🔭 astrophysics

Structure-motivated phantom-crossing dark energy: joint constraints from DESI DR2, supernovae and growth

This paper proposes a structure-motivated phantom-crossing dark energy model, where the dark energy density scales with the mass-weighted rms dark matter acceleration, and demonstrates that joint constraints from DESI DR2, supernovae, and growth data support a scenario linking the cosmic coincidence, the $w=-1$ crossing, and dark energy-matter equality to a single density maximum around $z \simeq 0.4\text{--}0.8$ .

Original authors: Zhijie Xu

Published 2026-10-19

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Zhijie Xu

Original paper licensed under CC BY 4.0 (https://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

For decades, astronomers have watched the universe expand at an ever-faster pace, a discovery that earned a Nobel Prize and reshaped our understanding of reality. To explain this acceleration, scientists proposed the existence of "dark energy," a mysterious force that pushes galaxies apart. In the standard model of cosmology, this force is treated as a constant, unchanging energy inherent to empty space itself. However, recent observations have begun to question whether this energy is truly constant or if it evolves over time, perhaps even shifting its nature in ways that defy current physics. At the same time, we know the universe is filled with invisible "dark matter" that clumps together to form the cosmic scaffolding for galaxies. The big question remains: is the force driving the universe apart connected to the way this invisible matter clumps together, or are they two completely separate phenomena?

A new study by Zhijie Xu at Pacific Northwest National Laboratory suggests a surprising link between these two cosmic mysteries. Instead of treating dark energy as a static background feature, the researcher proposes that it is a dynamic response to the formation of cosmic structure. The core idea is that as dark matter particles accelerate under gravity to form halos—the massive clouds that hold galaxies together—they generate a specific kind of motion. The study posits that dark energy is not a fixed property of the vacuum but is actually fueled by the intensity of this acceleration. As the universe evolves and these structures grow, the "push" of dark energy changes in direct response to the changing speed and violence of the dark matter's movement.

To test this idea, the researcher built a mathematical model that ties the density of dark energy to the average acceleration of dark matter particles and the growth of cosmic structures. This model was then put to the test against the most precise cosmological data available, including measurements from the Dark Energy Spectroscopic Instrument (DESI), observations of distant supernovae, and data on how fast galaxies are moving toward each other. The results were striking: this structure-driven model fit the observed data better than the standard constant-energy model and also outperformed other popular theories that allow dark energy to change over time. The data strongly favored a scenario where dark energy was slightly weaker in the past, grew stronger as cosmic structures formed, and is now beginning to weaken again as the universe expands.

The study also explored why the universe seems to be in a peculiar state right now, where the density of dark energy is roughly comparable to the density of matter. In many theories, this balance is a strange coincidence, but this model suggests it is a natural consequence of the timeline of cosmic history. The research identifies a specific moment in the past, roughly six billion years ago, when the smooth flow of matter in the early universe transitioned into the chaotic, clumpy motion seen in galaxies today. The model predicts that the mysterious "phantom" behavior of dark energy—where it acts more strongly than a simple constant—crosses a critical threshold right around this same era. This alignment suggests that the acceleration of the universe and the formation of galaxies are not separate events but parts of a single, interconnected process.

By analyzing the data, the researcher found that the history of dark energy is likely not a straight line but a curve. The energy density appears to have peaked in the recent cosmic past, around the time when the universe was about half its current age, and is now slowly declining. This peak coincides with the era when the universe shifted from being dominated by smooth, linear expansion to being dominated by the complex, non-linear clumping of matter. The study rules out the idea that dark energy is a simple, unchanging constant, and it also challenges models where the energy changes in a way that has no connection to the matter around it. Instead, the evidence points to a universe where the force driving expansion is intimately tied to the dance of the invisible matter that fills it.

The findings rely on a specific interpretation of how dark matter moves. The researcher used computer simulations to estimate the average acceleration of dark matter particles, finding a value that closely matches a known scale in galaxy physics. This connection allows the model to explain why the universe is accelerating now without needing to invent new, unexplained constants. The model suggests that the "push" of dark energy is essentially a reaction to the "strain" of the universe as it forms structures. When the universe was young and smooth, this push was negligible. As galaxies and clusters began to form, the acceleration of dark matter increased, and the dark energy response grew stronger. Now that the universe is expanding so rapidly that new structures are harder to form, the response is beginning to fade.

