Power-Law Inflation Models

• Power-law inflation is a model where the universe expands as a power-law function (a(t) ∝ t^p with p>1), providing exact analytic solutions for early-universe dynamics.
• Extensions include non-minimal couplings, non-canonical kinetic terms, and modified gravity frameworks that adjust predictions to match CMB observations.
• The framework features hidden symmetries, attractor dynamics, and mechanisms for a graceful exit, ensuring consistency with primordial fluctuation measurements.

Power-law inflation refers to a broad class of inflationary models in which the scale factor $a(t)$ evolves as a strict power of cosmic time: $a(t)\propto t^p$ with $p>1$. These models admit exact analytic solutions, are mathematically rich, and remain central in the study of alternative early-universe cosmologies and their generalizations. They include canonical single-field realizations, non-minimal and non-canonical extensions, and variants emerging in modified gravity, non-extensive thermodynamics, and higher-dimensional or memory-augmented frameworks. While the minimal version with exponential potentials is strongly constrained by current cosmological data, several extensions restore viability and deepen the theoretical landscape.

1. Fundamental Formulation and Exact Solutions

Power-law inflation is classically realized with a canonical scalar (inflaton) field $\phi$ and an exponential potential: $V(\phi) = V_0\, \exp(-\lambda\, \phi / M_{\rm Pl})$ in a spatially flat FLRW metric. The evolution equations,

$$\ddot\phi + 3H\dot\phi + V'(\phi) = 0, \qquad H^2 = \frac{1}{3M_{\rm Pl}^2} \left(\frac{1}{2}\dot\phi^2 + V(\phi)\right)$$

admit the exact power-law solution: $$a(t) \propto t^p, \qquad \phi(t) = \frac{2}{\lambda}\ln t + \phi_0, \qquad p = \frac{2}{\lambda^2}$$ with constant Hubble-flow parameters $\epsilon = \eta = 1/p$ and equation of state $w = -1 + 2/(3p)$, yielding an accelerated phase for $p > 1$ (i.e., $a(t)\propto t^p$0) (Yu et al., 30 May 2025, Chiba et al., 2024).

This solution generalizes in several directions, including non-minimal coupling, $a(t)\propto t^p$1 gravity, non-canonical kinetic terms, and the inclusion of additional fields or entropic corrections. Remarkably, the system possesses a "hidden" conformal Killing symmetry, ensuring classical integrability for the exponential-potential case (Chiba et al., 2024).

2. Perturbations, Observational Signatures, and Constraints

The spectral tilt and tensor-to-scalar ratio for standard single-field canonical power-law inflation are given by: $a(t)\propto t^p$2 The model yields exact scale invariance ($a(t)\propto t^p$3) in the de Sitter limit ($a(t)\propto t^p$4). However, Planck and BICEP/Keck data ($a(t)\propto t^p$5, $a(t)\propto t^p$6) require $a(t)\propto t^p$7 for $a(t)\propto t^p$8, which implies $a(t)\propto t^p$9, in tension with observational bounds (Oliveira et al., 17 Feb 2026, Yu et al., 30 May 2025, D'Amico et al., 2022).

Extensions and generalizations rectify this tension by:

• Dynamical Adjustment: Introduction of a continuous parameter family (beyond the classical $p>1$0 solution) relaxes the rigid $p>1$1–$p>1$2 relation, enabling models which fit precise cosmological data (Yu et al., 30 May 2025).
• Additional fields or modified gravity: Inclusion of non-minimal couplings, fractional calculus, or higher-derivative terms reduces $p>1$3 for a given $p>1$4 (Oliveira et al., 17 Feb 2026, Odintsov et al., 8 Sep 2025, Campo et al., 2015, Ito et al., 2019).
• Non-canonical kinetic terms: Power-law models with generalized kinetic structure adjust the consistency relation, again permitting compatibility with data (Unnikrishnan et al., 2013).
• Non-extensive entropic corrections: Modified thermodynamic relations, particularly with Tsallis, Rényi, and Sharma–Mittal entropy, allow viable $p>1$5 combinations over broader parameter domains (Khodam-Mohammadi, 2024).

