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?
 A: This is a classic example of why you shouldn't take popular science articles at face value. Note that the results reported in that article should actually be understood as a consequence of time-reversal symmetry, not a violation of it. Indeed, these results simply confirm exactly what would be expected from the assumptions of unitary quantum dynamics and time-reversal symmetry. Note, however, that exactly the same result would be obtained for purely classical dynamics [and this has been experimentally well established for some time].
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 $\lambda_{F/B}(0)$.
