报告题目：Stochastic thermodynamics

报 告 人：Erik Aurell, Royal Institute of Technology (KTH)

报告时间：2014年10月17日、21-24日下午3:30-5:30

报告地点：理科楼B315

报告摘要： According to a well-known quote by Einstein, thermodynamics is the only physical theory of universal content which will never be overthrown. Recent years have however seen unexpected generalizations of thermodynamics to single mesoscopic systems which are not in the thermodynamic limit. Although the established name has become "stochastic thermodynamics", an equally good or better term could have been "thermodynamic stochastics" since the main point is to introduce a thermodynamic point of view on systems described by stochastic laws, typically Markov processes. The main results of this science is a number of exact relations between fluctuating quantities which are called fluctuation relations. The simplest of these is the Jarzynski Equality stating that an exponential average of work (a non-equilibrium notion) equals the ratio of two partition functions (an equilibrium notion). This and other more refined relations have been used to determine the free energy differences of macromolecules in different conformations by measuring the work required to pull them apart many times, and then taking the proper average. The highest accuracy experimental tests of these relations, which are then rather tests that the premisses by which they have been derived hold, have on the other hand appeared in single-electron systems.

The lectures will cover the following topics:

1.Fluctuation relations as a consequence of a relation between forward and time-reversed path probabilities in stochastic systems

2.Optimized entropy production and a sharpening of the second law for finite-time processes

3.Entropy production at different powers of observation

4.Fluctuation relations in quantum systems: the simple results

5.Fluctuation relations in quantum systems: the less simple results