Abstract

Experiments featuring non-equilibrium glassy dynamics under temperature changes still await interpretation. There is a widespread feeling that temperature chaos (an extreme sensitivity of the glass to temperature changes) should play a major role but, up to now, this phenomenon has been investigated solely under equilibrium conditions. In fact, the very existence of a chaotic effect in the non-equilibrium dynamics is yet to be established. In this article, we tackle this problem through a large simulation of the 3D Edwards-Anderson model, carried out on the Janus II supercomputer. We find a dynamic effect that closely parallels equilibrium temperature chaos. This dynamic temperature-chaos effect is spatially heterogeneous to a large degree and turns out to be controlled by the spin-glass coherence length ξ. Indeed, an emerging length-scale ξ* rules the crossover from weak (at ξ ≪ ξ*) to strong chaos (ξ ≫ ξ*). Extrapolations of ξ* to relevant experimental conditions are provided.

Highlights

  • Experiments featuring non-equilibrium glassy dynamics under temperature changes still await interpretation

  • temperature chaos (TC) means that the spin configurations that are typical from the Boltzmann weight at temperature T1 are very atypical at temperature T2

  • Establishing the existence of TC for the SherringtonKirkpatrick model has been a real tour de force[12]

Read more

Summary

Introduction

Experiments featuring non-equilibrium glassy dynamics under temperature changes still await interpretation. An important lesson taught by spin glasses[1] regards the fragility of the glassy phase in response to perturbations such as changes in temperature—temperature chaos (TC)2–19—in the couplings[6,7,13,14] or in the external magnetic field[5,20,21] It is somewhat controversial[22,23,24,25,26,27] whether or not TC is the physical mechanism underlying the spectacular rejuvenation and memory effects found in spin glasses[28,29,30,31] and several other materials[32,33,34,35,36]. The mathematically rigorous analysis of TC in offequilibrium dynamics seems out of reach for even in the

Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call