Knowledge as Power: The Role of Microstate Information in Thermodynamic Free Energy

Overview

While it is incontrovertible that the thermodynamic entropy of closed systems increases or remains the same as time passes, the amount of entropy is itself reliant on the available knowledge of the system’s state. The very definitions of entropy in thermodynamics share a great deal of mathematical and conceptual parallels with similar ideas in information theory, suggesting the intimate relationship between information and energy in physical systems. Building on the ideas of Landauer’s principle, which set a lower theoretical bound on the energy consumption of computation, we consider a generalized Szilard-engine model, that is, a single particle in a confined space, coupled to a heat bath which maintains the system at a constant temperature. For this setup, boundaries may be put on the available phase space, which is physically analogous to such processes as piston compression. The isothermal expansion that follows constitutes work done by the system, resulting in energy extraction from what was previously a system at its maximal entropy state. The act of observation and information acquisition is thus necessarily a thermodynamic event, with the amount of information directly proportional to the increase in entropy, in order to remain consistent with the Second Law of thermodynamics. Fluctuation theorems are also used to show how this relation provides for a gradual rise in the information-processing capacity of systems when such a thermodynamic path is available, with a probability that increases with time, suggesting future paths of research concerning the emergence of information transfer hierarchies within complex systems of nature.

Competition history

• AJAS 2020 Category not listed

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

• AJAS None projects
• Every AJAS 2020 project