⚛️ phenomenology

Early vs late string networks from a minimal QCD Axion

This paper proposes a novel cosmological regime for minimal QCD axion dark matter where the Peccei-Quinn symmetry is restored by inflationary fluctuations but not thermally afterwards, leading to late-forming string networks that allow for axion masses up to $10^{-2}$ eV and the formation of massive miniclusters.

Original authors: Marco Gorghetto, Edward Hardy, Horia Nicolaescu, Alessio Notari, Michele Redi

Published 2026-07-21

📖 7 min read🧠 Deep dive

CC BY 4.0

Original authors: Marco Gorghetto, Edward Hardy, Horia Nicolaescu, Alessio Notari, Michele Redi

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. In the very first split second of its existence, this balloon inflated faster than the speed of light in a phase called "inflation." During this wild expansion, the universe was filled with a mysterious, invisible field that we call the "axion." Think of this field like a giant, cosmic compass needle. In some theories, this needle points in a specific direction everywhere at once, while in others, it gets jumbled up by the violent stretching of space.

Why do we care about this invisible needle? Because it might be the key to solving one of the biggest mysteries in physics: dark matter. Dark matter is the invisible "glue" that holds galaxies together; we can't see it, but we know it's there because of how it pulls on stars. For decades, scientists have been trying to figure out exactly what kind of "axion" makes up this dark matter. They usually imagine two main scenarios: either the axion needle was set in place before the universe expanded (the "pre-inflationary" story), or it got scrambled and reset after the expansion stopped (the "post-inflationary" story). These two stories lead to very different predictions about how heavy the axion is and how it clumps together.

Now, a team of physicists has discovered a third, surprising possibility that sits right in the middle of these two stories. They suggest that the axion needle might have been scrambled only on very large scales, leaving it somewhat orderly on smaller scales. This leads to a "late" version of the post-inflationary story, where the cosmic strings (knots in the field) that form the dark matter don't show up until after the QCD crossover (when the universe cools to the GeV scale). This new regime suggests that the axion could be much heavier and form much larger, denser clumps than previously thought, potentially changing how we hunt for dark matter today.

The Cosmic Knots That Arrived Late

To understand what this paper is about, let's picture the universe as a giant, stretchy rubber sheet. In the "post-inflationary" story, scientists thought that after the universe stopped inflating, this sheet would instantly snap into a new shape, creating a messy network of knots called "cosmic strings" and "domain walls." These knots are like the seams where different parts of the sheet meet. Usually, scientists thought these knots would form almost immediately after inflation ended, creating a chaotic web that would eventually settle down and produce the dark matter we see today.

But this paper suggests a different timeline. The authors, using a mix of math and computer simulations, propose that sometimes these knots don't form right away. Instead, the universe stays smooth for a while, and the knots only appear later, specifically after the universe has cooled down to the GeV scale (around the QCD crossover). They call this the "late string" regime.

Here is how it happens: Imagine the axion field is a ball rolling on a hilly landscape. During the rapid inflation of the universe, the ball gets kicked around randomly. In the old stories, scientists assumed the ball would either stay in a valley (broken symmetry) or get kicked so hard it lands in a new valley (restored symmetry) immediately. This paper finds a middle ground. The ball gets kicked just enough that, on a tiny scale, it looks like it's in a valley, but if you zoom out to a huge scale, it looks like it's been kicked all over the place.

When the universe stops inflating, the ball tries to settle down. But because of the way it was kicked, it sometimes rolls over the top of a hill and lands on the opposite side. This is called "overshooting." When the ball lands on the opposite side in some places but not others, it creates a mismatch. These mismatches are the cosmic knots (strings). The paper shows that if the universe is just the right size and the hills are just the right shape, these knots don't form until the universe has cooled to the GeV scale, rather than forming immediately after inflation.

The Simulation: Watching the Knots Form

To prove this, the authors ran computer simulations. They built a digital universe and watched how the axion field behaved. They found that in many cases, the field didn't just settle down smoothly. Instead, it oscillated, rolling back and forth over the top of its potential hill. This "overshooting" acted like a randomizer, scrambling the field's direction in a way that created knots.

The simulations showed that for a wide range of conditions, these knots form a network that is initially very sparse—like a few lonely threads in a vast ocean. This network then slowly grows as more of the universe's history "re-enters" the observable horizon. Eventually, it reaches a state where there is about one knot for every "Hubble patch" (a region of space roughly the size of our observable universe). The authors estimate that this process can happen anywhere from 25 to 50 "e-folds" (a measure of expansion) after the start of inflation.