This work does not claim to have solved the mystery of dark energy completely. The connection between the acceleration of dark matter and the energy of the vacuum is still a hypothesis that requires a deeper theoretical foundation. The researcher acknowledges that the model is built on a proposed interaction that has not yet been derived from a complete theory of gravity. However, the fact that this specific, structure-based law fits the observational data better than the standard models is a significant step forward. It suggests that the key to understanding the universe's acceleration may lie not in looking at empty space, but in watching how the invisible matter within it moves and changes.

The study also highlights the power of combining different types of astronomical data. By looking at the expansion of the universe, the brightness of exploding stars, and the growth of galaxy clusters all at once, the researcher was able to narrow down the possible histories of dark energy. The data consistently point to a single, distinct peak in the energy density, a feature that simpler models cannot easily reproduce. This peak occurs at a time when the universe was transitioning from a state where matter flowed smoothly to one where it was violently clumping together. The alignment of this transition with the behavior of dark energy provides a compelling narrative for the cosmic history we observe today.

Ultimately, this research offers a new way to think about the universe's expansion. It moves away from the idea of dark energy as a mysterious, unchanging background and toward a view where it is a dynamic participant in the cosmic story, responding to the formation of the very structures it helps to separate. While the full theoretical mechanism remains to be worked out, the observational evidence suggests that the acceleration of the universe is deeply rooted in the physics of structure formation. The next step for scientists will be to test these predictions with even more precise data, looking for the specific signatures of this interaction in the distribution of galaxies and the history of cosmic expansion.

Technical Summary: Structure-motivated phantom-crossing dark energy

Problem Statement
The standard cosmological model ( $\Lambda$ CDM) describes the accelerated expansion of the Universe via a cosmological constant, yet it leaves the cosmological constant problem (why the vacuum energy is so small) and the coincidence problem (why dark energy density $\rho_{DE}$ is comparable to matter density $\rho_m$ today) unresolved. Furthermore, recent data from DESI DR2 BAO, combined with CMB and supernovae, suggest that the dark energy equation of state $w$ may cross the phantom divide ( $w = -1$ ), a feature difficult to achieve smoothly in single minimally coupled scalar field theories. The paper seeks a phenomenological law for dark energy that is motivated by the formation of large-scale structure, potentially linking the amplitude and time dependence of $\rho_{DE}$ to the dynamics of dark matter.

Methodology
The author proposes a structure-motivated background law where the dark energy density is driven by the mass-weighted root-mean-square (rms) acceleration of dark matter particles, $\langle a_p^2 \rangle^{1/2}$ , and the linear growth factor $D(a)$ . The proposed density law is:
$\rho_{DE} = K \frac{\langle a_p^2 \rangle}{G} \left( \frac{D}{D_0} \right)^n$
where $K \approx 4$ is a dimensionless coefficient, $G$ is the gravitational constant, and $n$ is a free exponent.

The study proceeds through three distinct components:

Theoretical Framework: The author defines an $x_0 x_a$ parametrization where the equation of state $w$ is linear in the growth rate $f \equiv d \ln D / d \ln a$ , rather than the scale factor $a$ . This leads to a power-law density form $\rho_{DE} \propto a^{-n f_\times} D^n$ . They analyze the dynamical system in the $(\Omega_m, f)$ state space, showing that this closure creates an autonomous system with a heteroclinic orbit from a matter saddle to a dark energy attractor.
Structural Calibration: Using N-body simulations (Virgo SCDM and Illustris-1-Dark), the author characterizes the acceleration statistics of dark matter. They decompose the velocity and acceleration fields into linear bulk flows and nonlinear virial motions within halos. They identify a "bulk-virial crossover" epoch $a_*$ where the kinetic energy of random halo motions overtakes coherent bulk flow. They derive a relation between the halo acceleration scale $a_c$ and the velocity dispersion, linking the simulated acceleration to the Milgrom scale $a_0$ (associated with MOND) via the cascade rate of energy transfer.
Effective Model Construction: To provide a mechanistic basis, the author constructs an effective model within the framework of self-tuned $q$ -theory. They postulate a structural interaction energy term coupling the vacuum variable $q$ to the acceleration variance $\langle a_p^2 \rangle$ and a strain configuration factor. This interaction generates a residual dark energy that tracks the structural evolution adiabatically.

