Research Gallery

Why, you do not even know what will happen tomorrow. What is your life? You are a mist that appears for a little while and then vanishes. (James 4:14)

Mpemba effect:

Observation:

For a system of rugged energy landscape, what is the most time-efficient way of cooling it?

It turns out that under certain conditions, it is faster to first heat a system up and then cool it down than to cool it down directly! In my recent work, we explained this counter-intuitive effect in both cooling and heating processes. These effects occurs far away from the quasi-static limit, and can not be described by linear response theory. We described the non-equilibrium dynamics in the language of stochastic thermodynamics that is general for all non-equilibrium processes. We find the mathematical conditions for this counter-intuitive cooling effect. In addition, we predicted an inverse counter-intuitive heating effect as well -- to heat a system, it can be faster to first cool it down and then heat it up than heating it directly.

So what?

Non-equilibrium process can be very non-monotonic and very different from our near-equilibrium intuitions. Thus, it is a non-trivial question to find an optimal control protocol if the system is allowed to be driven far away from the quasi-static limit (which is very typical in real problems). To find such an true optimal control: One will need to sample the rarely visited regions in the state space and discover an optimal control protocol that is surprising and counter-intuitive.

Programmable Maxwell's Demon: