Ergodicity Breaking and Quantum Many-Body Scars

In Plain English

Usually, when you drop an ice cube into hot coffee, the ice rapidly melts as its temperature rises to match its surroundings—a process known as thermalization.

This drive toward equilibrium is taken as a given in classical physics. However, the quantum world can surprise us. In our collaborative research, we successfully engineered special states called "Quantum Many-Body Scars."

Returning to the analogy, these scars act exactly like a quantum version of ice that stubbornly refuses to melt, even when sitting in hot coffee. By avoiding thermalization, these states preserve the memory of their initial conditions, offering a pathway toward creating stable quantum memory systems.

Research Summary

The breakdown of thermalization in isolated quantum systems remains one of the most intriguing frontiers in many-body physics. Our research explores the phenomenon of quantum many-body scars (QMBS)—rare, non-thermal eigenstates embedded within an otherwise chaotic spectrum.

In recent collaborative efforts, we have theoretically and experimentally demonstrated that spinor gases driven by spin-flopping fields serve as excellent platforms for investigating ergodicity breaking. We observed that specific initial states remain non-thermal at weak driving strengths despite the majority of the system thermalizing, providing clear evidence of QMBS. As driving strength varies, we map the smooth transition from integrability to weak ergodicity breaking, shedding light on how quantum information can persist against the natural tendency toward thermal equilibrium.

Dynamical Phase Transitions and Spinor Phases

Continuing our collaboration with the group of Dr. Yingmei Liu, we have established a toolbox for the precise control, characterization and detection of nonequilibrium dynamics in spinor gases via their internal spinor phases. Extracting the phase evolution from the observed spin populations defines an order parameter that sharply identifies dynamical phase transitions over a wide range of conditions, and allows spin-dependent interactions to be inferred from a single experimental time trace. Controlling the initial spinor phases also gives experimental access to non-ergodic relaxation, in which states of similar energy in the nominally thermal part of the spectrum retain a memory of their initial state, directly connected to quantum scarring.

We extended these techniques to lattice-confined spinor gases subject to a priori unknown, time-dependent interactions, detecting dynamical phase transitions in real time from the system energy and the spinor phases. A new observable, the cutoff time, identifies the transition at holding times when commonly used order parameters still show transient, non-universal behavior.

Quenching spinor gases in an optical lattice has also allowed us to detect coherent spin-dependent three-body interactions, which are usually masked by much stronger two-body effects. Including them proves essential for correctly determining how atoms are distributed across the lattice sites, with applications in quantum sensing via spin singlets.