⚛️ phenomenology

Forbidden dark matter assisted by first-order phase transition and associated gravitational waves

This paper proposes a testable framework for light fermion dark matter where a strongly first-order phase transition in a dark $U(1)_D$ sector enables the correct relic abundance through kinematically forbidden or loop-suppressed annihilation channels, thereby naturally evading stringent CMB and indirect detection constraints while predicting associated gravitational waves.

Original authors: Satyabrata Mahapatra, Partha Kumar Paul, Narendra Sahu

Published 2026-08-14

📖 7 min read🧠 Deep dive

CC BY 4.0

Original authors: Satyabrata Mahapatra, Partha Kumar Paul, Narendra Sahu

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

The Invisible Ghost and the Cosmic Switch

Imagine the universe is a giant, bustling city. We can see the buildings, the cars, and the people—these are the normal things we know, like stars, planets, and you. But astronomers have discovered that this city is actually built on a foundation of invisible "ghosts" that make up about 85% of all the matter around us. We call this mysterious stuff Dark Matter. It doesn't shine, it doesn't reflect light, and it barely talks to normal matter, which is why we can't see it directly. We only know it's there because its gravity pulls on the visible stuff, like how a strong wind pushes a sailboat even if you can't see the wind itself.

For a long time, scientists thought these ghosts were heavy, slow-moving particles called WIMPs. But after years of searching with giant detectors deep underground, we haven't found them. This has led scientists to wonder: what if the ghosts are actually much lighter, like tiny, speedy fireflies? The problem is, if they are light and move fast, they should have annihilated (destroyed each other) too quickly in the early universe, leaving almost none left for us to find today. It's like a party where everyone leaves immediately after the music starts. So, how did these light ghosts survive to build the universe we see now? This is the puzzle a new paper tries to solve.

The Paper's Big Idea: A Forbidden Party

The researchers, Satyabrata Mahapatra, Partha Kumar Paul, and Narendra Sahu, propose a clever trick to save these light dark matter ghosts. They suggest a scenario where the dark matter particles are "forbidden" from destroying each other under normal conditions, but were allowed to do so just for a brief moment in the very early, hot universe.

Think of the dark matter particles as two people trying to high-five. In our current cold universe, they are too far apart to reach each other; the "high-five" is kinematically forbidden. It's like trying to touch a star from Earth with your hand—the distance is just too great. Because they can't high-five (annihilate), they don't disappear, and their numbers stay safe.

But here is the twist: in the very beginning, the universe was a scorching hot oven. The researchers suggest that a special event, called a First-Order Phase Transition, happened like a cosmic switch flipping. Imagine the universe was a room full of water (a liquid). Suddenly, the temperature dropped, and the water started turning into ice. But it didn't freeze smoothly; it bubbled and froze in sudden, violent chunks. This is a "strong first-order phase transition."

During this chaotic "freezing" moment, the dark matter particles gained their mass. Before this switch flipped, they were massless and could annihilate easily. After the switch, they became heavy, and the "high-five" became impossible because they didn't have enough energy to reach each other. This sudden change stopped the dark matter from disappearing, leaving just the right amount to fill the universe today.

The Rules of the Game

The paper builds a specific model to make this work. They imagine a hidden "dark sector" with its own rules, governed by a force similar to electricity but invisible to us. In this sector, there are dark matter particles (let's call them $\chi$ ), a dark force carrier ( $X_D$ ), and a dark Higgs-like particle ( $\phi$ ).

The authors point out a very specific rule in their model: the dark matter particles are strictly forbidden from destroying each other to make two $\phi$ particles at the most basic level. It's like a rule in a video game that says, "You cannot jump directly to the finish line." This process only happens if the particles take a complicated, roundabout path (a "loop" in physics terms), which makes it very slow and unlikely.

Instead, the main way they could destroy each other is by making one $X_D$ and one $\phi$ . However, the researchers carefully tune the masses of these particles so that this process is forbidden in our cold, current universe. The combined weight of the $X_D$ and $\phi$ is just slightly heavier than the two dark matter particles trying to make them. It's like trying to buy a candy bar that costs $1.05 when you only have $1.00. You can't do it.

But in the hot, early universe, the particles had extra "thermal energy" (like having a little bit of extra cash from a friend). This allowed them to briefly overcome the cost and annihilate. Once the universe cooled down and the phase transition happened, that extra energy vanished, the "candy bar" became too expensive, and the annihilation stopped. This "forbidden" mechanism ensures that the dark matter survives in the exact amount we observe today.