Key Contributions

Phenomenological Law: The introduction of a dark energy density law explicitly dependent on the variance of dark matter acceleration and the growth factor, motivated by the idea that structure formation sources the dark energy dynamics.
Resolution of Coincidences: The model demonstrates that the bulk-virial crossover ( $a_*$ ), the phantom crossing ( $a_\times$ , where $w=-1$ ), and the dark energy-matter equality ( $a_\Lambda$ ) occur in a closely ordered sequence ( $a_* \lesssim a_\times < a_\Lambda$ ). The model explains the proximity of the halo acceleration scale to the Milgrom scale $a_0$ as a consequence of the cascade rate and the equality horizon, without requiring new particle physics.
State Variable Formulation: The formulation of $w$ as a linear function of the growth rate $f$ ( $w = -1 - \frac{n}{3}(f - f_\times)$ ) provides a bounded equation of state that naturally crosses the phantom divide and remains finite over the entire cosmic history, unlike the CPL parametrization which diverges as $a \to \infty$ .
Effective Vacuum Mechanism: A conditional effective model is presented where a self-tuned vacuum, perturbed by a postulated structural interaction, reproduces the density law. The residual energy is identified as the direct interaction energy rather than a displacement of the vacuum minimum.

Results
The model is tested against DESI DR2 BAO data, the DESI compressed CMB likelihood, Pantheon+ supernovae, and SDSS growth rates ( $f\sigma_8$ ).

Fit Quality: The structural law improves upon $\Lambda$ CDM by $\Delta\chi^2 = -8.4$ to $-9.7$ and outperforms the CPL parametrization by $0.6$ to $1.9$ at an equal number of parameters.
Posterior Constraints: The data robustly constrain the history of dark energy to have a single density maximum at redshift $z \simeq 0.4-0.8$ , occurring before matter-dark energy equality. In the joint posterior, 97–100% of histories have $w_0 > -1$ and a single crossing of $w = -1$ .
Parameter Degeneracy: The exponent $n$ is found to be degenerate with the "clock" used to describe the decay of acceleration (scale factor $a$ , cosmic time $t$ , or growth factor $D$ ) and the halo assembly slope. The data constrain the combined logarithmic slope of the density rather than the specific decomposition into these components.
Predictive Power: Both the structural law and the linear- $w(f)$ model exhibit higher Bayesian evidence than CPL and better predict supernova data withheld from the fit. The structural law's slight improvement over the linear- $w(f)$ model is attributed to the curvature introduced by the halo assembly term.

Significance and Claims
The paper claims that the structural law offers a phenomenologically viable alternative to standard dark energy models that naturally accommodates phantom-crossing behavior and addresses the cosmic coincidence. The author emphasizes that the model's empirical success relies on a power-law response to growth near the phantom crossing.

However, the author maintains a modest stance regarding the theoretical underpinnings:

The coupling between structure and dark energy is postulated, not derived from a covariant action.
The calibration of the acceleration statistics relies on provisional simulation estimates (Virgo and Illustris) whose uncertainties are not fully propagated.
The results are conditional on the compressed CMB likelihood (which omits lensing) and the specific acceleration calibration.
The exponent $n$ is not yet a fixed physical constant but is degenerate with the assumed source model and clock; an independent calibration of $\langle a_p^2 \rangle(a)$ is required to interpret $n$ as a fundamental vacuum response exponent.

The work concludes that while the specific decomposition of the density law remains degenerate, the data robustly select a history with a single density maximum and a phantom crossing, a feature well-captured by structure-motivated models.

More like this