Higgs Inflation: SM Higgs as Inflaton

• Higgs Inflation is a class of models in which the Standard Model Higgs field, with a non-minimal coupling to gravity, acts as the inflaton driving the early universe's exponential expansion.
• The framework employs a conformal transformation to the Einstein frame, resulting in a plateau-like potential that supports slow-roll inflation with CMB-compatible predictions for spectral tilt and tensor-to-scalar ratio.
• Extensions including multifield dynamics, Higgs–portal couplings, and brane constructions address challenges of unitarity, quantum stability, and UV completion while linking cosmic inflation to observable particle physics phenomena.

Higgs inflation denotes a class of inflationary cosmological models in which the Standard Model (SM) Higgs field is identified as the inflaton responsible for the exponential expansion of the early universe. This paradigm leverages the existence of the Higgs as a fundamental scalar—and thus the only known elementary scalar field—to connect high energy cosmology with particle physics. Higgs inflation models are highly constrained by their theoretical structure, renormalization group running, Planck-scale stability, and their predictions for the cosmic microwave background (CMB) power spectrum.

1. Fundamental Structure of Higgs Inflation Models

The minimal scenario begins with the Standard Model Higgs sector, described by the Lagrangian

$$S_J = \int d^4x \sqrt{-g} \Bigl[ -\frac{1}{2}(M_P^2 + \xi h^2)R + \frac{1}{2}g^{\mu\nu}\partial_\mu h \partial_\nu h - V(h) \Bigr],$$

where $M_P$ is the reduced Planck mass, $h$ is the real mode of the Higgs doublet, $\xi$ is a dimensionless non-minimal coupling to the Ricci scalar $R$, and $V(h)$ is the standard Higgs potential $V(h) = \frac{\lambda}{4}(h^2 - v^2)^2$ with VEV $v \approx 246$ GeV and quartic $\lambda$.

The introduction of the $\xi h^2 R$ term is necessary to achieve slow-roll inflation compatible with observations; for the canonical potential, the quartic self-coupling $M_P$0 is too large to yield the observed amplitude of scalar perturbations in the absence of a large nonminimal coupling (Bezrukov, 2013, Rubio, 2018).

To study dynamics, a Weyl (conformal) transformation to the Einstein frame is performed, yielding a canonical gravity sector and a nontrivial inflaton kinetic function. The canonically normalized field $M_P$1 is related to $M_P$2 via

$M_P$3

At large field values $M_P$4, this leads to an exponentially flat Einstein-frame potential, generically of the form

$M_P$5

This potential supports slow-roll inflation with Planck-compatible scalar spectral tilt and negligible tensor-to-scalar ratio (Bezrukov, 2013, Rubio, 2018).

2. Variants and Generalizations

The generic Higgs inflation framework encompasses several model variants:

• Non-minimal coupling (canonical Higgs inflation): Requiring $M_P$6 for $M_P$7, produces $M_P$8, $M_P$9 for $h$0 e-folds (Bezrukov, 2013, Rubio, 2018, Greenwood et al., 2012).
• Running kinetic Higgs inflation: Field-dependent kinetic terms flatten the potential without non-minimal coupling, yielding higher $h$1 but requiring kinetic-sector modifications (Kamada et al., 2012, Takahashi, 2015).
• Higgs G-inflation (Galileon-type derivative coupling): Involves higher-derivative (Galileon-like) couplings, enhancing friction and supporting steeper potentials, with $h$2 for $h$3 (Kamada et al., 2010, Kamada et al., 2012).
• Higgs–Portal inflation: SM Higgs mixes with a singlet scalar, giving multifield inflation with reduced tensor amplitude and distinct mixing and decay signatures (Lebedev et al., 2011).
• Extension to brane-world and Gauss–Bonnet brane scenarios: Address unitarity and stability by embedding the Higgs sector in higher-dimensional (brane) setups (Cai et al., 2015, Escobar, 2012).

Generalized G-inflation unifies these in a single effective theory framework with second-order field equations (Kamada et al., 2012).