The Cosmic Echo: Gravitational Waves

Here is where the story gets really exciting. The paper suggests that this "switch flipping" (the phase transition) wasn't just a quiet event. It was violent and explosive, like bubbles forming in a boiling pot of water. When these bubbles of the new "frozen" state crashed into each other, they would have shaken the fabric of space-time itself, creating ripples called Gravitational Waves.

The authors calculate that these ripples would have a specific frequency, depending on how heavy the dark matter is. If the dark matter is very light (in the range of MeV to GeV), the ripples would be at a frequency that future space-based detectors, like LISA or DECIGO, might be able to hear. If the dark matter is heavier, the ripples would be at a different pitch, perhaps detectable by pulsar timing arrays (which listen to the rhythmic signals of spinning dead stars).

The paper finds a beautiful connection: the heavier the dark matter, the higher the temperature needed for the phase transition, and the higher the frequency of the gravitational waves. It's like a musical scale where the weight of the ghost determines the note of the cosmic song.

What the Paper Rules Out and Finds

The authors are very careful about what their model allows and what it bans. They explicitly argue against the idea that dark matter could simply annihilate into standard particles (like electrons or photons) in a simple, fast way. If that were true, the dark matter would have vanished long ago, or we would have seen it in experiments like the Cosmic Microwave Background (CMB) observations. Their model rules out these "easy" paths.

Instead, they find that the "forbidden" path is the only way to get the numbers right. They show that if the dark matter mass is between MeV and GeV, and the dark gauge coupling (how strongly they interact) is tuned correctly, the universe naturally ends up with the right amount of dark matter.

They also check their idea against a long list of real-world constraints. They look at data from experiments that hunt for dark matter hitting electrons or atoms (like XENONnT and PandaX-4T), and they check if the dark particles would have messed up the formation of elements in the early universe (Big Bang Nucleosynthesis). Their analysis suggests that there are specific "sweet spots" in the parameters—specific combinations of mass and force strength—where their model works perfectly without breaking any of these rules.

The Bottom Line

This paper doesn't claim to have found dark matter yet. Instead, it offers a new, testable map for where to look. It suggests that if dark matter is light and hidden in a "forbidden" state, we might not find it by smashing particles together in a collider alone. Instead, we might hear the echo of its birth in the form of gravitational waves.

The authors propose that the universe's history of dark matter is tightly linked to a violent phase transition that created a gravitational wave background. If future detectors like LISA, DECIGO, or µARES pick up these specific ripples, it would be a massive clue that this "forbidden" scenario is real. It turns the search for dark matter into a cosmic detective story where the clues are not just particles, but the very sound of the universe changing its shape.

Technical Summary: Forbidden Dark Matter Assisted by First-Order Phase Transition

Problem Statement
The paper addresses the challenge of explaining light fermion dark matter (DM) with masses in the MeV–GeV range. Standard thermal freeze-out scenarios for such light DM typically require annihilation cross-sections that are too large, leading to an under-abundant relic density. Furthermore, if these light DM particles annihilate into Standard Model (SM) particles or massless dark sector states, the resulting late-time energy injection creates severe tensions with Cosmic Microwave Background (CMB) constraints and indirect detection limits. While "forbidden" DM scenarios (where annihilation is kinematically suppressed at late times) offer a solution, they often struggle to generate the correct relic density without fine-tuning or face constraints if the final states are SM particles. Additionally, the mechanism required to generate masses for the dark sector often lacks a connection to observable cosmological signatures like gravitational waves.

Methodology
The authors propose a minimal $U(1)_D$ gauge extension of the Standard Model featuring a Dirac fermion DM candidate ( $\chi$ ) and a complex scalar singlet ( $\Phi$ ). The model relies on two interconnected mechanisms:

Symmetry Breaking and Mass Generation: The dark $U(1)_D$ symmetry is spontaneously broken by the vacuum expectation value (VEV) of $\Phi$ , generating masses for the dark gauge boson ( $X_D$ ) and the physical scalar ( $\phi$ ).
First-Order Phase Transition (FOPT): The scalar potential is constructed to undergo a strongly first-order phase transition in the early universe. This transition dynamically generates the masses for $X_D$ and $\phi$ , altering the kinematic landscape for DM annihilation.
Forbidden Annihilation Channels: The model exploits the $U(1)_D$ gauge structure to strictly forbid the tree-level annihilation $\bar{\chi}\chi \to \phi\phi$ . Consequently, the dominant annihilation channel is the tree-level process $\bar{\chi}\chi \to X_D \phi$ . The model parameters are tuned such that $2M_\chi < M_{X_D} + M_\phi$ , rendering this dominant channel kinematically forbidden at zero temperature but active during freeze-out via thermal assistance (Boltzmann tail).
Loop-Induced Processes: The sub-dominant channel $\bar{\chi}\chi \to \phi\phi$ arises only at the one-loop level (via box and triangle diagrams) and is intrinsically p-wave suppressed.

