On the potential of pseudo-scalar dark energy

This paper presents an extended analysis of pseudo-scalar dark energy models, demonstrating that while axion-like potentials require large anomaly coefficients to be viable, scenarios with quadratic, linear, or Ratra-Peebles potentials can successfully explain both dark energy and cosmic birefringence with a symmetry-breaking scale near the GUT scale.

The Gist

The Big Mystery: What is Dark Energy?

Imagine the universe is a giant balloon. For a long time, scientists thought this balloon was inflating at a steady, unchanging speed, driven by a mysterious force called "Dark Energy." The standard theory says this force is like a fixed weight glued to the balloon that never changes.

However, recent measurements (like those from the DESI telescope) suggest the balloon might be inflating a little differently than expected. It's not just a steady push; it might be a force that changes over time. This paper asks: What if Dark Energy isn't a fixed weight, but a rolling ball?

The Main Character: The "Ghost" Ball (Pseudo-Scalar Field)

The authors propose that Dark Energy is actually a "pseudo-scalar field." Think of this as a giant, invisible ball rolling across a hilly landscape that stretches across the entire universe.

• The Landscape: This is the "potential" (the shape of the hills and valleys).
• The Ball: This is the field itself.
• The Motion: As the ball rolls down the hill, it changes the way the universe expands.

But here is the special twist: This isn't just any ball. It's a "pseudo-scalar" ball, which means it has a special superpower: it can twist light.

The Superpower: Cosmic Birefringence (Twisting the Light)

Imagine you are looking at a distant lighthouse through a pair of polarized sunglasses. Usually, the light waves wiggle in a specific pattern.

• The Effect: If this "ghost ball" exists, as light from the early universe travels past it, the ball acts like a cosmic corkscrew. It twists the direction the light waves wiggle.
• The Name: Scientists call this Cosmic Birefringence (CB).
• The Clue: Recent data from the Planck satellite and the Atacama Cosmology Telescope suggests this twist is happening. The light from the early universe is indeed rotated by about 0.2 degrees.

The paper's goal is to see if a rolling ball model can explain both the changing expansion of the universe and this twisting of light.

The Experiments: Testing Different Landscapes

The authors tested five different shapes for the "landscape" the ball rolls on. They used a super-computer (modified version of the CLASS code) to simulate how the universe would look if the ball rolled on each of these shapes, and then compared the results to real telescope data.

Here are the five landscapes they tested:

1. The Axion Hill (The Wobbly Hill):

   • The Shape: A smooth, wavy hill (like a sine wave).
   • The Result: This works, but only if the ball has a very specific, strong "twisting power" (a high anomaly coefficient). It's like saying the ball only twists light if it's made of a very rare, special material. If the material is ordinary, this model fails to explain the light twist.

2. The Linear Slope (The Steady Ramp):

   • The Shape: A straight ramp going up or down.
   • The Result: This works well. The ball rolls down a straight slope, and it explains both the expansion and the light twist naturally. Interestingly, for the "uphill then downhill" version, the ball needs a little "kick" to start moving up the hill first.

3. The Quadratic Bowl (The Parabolic Valley):

   • The Shape: A classic U-shaped bowl.
   • The Result: This also works very well. The ball rolls down the sides of the bowl. It fits the data perfectly without needing any weird adjustments.

4. The Ratra-Peebles Hill (The Gentle Slope):

   • The Shape: A hill that gets flatter and flatter as you go down.
   • The Result: This is another strong contender. It behaves similarly to the linear slope and fits the observations nicely.

The "Kick" Factor

In some scenarios, the ball didn't just start rolling; it got a kick.

• Imagine the ball was sitting still for billions of years. Then, at a specific time (when the universe was young, but not too young), someone gave it a shove.
• This "kick" helps the ball start moving in a way that matches the data. The paper found that for some models, this kick is necessary to get the timing right for the light twist.

The Verdict: Is the Rolling Ball Real?

The authors ran a massive statistical analysis (using a method called MCMC, which is like running millions of simulations to find the best fit).

• The Score: When they compared their "Rolling Ball" models against the standard "Fixed Weight" model (Lambda-CDM), the Rolling Ball models won.
• The Confidence:
  • Looking only at how the universe expands, the Rolling Ball is about 3 times more likely (3-sigma) to be correct than the Fixed Weight.
  • When they added the "Light Twist" (Cosmic Birefringence) data, the confidence jumped to 4 times more likely (4-sigma).

The Conclusion

The paper concludes that Dynamical Dark Energy (a rolling ball) is a very strong candidate for what Dark Energy actually is.

• The Axion-like model works, but it requires a "special ingredient" (a high anomaly coefficient) to make the light twist happen.
• The Linear, Quadratic, and Ratra-Peebles models work beautifully with standard ingredients. They suggest the "energy scale" where this physics happens is close to the GUT scale (Grand Unified Theory), a massive energy level where the fundamental forces of nature might merge.

In short: The universe might not be powered by a static, unchanging force. Instead, it might be driven by an invisible field rolling down a cosmic hill, twisting the light of the early universe as it goes. The data we have right now strongly supports this rolling scenario over the old, static one.