The generic tensor power is a crucial discriminant: in standard and many generalized power-law models, unless correction mechanisms are present, the primordial gravitational wave background is too large.

3. Theoretical Generalizations and Alternative Realizations

Power-law inflation is robust under significant theoretical extensions:

• Non-minimal Coupling and Modified Gravity: For models where the scalar is non-minimally coupled to curvature, e.g., via $p>1$6, exact power-law attractors exist for $p>1$7 in a narrow range consistent with data ($p>1$8, $p>1$9), whereas the slow-roll regime is observationally excluded (Campo et al., 2015, Kundu et al., 2019).
• Non-canonical and k-inflation cases: Models with Lagrangians of the form $\phi$0 (for $\phi$1) can produce exact power-law solutions with modified consistency relations. Acceptable $\phi$2 demands $\phi$3, introducing a nontrivial sound speed and potentially small equilateral non-Gaussianity (Unnikrishnan et al., 2013).
• Memory/Non-local Corrections: Fractional calculus extensions yield non-local Friedmann and Klein–Gordon equations where the scalar spectral tilt is unchanged ($\phi$4), but $\phi$5 is suppressed for fractional order $\phi$6. Observationally viable power-law inflation arises for $\phi$7 and $\phi$8 (Oliveira et al., 17 Feb 2026).
• Warm and Non-extensive Entropy Inflation: Warm inflation and entropy-modified FLRW equations (Tsallis, Rényi, and Sharma–Mittal entropy) lead to modified power-law exponents and slow-roll observables, expanding the allowed model parameter space while ensuring a graceful exit. For Tsallis, viable inflation is found for $\phi$9 and $V(\phi) = V_0\, \exp(-\lambda\, \phi / M_{\rm Pl})$0, while for Rényi and Sharma–Mittal entropy, stringent smallness of $V(\phi) = V_0\, \exp(-\lambda\, \phi / M_{\rm Pl})$1 ($V(\phi) = V_0\, \exp(-\lambda\, \phi / M_{\rm Pl})$2) is required (Alhallak et al., 2022, Khodam-Mohammadi, 2024).
• Tachyonic and Cuscuton Fields: Non-canonical kinetic structures (e.g., tachyon with $V(\phi) = V_0\, \exp(-\lambda\, \phi / M_{\rm Pl})$3 (Souza et al., 2013) or cuscuton backgrounds (Ito et al., 2019)) admit exact power-law solutions, with reduced or tunable tensor-to-scalar ratios and robust control over tilt, even for steep potentials.
• Anisotropic and Multi-field Realizations: Power-law inflation within frameworks incorporating gauge fields or multiple scalar fields yields anisotropic generalizations (with controlled statistical anisotropy), plateau-like potentials, or variable consistency relations; dynamical systems analyses confirm late-time attractor behavior (Kanno et al., 2010, Alhallak et al., 2016).
• $V(\phi) = V_0\, \exp(-\lambda\, \phi / M_{\rm Pl})$4 Gravity and Modified Theories: In $V(\phi) = V_0\, \exp(-\lambda\, \phi / M_{\rm Pl})$5 gravity, naive analytic solutions are ruled out by coherence arguments and data, but rectified frameworks, which avoid premature slow-roll truncations, admit quasi–de Sitter expansion compatible with the latest CMB/ACT bounds for small negative $V(\phi) = V_0\, \exp(-\lambda\, \phi / M_{\rm Pl})$6 ($V(\phi) = V_0\, \exp(-\lambda\, \phi / M_{\rm Pl})$7) (Odintsov et al., 8 Sep 2025).

4. Graceful Exit and Multi-Phase Dynamics

A key issue with minimal power-law inflation is the absence of a graceful exit. Viable model-building requires:

• Potential Deformation: Modified or hybrid potentials, such as two-branch forms $V(\phi) = V_0\, \exp(-\lambda\, \phi / M_{\rm Pl})$8, generate a dynamical end to inflation followed by reheating (Unnikrishnan et al., 2013).
• Smooth Transitions: A dynamical transition from slow-roll to power-law regimes permits resolution of trans-Planckian issues while matching large-scale CMB observations; explicit interpolating kinetic structures achieve this transition with finite total e-folds and controlled mode evolution (Odintsov et al., 27 Jan 2026, Odintsov et al., 6 Apr 2025).

