Single field slow-roll inflation with step uplift to $n_s=1$

This paper proposes a single-field slow-roll inflation model featuring a sudden potential step at the end of inflation, which allows standard models like chaotic and Starobinsky inflation to produce a scale-invariant primordial spectrum ($n_s=1$) consistent with early dark energy solutions to the Hubble tension.

The Gist

The Big Problem: The Universe's Speed Limit is Broken

Imagine the universe is a giant city. Astronomers have two ways to measure how fast this city is expanding (the Hubble Constant, or $H_0$).

1. The "Old Map" (Planck): Looking at the baby picture of the universe (the Cosmic Microwave Background), they calculate the speed is about 67 km/s.
2. The "New GPS" (SH0ES): Looking at nearby stars and supernovas, they calculate the speed is about 73 km/s.

These numbers don't match. The difference is so big that it's like two GPS apps telling you to take two completely different routes to the same destination. This is called the Hubble Tension.

The Clue: A Perfectly Flat Spectrum

Recently, scientists realized that if the universe is actually expanding at the faster speed (73 km/s), the "baby picture" of the universe must look slightly different than we thought. Specifically, the ripples in the early universe (primordial perturbations) should be perfectly flat across all sizes.

In physics terms, this means the "spectral index" ($n_s$) should be exactly 1.

• Current models say $n_s \approx 0.965$ (a slightly tilted ramp).
• The new data suggests $n_s = 1$ (a perfectly flat floor).

The problem? The most famous theories of how the universe started (Inflation) naturally produce that "tilted ramp" ($0.965$). They struggle to produce the "perfectly flat floor" ($1.0$) without breaking the rules of physics or adding too many complicated ingredients.

The Solution: The "Sudden Stop" Trick

The authors of this paper propose a clever workaround. They say: "What if the universe started inflating exactly as the famous theories predict, but then hit a giant speed bump and stopped abruptly?"

Here is the analogy:

1. The Long, Gentle Slope (The Inflation)

Imagine a skier (the "inflaton" field) gliding down a very long, smooth, gentle hill. This is the standard "slow-roll inflation."

• As the skier glides, they create ripples in the snow (the universe's structure).
• If they glide for a long time, the ripples are slightly tilted ($n_s = 0.965$).
• To get a perfect flat ripple ($n_s = 1$), the skier would usually need to glide for way longer than the universe allows.

2. The Giant Step (The "Step Uplift")

The authors suggest that near the bottom of the hill, there is a sudden, sharp step or a cliff edge.

• The skier glides down the gentle slope for a long time (creating the deep slow-roll region).
• Suddenly, they hit the step. The slope becomes vertical.
• BAM! The skier falls off the edge. Inflation stops instantly.

3. Why This Changes the Math

Here is the magic trick:

• The "tilt" of the ripples depends on how long the skier was gliding before they stopped.
• In the old models, the skier stopped when the hill got too steep naturally.
• In this new model, the skier glided for a very long time (deep in the slow-roll region) but was forced to stop early by the step.
• Because the skier spent so much time on the gentle part of the hill, the ripples they left behind look almost perfectly flat ($n_s \approx 1$), even though the inflation ended "early" in terms of the total distance traveled.

Testing the Idea: Two Famous Models

The authors tested this "Sudden Stop" idea on two of the most popular inflation models:

1. Chaotic Inflation (The Steep Hill):

   • Old Problem: This model usually predicts a tilt that is too steep and creates too much gravitational wave noise (which we haven't seen yet).
   • New Fix: With the "step," the skier glides longer on the smooth part. This flattens the ripples to $n_s \approx 1$ and reduces the gravitational wave noise to a level that matches current observations.

2. Starobinsky Inflation (The Smooth Bowl):

   • Old Problem: This model is already very good, but it predicts $n_s = 0.967$.
   • New Fix: By adding the step, the model can push the value up to $n_s \approx 0.996$ or even $1.0$, making it perfectly compatible with the new "fast expansion" data.

