The work addresses how information stored in environmental memory - hidden degrees of freedom that induce non-Markovian dynamics - can be turned into extractable work. Building on information-thermodynamic bounds that relate extractable work to mutual information gained by measurements, it shows that a single instantaneous measurement cannot access information stored in the bath; temporal correlations make the environment opaque unless probes are timed to exploit information backflow. Formal considerations using outcome-resolved relative entropy clarify why repeated or time-delayed measurements are required to capture excess free energy residing in hidden variables and thereby relax the usual Markovian limits on work extraction.
Experiments use time-resolved position measurements of a 2.7 μm silica particle optically trapped in a viscoelastic micellar solution (8 mM cetylpyridinium chloride / sodium salicylate) with a structural relaxation time of about 6 s. A minimal protocol of two consecutive binary (left/right) measurements with controlled delay retrieves information returned from the bath and alters relaxation dynamics. Quantitative analysis shows that this information backflow can be independently identified from initial-state effects and, when exploited, increases extracted work; in specific regimes the extracted work exceeds the energy stored in the observable degree of freedom alone. The results establish environmental memory as an accessible thermodynamic resource and demonstrate how non-Markovian dynamics can be harnessed to boost information-engine performance.
Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.