The total number of e-folds ($V(\phi) = V_0\, \exp(-\lambda\, \phi / M_{\rm Pl})$9) and field excursions remain consistent with both standard and Swampland conjectures depending on model details (Odintsov et al., 6 Apr 2025, D'Amico et al., 2022).

5. Physical Significance, Symmetry, and Attractor Structure

Power-law inflation exhibits exceptional mathematical tractability due to its underlying hidden symmetries:

• The Eisenhart lift uncovers a conformal Killing vector in minisuperspace field space unique to the single-exponential potential case, ensuring integrability and the existence of additional conserved charges (Chiba et al., 2024).
• This integrability explains why the power-law class admits exact analytic solutions for all background functions ($$\ddot\phi + 3H\dot\phi + V'(\phi) = 0, \qquad H^2 = \frac{1}{3M_{\rm Pl}^2} \left(\frac{1}{2}\dot\phi^2 + V(\phi)\right)$$0), forming a continuous family of late-time attractors (Yu et al., 30 May 2025).
• Dynamical systems analyses across models (single-field, $$\ddot\phi + 3H\dot\phi + V'(\phi) = 0, \qquad H^2 = \frac{1}{3M_{\rm Pl}^2} \left(\frac{1}{2}\dot\phi^2 + V(\phi)\right)$$1, multi-field, anisotropic) universally support the presence of stable inflationary attractor manifolds (Oliveira et al., 17 Feb 2026, Kanno et al., 2010).

6. Observational and Model-building Perspective

Table: Comparison of Key Power-law Inflationary Extensions

Model Extension	Spectral Tilt $$\ddot\phi + 3H\dot\phi + V'(\phi) = 0, \qquad H^2 = \frac{1}{3M_{\rm Pl}^2} \left(\frac{1}{2}\dot\phi^2 + V(\phi)\right)$$2	Tensor Ratio $$\ddot\phi + 3H\dot\phi + V'(\phi) = 0, \qquad H^2 = \frac{1}{3M_{\rm Pl}^2} \left(\frac{1}{2}\dot\phi^2 + V(\phi)\right)$$3	Planck/ACT Compatibility
Canonical, $$\ddot\phi + 3H\dot\phi + V'(\phi) = 0, \qquad H^2 = \frac{1}{3M_{\rm Pl}^2} \left(\frac{1}{2}\dot\phi^2 + V(\phi)\right)$$4	$$\ddot\phi + 3H\dot\phi + V'(\phi) = 0, \qquad H^2 = \frac{1}{3M_{\rm Pl}^2} \left(\frac{1}{2}\dot\phi^2 + V(\phi)\right)$$5	$$\ddot\phi + 3H\dot\phi + V'(\phi) = 0, \qquad H^2 = \frac{1}{3M_{\rm Pl}^2} \left(\frac{1}{2}\dot\phi^2 + V(\phi)\right)$$6	Excluded for $$\ddot\phi + 3H\dot\phi + V'(\phi) = 0, \qquad H^2 = \frac{1}{3M_{\rm Pl}^2} \left(\frac{1}{2}\dot\phi^2 + V(\phi)\right)$$7 ($$\ddot\phi + 3H\dot\phi + V'(\phi) = 0, \qquad H^2 = \frac{1}{3M_{\rm Pl}^2} \left(\frac{1}{2}\dot\phi^2 + V(\phi)\right)$$8)
Non-minimal coupling	$$\ddot\phi + 3H\dot\phi + V'(\phi) = 0, \qquad H^2 = \frac{1}{3M_{\rm Pl}^2} \left(\frac{1}{2}\dot\phi^2 + V(\phi)\right)$$9 (exact: red tilt for $$a(t) \propto t^p, \qquad \phi(t) = \frac{2}{\lambda}\ln t + \phi_0, \qquad p = \frac{2}{\lambda^2}$$0)	Model-dependent	Viable in narrow range ($$a(t) \propto t^p, \qquad \phi(t) = \frac{2}{\lambda}\ln t + \phi_0, \qquad p = \frac{2}{\lambda^2}$$1) (Campo et al., 2015)
Non-canonical kinetic	$$a(t) \propto t^p, \qquad \phi(t) = \frac{2}{\lambda}\ln t + \phi_0, \qquad p = \frac{2}{\lambda^2}$$2	$$a(t) \propto t^p, \qquad \phi(t) = \frac{2}{\lambda}\ln t + \phi_0, \qquad p = \frac{2}{\lambda^2}$$3	Viable for large $$a(t) \propto t^p, \qquad \phi(t) = \frac{2}{\lambda}\ln t + \phi_0, \qquad p = \frac{2}{\lambda^2}$$4 ($$a(t) \propto t^p, \qquad \phi(t) = \frac{2}{\lambda}\ln t + \phi_0, \qquad p = \frac{2}{\lambda^2}$$5)
Fractional memory	$$a(t) \propto t^p, \qquad \phi(t) = \frac{2}{\lambda}\ln t + \phi_0, \qquad p = \frac{2}{\lambda^2}$$6	$$a(t) \propto t^p, \qquad \phi(t) = \frac{2}{\lambda}\ln t + \phi_0, \qquad p = \frac{2}{\lambda^2}$$7	Compatible for $$a(t) \propto t^p, \qquad \phi(t) = \frac{2}{\lambda}\ln t + \phi_0, \qquad p = \frac{2}{\lambda^2}$$8; $$a(t) \propto t^p, \qquad \phi(t) = \frac{2}{\lambda}\ln t + \phi_0, \qquad p = \frac{2}{\lambda^2}$$9 (Oliveira et al., 17 Feb 2026)
Warm, non-extensive	Model-dependent, Tsallis $\epsilon = \eta = 1/p$0, Rényi: $\epsilon = \eta = 1/p$1 (tiny $\epsilon = \eta = 1/p$2)	Model-dependent	Viable for tuned entropic parameters (Khodam-Mohammadi, 2024)
Cuscuton-dressed	$\epsilon = \eta = 1/p$3	$\epsilon = \eta = 1/p$4	$\epsilon = \eta = 1/p$5 suppressed; fully viable (Ito et al., 2019)
$\epsilon = \eta = 1/p$6 gravity, rectified	$\epsilon = \eta = 1/p$7	$\epsilon = \eta = 1/p$8	Admits fit for small negative $\epsilon = \eta = 1/p$9 (Odintsov et al., 8 Sep 2025)