3. Primordial Perturbations and Predictions

All viable Higgs inflation models are assessed by their predictions for the primordial curvature and tensor perturbations. For the canonical plateau potential, the slow-roll parameters are derived as

$h$4

leading to

$h$5

For $h$6, $h$7 and $h$8 (Bezrukov, 2013, Greenwood et al., 2012, Okada et al., 2015).

In non-canonical and Galileon-type (G-inflation) models, $h$9 may be substantially enhanced ($\xi$0), with distinctive modifications to the consistency relation between $\xi$1 and the tensor tilt $\xi$2; e.g., $\xi$3 instead of $\xi$4 of canonical single-field inflation (Kamada et al., 2010).

The amplitude of scalar perturbations $\xi$5 and the normalization of the power spectrum fix the required nonminimal coupling: $\xi$6 so for $\xi$7, $\xi$8 (Bezrukov, 2013, Rubio, 2018).

Extensions allowing general initial (non-Bunch–Davies) quantum states for perturbations can amplify $\xi$9 by a factor $R$0, potentially elevating $R$1 to the current observational bounds ($R$2), while leaving $R$3 largely unchanged and still compatible with non-Gaussianity constraints (Zeynizadeh et al., 2015).

4. Quantum Stability, Unitarity, and UV Embedding

Stability of the scalar potential up to the inflationary scale is essential: SM RG running can drive $R$4 negative at high scales, compromising inflationary dynamics. This imposes tight constraints on the top mass $R$5, Higgs mass $R$6, and, if necessary, mandates the addition of new states (such as fermionic singlets or dark matter multiplets) to guarantee vacuum stability (Enqvist et al., 2014, Okada et al., 2015).

The unitarity cutoff for the non-minimally coupled Higgs model is $R$7, which can drop below the inflationary scale for large $R$8. However, the relevant cutoff during inflation is field-dependent and can be raised to $R$9, avoiding strong coupling in the background field (Atkins et al., 2010, Bezrukov, 2013).

Alternative frameworks, such as asymptotically safe gravity, have been proposed to resolve unitarity and provide a UV completion. Here the gravitational couplings flow to a fixed point, $V(h)$0 and $V(h)$1 in the UV, removing the dangerous operators at high scale and eliminating the need for new states beyond the SM plus gravity (Xianyu et al., 2014, Atkins et al., 2010). These scenarios are highly predictive, with tensor amplitudes further suppressed ($V(h)$2) (Xianyu et al., 2014).

5. Extensions: Multifield, Portal, and Brane Constructions

Multifield dynamics: The full electroweak Higgs is a complex SU(2) doublet. Due to SO(4) symmetry, multifield effects (arising from Goldstone bosons) damp rapidly and reduce to effective single-field inflation before observable modes exit the horizon (Greenwood et al., 2012).

Higgs–Portal and hidden sector: By coupling the Higgs to a real singlet via a portal interaction, inflation can occur along a mixed Higgs–singlet direction, with the tensor-to-scalar ratio $V(h)$3 and $V(h)$4 mirroring the plateau values. Vacuum stability, unitarity, and phenomenological constraints on mixing and decay signatures at colliders play central roles (Lebedev et al., 2011).

Braneworld and Gauss–Bonnet generalizations: In 5D braneworlds and Gauss–Bonnet braneworlds, the Planck mass (and inflationary dynamics) become functions of the extra-dimensional geometry. For specific choices of extra dimension scale $V(h)$5, it is possible to achieve $V(h)$6 and avoid the unitarity problems of large $V(h)$7, with tensor modes further suppressed, $V(h)$8 (Cai et al., 2015).

