What is nonequilibrium?

Abstract: Lecture notes on elements of nonequilibrium statistical mechanics: (1) a characterization of the nonequilibrium condition, largely by contrast to equilibrium; (2) a retelling of some of the great performances of the more distant past, including the perspectives of Boltzmann and Onsager; and (3) more recent methods and concepts, from local detailed balance and the identification of entropy fluxes to dynamical fluctuation theory, and the importance of dynamical activity.

• The paper introduces a formal characterization of nonequilibrium physics via pathspace measures, breaking conventional thermodynamic bounds.
• It demonstrates explicit models showcasing kinetic effects like negative heat capacity and population inversion in multilevel systems.
• The analysis emphasizes the role of time-reversal symmetry breaking and frenetic contributions in driving persistent currents and dissipation.

What Is Nonequilibrium? – A Formal Overview

Introduction and Scope

"What is nonequilibrium?" (2601.16716) provides a comprehensive, technical exploration of nonequilibrium physics, establishing it as a foundational, rapidly evolving field with deep connections to statistical physics, stochastic thermodynamics, complexity science, and interdisciplinary domains. The notes delineate the field’s scope, theoretical underpinnings, phenomenology, and unresolved challenges, arguing that the nonequilibrium condition cannot be characterized as a mere complement to equilibrium, nor can it be fully subsumed by extensions of thermodynamic concepts such as entropy or temperature. Instead, nonequilibrium physics requires consideration of dynamical ensembles, path-probabilities, the breaking of time-reversal invariance, and novel kinetic effects, including nondissipative (frenetic) contributions.

Contrasts Between Equilibrium and Nonequilibrium

The treatment begins with a careful contrast between equilibrium—characterized by time-invariant, detailed balanced states, and symmetry under time-reversal—and stationary nonequilibrium—characterized by persistent currents, temporal cycles, and generically broken time-reversal symmetry. Notably, stationary and stable nonequilibrium systems can appear time-invariant for observables, yet their physical nature is fundamentally distinguished by persistent entropy production, currents, and often nonlocal correlations.

The absence of a Gibbs-type variational structure governing static fluctuations in nonequilibrium, and the lack of a universally meaningful "nonequilibrium free energy," are emphasized. Instead, nonequilibrium systems are fundamentally characterized by their dynamics in trajectory-space, requiring probabilistic formalism over spacetime configurations.

Dissipation and Time-Reversal Symmetry Breaking

The central role of dissipation is clarified, identifying it as a consequence of time-reversal symmetry breaking at the level of typical trajectories in driven steady states. However, the manuscript is careful to stress that dissipation is neither a universal nor a sufficient descriptor: time-symmetric systems can still display dissipation if memory or coupling to hidden reservoirs is present, and counterexamples (e.g., superconductors, superfluids) can support sustained currents without dissipation.

The text provides two fundamental observations:

• Steady currents imply heat dissipation (under weak conditions).
• Heat flow in open systems (coupled to baths) is typically a signature of underlying steady-state currents.

Multi-Scale Organization: Micro, Meso, Macro

Three levels of description—microscopic, mesoscopic, and macroscopic—are formally distinguished, where mesoscopic physics (stochastic thermodynamics) bridges the scale and exhibits pronounced fluctuations, as in colloidal or biomolecular systems. Nonequilibrium, particularly far from equilibrium, presents challenges such as nonlocality, long-range correlations, and the irrelevance or inapplicability of standard thermodynamic quantities.

Statistical Fluctuations: Static Versus Dynamic

For equilibrium systems, static fluctuation theory is governed by the Gibbs formalism and large deviation principles, with relative entropy as the rate function. In nonequilibrium, such tidy characterizations are unavailable; instead, dynamical ensembles and trajectory-space measures become central. The path-space action and the absence of detailed balance play crucial roles in the emergence of new fluctuation structures and in the breakdown of classic fluctuation-dissipation theorems.

Nonequilibrium Phenomenology and Exemplars

The phenomenology covered is broad, with applications ranging from molecular motors, ion channels in cell membranes, and Rayleigh-Bénard convection to turbulence, climate science, and the emergence of life. Prototypical paradigms such as heat conduction under thermal gradients (Figure 1) and stationary ionic flow across biomembranes (Figure 2) are provided.

Figure 1: Energy transport in a rod under nonequilibrium boundary conditions, driven by temperature gradients.

Figure 2: Schematic of membrane transport, with ion fluxes and local dissipation illustrating nonequilibrium steady-state processes.

These examples highlight the ubiquity of nonequilibrium in both natural and engineered systems and motivate theoretical formalisms capable of describing currents, entropy production, and emergent order away from equilibrium.

Historical Foundations and Current Theoretical Frameworks

The text situates nonequilibrium statistical mechanics in historical context—Boltzmann, Onsager, Prigogine, and others are discussed. The development of the H-theorem, the Onsager reciprocity relations, fluctuation theorems (Gallavotti-Cohen, Crooks, Jarzynski), and stochastic thermodynamic extensions are surveyed with the acknowledgment of persistent theoretical gaps:

• The absence of a nonequilibrium counterpart to the Gibbs ensemble or Landau theory for phase transitions.
• The lack of a general theory for nonequilibrium nucleation, metastability, or the supposed "maximum entropy production" principles.