Power-law inflation therefore serves as a unifying scheme, underlying multiple ostensibly distinct inflationary phenomena. In its generalized forms, it can be made compatible with all currently available observational constraints, offers a well-defined framework for model exploration (both canonical and non-canonical), and provides clear targets for future CMB and primordial gravitational wave experiments (Yu et al., 30 May 2025, Oliveira et al., 17 Feb 2026, Odintsov et al., 8 Sep 2025).

7. Future Directions and Open Challenges

Current and forthcoming advances focus on:

• Precision Testing: Exploiting the sharp predictions for $w = -1 + 2/(3p)$0 and $w = -1 + 2/(3p)$1 in the various extensions, particularly in the context of upcoming CMB-S4, Simons Observatory, and space-based missions.
• Primordial Non-Gaussianity: Examining non-canonical scenarios for potentially detectable non-Gaussian signals, especially equilateral-type $w = -1 + 2/(3p)$2 (Unnikrishnan et al., 2013).
• Quantum Corrections and Non-Perturbative Effects: Studies show the importance of secular loop effects and IR sensitivity in power-law backgrounds, with non-trivial implications for the predictivity of quantum field theory in inflation (Glavan et al., 2019).
• Multi-phase and Anisotropic Inflation: Probing the dynamical transition regions, late-time attractors, and possible small signatures of statistical anisotropy (Odintsov et al., 27 Jan 2026, Kanno et al., 2010).
• Thermodynamic Foundations: Non-extensive and generalized entropy approaches suggest deeper connections between gravity, horizon thermodynamics, and inflationary cosmology (Khodam-Mohammadi, 2024).

A plausible implication is that power-law inflation, in its modern generalized forms, functions as a flexible theoretical template that accommodates the data while exploring the interface of fundamental symmetries, dynamical attractor structure, and quantum/thermodynamic effects in the early universe.