⚛️ phenomenology

Gravitational waves from supercooled phase transitions and pulsar timing array signals

This paper proposes that a supercooled first-order phase transition in a hidden sector with spontaneously broken $U(1)_X$ gauge symmetry can generate a gravitational wave background strong enough to explain recent Pulsar Timing Array observations while remaining consistent with Big Bang Nucleosynthesis constraints.

Original authors: Jinzheng Li, Pran Nath

Published 2026-02-17

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Jinzheng Li, Pran Nath

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

Imagine the universe as a giant, expanding balloon. For decades, physicists have been trying to listen to the "hum" of the universe's earliest moments. Recently, a group of astronomers using giant radio telescopes (called Pulsar Timing Arrays) finally heard a low-frequency rumble—a background noise of gravitational waves. It's like hearing the distant roar of a crowd, but we don't know who is shouting or what they are shouting about.

This paper by Jinzheng Li and Pran Nath proposes a solution to that mystery. They suggest the noise comes from a cosmic "snap" that happened in a hidden, invisible part of the universe.

Here is the story of their discovery, broken down into simple concepts:

1. The Hidden Room and the "Snap"

Think of our visible universe (stars, planets, us) as a living room. But imagine there is a hidden room right next to it that we can't see. This hidden room has its own rules and particles.

The authors propose that in this hidden room, something dramatic happened billions of years ago. Imagine the hidden room was filled with super-hot steam (energy). Suddenly, the temperature dropped, and the steam tried to turn into ice. But instead of freezing smoothly, it got "stuck" in a super-cooled state, like water that stays liquid even below freezing.

Eventually, the "ice" (a new, stable state) suddenly formed in bubbles. These bubbles expanded and crashed into each other, creating a massive cosmic shockwave. This crash sent ripples through space-time itself—these are the gravitational waves we are hearing today.

2. The "Super-Cooled" Trick

Why didn't this happen earlier? The paper explains that for the waves to be loud enough to be heard by our telescopes today, the "snap" had to happen very late and very suddenly.

Think of it like popping a balloon. If you pop it slowly, it makes a quiet fizz. If you super-cool the rubber and then pop it, it makes a loud BANG. The authors show that this "super-cooled" snap is the only way to get a signal strong enough to match what the telescopes are hearing, without breaking the rules of how the early universe evolved (specifically, how the first atoms formed).

3. The Two-Temperature Problem (The Thermostat Analogy)

This is the most clever part of the paper. Usually, scientists assume the hidden room and the living room (our visible universe) have the same temperature, like two rooms connected by an open door.

But the authors realized: What if the hidden room was much colder than the living room?

Imagine the living room is a sauna (hot), and the hidden room is a freezer (cold).

The Old Way: Scientists assumed the freezer and sauna were always the same temperature. If they used this assumption, the math said the "snap" would happen too early or be too weak to explain the radio telescope signals.
The New Way: The authors tracked the "thermostats" of both rooms separately. They found that because the hidden room was colder, the "snap" happened at just the right time and with just the right force to create the specific low-frequency hum the telescopes detected.

If you ignore the fact that the hidden room was colder, your prediction for the sound is off by a factor of ten thousand! It's like trying to predict the sound of a drumbeat while ignoring that the drummer is wearing heavy boots.

4. The "Bubble" Math

When these bubbles of "ice" formed and crashed, they created the sound.

Old Math: Scientists used a simple ruler to measure how fast the bubbles formed. This ruler said, "This is impossible; the bubbles would form too slowly."
New Math: The authors used a more precise tool (measuring the average distance between bubbles). This tool showed that even though the process was slow, it was still fast enough to create a loud signal. It's like realizing that even if a snail is slow, if it's moving on a giant track, it can still finish the race in time.

5. Why This Matters

The paper solves a puzzle with three pieces:

The Signal: It explains the mysterious "hum" heard by the radio telescopes.
The Rules: It doesn't break the laws of physics regarding how the early universe formed atoms (Big Bang Nucleosynthesis).
The Future: It predicts that if we build better space-based detectors (like LISA or Taiji), we might hear this same "snap" again, but at a higher pitch, confirming that our hidden room theory is correct.

The Bottom Line

The universe has a secret history. Just like a house might have a hidden basement where a party happened long ago, our universe has a hidden sector where a massive, super-cooled phase transition occurred. This event created a cosmic "thunderclap" that is still echoing through space today. By carefully tracking the temperature differences between the visible world and this hidden world, these scientists have finally figured out the source of the noise.

1. Problem Statement

Recent observations by Pulsar Timing Array (PTA) collaborations (NANOGrav, EPTA, PPTA) have detected a stochastic gravitational wave background (SGWB) in the nano-Hertz frequency range ( $10^{-10} - 10^{-8}$ Hz). The Standard Model (SM) cannot explain this signal because the electroweak phase transition is a smooth crossover, not a first-order transition.

To explain the PTA signal via a First-Order Phase Transition (FOPT) in the early universe, two conflicting requirements must be met:

Frequency: The transition must occur at a low enough temperature ( $T \gtrsim 1$ MeV) to satisfy Big Bang Nucleosynthesis (BBN) constraints, yet produce GWs in the nano-Hz range. Standard FOPTs with rapid transition rates ( $\beta/H \gg 1000$ ) typically produce GWs at much higher frequencies (mHz to Hz).
Amplitude: To generate sufficient GW power at low frequencies, the transition must be supercooled (occurring at a temperature $T_*$ much lower than the critical temperature $T_c$ ), requiring a very slow transition rate (small $\beta/H$ ).

