Multifield dark energy: Interplay between curved field space and curved spacetime

This paper demonstrates that despite the theoretical interplay between curved field space and spatial curvature in multifield exponential quintessence models, observationally viable late-time acceleration requires a sufficiently flat potential, as non-geodesic scaling mechanisms cannot coexist with background fluids and the axion effectively decouples, thereby preserving the tension between cosmic acceleration and quantum gravity expectations.

The Gist

Imagine the universe is a giant, expanding balloon. For a long time, scientists thought the air inside was just "stuff" (matter and radiation) that was slowly slowing the balloon down due to gravity. But then, we discovered the balloon is actually speeding up its expansion. Something invisible is pushing it outward. We call this mysterious pusher Dark Energy.

For decades, the simplest explanation was a constant "push" (like a cosmological constant). But this paper asks: What if Dark Energy isn't a constant, but a dynamic field that changes over time?

Specifically, the authors investigate a scenario inspired by String Theory (the theory that tries to unify all physics). In this scenario, Dark Energy isn't just one thing; it's a duo:

1. The Modulus (The Roller): A heavy ball rolling down a hill.
2. The Axion (The Spin): A spinning top attached to the ball.

They are moving on a curved surface (like a saddle or a bowl), not a flat floor. Also, the universe itself might be slightly curved (like a sphere or a saddle shape), not perfectly flat.

Here is the story of what they found, broken down with simple analogies:

1. The Setup: A Roller and a Spin on a Curved Track

Imagine a skateboarder (the Modulus) trying to roll down a hill to power a machine (the universe's expansion).

• The Problem: In String Theory, the hill is usually very steep. If you roll down a steep hill, you go too fast, and the machine doesn't run smoothly. To get the smooth, slow push needed for the universe to accelerate, you usually need a very gentle, flat hill. But String Theory says the hills are steep!
• The Hope: The authors wondered if two things could save the day:
  1. The Spin (Axion): Maybe if the skateboarder spins wildly while rolling, the "centrifugal force" of the spin could counteract the steepness of the hill, allowing for a smooth ride even on a steep slope.
  2. The Curved Track (Space): Maybe if the universe itself is curved (like a bowl), it could help the skateboarder maintain the right speed.

2. The Investigation: Can the Spin and the Curve Save Us?

The authors did a massive amount of math (using "dynamical systems" which is like mapping every possible path a skateboarder could take) to see if these two tricks could work together to make the universe accelerate without needing a flat hill.

The Big Discovery:
They found that while these tricks work individually in theory, they cannot work together in the real universe we live in.

• The Spin Trick: The spinning axion can help the system accelerate on a steep hill, but only if the universe is empty of other stuff (like matter and radiation).
• The Curve Trick: A curved universe can help, but only if the skateboarder is just rolling straight (not spinning).

The Conflict:
In our real universe, we are full of matter (galaxies, dust, etc.). The authors found that as soon as you add this "background stuff" (matter/radiation), the Spin Trick fails. The axion (the spin) gets left behind; it doesn't keep up with the rest of the universe. It becomes "dynamically subdominant," meaning it's too weak to matter.

So, the universe effectively becomes a single-field system again. The skateboarder is just rolling straight, and the spin is negligible.

3. The Result: The Hill Must Still Be Gentle

Because the "Spin Trick" and the "Curved Universe Trick" cancel each other out in a universe full of matter, the system reverts to the old rules.

• The Conclusion: To get the universe to accelerate today, the hill (the potential energy) must still be relatively flat.
• The Limit: They calculated that the steepness of the hill (a parameter called $\lambda$) must be less than about 0.75.
• The Tension: This creates a conflict with String Theory. String Theory naturally predicts hills that are much steeper (around $\lambda \approx 1$ or higher). The universe seems to be saying, "I need a gentle slope," while String Theory says, "I only have steep slopes."