6. Observational and Phenomenological Constraints

CMB measurements by Planck and BICEP/Keck tightly constrain the spectral tilt and tensor amplitude: $V(h)$9, $V(h) = \frac{\lambda}{4}(h^2 - v^2)^2$0 (95\% CL). Canonical Higgs inflation predictions are in excellent agreement: $V(h) = \frac{\lambda}{4}(h^2 - v^2)^2$1, $V(h) = \frac{\lambda}{4}(h^2 - v^2)^2$2 for $V(h) = \frac{\lambda}{4}(h^2 - v^2)^2$3 (Malekpour et al., 2024, Bezrukov, 2013, Greenwood et al., 2012). The amplitude of scalar perturbations fixes the ratio $V(h) = \frac{\lambda}{4}(h^2 - v^2)^2$4 and thus, given the measured Higgs mass and SM couplings, tightly constrains viable parameter space.

Vacuum meta(stability) remains crucial. The conventional SM running generically destabilizes $V(h) = \frac{\lambda}{4}(h^2 - v^2)^2$5 at $V(h) = \frac{\lambda}{4}(h^2 - v^2)^2$6 GeV unless the top mass is lower than the central experimental value, or extra states are introduced (Enqvist et al., 2014, Okada et al., 2015). Precise measurements of $V(h) = \frac{\lambda}{4}(h^2 - v^2)^2$7, $V(h) = \frac{\lambda}{4}(h^2 - v^2)^2$8, and $V(h) = \frac{\lambda}{4}(h^2 - v^2)^2$9 feed directly into the stability analysis.

Reheating occurs rapidly via Standard Model couplings; the reheat temperature is typically $v \approx 246$0 GeV, and the large coupling between the Higgs and gauge bosons guarantees efficient preheating and thermalization (Rubio, 2018). This connects the inflationary scenario directly to the baryogenesis problem and dark matter genesis, particularly in models with portal or seesaw-type extensions (Okada et al., 2015).

7. Current Directions and Open Problems

Higgs inflation provides a concrete, minimalistic bridge between particle physics and the early Universe, but several open questions are active areas of research:

• Quantum corrections, frame dependence, and subtraction scheme ambiguities: Loop effects, RG improvement, and the choice of renormalization prescription can impact inflationary predictions, affect stability, and alter required values of $v \approx 246$1 (Bezrukov, 2013).
• Unitarity and UV completion: Whether the required large nonminimal couplings can be consistently embedded in a UV-complete theory, and whether strong coupling is avoided, remains unsettled. Asymptotic safety, strong coupling “self-healing,” and extended field content are all under scrutiny (Atkins et al., 2010, Xianyu et al., 2014).
• Initial state effects and non-Bunch–Davies vacua: Deviations from the minimal initial quantum state can raise $v \approx 246$2 into observable ranges while preserving $v \approx 246$3, adding an extra model-dependent degree of freedom (Zeynizadeh et al., 2015).
• Braneworld and extra-dimensional embeddings: Modifications to Friedmann dynamics through higher-dimensional or Gauss–Bonnet corrections can allow for natural $v \approx 246$4, avoiding unitarity problems, though with typically unobservable primordial tensor signals (Cai et al., 2015).
• Phenomenological signatures at colliders: Portal and multifield scenarios predict deviations in Higgs couplings, mixing angles, and exotic decays potentially testable at the LHC or high-luminosity upgrades (Lebedev et al., 2011).

Future experimental results from CMB B-mode searches, collider precision Higgs and top measurements, and possible detection of gravitational waves from cosmic defects or domain walls will further probe or constrain the landscape of Higgs inflation models.

---

Key references:

• Minimal and non-minimal Higgs inflation (Bezrukov, 2013, Rubio, 2018, Greenwood et al., 2012)
• G-inflation and generalized formulations (Kamada et al., 2010, Kamada et al., 2012)
• Higgs–portal and multifield scenarios (Lebedev et al., 2011, Greenwood et al., 2012)
• Asymptotic safety and UV completions (Atkins et al., 2010, Xianyu et al., 2014)
• Quantum initial condition effects (Zeynizadeh et al., 2015)
• Brane and higher-dimensional versions (Cai et al., 2015, Escobar, 2012)
• Recent model developments in unimodular gravity (Malekpour et al., 2024)
• Quantum moment/multifield Higgs inflation (Bojowald et al., 2020)