⚛️ general relativity

Phase-space structure of cosmological thermogravity

This paper analyzes the homogeneous cosmological dynamics of an emergent gravity model with non-conserved stress-energy, mapping the phase-space structure of both single-fluid and mixed-component scenarios to reveal distinct late-time behaviors including asymptotic de Sitter expansion, recollapse, and critical trajectories that generalize known analytic solutions.

Original authors: Charlotte Brereton, João Magueijo, Mar Alí Caballo

Published 2026-10-08

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Charlotte Brereton, João Magueijo, Mar Alí Caballo

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Gravity is often thought of as the invisible force that keeps our feet on the ground and the planets in their orbits, but in modern physics, it is also viewed through the lens of thermodynamics, the study of heat and energy. This perspective suggests that gravity might not be a fundamental force at all, but rather an emergent phenomenon, much like temperature arises from the motion of countless atoms. In this view, the equations governing the universe are similar to the laws of thermodynamics that describe how heat flows. A recent line of research has explored a specific version of this idea where the usual rules of energy conservation are slightly bent. In our everyday experience, energy is never created or destroyed, only changed in form. However, this new theoretical framework allows for a scenario where energy can be subtly generated or lost, driven by a preferred direction in time and space. This concept is significant because it offers a potential explanation for why the universe is expanding at an accelerating rate without needing to invoke a mysterious, invisible energy filling empty space, a problem that has long puzzled cosmologists.

Researchers Charlotte Brereton, João Magueijo, and Mar Alí Caballo have taken a deep dive into the mathematical landscape of this theory to understand exactly how the universe would behave under these rules. They focused on the "phase space" of the cosmos, which is a way of mapping every possible state the universe could be in, rather than just looking at a single timeline. Imagine a vast, multi-dimensional map where every point represents a different version of the universe with specific amounts of matter and expansion rates. By plotting the evolution of the universe on this map, the team discovered that the behavior of the cosmos is far more varied than previously thought. They found that the universe does not have just one destiny; instead, it can follow one of three distinct paths depending on its initial conditions.

In the simplest version of their model, where the universe is filled only with a single type of dust-like matter, the researchers found that the entire history of the cosmos is controlled by a single number. If this number falls within a certain range, the universe expands forever, eventually settling into a smooth, steady growth that looks like the current acceleration we observe. If the number is different, the universe expands for a while but then stops and collapses back in on itself in a "big crunch." Between these two outcomes lies a very specific, critical path. This special path corresponds to a solution that was previously identified as a benchmark for testing the theory. The researchers showed that this benchmark solution is not a typical outcome but rather a razor-thin dividing line between a universe that expands forever and one that eventually dies. It is a delicate balance, and most universes in this model would naturally drift away from it toward one of the two more common fates.

The study becomes even more intricate when the researchers added a second ingredient to the mix: normal matter, like the atoms that make up stars and people, which obeys the standard rule of energy conservation, alongside "cold dark matter," which does not. This is a more realistic scenario because it separates the behavior of the visible universe from the invisible dark matter that dominates the cosmos. In this three-dimensional map, the researchers discovered a complex structure of stable regions. They found that for a wide range of conditions, the universe naturally evolves toward a state where it expands forever, with the dark matter eventually dominating the material sector relative to the normal matter. However, they also uncovered a subtle and surprising quirk at the edge of this behavior. When the amount of dark matter is reduced to almost nothing, the behavior of the universe changes abruptly. If there is even a tiny, non-zero amount of dark matter, the universe eventually settles into a smooth, accelerating expansion. But if the dark matter is removed entirely from the start, the universe follows a completely different path, remaining dominated by normal matter and never reaching that smooth, accelerating state.

This discontinuity reveals that the way the universe behaves at the very end of its life depends critically on whether dark matter is present, no matter how small the amount. The researchers found that the transition between these two behaviors is not smooth but singular, meaning the limit of "almost no dark matter" is not the same as "no dark matter at all." This finding clarifies that the accelerating expansion seen in our universe, if explained by this theory, relies on the presence of dark matter violating energy conservation. Without it, the theory reverts to a different kind of solution that requires a different mathematical adjustment to match observations. The work provides a comprehensive map of how this theory of gravity could play out, showing that while the specific solution used in earlier studies is a valid path, it is just one of many possibilities, and the true nature of our universe's future depends on the precise balance of matter and the fundamental rules governing energy.

Technical Summary: Phase-space structure of cosmological thermogravity

Problem Statement
This paper investigates the global dynamical structure of "thermogravity," an emergent gravity model derived from a non-degenerate thermodynamic cycle. Unlike standard general relativity, this theory features trace-free Einstein equations and a preferred-frame violation of stress-energy conservation. While a specific analytic solution for a flat, dust-filled universe (Eq. 1 in Ref. [1]) has been established and used as a benchmark for observational tests, the broader phase-space behavior of the theory remains unexplored. Specifically, the authors seek to determine how special the known analytic solution is within the wider solution space and to understand the global asymptotics of both universal non-conservation (where all matter violates conservation) and phenomenologically cleaner models (where only Cold Dark Matter, CDM, violates conservation while baryons and photons remain standard).

