⚛️ phenomenology

Cosmological Dynamics of Multi-Axion Quintessence

This paper investigates the cosmological dynamics of a two-axion quintessence model, demonstrating that multiple axions—particularly with cross-interactions—can generate novel dark energy equation of state behaviors and relax the requirement for super-Planckian decay constants, thereby offering alternative explanations for DESI observations that deviate from standard single-field thawing quintessence.

Original authors: Supakorn Katewongveerachart, David J. E. Marsh

Published 2026-02-03

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Supakorn Katewongveerachart, David J. E. Marsh

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 Big Picture: What is the Universe Doing?

Imagine the universe is a giant balloon. For a long time, scientists thought this balloon was being inflated by a constant, unchanging force (like a steady hand pumping air). This force is called the Cosmological Constant.

However, new data from a telescope project called DESI suggests the balloon isn't just inflating at a steady rate; the speed of inflation might be changing. This implies the "dark energy" pushing the universe apart isn't a static constant, but something that moves and evolves over time.

The Main Character: The "Axion"

To explain this changing force, scientists look at a theoretical particle called an axion.

The Analogy: Think of an axion as a ball sitting on a wavy, rolling hill.
The Physics: If the ball rolls very slowly down the hill, it creates a pressure that pushes the universe apart (accelerated expansion).
The Problem: For a single ball (a single axion) to do this job, the hill has to be incredibly flat and huge. In physics terms, the "decay constant" (the size of the hill) needs to be almost as big as the Planck scale (the fundamental limit of size in the universe). This is hard to justify with current string theory models.

The New Idea: Two Balls Instead of One

The authors of this paper ask: What if we don't just have one ball, but two?
In the "string axiverse" (a theory from string theory), there aren't just one or two axions; there could be many. The paper investigates a simple "toy model" with two axions to see if they can work together to explain the universe's expansion.

They looked at two scenarios:

Non-interacting: Two balls rolling on their own separate hills.
Interacting: Two balls tied together or rolling on a complex, shared hill where they bump into each other.

What They Found (The Results)

1. The "Two-Ball" Advantage

When the two axions don't interact, they can actually do a better job than a single axion.

The Analogy: Imagine trying to push a heavy car. One person pushing might need superhuman strength (a huge decay constant). But if two people push together, they can get the car moving even if neither of them is quite as strong.
The Result: The two-ball model allows for "smaller" hills (lower decay constants) while still driving the universe's expansion. This makes the theory more flexible and potentially more realistic.

2. The "Tied-Together" Surprise

When the two axions do interact (they are linked), things get weird and wild.

The Analogy: Imagine two dancers tied together by a rope. If one spins, the other has to spin too, but their movements can create complex, swirling patterns that neither could do alone.
The Result: This interaction creates strange behaviors that a single axion (or two separate ones) cannot do.
- Oscillating but Negative: Usually, if a ball rolls back and forth (oscillates) on a hill, it acts like normal matter (dust), not dark energy. But in this two-ball model, the balls can oscillate rapidly, yet the average effect still acts like dark energy, pushing the universe apart.
- Changing Rules: The "equation of state" (a number that describes how the dark energy behaves) can change in ways that are impossible for a single axion. For example, the rate of change can flip signs or become extremely large.

3. The DESI Connection

The DESI data currently favors a specific type of behavior called "thawing quintessence" (where the dark energy starts frozen and slowly starts moving).

The Twist: The single-axion model fits this "thawing" behavior perfectly.
The Complication: The interacting two-axion model produces so many weird, exotic behaviors that it actually moves away from the specific pattern DESI likes. While the two-ball model is theoretically interesting, the "tied-together" version makes it harder to match the current telescope data compared to the simple single-ball model.

The Bottom Line

The paper explores whether having two axion fields instead of one helps explain the universe's acceleration.

Good news: Two non-interacting axions can work together to allow for more realistic physical parameters.
Mixed news: Two interacting axions create fascinating, complex dynamics (like oscillating fields that still act like dark energy), but these complex behaviors often stray from the specific patterns currently observed by the DESI telescope.

In short, adding a second axion opens up a whole new playground of physics, but it doesn't necessarily make it easier to match the current "snapshot" of our universe we see today. The authors suggest that looking at even more axions (N >> 1) might be the next step to fully understand this cosmic mystery.