The Takeaway

This paper suggests that the universe might not need a completely new, weird theory to solve the Hubble Tension. Instead, the famous theories we already love might just need a tiny modification: a sudden "step" in the energy landscape that forces inflation to end abruptly.

It's like realizing that a car doesn't need a new engine to drive faster; it just needs to hit a specific gear shift at the right moment. This "step" allows the universe to look perfectly flat ($n_s=1$) while still expanding at the faster speed ($H_0 \approx 73$) that our new measurements demand.

Technical Summary

1. The Problem

The paper addresses a growing tension in modern cosmology regarding the spectral index of primordial scalar perturbations ($n_s$) and the Hubble constant ($H_0$).

• The Hubble Tension: There is a significant discrepancy ($\gtrsim 5\sigma$) between the value of $H_0$ inferred from Cosmic Microwave Background (CMB) data under the standard $\Lambda$CDM model ($H_0 \approx 67.4$ km/s/Mpc) and local measurements using Cepheid-calibrated supernovae (SH0ES, $H_0 \approx 73$ km/s/Mpc).
• Early Dark Energy (EDE) Resolution: A leading solution to this tension involves Early Dark Energy models, particularly those with an Anti-de Sitter (AdS) well (AdS-EDE). These models suppress the sound horizon, allowing for a higher $H_0$ ($\sim 73$ km/s/Mpc) while fitting CMB data.
• The $n_s$ Conflict: Crucially, when fitting cosmological parameters assuming an AdS-EDE scenario with $H_0 \approx 73$ km/s/Mpc, the best-fit value for the spectral index shifts to $n_s \approx 1$ (scale-invariant Harrison-Zeldovich spectrum), with $|n_s - 1| \sim \mathcal{O}(0.001)$.
• The Inflationary Dilemma: Standard single-field slow-roll inflation models (e.g., Chaotic, Starobinsky) predict $n_s - 1 \approx -O(1)/N_*$, where $N_*$ is the number of e-folds before the end of inflation. For CMB modes exiting the horizon at $N_* \approx 60$, these models predict $n_s \approx 0.965$. Achieving $n_s \approx 1$ in standard slow-roll models would require $N_* \gg 60$ (e.g., $N_* > 200$), which contradicts the standard assumption that inflation lasts only $\sim 60$ e-folds after the CMB scales exit the horizon.

2. Methodology

The authors propose a mechanism to reconcile standard single-field slow-roll inflation with the observation of $n_s \approx 1$ without abandoning the slow-roll paradigm or introducing multi-field waterfall instabilities.

• Step-Modified Potential: They introduce a "step" in the inflaton potential ($V_s(\phi)$) that abruptly ends inflation. The potential is defined as:
  $$V_s(\phi) = f_{step}(\phi) V_o(\phi)$$
  where $V_o(\phi)$ is the original slow-roll potential (e.g., monomial or Starobinsky), and $f_{step}$ is a smooth step function (using a hyperbolic tangent) that drops the potential sharply at a specific field value $\phi_c$.
• Decoupling $N_*$ and $\Delta N$:
  • $\Delta N \approx 60$: The number of e-folds generated after the CMB modes exit the horizon but before the step triggers the end of inflation. This remains $\approx 60$.
  • $N_{*,o}$: The total number of e-folds the inflaton would have rolled in the original potential $V_o$ before reaching the original slow-roll end point.
  • Mechanism: The step is placed deep within the slow-roll region of the original potential. The inflaton rolls slowly for a long time ($N_{*,o} \gg 60$), generating the CMB modes. However, when it reaches $\phi_c$, the step causes a sudden violation of the slow-roll condition ($\epsilon_V \approx 1$), ending inflation abruptly.
• Spectral Index Calculation: The spectral index is determined by the slope of the potential at the time the CMB modes exit. Since the CMB modes exit while the field is still in the deep slow-roll region of $V_o$ (where $N_{*,o}$ is large), the prediction becomes:
  $$n_s - 1 \approx -\frac{O(1)}{N_{*,o}}$$
  By choosing $N_{*,o} \gg 60$ (e.g., $N_{*,o} \sim 200-500$), the model naturally yields $n_s \approx 1$, even though the total duration of inflation relevant to the CMB is only $\sim 60$ e-folds.

