Single-particle universality of the many-body spectral form factor
ORAL
Abstract
The dynamics of many-body systems is extraordinarily rich, including phenomena such as chaos and thermalization. A cornerstone of our understanding of the late-time dynamics of interacting quantum systems is random matrix theory (RMT), detected through measures of spectral statistics such as the Spectral Form Factor (SFF). In general, computing the SFF is a numerically difficult problem and it remains an outstanding challenge to design toy models which effectively probe different dynamical regimes. We introduce a model of free fermions in the presence of spatially correlated disorder which is designed to study random matrix statistics at the single-particle level and for which the SFF can be computed exactly. In contrast to standard RMT predictions, which predict linear growth of the SFF in time, our model grows through a series of exponential ramps. After analyzing the exactly solvable point, we will introduce interactions and discuss the crossover to a linear ramp, that is, the crossover from single-particle to many-body random matrix universality. The structure of the crossover is rich, including a numerical regime in which the SFF exhibits scale invariance consistent with predictions of an ergodicity-breaking transition.
*M.O.F. acknowledges support from the Faculty of Science at the University of Victoria through Thomas E. Baker. L.V. acknowledges support from the Slovenian Research and Innovation Agency (ARIS), Research core funding Grants No.~P1-0044, N1-0273, J1-50005 and N1-0369, as well as the Consolidator Grant Boundary-101126364 of theEuropean Research Council (ERC). T. N. I. was supported by JST PRESTO Grant No.~JPMJPR2112, JSPS KAKENHI Grant No.~JP21K13852, and the Boston University CMT visitors program. This research was supported in part by the International Centre for Theoretical Sciences (ICTS) via participation in the program ``Stability of Quantum Matter in and out of Equilibrium at Various Scales'' (code: ICTS/SQMVS2024/01).