4. The "Degeneracy" Analogy

The paper also mentions a "degeneracy." Imagine you are trying to figure out why a car is moving fast. Is it because the engine is powerful (steep hill)? Or is it because the road is downhill (curved space)?
The authors found that in this model, it's very hard to tell the difference between the "engine power" (the field dynamics) and the "road slope" (spatial curvature). They look the same to our telescopes. This makes it hard to prove which one is actually doing the work, but it doesn't change the final verdict: the engine needs to be tuned down (flat potential) to match observations.

Summary for the General Audience

• The Dream: Scientists hoped that a "spinning" dark energy field and a "curved" universe could work together to allow for a steep, String Theory-friendly universe that still accelerates nicely.
• The Reality Check: The math shows that in a universe filled with matter (like ours), the spinning part gets "drowned out" and stops helping. The curvature of space doesn't help enough to fix the steepness problem.
• The Verdict: The universe still requires a gentle slope for Dark Energy to work. This keeps the tension alive between what we see in the sky (gentle slopes) and what String Theory predicts (steep slopes).

In short: The universe is picky. Even with a fancy two-field setup and a curved stage, it still demands a gentle slope for Dark Energy to do its job. The "magic tricks" of String Theory don't quite work in our specific cosmic neighborhood.

Technical Summary

1. Problem Statement

The paper addresses the tension between the observed late-time cosmic acceleration and theoretical expectations from UV-complete theories (specifically String Theory and the Swampland Conjectures).

• The Conflict: Swampland conjectures suggest that scalar potentials in quantum gravity should be steep ($\lambda \gtrsim O(1)$), whereas standard quintessence models require shallow potentials ($\lambda < \sqrt{2}$) to sustain acceleration.
• The Proposed Loopholes: Two mechanisms have been proposed to reconcile this:
  1. Multifield Dynamics: In string compactifications, a scalar modulus is accompanied by an axion. The curvature of the field space allows for non-geodesic trajectories (turning motions) that can sustain acceleration even on steep potentials.
  2. Spatial Curvature: A non-zero spatial curvature ($\Omega_k \neq 0$) in the FLRW metric can alter the kinematic conditions for acceleration, potentially enlarging the viable parameter space.
• The Gap: Previous studies often treated these mechanisms in isolation (single-field with curvature, or multifield in flat space). This paper investigates the simultaneous interplay of a curved field space (modulus-axion system) and a curved spacetime (FLRW with $\Omega_k \neq 0$) to determine if their combination can resolve the tension for steep potentials.

2. Methodology

The authors employ a comprehensive three-pronged approach:

A. Theoretical Setup

• Model: A minimal two-field system consisting of a scalar modulus ($\phi_1$) and an axion ($\phi_2$) evolving on a curved field space with metric $ds^2_{field} = d\phi_1^2 + f^2(\phi_1)d\phi_2^2$.
• Potential: Exponential potential $V(\phi_1) = V_0 e^{-\lambda \phi_1}$ and coupling $f(\phi_1) = f_0 e^{-\nu \phi_1}$.
• Background: A curved FLRW universe containing radiation, matter, and the scalar fields.
• Lagrangian:
  $$\mathcal{L} = -\frac{1}{2}\partial_\mu \phi_1 \partial^\mu \phi_1 - \frac{1}{2}f^2(\phi_1)\partial_\mu \phi_2 \partial^\mu \phi_2 - V(\phi_1)$$

B. Dynamical Systems Analysis

• The authors convert the equations of motion into an autonomous system using dimensionless variables ($x_1, x_2, y, \Omega_\alpha$).
• They perform a global phase-space analysis, identifying all fixed points (equilibrium solutions) and their stability properties (attractors, saddles, repellers).
• Key fixed points analyzed include:
  • Geodesic (G): Single-field behavior ($x_2=0$).
  • Non-Geodesic (NG): Turning trajectories ($x_2 \neq 0$).
  • Scaling (S): Solutions where the scalar energy density tracks the background fluid.
• They analyze invariant manifolds (subspaces where the flow remains confined) to understand the structural constraints of the system.

C. Numerical and Observational Constraints

• Numerical Scan: Forward integration of the full multifield system from the radiation-matter equality epoch to the present day, scanning initial conditions and parameters ($\lambda, \nu$).
• Observational Analysis: Implementation of the model in the Boltzmann solver CLASS (modified to include the dynamical curvature variable).
• Data: Confronted with Planck 2018 distance priors, Pantheon+ SNe Ia, BAO, and Cosmic Chronometers using MCMC (MontePython).

