2017/05/16 by Cato Sandford, Sandford, Cato, Daniel Seeto +3
Neuroscience · Physics and Astronomy · #Advanced Thermodynamics and Statistical Mechanics #FOS: Physical sciences #Neural dynamics and brain function #Statistical Mechanics (cond-mat.stat-mech) #Statistical Mechanics and Entropy #stochastic dynamics and bifurcation
paper · pdf · doi:10.48550/arxiv.1705.05537
openalex publication_date 2017/05/16 · openalex created_date 2022/09/30 · openalex updated_date 2026/07/28
The Gibbs Mixing Paradox is a conceptual touchstone for understanding\nmixtures in statistical mechanics. While debates over the theoretical\nsubtleties of particle distinguishability continue to this day, we seek to\nextend the discussion in another direction by considering devices which can\nonly distinguish particles with limited accuracy. We introduce two illustrative\nmodels of sorting devices which are designed to separate a binary mixture, but\nwhich sometimes make mistakes. In the first model, discrimination between\nparticle types is passive and sorting is driven, while the second model is\nbased on an active proofreading network, where both discrimination and sorting\nhave a tunable active component. We show that the performance of these devices\nmay be enhanced out of equilibrium, and we further probe how the quality of\nparticle sorting is maintained by trade-offs between the time taken and the\nenergy dissipated. Considering these examples, we demonstrate how increasing\nthe similarity between particles gradually increases the work required to sort\nthem, eliminating the paradox, while preserving the limits imposed by standard\nequilibrium statistical mechanics.\n