Antonio PICANO : Disorder and Fluctuations to Control and Probe Ultrafast Phase Transitions
Antonio PICANO, JEIP, Collège de France
Le résumé :
With recent experimental developments in time-control of quantum materials, we can drive phase transitions on femtosecond timescales and generate states absent in equilibrium. Despite relevant results, such as the stabilization of superconducting-like states above the equilibrium critical temperature or the realization of ultrafast insulator-to-metal transitions, our understanding of how these phenomena occur remains incomplete.
A key limitation is that conventional theories assume coherent, homogeneous atomic motion, with all atoms moving in a concerted manner toward new equilibrium positions, whereas recent experiments show that disorder often plays a fundamental role in the dynamics. Ultrafast disordering observed after photo-excitation challenges the conventional picture of photo-induced transitions, where symmetry breaking takes place along a single collective coordinate.
Here we report the microscopic theories and computational methods that we developed to simulate the full thermalization dynamics of strongly correlated electrons and coupled electron–phonon systems, with the goal of understanding the role of disorder in ultrafast phase transitions. We demonstrate that it can be turned from a limiting factor that prevents our comprehension of ultrafast phase transitions into a tool to: control and stabilize long-lived hidden phases that are inaccessible at equilibrium with properties on demand; design multi-pulse excitation protocols that exploit incoherent disordered dynamics to increase the energy efficiency of ultrafast switching mechanisms; and develop ultrafast spectroscopic techniques to extract quantitative information about the transient disordered state of a system in real-time through stochastic resonance, a hallmark of nonlinear stochastic dynamics.
These findings can impact technologically relevant systems, such as chalcogenide glasses for data storage, where coherent control is not possible, or resistive-switching-based neuromorphic memories based on memristive technology.