3. Key Contributions and Results

A. Structural Absence of a Non-Geodesic Scaling Solution

A central theoretical finding is that no stable fixed point exists that combines non-geodesic motion (turning) with scaling behavior in the presence of a background fluid.

• Reasoning: The radial field ($\phi_1$) can scale with the background due to its exponential potential. However, the axion ($\phi_2$) has no potential; its energy density is governed by a conserved current.
• Result: The axion energy density scales as $\rho_{axion} \propto a^{-6}$ (stiff fluid) or decays even faster with coupling. It cannot track the background fluid ($a^{-3}$) unless parameters are fine-tuned to a specific bifurcation surface ($\lambda = 2\nu$).
• Implication: In the presence of matter/radiation, the axion remains dynamically subdominant during the matter era. The "turning" mechanism does not activate until the scalar field dominates the universe.

B. Decoupling of Curvature and Turning

The study reveals a dynamical hierarchy:

1. Spatial Curvature: Does not significantly relax the acceleration condition. While it modifies the kinematic bounds slightly, it cannot compensate for steep potentials ($\lambda \gtrsim 1$) to produce acceleration.
2. Non-Geodesic Parameter ($\nu$): The turning rate $\nu$ is effectively decoupled from the background evolution during the observable epoch (thawing regime). Because the axion is subdominant, the system behaves effectively as a single-field model ($x_2 \approx 0$) until late times.
3. Potential Slope ($\lambda$): This remains the dominant control parameter. The system behaves as a standard thawing quintessence model.

C. Analytical and Numerical Bounds on $\lambda$

• Thawing Approximation: Analytical derivation shows that for the system to transition from matter domination to acceleration without overshooting into a kinetic-dominated (kination) phase, the slope must satisfy $\lambda \lesssim 1.6 - 1.7$.
• Numerical Scan: Confirms that for $\lambda > 1.7$, the kinetic energy grows too rapidly, preventing the onset of acceleration.
• Observational Constraints: The MCMC analysis yields a tight upper bound:
  $$\lambda \lesssim 0.75 \quad (95\% \text{ C.L.})$$
  This bound holds even when allowing for dynamical spatial curvature and the full two-field dynamics. The parameter $\nu$ remains unconstrained by background data.

D. Degeneracies

The authors identify a degeneracy between the potential slope $\lambda$ and the initial energy partition between the scalar fields and spatial curvature. Variations in $\lambda$ shift the redshift of matter-radiation equality ($z_{rm,eq}$) in a way that mimics changes in curvature. However, this degeneracy does not allow steep potentials to evade the acceleration constraint.

4. Significance and Conclusions

• Resolution of the Loophole: The paper demonstrates that the combined effects of curved field space and curved spacetime do not enlarge the viable parameter space for exponential quintessence. The mechanisms that allow for acceleration on steep potentials (non-geodesic turning) are structurally suppressed in the presence of a background fluid.
• Persistence of Tension: The tension between Swampland conjectures (predicting $\lambda \sim O(1)$) and observational data (requiring $\lambda \lesssim 0.75$) persists even in this minimal multifield setup. The "turning" mechanism is a late-time asymptotic feature that does not assist in the transition from matter domination to acceleration.
• Effective Single-Field Limit: Despite being a two-field model, the observationally viable regime is effectively a single-field thawing model. The axion and spatial curvature are dynamically subdominant during the epochs probed by current data.
• Future Directions: The authors suggest that to reconcile steep potentials with data, one must look beyond the minimal setup, potentially involving perturbative effects (isocurvature modes, sound speed modifications) or more complex potentials, as background geometric probes alone cannot support the steep slopes favored by UV completions.

In summary, the paper provides a rigorous dynamical and statistical proof that spatial curvature and field-space curvature cannot jointly rescue steep exponential potentials from observational exclusion, reinforcing the challenge of constructing viable dark energy models within the Swampland paradigm.