Non-equilibrium temperatures in steady-state systems with conserved energy
Eric Bertin, Olivier Dauchot, and Michel Droz

TL;DR
This paper investigates non-equilibrium lattice models with conserved energy, defining a statistical temperature T_{th} and comparing it to the fluctuation-dissipation temperature T_{FD}, revealing differences and proposing a new parameter for deviation from equilibrium.
Contribution
It introduces an exactly solvable class of non-equilibrium models and establishes the physical relevance of the temperature T_{th} over T_{FD} in steady states.
Findings
T_{th} equals the microcanonical temperature in the models.
T_{FD} differs from T_{th} when the fluctuation-dissipation relation is linear.
A new parameter describes deviations from equilibrium, supported by mean-field predictions.
Abstract
We study a class of non-equilibrium lattice models describing local redistributions of a globally conserved quantity, which is interpreted as an energy. A particular subclass can be solved exactly, allowing to define a statistical temperature T_{th} along the same lines as in the equilibrium microcanonical ensemble. We compute the response function and find that when the fluctuation-dissipation relation is linear, the slope T_{FD}^{-1} of this relation differs from the inverse temperature T_{th}^{-1}. We argue that T_{th} is physically more relevant than T_{FD}, since in the steady-state regime, it takes equal values in two subsystems of a large isolated system. Finally, a numerical renormalization group procedure suggests that all models within the class behave similarly at a coarse-grained level, leading to a new parameter which describes the deviation from equilibrium. Quantitative…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
