equilibrium solid state. This usually happens when the relaxation time, $\tau$, reaches a timescale of 1000 seconds, and $\tau=1000$ s is pragmatically used as a definition of the glass transition temperature $T_g$. However, if the glass is studied on a long enough time scale then relaxation is still seen as physical aging. Aging is a non-linear signature of the alpha relaxation in which the relaxation dynamics changes as a function of how far the system has relaxed. If the system is studied well below $T_g$ then equilibrium will not be achieved, but just below or around $T_g$ it is possible to systematically monitor the non-linear relaxation all the way to equilibrium. We have developed a micro crystat which is optimized for making fast changes in temperature and keeping temperature stable over days and even weeks. Combining this micro cryostat with a small dielectric cell it is possible to monitor non-linear relaxation in a dynamical range of more than 4 decades from 10 seconds to a $10^5$ seconds. The aging is monitored after a fast temperature jump. This means that the aging itself is isotherm, and the data therefore directly shows, how the relaxation-rate changes as volume and structure change on the isotherm. We have studied several molecular liquids and find that the data to a very large extend can be understood in terms of a TNM formalism. This implies time-aging-time superposition and suggests a simple picture where the out of equlibrium ``states'' correspond to equilibrium states - at an other temperature. If the alpha relaxation is dynamically heterogeneous as it is commonly believed, then the aging results show that fast and slow ``modes'' of the relaxation are governed in the same way by structure and volume. We hypothesize that aging according to TNM formalism is an intrinsic property of Roskilde Simple liquids.