Dark Matter Production from Bubble Collisions during a First-Order Phase Transition at the End of Inflation

This paper demonstrates that a first-order phase transition at the end of inflation, occurring in a spectator scalar sector dominated by Coleman-De Luccia tunneling, can successfully generate the observed dark matter abundance through a combination of direct particle production from bubble collisions and the subsequent decay of spectator particles.

The Gist

The Big Picture: A Cosmic "Pop" at the End of the Universe's Stretch

Imagine the very early universe as a giant balloon being blown up incredibly fast. This rapid stretching is called inflation. Usually, scientists think that the "stuff" that makes up Dark Matter (the invisible glue holding galaxies together) was created after the balloon stopped blowing up, when things cooled down.

This paper asks a different question: What if Dark Matter was created while the balloon was still being blown up, right at the very last second?

The authors propose a scenario where the universe undergoes a sudden "phase transition" (like water suddenly turning into ice) just as inflation is ending. This transition happens through bubble collisions.

The Story in Three Acts

Act 1: The Sleeping Giant (The Spectator Field)

Imagine the universe is filled with a calm, invisible field (let's call it the "Spectator"). For most of the inflation period, this field is happy and stable. It's like a ball sitting quietly at the bottom of a deep valley.

However, the field is connected to the "inflaton" (the engine driving the balloon's expansion). As the balloon stretches, the shape of the valley changes. The bottom of the valley slowly rises, turning into a hill. The ball is now perched precariously on a hilltop, ready to roll down, but it's stuck there for a long time.

Act 2: The Bubble Explosion (The Phase Transition)

Eventually, near the very end of inflation, the hill becomes unstable. The ball can't stay put anymore. It doesn't roll down smoothly; instead, it "tunnels" through the barrier and creates a bubble of the new, lower-energy state.

Think of this like popping a bubble in a pot of boiling water, but instead of water, it's the fabric of space itself.

• The Runaway: Because there is no "friction" (no hot gas or plasma) in the empty vacuum of inflation, the walls of these bubbles don't slow down. They accelerate to nearly the speed of light.
• The Crash: These super-fast bubbles expand until they crash into each other. Imagine two cars driving at light speed slamming into each other. The energy from that crash is massive.

Act 3: The Dark Matter Factory

When these bubbles collide, the energy stored in their walls is released. This energy acts like a giant particle accelerator.

1. Direct Creation: The crash directly smashes out particles of Dark Matter.
2. Indirect Creation: The crash also creates a lot of "spectator" particles (the field that was rolling down the hill). These particles are unstable and quickly decay (break apart) into more Dark Matter.

The authors calculated that if the timing is just right, this process could create exactly the amount of Dark Matter we see in the universe today.

The Tricky Parts (Why it's hard to do)

The paper highlights three major hurdles that make this scenario very specific and difficult to pull off:

1. The "Goldilocks" Timing

• Too Early: If the bubbles form too early in inflation, they grow so big they would tear the universe apart or create huge, uneven patches that we don't see today.
• Too Late: If they form too late, the universe expands so fast that the bubbles get separated before they can crash into each other. No crash means no Dark Matter.
• Just Right: The transition must happen in a tiny window right at the end of inflation, where the bubbles form, crash, and finish the job before the universe stretches them apart.

2. The Tunneling Rules (Gravity's Quirks)
In normal physics, a ball rolling over a hill is easy to calculate. But in the expanding universe (de Sitter space), gravity changes the rules.

• Sometimes, instead of a bubble forming (a localized event), the entire universe might just fluctuate over the hill at once. This is called the "Hawking-Moss" transition.
• The authors had to prove that in their scenario, the "bubble" way of tunneling is the one that actually happens, not the "whole universe" way. If the whole universe jumps the hill, there are no bubbles to crash, and no Dark Matter is made.

3. The "Clean" Collision
For the math to work, the bubbles must crash in a very specific way.

• If they crash too softly, they just bounce off.
• If they crash too violently, they might create too much heat or mess up the inflation.
• The authors found a "sweet spot" where the collisions are violent enough to create particles but controlled enough to leave the universe intact.

The Result: A Hidden Signal

The paper concludes that while this mechanism can work, it only works in a very narrow, specific set of conditions (a "restricted region of parameter space").

What about Gravitational Waves?
When bubbles crash, they should create ripples in space-time called gravitational waves. The authors calculated what these ripples would look like today.

• The Bad News: Because the universe expanded so much after the crash, these ripples have been stretched out and weakened.
• The Conclusion: The signal is likely too faint for our current or even future planned detectors (like LISA or TianQin) to hear. It's like trying to hear a whisper from across the galaxy after the wind has blown for billions of years.

Summary Analogy

Imagine a giant, silent balloon being inflated. Just before you stop blowing, a tiny, hidden mechanism inside the balloon triggers a chain reaction.

• Tiny bubbles form inside the rubber.
• They zoom around at light speed and slam into each other.
• The sound of the crash (energy) creates a new type of invisible dust (Dark Matter) that fills the balloon.
• The authors figured out the exact recipe for the rubber and the air pressure so that this happens once, creating just the right amount of dust, without popping the balloon.

However, because the balloon kept expanding for so long after the crash, the "sound" of that event is now too quiet for us to hear, even with the best microscopes we have.

