What rejuvenation and memory tell us about non-equilibrium spin glass dynamics

The spin glass problem is interesting because of its unique dynamical properties [1], the most famous of these being aging, rejuvenation, and memory. Aging is a relaxation of the susceptibility at fixed temperature [2]. Rejuvenation occurs when the sample is cooled after an aging process and returns to a reference cooling curve taken without any aging procedures – seemingly “forgetting” that it aged [2]. In the memory effect, a sample starts from a temperature below the aging temperature and upon warming the sample “remembers” its previous cooling history [2]. In this talk, we discuss our ac susceptibility experiments studying rejuvenation and memory in a single crystal of CuMn with 7.92 at.% Mn (grown by Deborah Schlagel at Ames National Laboratory). To do this, a “double waiting time” experiment is conducted where the sample is first aged at high temperature (T₁) for some time (t₁). The temperature is then lowered by 4 kelvin (T₂) and aged again for a second time (t₂). Our results show that the amount of memory retained upon heating back to T₁ depends on the aging time at each temperature – the longer t₁ is, the more memory is retained, but the longer t₂ is, the less memory is retained. While seemingly contradictory, we explain our results using the framework of temperature chaos – or the temperature change required to make the energy landscape unrecognizable [3]. At each waiting temperature, correlated regions develop, but due to temperature chaos, they are independent. Thus, waiting longer at T₁ will improve memory, but waiting longer at T₂ will reduce it.



[1] Stein, Daniel L., and Charles M. Newman. “Spin glasses and complexity.” Vol. 4. Princeton University Press (2013).

[2] Jonason, K., et al. "Memory and chaos effects in spin glasses." Physical Review Letters 81.15 (1998): 3243.

[3] Bray, Alan J., and Michael A. Moore. "Chaotic nature of the spin-glass phase." Physical review letters 58.1 (1987): 57.