Formal Examples and Model Systems

Agitated Molecular Systems and Nonequilibrium Switches

Explicit model constructions include multilevel (three-level) "ladder" molecular switches, with rates coupling energy level populations to bath temperatures and kinetic switching between configurations:

Figure 6: Left, three-level ladders illustrating transitions; right, an energy landscape with potential barrier $\Delta$; the model enables experimental realization as a colloidal particle in a flashing potential.

Stationary populations and nonequilibrium heat capacities are computed exactly. For large switching rates or high barriers, the system approaches equiprobability—an instance of population inversion and the emergence of negative heat capacity, contrasting equilibrium behavior.

Figure 3: Stationary distribution plots for the three-level ladder, as a function of key kinetic parameters and temperature.

Figure 4: Nonequilibrium heat capacity showing a negative branch under high kinetic barriers or switching rates, indicative of nontrivial kinetic-dominated thermodynamic anomalies.

Ratchets, Parrondo Games, and Active Rowers

The text presents ratchet mechanisms, including flashing potentials and discrete-time analogs (Parrondo games), in which net currents are generated via the alternation of individually unbiased or equilibrium games—demonstrating the principle that nonequilibrium protocols can break detailed balance to generate directed motion.

A specific class of models ("rowers") illustrates the effects of boundary-driven kinetic asymmetry on persistent current generation in multi-state oscillators:

Figure 5: Graph representation and transition rates for a rower with $n=3$, capturing cyclic motion and energy exchange with the environment.

Figure 7: Example trajectory for a single rower, demonstrating alternate switching, energy uptake, and dissipation sequences over time.

Central Theoretical Structures

Pathspace Action and Local Detailed Balance

Pathspace actions are defined for Markov jump processes, with explicit decomposition into time-antisymmetric (entropy flux) and time-symmetric (frenetic or dynamical activity) components. Under local detailed balance, the log ratio of forward and reverse rates is tied to the entropy change contributed by each transition to the reservoirs.

The decomposition

$$\mathcal{A}(\omega) = \mathcal{F}(\omega) - \frac{1}{2} \mathcal{S}(\omega)$$

leads to refined fluctuation-response relations, and, in the linear regime, recovers the Kubo formula as a special case. Far from equilibrium, the time-symmetric component (frenesy) becomes dominant, acting as an order parameter for stabilization, selection, or the emergence of new steady states.

Macroscopic Fluctuation Theory

Spatiotemporally extended systems in nonequilibrium possess dynamical large deviation principles for densities and currents, parameterized via convex Lagrangians. The zero--cost flows minimize these, leading to deterministic hydrodynamic equations, while excitation spectra and the structure of fluctuations encode the stability, possible phase coexistence, and emergence of long-range correlations.

Strong Results and Contradictory Phenomena

Bold Highlights:

• The emergence of negative heat capacity and energy population inversion in kinetic-dominated regimes for simple multilevel models.
• Explicit demonstrations that heat current directionality cannot be fully predicted from static free energy differences in nonequilibrium (blowtorch theorem).
• The construction of ratchets and Parrondo games as examples where nonequilibrium protocols create currents otherwise absent in equilibrium.
• The failure of equilibrium fluctuation-response relations (Sutherland-Einstein, FDR) in the presence of nonconservative driving or dominant kinetic effects.

Implications and Open Problems

Theoretical Frontiers

Major open questions identified:

• Nonequilibrium analogs of equilibrium phase transitions and the delineation of universality classes under driving.
• The dynamical origin and role of long-range correlations and critical phenomena in driven systems.
• The extension of thermodynamic characterization (e.g., entropy functionals, temperature) to open, small, or far-from-equilibrium systems.
• The role of frenesy in stabilization, pattern formation, and control—especially in biological and active matter contexts.

Practical Perspectives

This framework enables:

• Design and analysis of nanodevices, molecular motors, and stochastic engines operating under sustained nonequilibrium.
• Controlled manipulation of heat and particle currents at mesoscopic scales.
• New insights into the statistical mechanics of biological processes, including homeostasis, molecular switching, and pattern formation at the cell or tissue level.

Future Trajectories

Progress in nonequilibrium physics is expected to rely heavily on:

• Quantification and exploitation of nondissipative (frenetic) contributions to pathspace measures.
• The continued development of macroscopic large deviation theory to accommodate multi-reservoir, driven, and correlated media.
• Formulation of useful nonequilibrium entropy functionals and pseudo-potentials adapted to complex, fluctuating environments.

Conclusion

The paper synthesizes a rigorous, multiscale perspective on nonequilibrium physics, moving beyond traditional thermodynamic quantities and equilibrium-centric paradigms. By emphasizing the importance of pathspace probabilities, explicit kinetic mechanisms, and the necessity of accounting for time-symmetric (frenetic) activity, the document outlines both the achievements and considerable challenges in building a predictive, unifying nonequilibrium statistical mechanics. The interplay between dissipation and dynamical activity—not only as a source of entropy production but as an organizer of patterns and function—emerges as a central theme, especially pertinent for systems at the convergence of physics, chemistry, and biology.