Methodology
The authors employ dynamical-system methods to classify the cosmological asymptotics of the theory. They restrict the analysis to spatially flat ( $K=0$ ) homogeneous universes, setting $8\pi G = 1$ and assuming a positive parameter $L^2$ . Time is parameterized by the number of e-foldings, $N \equiv \ln a$ .

Universal Dust Model: The system is reduced to a two-dimensional autonomous system using the variables $\Omega = \rho/(3H^2)$ and $u = (HL^2)^{-1}$ . The authors derive an exact first integral to analytically classify all trajectories.
Conserved Baryons / Non-conserved CDM Model: The system is extended to a three-dimensional phase space by introducing separate density parameters for baryons ( $\Omega_b$ ) and CDM ( $\Omega_c$ ), while maintaining standard conservation for baryons and non-conservation for CDM. The authors analyze fixed points, eigenvalues, and stability sheets, paying particular attention to the limit where CDM abundance tends to zero.

Key Contributions and Results

Universal Dust Model Classification:
- The phase space is exactly integrable. Every orbit is labeled by a single constant $c$ , derived from a first integral.
- The solution space separates into three distinct classes based on $c$ $c$ :
  1. $c > -1/6$ : Trajectories terminate on an asymptotically de Sitter fixed line ( $\Omega=0, 0 < u < 3$ ).
  2. $c = -1/6$ : This is the critical trajectory corresponding to the analytic solution found in Ref. [1]. It acts as a separatrix between expansion and recollapse, approaching the non-hyperbolic endpoint $(0, 3)$ algebraically rather than exponentially.
  3. $c < -1/6$ : Trajectories reach $H=0$ at finite time with finite scale factor and density, leading to a turnaround and recollapse (Big Crunch).
- The known analytic solution is thus identified not as a generic future attractor, but as a critical separatrix.
Conserved Baryons and Non-conserved CDM:
- The phase space becomes three-dimensional, featuring a line of matter repellers and a line of vacuum fixed points $D(u_0) = (0, 0, u_0)$ .
- For $0 < u_0 < 3$ , each vacuum point possesses a two-dimensional stable sheet. Trajectories on these sheets exhibit late-time behavior where CDM decays slower than baryons, eventually dominating the matter sector.
- The endpoint $u_0 = 3$ is critical; it asymptotically recovers the single-fluid separatrix behavior of the universal model, with algebraic approach rates.
The Singular Baryonic Limit:
- The authors identify a non-uniform limit as CDM abundance ( $\Omega_c$ ) approaches zero.
- If $\Omega_c > 0$ (arbitrarily small), the trajectory eventually enters a stable sheet ending at a finite $u_0$ , corresponding to a solution with a cosmological-constant-like integration constant ( $\Lambda_{eff} \propto 1/u_0^2$ ).
- If $\Omega_c = 0$ exactly, the trajectory remains on the conserved baryonic plane ( $\Omega_b = 1$ ) and runs to $u \to \infty$ , never reaching a finite vacuum fixed point.
- This reveals that the operation of taking the late-time limit and then setting CDM to zero is not equivalent to setting CDM to zero from the outset. The finite- $u_0$ solutions on the baryonic boundary require the standard integration constant freedom of trace-free gravity.

Significance and Claims
The paper claims that the homogeneous thermogravity equations possess a significantly richer global structure than suggested by the single analytic solution previously cited. The known solution is shown to be a critical orbit separating distinct cosmological fates (de Sitter expansion vs. recollapse).

In the phenomenologically cleaner model (conserved baryons, non-conserved CDM), the analysis reveals a family of stable sheets feeding a line of vacuum fixed points. The authors emphasize that the critical endpoint $u_0=3$ generalizes the special single-fluid solution, providing a natural phase-space continuation when conserved baryons are restored.

Crucially, the study highlights a conceptual distinction regarding the cosmological constant. In the absence of CDM, finite- $u_0$ attractors require a cosmological-constant-like integration constant. However, the "zero-integration-constant" baryonic solution behaves fundamentally differently, remaining at $\Omega_b=1$ and failing to reach the vacuum line. The authors assert that this singular limit must be distinguished from the theoretical statement that vacuum energy does not gravitate in the underlying theory.

Finally, the phase-space map serves as a global guide for numerical solutions (such as those in Ref. [21] which include radiation) by identifying critical surfaces and integration-constant freedoms that must be controlled when specifying cosmological boundary conditions. The authors note that extending this analysis to include radiation, spatial curvature, and perturbations is straightforward in principle.

More like this