The authors solve the Boltzmann equations for the DM yield, incorporating detailed balance for the forbidden channels and including both tree-level and loop-level contributions. They analyze two distinct regimes:

Regime A: $\bar{\chi}\chi \to X_D \phi$ is forbidden, but $\bar{\chi}\chi \to \phi\phi$ is kinematically allowed.
Regime B: Both $\bar{\chi}\chi \to X_D \phi$ and $\bar{\chi}\chi \to \phi\phi$ are kinematically forbidden.

The study also calculates the stochastic gravitational wave (GW) background generated by the FOPT, evaluating the signal-to-noise ratio (SNR) for various future detectors (LISA, DECIGO, BBO, $\mu$ ARES, PTA arrays) and cross-referencing the parameter space with constraints from direct detection, Big Bang Nucleosynthesis (BBN), and Higgs invisible decays.

Key Contributions

Dual-Role Symmetry Breaking: The paper demonstrates that the same scalar field responsible for generating DM masses also drives a strong FOPT, linking the DM relic density mechanism directly to GW production.
Tree-Level Suppression via Gauge Structure: By utilizing specific $U(1)_D$ charge assignments, the model naturally eliminates the dangerous tree-level $\bar{\chi}\chi \to \phi\phi$ annihilation, forcing the relic density to be determined by the interplay between a forbidden s-wave process and a loop-induced p-wave process.
Natural Evading of Constraints: The scenario naturally evades CMB and indirect detection bounds. The forbidden $\bar{\chi}\chi \to X_D \phi$ channel is exponentially suppressed at late times due to the kinematic threshold, while the loop-induced $\bar{\chi}\chi \to \phi\phi$ channel is suppressed by p-wave velocity factors ( $v^2$ ).
Mass-Temperature Correlation: The authors identify a robust correlation where heavier DM masses require higher phase transition temperatures to maintain the forbidden kinematic condition ( $M_{X_D} + M_\phi > 2M_\chi$ ). This shifts the peak frequency of the resulting GW signal to higher values for heavier DM.

Results

Relic Abundance: The model successfully reproduces the observed DM relic density ( $\Omega_{DM}h^2 \approx 0.12$ ) for DM masses ranging from the MeV to GeV scale with order-one couplings.
Benchmark Points: Several benchmark points (BP0–BP10) are identified. The analysis shows that in Regime A, the dominant channel shifts from the forbidden tree-level process to the allowed loop-level process as the mass splitting increases or the DM mass changes. In Regime B, the milder exponential suppression of the forbidden $\bar{\chi}\chi \to \phi\phi$ channel allows it to dominate over the heavily suppressed $\bar{\chi}\chi \to X_D \phi$ channel.
Experimental Viability: After applying constraints from direct detection (XENONnT, PandaX-4T, SuperCDMS), BBN, and Higgs invisible decays, a subset of benchmark points (BP0, BP5, BP6, BP8, BP10) remains viable.
Gravitational Waves: The FOPT produces a stochastic GW background detectable by upcoming observatories.
- MeV-scale DM (e.g., BP8) corresponds to phase transitions in the MeV range, producing GW signals in the nHz– $\mu$ Hz band, potentially accessible to Pulsar Timing Arrays (PTAs) and $\mu$ ARES.
- GeV-scale DM (e.g., BP5, BP6) requires higher transition temperatures, shifting GW peaks to $\sim 10^{-4}$ Hz, within the sensitivity windows of LISA and DECIGO.
Correlation: A clear correlation is established between the DM mass and the GW peak frequency: heavier DM necessitates a higher nucleation temperature, shifting the GW signal to higher frequencies.

Significance
The paper claims that this framework offers a "simple yet testable" solution to the light DM problem that simultaneously addresses the relic density puzzle and predicts observable cosmological signatures. Its significance lies in the tight linkage between microscopic particle physics parameters (DM mass, gauge coupling) and macroscopic cosmological observables (GW spectrum). The authors argue that this setup provides complementary discovery paths: terrestrial experiments can probe the DM-SM portal and direct detection, while space-based interferometers and PTAs can probe the dark sector dynamics via gravitational waves. The model suggests that a detection of a stochastic GW background could not only confirm a phase transition but also constrain the mass and coupling of the dark matter candidate, offering a unique consistency check for the "forbidden" DM paradigm.