The Dark Sector on a Seesaw: How a Scalar Spring Solves Cosmological Puzzles ✛
Context
The Hubble tension is one of the sharpest contradictions in modern cosmology. Local measurements, such as SH0ES, give H0 ≈ 73 km/s/Mpc, while indirect estimates from the cosmic microwave background within ΛCDM point to 67.4 km/s/Mpc. This discrepancy does not disappear with new data; it only grows: recent SPT and ACT results push it to 6σ. Attempts to solve the problem with post-recombination physics fail because baryon acoustic oscillations (BAO) tightly fix the product rd·H0. The remaining option is to modify physics before recombination, and one promising direction is early dark energy, particularly the Hot NEDE model, where a phase transition in the dark sector after nucleosynthesis reduces the sound horizon. This work joins the line of research on interacting dark sectors, such as SIDR and Cold NEDE, but proposes a new mechanism—the decoupling of dark matter and radiation—which can fully relieve the tension.
Methods
The authors build a microscopic model of the dark sector based on a SU(N) gauge symmetry, spontaneously broken to SU(N−1) via the Higgs mechanism. A fermion multiplet χ is added, serving as dark matter. After the phase transition, a mass splitting emerges between the neutral and charged components of the multiplet, leading to suppressed interaction as the universe cools—a process dubbed Dark Radiation Matter Decoupling (DRMD). For the analysis, a simplified three-parameter version of the model (ΔNeff, fidm, zdec) is implemented in a modified CLASS code. A Markov chain Monte Carlo analysis is performed with Planck 2018, DESI BAO DR2, and Pantheon+ data, along with a profile likelihood to account for projection effects.
Results
The DRMD model achieves agreement between combined CMB, BAO, and supernova data and the local H0 measurement from SH0ES at the 1.4σ level, whereas ΛCDM gives 5.7σ and SIDR gives 3.0σ. When the SH0ES prior is added, the χ² improvement is −33.3 compared to ΛCDM. Key parameters: ΔNeff ≈ 0.8, the fraction of interacting dark matter fidm < 4.4% (95% C.L.), and the decoupling occurs at zdec ≈ 2200, close to matter-radiation equality. The profile likelihood shows an almost flat dependence on H0 in the range 0.69–0.73, indicating effective removal of the tension. The model also predicts suppression of the matter power spectrum on small scales, consistent with observations and not increasing S8.
Implications
This work adds a fundamentally new element to the toolkit for solving the Hubble tension: the dynamic decoupling of dark matter and dark radiation. Unlike previous SIDR models, where the interaction is constant, here it disappears as the universe cools, allowing both an increase in H0 and consistency with CMB and BAO. This is the first model based on microphysics with gauge symmetry that achieves full resolution of the tension without exotic potentials. The work aligns with dark sector research but stands out because the phase transition and mass splitting naturally predict the decoupling, rather than being introduced ad hoc.
Future development
The DRMD model opens several avenues for testing. First, predictions for the matter power spectrum on small scales can be checked by future surveys such as DESI and Euclid. Second, the phase transition in the dark sector should generate gravitational waves that could be detected by pulsar timing arrays. Third, refining the decoupling dynamics will require a more detailed analysis of kinetic and chemical equilibrium in the dark sector, including possible Sommerfeld corrections. Finally, the model can be extended by adding massless fermions to enhance the effect.
Impact
The results will impact cosmology, particle physics, and gravitational-wave astronomy, as the model links observational data to specific dark sector microphysics and predicts testable signals.
Next steps
Immediate next steps include analyzing the model with new ACT and SPT data, as well as studying the full version of the model including phase transition effects on the power spectrum and gravitational waves.
Key open problems
This work directly addresses the unresolved nature of dark matter and dark energy, and the question of what new particles and interactions might exist beyond the Standard Model. It also touches on the problem of initial conditions for phase transitions in the early universe and their observational signatures.
The knowledge machine's view
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
The paper proposes the DRMD model, in which the dark sector undergoes a phase transition that produces self-interacting dark radiation, and the interaction of dark matter with this radiation switches off as it cools. Analysis of Planck, DESI, and Pantheon+ data shows agreement with the local value of H0 at the 1.4σ level, compared to 5.7σ in ΛCDM.
Where its strength is
The strength of the work lies in constructing a microscopic model with a concrete mechanism for the decoupling of the interaction, which naturally resolves the Hubble tension without increasing S8, setting it apart from many alternatives.
Where it could go next
- It might be interesting to compare DRMD predictions with early dark energy models, which also use a phase transition before recombination, to understand how general the mechanism of reducing the sound horizon is.based on: A Dark Deal: How the Universe Reconciles Its Clocks arXiv:2605.01904
- A related line of work uses axions to reconcile the two measurements of the Universe; it may be worth examining how axion models fare with the same data and whether the approaches can be combined.based on: Axions reconcile two measurements of the Universe arXiv:2606.09427
- It would be interesting to investigate how DRMD affects the signal from fast radio bursts, which are used for independent H0 measurements, and whether such data can be used to further test the model.based on: In the Rhythm of the Universe: Radio Bursts vs. Hubble Tension arXiv:2606.22390
Worth trying
- One could test whether DRMD produces a characteristic signature in the power spectrum, distinguishable from other phase-transition models, using data from future surveys.
- It is worth exploring how the decoupling depends on the parameters of the gauge group and whether observational constraints on N can be derived.
- DRMD might predict specific features in the distribution of dark matter on small scales that could be sought in lensing data.
Why this may matter
If DRMD is confirmed, it would be an important step toward understanding the physics of the dark sector and resolving the Hubble tension without fine-tuning parameters, potentially pointing to a new fundamental symmetry.
Nearby works in our archive
- Dark-Sector Seesaw: How a Scalar Spring Solves Cosmological Puzzlessimilarity 0.61
- Why Has Cosmology Cracked?similarity 0.60
- Axions reconcile two measurements of the Universesimilarity 0.56
- In the Rhythm of the Universe: Radio Bursts vs. Hubble Tensionsimilarity 0.56
- A Dark Deal: How the Universe Reconciles Its Clockssimilarity 0.56
- The Cosmic Tuning Fork: Why the Universe Plays Two Different Notessimilarity 0.56
The closest work in the archive sits at 0.61 by meaning — the usual level for our corpus.
The mass splitting between dark matter components in the DRMD model is only about 0.1% of the mass, yet this tiny difference determines when dark matter stops interacting with radiation, which is critical for solving the Hubble tension.
The idea of a dark sector with its own particles and interactions echoes the concept of 'dark life' from science fiction, for example in Liu Cixin's novel 'The Dark Forest', where dark matter plays a key role in cosmology.
Key numbers
- H0 (local value): 73.04 ± 1.04 km/s/Mpc
- H0 (ΛCDM prediction): 67.36 ± 0.54 km/s/Mpc
- DRMD agreement with SH0ES: 1.4σ
- χ² improvement with SH0ES: -33.3
- ΔNeff (best fit): ≈ 0.8