Linear response in aging glassy systems, intermittency and the Poisson statistics of record fluctuations

Abstract

We study the intermittent behavior of the energy decay and linear magnetic response of a glassy system during isothermal aging after a deep thermal quench using the Edward-Anderson spin glass model as a paradigmatic example. The large intermittent changes in the two observables are found to occur in a correlated fashion and through irreversible bursts, `quakes', which punctuate reversible and equilibrium-like fluctuations of zero average. The temporal distribution of the quakes it found to be a Poisson distribution with an average growing logarithmically on time, indicating that the quakes are triggered by record sized fluctuations. As the drift of an aging system is to a good approximation subordinated to the quakes, simple analytical expressions (Sibani et al. Phys Rev B 74, 224407, 2006) are available for the time and age dependence of the average response and average energy. These expressions are shown to capture the time dependencies of the EA simulation results. Finally, we argue that whenever the changes of the linear response function and of its conjugate autocorrelation function follow from the same intermittent events a fluctuation-dissipation-like relation can arise between the two in off-equilibrium aging.