3. Key Contributions

• Novel Mechanism for $n_s=1$: The paper demonstrates that a scale-invariant spectrum ($n_s=1$) can be achieved in single-field slow-roll inflation without requiring exotic multi-field dynamics or non-standard exit mechanisms, provided the potential has a sharp step that terminates inflation early relative to the deep slow-roll trajectory.
• Compatibility with AdS-EDE: The model provides a concrete inflationary realization for the $n_s \approx 1$ requirement emerging from AdS-EDE solutions to the Hubble tension.
• Application to Popular Models: The authors explicitly apply this step-uplift mechanism to two of the most successful inflationary models:
  1. Chaotic Inflation (Monomial): Specifically $V \sim \phi^n$.
  2. Starobinsky Inflation ($R^2$ inflation): $V \sim (1 - e^{-\sqrt{2/3}\phi})^2$.

4. Results

The authors perform numerical and analytical checks on their proposed scheme:

• Chaotic Inflation ($V \sim \phi^n$):
  • Standard chaotic inflation ($n=2$) predicts $r \approx 0.13$, which is ruled out by BICEP/Keck ($r < 0.036$).
  • With the step modification, the model can achieve $n_s \approx 0.991$ (for $N_{*,o} > 200$).
  • Crucially, the tensor-to-scalar ratio $r$ is suppressed to $r \approx 0.035$, bringing it within the $2\sigma$ consistency range of EDE cosmologies and current observational limits.
  • For smaller $n$ (e.g., $n=0.1$), the model yields $n_s > 0.99$ and a highly suppressed $r \approx 0.0043$.
• Starobinsky Inflation:
  • Standard Starobinsky inflation predicts $n_s \approx 0.967$ and $r \approx 0.0033$ (consistent with $\Lambda$CDM).
  • With the step uplift, the model achieves $n_s \gtrsim 0.996$ (consistent with $n_s=1$) while maintaining a very low $r \lesssim 4.8 \times 10^{-5}$.
  • This result is highly consistent with combined datasets (Planck + SPT + ACT) under the AdS-EDE framework.
• Step Parameters: The step is characterized by a location $\phi_c$ and a width $d$. The authors show that if the step is sufficiently sharp ($d \ll 1$), the contribution of the step to the e-fold count ($\Delta N_{step}$) is negligible, ensuring the validity of the slow-roll approximation for the CMB modes.

5. Significance

• Preservation of Standard Models: This work suggests that the shift in observational preference toward $n_s = 1$ (driven by the Hubble tension) does not necessarily invalidate well-motivated single-field inflation models. Instead, it implies that these models may simply possess a "step" feature that terminates inflation earlier than the standard slow-roll end point.
• Phenomenological Implications: The step-like potential is a versatile tool. While this paper uses a large step to end inflation, similar features at different scales could explain other phenomena, such as oscillations in the primordial power spectrum (large scales) or the generation of Primordial Black Holes (small scales).
• Future Observables: The model predicts a nearly scale-invariant spectrum with potentially detectable oscillations or specific non-Gaussian signatures depending on the sharpness of the step, offering new targets for future CMB and Large Scale Structure surveys (e.g., DESI, CMB-S4).

In conclusion, the paper provides a robust theoretical framework where the "step uplift" allows standard single-field inflation to naturally accommodate the $n_s \approx 1$ requirement of AdS-EDE cosmologies, thereby resolving the tension between inflationary model predictions and the emerging cosmological data favoring a higher Hubble constant.