Speaker
Description
Thermodynamic uncertainty relations show that high precision in thermodynamic processes requires a physical cost, such as entropy production or dynamical activity. In this talk, I will discuss recent developments of these relations from two viewpoints. First, I will introduce fundamental precision limits in finite-dimensional quantum thermal machines. Conventional thermodynamic uncertainty relations suggest that precision can be improved by increasing entropy production. However, in realistic finite-dimensional systems, physical constraints such as the dimension of the system and environment and the energy bandwidth limit the achievable precision. I will explain dynamics-independent bounds on the fluctuations and expectation values of observables, and discuss their implications for quantum thermal machines such as quantum batteries. Second, I will present a replica Markov process approach to thermodynamic trade-off relations. By considering several independent copies of the same Markov process, nonlinear quantities of probability distributions can be treated within a stochastic thermodynamic framework. This method leads to bounds on entropic quantities such as Tsallis and Rényi entropies in terms of dynamical activity. These results provide a new way to extend thermodynamic uncertainty relations to information-theoretic measures.