Technical Summary

Technical Summary: On the potential of pseudo-scalar dark energy

Problem and Motivation
Recent cosmological surveys, including results from the DESI Collaboration, suggest potential deviations from the concordance $\Lambda$CDM model, specifically hinting at a dynamical dark energy (DE) equation of state ($\omega > -1$) at low redshifts. Concurrently, analyses of Cosmic Microwave Background (CMB) polarization data from Planck and the Atacama Cosmology Telescope (ACT) indicate a non-zero rotation of the polarization plane, a phenomenon known as cosmic birefringence (CB). While scalar fields are standard candidates for dynamical DE, pseudo-scalar fields (axion-like particles, or ALPs) offer a unique dual capability: they can drive DE evolution and, through their coupling to electromagnetism, induce the observed CB. This work investigates whether specific pseudo-scalar DE models can simultaneously explain the background expansion history and the CB signal.

Methodology
The authors perform a comprehensive statistical analysis of four distinct pseudo-scalar DE potentials:

1. Axion-like potential: $V(\phi) = m^2f^2[1 - \cos(\phi/f)]$
2. Linear potential: $V(\phi) = \mu^4 \phi/f$ (both positive and negative slopes)
3. Quadratic potential: $V(\phi) = \frac{1}{2}m^2\phi^2$
4. Ratra-Peebles potential: $V(\phi) = \mu^4 f / \phi$

The study utilizes a modified version of the CLASS code to solve the background evolution equations for the scalar field $\phi$ and the perturbed Klein-Gordon equation for density fluctuations. The theoretical predictions for the CMB angular power spectra, distance measures (luminosity, angular diameter, and volume-averaged distances), and the CB angle ($\beta$) are computed.

The statistical analysis employs a Monte Carlo Markov Chain (MCMC) sampler (via the Cobaya code) to explore the parameter space. The likelihood function combines four distinct datasets:

• CMB Power Spectra: Planck PR4 NPIPE high-$\ell$ (TT, EE, TE) and low-$\ell$ likelihoods.
• Supernovae: The Pantheon+ dataset (1701 light curves).
• Baryon Acoustic Oscillations (BAO): DESI DR2 data.
• Cosmic Birefringence: The ACT DR6 measurement of the rotation angle, $\beta_{\text{data}} = 0.215 \pm 0.074$ degrees.

The analysis is conducted using both Bayesian (posterior distributions) and frequentist (profile likelihood) approaches to ensure robustness. Non-linear clustering effects from dynamical DE are deemed negligible for the CMB lensing and are not included.

Key Contributions and Results
The paper derives constraints on the parameters of the potentials (energy scale $f$, mass/slope $\mu$, initial displacement $\phi_{\text{ini}}$, and initial velocity kick $\phi'_{\text{kick}}$) and the anomaly coefficient $c_{\phi\gamma}$.

• Axion-like Potential: This model is viable only if the anomaly coefficient is large. For a standard coefficient ($c_{\phi\gamma}=1$), the model fails to fit both cosmological data and the CB angle simultaneously because the required energy scale $f$ leads to a trans-Planckian field excursion to match $\beta$. The analysis finds a preference for $c_{\phi\gamma} \approx 11.5$ (pure thawing) or $\approx 20$ (freezing and thawing scenarios).
• Linear, Quadratic, and Ratra-Peebles Potentials: These models successfully explain both DE evolution and CB with a standard anomaly coefficient ($c_{\phi\gamma} = 1$).
  • Linear (Positive Slope): Requires a "kicked" initial velocity to roll up the potential before rolling down, fitting the data but implying a somewhat fine-tuned coincidence for the observed small CB angle.
  • Linear (Negative Slope), Quadratic, and Ratra-Peebles: These scenarios allow the field to roll down naturally. They favor a symmetry-breaking scale $f$ close to the Grand Unified Theory (GUT) scale (significantly below the Planck mass) and require specific initial displacements to satisfy the DE density constraints.
• Equation of State Evolution: All viable models predict an equation of state that is frozen at $\omega \approx -1$ at high redshifts (due to Hubble dragging) and evolves toward $\omega \approx -0.9$ at low redshifts ($z \lesssim 2$), consistent with DESI hints.
• Model Comparison: The inclusion of pseudo-scalar DE improves the fit over the standard $\Lambda$CDM model.
  • Considering only background expansion data (CMB, SNIa, BAO), the preference for dynamical DE is at the $\sim 3\sigma$ level ($\Delta \text{AIC} \approx 6.9 - 11.5$).
  • Including the CB measurement increases the statistical significance to $\sim 4\sigma$, as $\Lambda$CDM predicts zero birefringence.

Significance
The authors conclude that pseudo-scalar fields constitute a compelling realization of dynamical dark energy that can naturally account for the observed cosmic birefringence. The study demonstrates that while axion-like potentials require large anomaly coefficients to be viable, other potentials (linear, quadratic, Ratra-Peebles) are fully consistent with current data using standard coupling strengths. The work provides a comprehensive overview of the parameter space for these models, highlighting that the symmetry-breaking scale for viable non-axion models is likely close to the GUT scale. The authors note that forthcoming data from DESI, Euclid, and CMB experiments with absolute polarization calibration will be crucial for further constraining or ruling out these scenarios.