Crucially, the paper suggests that this "late" formation changes everything about the dark matter. In the standard story, the knots form early, and the dark matter is relatively light and spread out. In this new "late string" story, the knots form so late that they are attached to "domain walls" (like sheets of fabric) that only break apart when the universe is already at the GeV scale. When these walls finally snap, they release a burst of axions.

The Heavyweight Champion and the Giant Clumps

This late arrival has a massive consequence: it allows the axion to be much heavier than we thought. In the standard models, if the axion is too heavy, it would create too much dark matter and crash the universe. But because these "late" knots release their energy later, the math works out differently. The paper suggests that in this specific regime, the axion could have a mass up to $10^{-2}$ eV (electron-volts). This is much heavier than the ultra-light axions predicted by older theories.

Even cooler, this scenario predicts that the dark matter won't be spread out evenly. Instead, it will clump together into massive, dense balls called "miniclusters." In the standard story, these clumps are tiny, like small asteroids. But in this "late string" world, the clumps can be as massive as $O(10)M_{\odot}$ (about 10 times the mass of our Sun). Imagine dark matter not as a fog, but as giant, invisible planets floating in space.

What This Means for Us

The authors are careful to note that this is a new regime they have identified, not a proven fact yet. They used simulations to show that the "overshooting" mechanism works and that it leads to these late-forming knots. They also point out that there are still big questions, especially regarding "isocurvature perturbations." These are ripples in the dark matter that could mess up the patterns we see in the cosmic microwave background (the afterglow of the Big Bang). The paper suggests that in this late regime, these ripples might be suppressed, making the theory viable, but more work is needed to be sure.

The bottom line is that this paper opens a new door. It suggests that the axion, our best candidate for dark matter, might be heavier, clumpier, and formed later than we ever imagined. If this is true, it means that experiments looking for dark matter might need to tune their dials to look for these heavier, "late-blooming" axions. It turns the search for the universe's missing mass into a hunt for giant, invisible clumps that formed just in time to shape the cosmos we see today.

Technical Summary: Early vs Late String Networks from a Minimal QCD Axion

Problem Statement
The cosmological dynamics of the QCD axion are traditionally categorized into two regimes: the "pre-inflationary" scenario, where the Peccei-Quinn (PQ) symmetry breaks before inflation and remains broken, and the "post-inflationary" scenario, where the symmetry is restored after inflation (typically by thermal effects) and breaks subsequently. These regimes yield distinct observational signatures, particularly regarding topological defects (strings and domain walls) and the resulting dark matter (DM) abundance and spatial distribution (miniclusters).

This paper investigates a "minimal" QCD axion scenario where the PQ symmetry is not restored by thermal effects ( $T_{max} < f_a$ ) but may be effectively restored on average over large spatial scales due to inflationary fluctuations ( $H_I \gtrsim f_a$ ). The authors identify a novel intermediate regime where the PQ symmetry is restored only on scales larger than the Hubble radius at the end of inflation. This leads to the formation of an initially underdense string network that re-enters the horizon significantly later than in standard post-inflationary scenarios, potentially after the QCD crossover.

Methodology
The authors employ a combination of analytic stochastic formalism, numerical simulations, and cosmological modeling:

