Thermodynamic Phase Space Reduction

Updated 17 January 2026

Thermodynamic phase space reduction is a mathematical approach that projects high-dimensional systems onto lower-dimensional manifolds, retaining essential variables like entropy and energy.
It employs constraint-based, coarse-graining, and geometric techniques—applied in contexts from black holes to active matter—to ensure accurate macroscopic descriptions.
Key methodologies include symplectic/contact reduction, Hamilton-Jacobi projection, and Bogoliubov transformations, which bridge microscopic details with observable thermodynamic laws.

Thermodynamic phase space reduction refers to a suite of mathematical and physical methodologies aimed at projecting, constraining, or coarse-graining the high-dimensional phase spaces inherent to thermodynamic systems onto lower-dimensional manifolds that retain the essential macroscopic or collective variables. This concept appears in statistical mechanics, non-equilibrium dynamics, black hole thermodynamics, quantum statistical frameworks, and dynamical systems far from equilibrium. Reduction can occur via explicit elimination of variables (constraints), coarse-graining to focus on slow degrees of freedom, projection onto equilibrium submanifolds, or geometrical procedures (symplectic/contact reduction). The purpose is typically to elucidate the collective thermodynamic behavior, construct self-consistent thermodynamic laws, and resolve inconsistencies—especially in models with constraints or nontrivial microscopic structure.

1. Geometric and Constraint-Based Reduction in Thermodynamic Phase Space

A central motif is the reduction of the high-dimensional contact manifold describing a thermodynamic system—whose local coordinates involve all extensive and intensive variables—to a physically meaningful, lower-dimensional submanifold. For black holes and other constrained thermodynamic systems, the reduction is effected by algebraic or geometric constraints (e.g., regularity or horizon conditions), which relate otherwise independent thermodynamic variables.

General procedure (see (Li et al., 11 Jan 2026, Ma et al., 13 Jul 2025)):

Begin with the full set of extensive variables $E^a = \{S, Q, J, ...\}$ and a fundamental relation $M = M(E^a)$ .
Write the first law in the full phase space, with all conjugate variables obtained as derivatives of $M$ .
Impose the constraint $f(E^a) = 0$ , which relates (and thus eliminates) variables, reducing the phase space.
Compute all thermodynamic quantities in the full unconstrained space, then substitute the constraint to obtain the physically relevant, reduced potentials and conjugates.
Only after all differentials and Legendre transforms are taken, restrict to the constraint surface.

The ordering of operations is crucial: restricting before differentiating generally leads to inconsistencies, such as the mismatch between the geometric (surface-gravity) and thermodynamic temperature in regular black holes. Proper reduction projects the contact form and produces a reduced symplectic structure, ensuring the first law remains consistent on the reduced manifold (Li et al., 11 Jan 2026).

2. Symplecto-Contact Reduction and the Generation of Macroscopic Thermodynamics

The process of reduction can be formalized using symplectic and contact geometry. In statistical mechanics, one starts from the $6N$-dimensional phase space (SPS), passes through the kinetic theory phase space (KTPS), and ultimately arrives at the thermodynamic phase space (TPS) via:

Step 1: Marsden–Weinstein symplectic reduction via a moment map associated with a finite set of coarse observables (e.g., energy, volume).
Step 2: Construction of the TPS as a $(2n+1)$ -dimensional contact manifold, with equilibrium states as Legendrian submanifolds specified by the maximization of relative information entropy subject to observational constraints.

The reduced entropy functional acts as a generating function for these Legendrian submanifolds, encoding equilibrium thermodynamics. The necessity of the Maxwell construction in phase transitions (e.g., the van der Waals loop) corresponds to the need to select a continuous graph-selector from a multi-branched equilibrium Legendrian (Lim et al., 2022).

3. Coarse-Grained Phase Space Reduction in Nonequilibrium Statistical Mechanics

In non-equilibrium systems, especially active matter displaying motility-induced phase separation (MIPS), phase space reduction is implemented by coarse-graining spatially into cells and projecting the dynamics onto a small set of slow collective variables, typically local density ( $\rho$ ) and kinetic energy ( $E$ ). This yields a reduced, tractable $(\rho,E)$ phase space where the joint stochastic dynamics can be directly measured.

For each phase-space point $(\rho, E)$ , one reconstructs:

The nonequilibrium potential $U_{neq}(\rho, E) = -\ln P_{ss}(\rho, E)$ ,
The probability current $\mathbf{J}_{ss}(\rho, E)$ ,
And the decomposition of forces into "landscape" (gradient) and "curl" (non-gradient) components.

