# Does this experimental confirmation of thermodynamic irreversibility at the quantum level mean that information can be created/destroyed over time?

According to this article, https://phys.org/news/2015-12-physicists-thermodynamic-irreversibility-quantum.html, physicists recently confirmed thermodynamic irreversibility at the quantum level. Does this mean we can say that the quantity of information changes over time? That we can't reconstruct a broken egg-shell or burned paper? Is time reversal symmetry violated?

Now if that is true, why did Hawking think that the information paradox is a problem?

And how is the macro level reversible and the micro level irreversible?? One possible solution is quantum particles escaping to higher tiny dimensions, is that confirmed?

• Well, we already know that the laws of classical physics are time-reversible and yet thermodynamics, purely classically, is not, so this is really just showing that QM has the same problem. That's important but it just shows that you can't escape from the problem by waving some quantum magic wand over it. – tfb Apr 6 '18 at 9:39
• How is the macro level reversible and the micro level irreversible ?? the only solution is quantom particles escaping to higher tiny dimensions – Dhia Hassen Apr 6 '18 at 9:55
• No, it's not, because this is true for classical mechanics as well. – tfb Apr 6 '18 at 13:56
• excuse me , you classical mechanics irreversible ?! i really dont think so – Dhia Hassen Apr 6 '18 at 17:52
• Classical mechanics is time symmetric: classical thermodynamics is not. This means that the explanation for that irreversibility cannot involve QM, let alone particles escaping to tiny higher dimensions. – tfb Apr 6 '18 at 17:56

The experiments compare two processes in which an externally controlled parameter $\lambda(t)$ (in this case, a radio-frequency field) is changed in order to do some work on a quantum system. In one process (forward) the parameter changes as $\lambda(t) = \lambda_F(t)$, $0<t<\tau$. In the other process (backward), the parameter changes in the time-reversed fashion $\lambda(t) = \lambda_B(t) = \lambda_F(\tau-t)$. This is not the same as literally reversing time and running the evolution backward [if an experimental group ever manages to run time backwards, I guarantee you they would publish it in Nature, not PRL ;-) ]. In particular, the initial state of the backward process is not the final state of the forward process. Rather, each process starts with the system in thermal equilibrium corresponding to the initial value of the control parameter $\lambda_{F/B}(0)$.