Stochastic Formalism during Inflation:
- The evolution of the complex PQ scalar field $\Phi$ during inflation is modeled using the Fokker-Planck (FP) equation.
- The authors solve for the probability distribution $P(\Phi)$ to determine the number of e-folds ( $N_{11}$ ) required for the field to reach an equilibrium distribution where the radial mode $\rho$ is displaced from the vacuum or randomized.
- They define a critical fraction $F(N)$ representing the portion of the field that has spread to the opposite side of the potential (either via inflationary fluctuations or post-inflationary overshoot). They posit that a string network reaches the scaling solution ( $\xi \sim 1$ ) when $F(N)$ reaches a critical threshold (approximately $0.35$).
Post-Inflationary Dynamics and Overshoot:
- The paper analyzes the relaxation of the radial mode $\rho$ immediately after inflation.
- A key mechanism identified is "overshoot": if the radial mode is displaced to values $\rho \gg f_a$ during inflation, it may oscillate over the top of its potential as it relaxes to the minimum. This oscillation can randomize the angular phase $\theta$ even if the field was initially homogeneous, leading to string formation in parameter space where inflationary fluctuations alone would not suffice.
Numerical Simulations:
- The authors perform lattice simulations of the PQ field evolution after inflation, starting from initial conditions generated by the stochastic inflationary process.
- They utilize a "fat-string" trick (time-dependent quartic coupling $\lambda$ ) to resolve string cores on manageable lattice sizes ( $2048^3$ ).
- Simulations track the string density $\xi$ as a function of the fraction $F$ and the number of e-folds re-entering the horizon, validating the analytic prediction that $\xi$ is determined by $F$ .
Phenomenological Mapping:
- The authors map the parameter space of the Hubble scale during inflation ( $H_I$ ) and the quartic coupling ( $\lambda$ ) to the temperature ( $T_{PQ}$ ) at which the string network re-enters the horizon.
- They calculate the resulting axion relic abundance from both misalignment and the annihilation of the late string-wall network.
- Isocurvature perturbations are estimated by analyzing super-horizon correlations of the axion field, considering the potential suppression effects of overshoot and string dynamics.

Key Contributions and Results

Identification of the "Late String" Regime: The paper establishes a regime where $H_I \gtrsim f_a$ but $T_{max} < f_a$ . In this regime, the PQ symmetry is restored on average only on scales larger than the observable universe at the end of inflation. Consequently, the resulting string network is initially underdense and re-enters the horizon late ( $T_{PQ} \lesssim 1$ GeV), potentially after the QCD crossover.
The Overshoot Mechanism: The authors demonstrate that even when inflationary fluctuations are insufficient to randomize the field (i.e., the field does not reach equilibrium during inflation), the post-inflationary oscillation of the radial mode over the potential barrier ("overshoot") can efficiently randomize the phase and generate a string network. This extends the viable parameter space for post-inflationary axion scenarios to smaller values of $\lambda$ and specific ranges of $H_I/f_a$ .
Dark Matter Abundance and Miniclusters:
- In the late string regime, the axion dark matter abundance is dominantly produced by the annihilation of the string-wall network after the QCD crossover, rather than by the scaling solution of early-forming strings.
- This mechanism allows the observed dark matter abundance to be reproduced for axion decay constants as low as $f_a \sim 10^8$ GeV (the astrophysical bound), corresponding to axion masses up to $m_a \sim 10^{-2}$ eV.
- The collapse of the late string-wall network produces axion miniclusters with masses significantly larger than in standard post-inflationary scenarios, potentially reaching $O(10) M_\odot$ .
Isocurvature Constraints: The paper analyzes isocurvature perturbations. While the late decay of domain walls produces a white-noise isocurvature spectrum (constrained by Lyman- $\alpha$ and CMB), the misalignment component inherits super-horizon correlations. The authors suggest that the overshoot mechanism and string dynamics may suppress these correlations, potentially relaxing isocurvature bounds, though this requires further verification.
Simulation Validation: Numerical simulations confirm that the string density $\xi$ in an underdense network is approximately determined by the fraction $F$ of the field that has crossed the potential, regardless of whether the crossing is due to inflationary fluctuations or overshoot.

Significance
The paper claims to reveal a new, robust regime of minimal QCD axion cosmology that bridges the gap between pre- and post-inflationary scenarios. Its primary significance lies in:

Lowering the Axion Mass Scale: It demonstrates that the QCD axion can constitute the entirety of dark matter at decay constants near the astrophysical lower bound ( $f_a \sim 10^8$ GeV), making it accessible to direct detection experiments targeting higher masses (e.g., IAXO).
Novel Phenomenology: It predicts distinct observational signatures, including massive miniclusters and a specific gravitational wave spectrum from late string-wall annihilation, differentiating it from standard thermal post-inflationary scenarios.
Theoretical Completeness: It challenges the binary classification of axion cosmology, showing that the interplay between inflationary fluctuations and post-inflationary radial mode dynamics creates a rich landscape of possibilities, particularly for minimal models without additional symmetry-breaking biases.

The authors remain modest regarding quantitative precision, noting that the exact boundaries of the parameter space and the efficiency of correlation suppression depend on details (such as the initial field value prior to visible inflation and the precise dynamics of the string-wall network) that require further numerical investigation.

The Cosmic Knots That Arrived Late

The Simulation: Watching the Knots Form

The Heavyweight Champion and the Giant Clumps

What This Means for Us

More like this