Technical Summary

Technical Summary: Dark Matter Production from Bubble Collisions during a First-Order Phase Transition at the End of Inflation

Problem Statement
The particle nature and cosmological origin of dark matter (DM) remain open questions. While thermal freeze-out and freeze-in are standard benchmarks, the lack of positive signals in direct detection and collider experiments has motivated interest in non-thermal production mechanisms. Specifically, cosmological first-order phase transitions (FOPTs) offer rich settings for DM genesis. However, most existing studies focus on FOPTs occurring after reheating. This paper investigates a distinct scenario: a FOPT occurring during inflation, specifically at its end, within a spectator scalar sector. The central challenge is to determine if such a transition can generate the observed DM abundance via bubble collisions without disrupting the inflationary background or failing to complete due to the rapid expansion of the universe.

Methodology
The authors construct a model involving an inflaton field $\phi$ driving inflation and a real scalar spectator field $\sigma$ undergoing a FOPT. The spectator potential depends on the inflaton background, $V(\phi, \sigma)$, allowing the tunneling rate to evolve during inflation. Dark matter is modeled as a fermion $\psi$ coupled to the spectator field via a Yukawa interaction.

The analysis proceeds through three main stages:

1. Vacuum Decay in de Sitter Space: Unlike flat-space approximations, the authors analyze vacuum decay directly in a de Sitter background. They distinguish between the Coleman–De Luccia (CDL) bounce (localized bubble nucleation) and the Hawking–Moss (HM) instanton (homogeneous transition). They impose strict criteria to ensure the CDL channel dominates and that the solution corresponds to a physically meaningful localized bubble (small type-A bounces), avoiding regimes where the CDL solution disappears or where HM transitions dominate.
2. Cosmological Viability Constraints: The nucleation rate $\Gamma$ must be sufficiently suppressed during the early stages of inflation to avoid "big bubbles" that would violate observational constraints (e.g., CMB isotropy, $\mu$-distortions, Big Bang Nucleosynthesis). Conversely, near the end of inflation, the rate must rise rapidly to ensure percolation and efficient bubble collisions within a Hubble time. This requires a specific time dependence of the tunneling exponent and a dimensionless parameter $\beta/H \gtrsim 10$.
3. Particle Production and Evolution: The authors compute DM production from two sources:
   • Direct Production: Energy released during bubble collisions excites fields coupled to the spectator sector.
   • Indirect Production: Spectator field particles ($h$) produced in collisions subsequently decay into DM ($h \to \psi\bar{\psi}$).
     The post-collision evolution is modeled using Boltzmann equations. The authors assume a regime where elastic self-scatterings ($hh \to hh$) efficiently redistribute momenta, while number-changing processes and competing sink channels (involving inflaton fluctuations) remain inefficient compared to the Hubble expansion. The final relic abundance is calculated by tracking the dilution of produced particles from the end of inflation to the present day, assuming instantaneous reheating.

Key Contributions and Results

• Parameter Space Identification: The study identifies a restricted region of parameter space where the phase transition is cosmologically viable. This region is defined by the simultaneous satisfaction of:
  • The existence of a dominant CDL bounce (sub-dominant HM transition).
  • Suppression of early-time nucleation to avoid large-scale inhomogeneities.
  • Rapid completion of the transition near the end of inflation ($\beta/H \gtrsim 10$).
  • Sub-dominant backreaction on the inflationary background.
• DM Abundance Mechanism: In the viable regime, the observed DM abundance can be accommodated. The mechanism relies on the spectator field particles ($h$) produced in bubble collisions, which then decay into DM. The authors find that the final abundance receives contributions from both direct production in collisions and the subsequent decay of the spectator population.
• Mass Flexibility: The analysis demonstrates that this mechanism can produce viable DM candidates across a broad range of masses (spanning several orders of magnitude), rather than being confined to a narrow mass window. This flexibility arises from the variation in inflationary Hubble scales and phase transition parameters.
• Gravitational Wave (GW) Signal: The authors estimate the stochastic GW background generated by the bubble collisions. They find that, due to the requirement that the spectator sector remains sub-dominant (to avoid disrupting inflation) and the subsequent redshift and spectral deformation caused by the remaining inflationary expansion, the predicted GW amplitude is significantly suppressed. Consequently, the signal in their benchmark points remains below the projected sensitivities of future GW experiments (e.g., TianQin, DECIGO, LISA).

Significance and Claims
The paper claims to provide a unified analysis of DM production from an inflationary FOPT, explicitly addressing the complexities of vacuum decay in de Sitter space which are often overlooked in flat-space approximations. The authors emphasize that a viable transition in this context is not a generic consequence of a metastable potential but requires a carefully selected parameter region where the tunneling history is highly time-dependent (negligible early, rapid late).

The authors maintain a modest tone regarding their results. They state that their treatment of post-collision particle phase-space distributions is approximate, and thus the relic density estimates are accurate only to an order-of-magnitude level. They acknowledge that their numerical scan covers only a limited region of the full parameter space and that the conclusion regarding the undetectability of the GW signal applies specifically to the class of scenarios explored, not necessarily the entire model space. The work serves as a proof-of-concept that inflationary FOPTs can generate DM, provided strict consistency conditions regarding the tunneling mechanism and cosmological evolution are met.