This mapping reveals both the dynamical (via the emergence and splitting of potential wells, flux tearing, and rotational currents) and thermodynamic (via scaling of the entropy production rate $e_p$ ) origins of phase separation. The transition from a single to a bimodal $U_{neq}$ and the associated change in $e_p$ scaling signals the onset of phase separation and the creation of new dissipation channels in the reduced space (Su et al., 2022).

4. Reduction in Dynamical Systems: Thermodynamic Phase Space Contraction

In dissipative dynamical systems driven far from equilibrium, such as the Malkus-Lorenz waterwheel, thermodynamic phase-space reduction is realized as the progressive contraction of the accessible phase space due to entropy production. Irreversible losses render the asymptotic attractor lower-dimensional ( $D_H < N$ for an $N$ -dimensional system), quantifying the loss of phase-space volume.

The thermodynamic efficiency $\eta$ or entropy production rate $\dot{S}_{gen}$ can be explicitly written in terms of the remaining phase-space variables, and bifurcations manifest as sudden reductions in attractor volume, coincident with drops in efficiency. This phase-space contraction provides a geometric measure of irreversible dissipation and the transition to more complex or self-organized steady states (López et al., 2022).

5. Reduction in Quantum Statistical Frameworks

In quantum thermodynamics, phase-space reduction is prominent in frameworks like thermofield dynamics (TFD), where the system's Hilbert space is doubled to encode thermal effects. Reduced phase-space observables such as the single-particle density matrix (1-RDM) or Wigner distribution are obtained by tracing out degrees of freedom and performing a Bogoliubov back-transformation:

The 1-RDM is reconstructed from the reduced two-particle density matrix (2-RDM) with subtleties in the presence of entanglement,
Approximate schemes such as neglecting real–tilde correlations or moment-expansion bypass the need for full 2-RDM construction,
Computationally efficient reduction methods are essential for scalability to high dimensions (Błasiak et al., 27 May 2025).

Such reduction quantifies thermal and quantum correlations, and the quality of approximate phase-space reduction can be assessed by the accuracy of distribution moments and functionals.

6. Contact Hamilton-Jacobi and the Projection to Equilibrium Manifolds

The Hamilton-Jacobi theory formulated on contact manifolds (thermodynamic phase space) yields another reduction protocol. Here, the dynamical equations governing thermodynamic transformations are projected from the full $(2n+1)$ -dimensional phase space onto $n$ -dimensional equilibrium submanifolds—Legendre submanifolds characterized by vanishing of the contact form.

This reduction is implemented via:

Defining a principal function $S(q^i, t)$ on the equilibrium manifold,
Deriving reduced dynamics through the contact Hamilton–Jacobi equation,
Realizing that characteristics of the reduced PDE describe all possible quasi-static thermodynamic trajectories on the equilibrium manifold.

This scheme ensures the consistent reduction and dynamical closure on the physically admissible, equilibrium states of the system (Ghosh, 2022).

7. Thermodynamic Phase Space Reduction in Black Hole Thermodynamics

Dimensional reduction in black hole thermodynamics is essential when additional constraints (e.g., regularity conditions for regular black holes) relate black hole parameters and reduce the number of independent thermodynamic degrees of freedom. In the "alternative phase space" for AdS black holes, one can fix the pressure and treat the square of the charge as an independent variable, leading to van der Waals-type critical behavior and a two-variable thermodynamic description (Xu, 2020).

The application of reduction clarifies the physical interpretation of thermodynamic quantities, yields universal critical exponents for curvature invariants, and substantiates conclusions regarding microscopic interaction character (e.g., repulsiveness from positive Ruppeiner curvature).

References:

(Su et al., 2022): Dynamic and Thermodynamic Origins of Motility-Induced Phase Separation
(Li et al., 11 Jan 2026): Reductions of Thermodynamic Phase Space for Rotating Regular Black Holes
(Ma et al., 13 Jul 2025): Regular black holes and reductions of thermodynamic phase spaces
(Lim et al., 2022): Nonequilibrium thermodynamics as a symplecto-contact reduction and relative information entropy
(Ghosh, 2022): Hamilton-Jacobi approach to thermodynamic transformations
(Xu, 2020): Analytic phase structures and thermodynamic curvature for the charged AdS black hole in alternative phase space
(Błasiak et al., 27 May 2025): Reduced Density Matrices and Phase-Space Distributions in Thermofield Dynamics
(López et al., 2022): The thermodynamic efficiency of the Lorenz system
(Collins et al., 2010): Phase space structure and dynamics for the Hamiltonian isokinetic thermostat