AG-2026.03-349·gr-qc
Logarithmically enhanced hyperbolic square-root deformation of Starobinsky inflation
Authors
- Andrei Galiautdinov
Abstract
We propose an enhanced hyperbolic square-root (HSQRT) deformation of the Starobinsky model in the context of $f(R)$ gravity. The original HSQRT construction provided a globally regular modification of $R^2$ inflation, curing the strong-coupling singularity at negative curvatures while preserving the exponential slow-roll plateau at large positive curvatures. Motivated by recent cosmological data (ACT DR6 and DESI), we introduce a structurally minimal, rational-logarithmic enhancement. This enhancement modifies the deep UV asymptotic regime while preserving global tachyon-free stability, ghost freedom, and the recovery of general relativity at low curvatures. In the Einstein frame, the scalaron dynamics is described by a globally defined, Padé-regulated effective potential, $V_{\text{Padé}}(φ) = V_{\text{Staro}}(φ) \left(1+ \frac{4β}{\frac{2}{3}κ^2φ^2 + 1}\right)^{-1} + \frac{1}{4ακ^2} \left[1 - \left( 1 + \sqrt{\frac{2}{3}}κφ\right) e^{-\sqrt{\frac{2}{3}}κφ} \right] e^{-\sqrt{\frac{2}{3}}κφ}$, which exhibits the inverse-power asymptotic form at high curvatures, $V(φ) \simeq \frac{1}{8ακ^2} \left( 1 - \frac{6β}{κ^2φ^2} \right)$, with the strength of the deviation from the baseline HSQRT geometry governed by dimensionless parameter $β$. We derive analytic expansions for the slow-roll observables, yielding a leading-order spectral index $n_s \simeq 1 - 3/(2N)$ and a tensor-to-scalar ratio $r \simeq 2\sqrt{3β}/N^{3/2}$. For standard e-fold durations ($N \in [50, 60]$), this model drives the spectral index directly into the newly favored observational window ($n_s \simeq 0.970\text{--}0.975$) and predicts an exceptionally small spectral running $α_s \simeq -3/(2N^2) \in [-0.00060, -0.00042]$, establishing a viable target for next-generation CMB observatories.
Submitted
16 March 20261 month ago
Version
v1
License
CC-BY-4.0
DOI
10.48550/arXiv.2603.14743
Chat with this PDF
Ask questions, probe assumptions, request a plain-English summary. Answers cite sections from the preprint itself.
Community
Questions and answers about this paper from other readers. No formal peer review — just a place to think out loud.