Technical Summary: Cosmological Dynamics of Multi-Axion Quintessence

Problem Statement
Recent results from the Dark Energy Spectroscopic Instrument (DESI) suggest a departure from the standard cosmological constant ( $\Lambda$ ) model, favoring a dark energy (DE) equation of state with $w_0 > -1$ and $w_a < 0$ (the "thawing quintessence" regime) with $2\text{--}3\sigma$ significance. While single-axion models naturally align with this thawing behavior, they face theoretical challenges: achieving the necessary effective decay constant ( $f_a \sim M_{\text{Pl}}$ ) to drive current acceleration is difficult within string theory landscapes (specifically Type IIB Calabi-Yau compactifications) without violating the Weak Gravity Conjecture. Furthermore, single-field models struggle to explain the DESI preference without exacerbating the $H_0$ tension. The authors investigate whether a multi-axion framework, specifically a two-axion model, can resolve these issues by engineering larger effective decay constants or generating novel cosmological dynamics that differ from single-field predictions.

Methodology
The authors employ a numerical study of a two-axion quintessence model within a Friedmann-Robertson-Walker (FRW) background. The theoretical framework includes:

Potential Forms: They analyze both non-interacting axions (sum of independent cosine potentials) and interacting axions where the potential arises from a joint instanton-generated term, $V(\phi_1, \phi_2) = \sum \Lambda_i^4 [1 - \cos(\sum Q_{ij}\phi_j/f_j)]$ . The kinetic terms are diagonalized to canonical form, resulting in effective decay constants and cross-coupling terms.
Numerical Evolution: The equations of motion are solved from the matter-radiation equality epoch ( $z_{\text{eq}} \approx 3400$ ) to the present day ( $a=1$ ). The system is evolved using dimensionless variables scaled by an effective mass $M$ and the Planck mass $M_{\text{Pl}}$ .
Parameter Sampling: A Monte Carlo approach is used to sample the parameter space with log-uniform priors for masses ( $m_a \in [10^{-33}, 10^{-32}]$ eV) and decay constants ( $f_a/M_{\text{Pl}}$ ), and uniform priors for initial misalignment angles.
Selection Criteria: The generated samples are filtered to retain only those models yielding a present-day Hubble parameter $0.6 < h < 0.8$ and a quintessence energy density fraction $0.6 < \Omega_\phi < 0.8$ , ensuring consistency with current cosmological inferences.

Key Contributions and Results
The study computes the prior probability distribution for the DE equation of state parameters $(w_0, w_a)$ conditioned on the fundamental model parameters. Key findings include:

Decay Constant Requirements: In the non-interacting two-axion case, the model allows for smaller individual decay constants compared to the single-axion case, which strongly prefers $f_a \sim M_{\text{Pl}}$ . However, in the interacting case, at least one decay constant must still be large (near $M_{\text{Pl}}$ ), suggesting that interactions do not easily alleviate the need for super-Planckian scales in the same way non-interacting "N-flation" scenarios might.
Equation of State Diversity: The two-field models explore regions of the $(w_0, w_a)$ $(w_{0}, w_{a})$ plane significantly distinct from the single-field "thawing quintessence" region.
- Sign of $w_a$ : Unlike single-field models which typically predict $w_a < 0$ , the two-field models can produce $w_a > 0$ .
- Magnitude of $w_a$ : Interacting models exhibit a much wider range of $|w_a|$ values, including extreme variations not seen in single-field scenarios.
Novel Dynamics: The authors identify specific dynamical regimes where both fields are in an oscillating phase yet the time-averaged equation of state remains negative ( $w < -1/3$ $w < - 1/3$ ), allowing them to drive accelerated expansion.
- In some interacting cases, the fields oscillate rapidly, but the equation of state does not average to zero (matter-like); instead, it undergoes secular evolution, sometimes becoming more negative over time.
- The interaction induces a "rotating eigenbasis" effect, where the trajectory in field space causes complex behaviors, such as the equation of state drifting toward more negative values before potentially reversing.

Significance and Claims
The paper claims that while multi-axion models introduce rich dynamical possibilities, they do not necessarily provide a "better" fit to the specific DESI thawing preference than the single-field model; in fact, the broadened parameter space of interacting models makes them statistically less likely to align with the specific DESI inference compared to the constrained single-field prior.

The primary significance lies in demonstrating that:

Interactions alter priors: Cross-interactions between axions significantly change the prior probability distribution of $(w_0, w_a)$ , moving support away from the thawing region preferred by current data.
Exotic behavior is possible: Multi-axion systems can sustain accelerated expansion even when fields are oscillating, a behavior impossible in single-field harmonic potentials.
Relevance to broader cosmology: These novel equations of state and dynamics may be relevant to other cosmological epochs, such as early dark energy or inflation/reheating scenarios involving multiple axions.

The authors conclude that further exploration of models with $N \gg 1$ axions is warranted to fully understand the cosmological implications of the "string axiverse."