Key Challenges:

BBN Constraints: Late transitions risk interfering with light element abundances.
Entropy Injection: Energy release must not spoil the thermal history or violate constraints on the effective number of neutrino species ( $\Delta N_{\text{eff}}$ ).
Completion: Supercooled transitions risk being too slow to complete before the universe expands too much.
Thermal History: Previous analyses often ignored the distinct thermal evolution of a "hidden sector" relative to the visible sector, leading to inaccurate predictions.

2. Methodology

The authors propose a Hidden Sector Model coupled to the Standard Model via kinetic mixing of gauge fields.

Model Structure:
- Symmetry: A hidden $U(1)_X$ gauge symmetry with a complex scalar $\Phi$ and a Dirac fermion $q$ .
- Coupling: The hidden sector couples to the SM hypercharge field $B_\mu$ via a kinetic mixing parameter $\delta$ .
- Potential: The scalar potential $V_{\text{hid}}$ includes zero-temperature terms and finite-temperature corrections (Coleman-Weinberg potential + thermal loops). This structure allows for a first-order phase transition where the scalar acquires a vacuum expectation value (VEV).
Thermal Evolution Analysis:
- The authors explicitly track the temperature ratio $\xi(T) = T_h / T_v$ between the hidden ( $T_h$ ) and visible ( $T_v$ ) sectors.
- They utilize energy conservation equations to derive the evolution of $\xi(T)$ . For small kinetic mixing ( $\delta \lesssim 10^{-10}$ ), they assume separate entropy conservation for each sector, which is computationally efficient and accurate.
- They calculate the Hubble parameter $H(T_h)$ using the total energy density of both sectors, which significantly impacts the percolation condition.
Phase Transition Dynamics:
- Percolation Temperature ( $T_{h,p}$ ): Defined as the temperature where 71% of the universe remains in the false vacuum.
- Transition Rate Characterization: The authors argue that the traditional parameter $\beta/H$ (inverse time scale) is insufficient for supercooled transitions. Instead, they use the mean bubble separation $R_*$ , calculated directly from the bubble nucleation rate and number density.
- Regime Classification: They analyze four cases of transition behavior, focusing on Case 3 (successful completion with a permanent barrier), which allows for the slow rates needed for PTA signals.
Gravitational Wave Calculation:
- The GW spectrum is calculated from three sources: bubble collisions, sound waves, and turbulence.
- They determine the transition strength $\alpha$ and the bubble wall velocity $v_w$ . For supercooled transitions matching PTA data, the scenario is found to be in the runaway regime (detonation type), where bubble collisions dominate the GW signal.

3. Key Contributions

Resolution of the $\beta/H$ Constraint: The paper demonstrates that the traditional constraint $\beta/H > 3$ (used to exclude slow transitions) is an artifact of the approximation $R_* H \approx (\beta/H)^{-1}$ . By using the exact mean bubble separation $R_*$ , a large portion of the parameter space with negative or very small $\beta/H$ values becomes viable for generating PTA signals.
Critical Role of Thermal History: The authors show that ignoring the evolution of the temperature ratio $\xi(T)$ leads to errors of up to four orders of magnitude in the predicted GW power spectrum. The evolving temperature ratio affects the Hubble rate, the percolation probability, and the final transition strength $\alpha_{\text{tot}}$ .
Resolution of $\Delta N_{\text{eff}}$ Tension: They propose a mechanism where the hidden sector undergoes "cannibalism" (number-changing processes like $3\phi \to 2\phi$ ) as particles become non-relativistic. This allows the hidden sector to decouple and cool rapidly, satisfying BBN constraints ( $\Delta N_{\text{eff}} < 0.3$ ) while maintaining a high temperature ratio $\xi_p \sim O(1)$ during the phase transition to maximize GW production.
Benchmark Points: The study provides five specific benchmark points (BP1–BP5) that successfully match NANOGrav 15-year data while satisfying all cosmological constraints.

4. Results

GW Spectrum: The calculated GW power spectra for the benchmark points peak in the nano-Hertz range, aligning with the NANOGrav, EPTA, and PPTA observations.
Dominant Source: In the nano-Hz region, bubble collisions are the dominant source of GWs (by several orders of magnitude), while sound waves and turbulence are highly suppressed due to the runaway nature of the transition.
Parameter Space:
- Transition strength $\alpha_{\text{tot}}$ ranges from 0.15 to 1.96.
- Temperature ratio at percolation $\xi_p$ ranges from 0.53 to 0.71.
- The transition occurs at hidden sector temperatures $T_{h,p} \sim 0.6 - 2.0$ GeV, well above the BBN limit ( $\sim 1$ MeV).
Future Detectability: The predicted spectra fall within the sensitivity ranges of future space-based detectors (LISA, Taiji, TianQin, BBO, DECIGO), offering a pathway for cross-verification.

5. Significance

This work provides a robust theoretical framework linking the recent PTA detections to Beyond Standard Model (BSM) physics involving a hidden sector. It highlights that:

Supercooled phase transitions are a viable and necessary mechanism to explain nano-Hz GWs without violating BBN constraints.
Precise thermal modeling (tracking the separate evolution of hidden and visible sectors) is non-negotiable for accurate GW predictions.
Re-evaluation of constraints: The reliance on the mean bubble separation $R_*$ rather than the simplified $\beta/H$ parameter opens up a vast, previously excluded region of parameter space for early universe cosmology.

The paper concludes that supercooled phase transitions in hidden sectors, properly accounting for thermal history and cannibalism, offer a compelling explanation for the PTA signals and predict observable signatures for next-generation gravitational wave observatories.