Testing $α$-attractor P-model of inflation by Cosmic Microwave Background radiation

Abstract: In a recently proposed approach to testing models of inflation by Cosmic Microwave Background (CMB) radiation the reheating temperature is directly expressed in terms of the CMB observables. Its model independent bounds translate in a given model into narrow ranges of those observables. In that approach we analyse the polynomial class of the $α$-attractor inflaton potential models (P-models), in a broad range of polynomials and with the inflaton decays and fragmentation in the reheating period taken into account. The predictions for the CMB observables, the scalar spectral index $n_s$ and tensor-to-scalar ratio $r$, are compared with the Planck and Planck combined with ACT data. Both can be accommodated by that class of the $α$ attractor models. The sensitivity of the results of that comparison to the reheating temperature and to the upper bound on the ratio $r$ is clearly demonstrated.

• The paper introduces a novel method connecting the reheating temperature Tₓ with CMB observables to constrain α-attractor polynomial inflation models.
• The analysis distinguishes between integer and fractional exponent models while accounting for inflaton fragmentation effects and both perturbative and non-perturbative reheating regimes.
• Key implications include more precise predictions for nₛ and r, setting tighter bounds that will be further tested by future high-precision CMB polarization experiments.

Testing $\alpha$-Attractor P-Model Inflation via CMB: Reheating Constraints and Fragmentation Effects

Introduction

This work undertakes a rigorous analysis of the polynomial $\alpha$-attractor class of inflationary models ("P-models"), focusing on novel methods to constrain their parameter space by exploiting Cosmic Microwave Background (CMB) observables. Unlike approaches relying on an assumed range of inflationary e-foldings, this analysis directly links the reheating temperature $T_\times$—the temperature at the onset of radiation domination—with the CMB spectral index $n_s$ and tensor-to-scalar ratio $r$. The study encompasses both perturbative and non-perturbative reheating regimes, accounting for effects from inflaton fragmentation, and systematically evaluates the compatibility of various P-models with Planck, BICEP/Keck, ACT, and combined data.

Analytical Framework: Connecting CMB Observables and Reheating

The inflationary observables $n_s$, $r$, and $A_s$ are computed in terms of the polynomial potential parameters and the inflaton value at horizon exit $\phi_{k_*}$. The P-model potential,

$$V(\phi,\alpha,n,\Lambda_{inf}) = \Lambda_{inf}^4 \frac{\phi^{2n}}{\phi^{2n} + (\sqrt{3\alpha/2})^{2n}},$$

is analyzed for a range of integer and fractional $\alpha$0. The traditional approach of specifying the number of e-foldings $\alpha$1 is replaced by solving for $\alpha$2 via a consistency condition that traces the comoving pivot scale through inflation, reheating, and radiation era. In particular,

$\alpha$3

with $\alpha$4 set by slow-roll violation, yields precise relations between $\alpha$5, $\alpha$6, $\alpha$7, and the model parameters.

An essential feature of this approach is its utilization of the model-independent bounds on the reheating temperature,

$\alpha$8

with the lower limit from BBN and the upper limit from physical consistency. This translates into sharp constraints on the allowed regions in the $\alpha$9 plane for each $T_\times$0.

Model Behavior Across Polynomial Exponents

$T_\times$1 (Quadratic Potential) and $T_\times$2 (Quartic Potential)

For quadratic $T_\times$3 and quartic $T_\times$4 P-models, fragmentation is subdominant since oscillations in these potentials do not lead to efficient non-perturbative particle production under minimal couplings. In the quartic case, due to $T_\times$5 during both inflaton oscillations and radiation domination, the prediction for $T_\times$6 is independent of $T_\times$7, yielding a unique trajectory in $T_\times$8 space.

For $T_\times$9, $n_s$0 increases with $n_s$1 for fixed $n_s$2; the range of $n_s$3 over allowed $n_s$4 is approximately $n_s$5. The $n_s$6 model predicts a nearly vertical line in the $n_s$7 plane, reflecting the absence of $n_s$8 dependence.

Figure 1: Curves of constant reheating temperature $n_s$9 on the $r$0 plane for $r$1-attractor P-models with $r$2 (left) and $r$3 (right).

Both models accommodate the P-BK-D data robustly, but compatibility with P-BK-D-ACT is more selective, particularly at low $r$4.

$r$5 and $r$6: Effects of Inflaton Fragmentation

For higher polynomials ($r$7), the inflaton oscillates in a potential shallower than quadratic, yielding a stiffer equation of state and facilitating fragmentation. Numerical results indicate that after rapid transfer of energy to relativistic inflaton quanta, $r$8 shifts quickly to $r$9, capping the $n_s$0-dependence. Thus, for each $n_s$1, $n_s$2 falls within a narrow range dictated by the model-independent reheating bounds and fragmentation thresholds.

Figure 2: Curves of constant reheating temperature $n_s$3 on the $n_s$4 plane for $n_s$5 (left) and $n_s$6 (right), with and without fragmentation.

Fragmentation induces a stepwise change in the expansion history, and for $n_s$7 the allowable range for $n_s$8 for a given $n_s$9 is reduced to $r$0--$r$1, increasingly so for larger $r$2. Experimental data requires $r$3 for viability, and upcoming CMB polarization measurements reaching $r$4 will offer strong exclusion potential for such models.

$r$5 and $r$6: Fractional Powers and Enhanced Sensitivity

Fractional exponents ($r$7) exhibit distinct fragmentation dynamics: temporary increases in $r$8 due to particle production are reversed as coherent oscillations re-dominate. Numerical lattice simulations reveal transient behaviors in the energy density and equation-of-state,

Figure 3: Evolution of inflaton energy density components and equation of state $r$9 for $A_s$0 (left) and $A_s$1 (right) during reheating with fragmentation.

Incorporating fragmentation yields substantial shifts in the predicted $A_s$2, especially at low $A_s$3. For instance, $A_s$4 models display a variation in $A_s$5 of almost $A_s$6 across the allowed $A_s$7 (compared to $A_s$8 for $A_s$9), allowing for possible lower bounds on the reheating scale from CMB data that can exceed those from BBN.

Figure 4: Curves of constant reheating temperature $\phi_{k_*}$0 on the $\phi_{k_*}$1 plane for $\phi_{k_*}$2 (left) and $\phi_{k_*}$3 (right), contrasting cases with and without fragmentation.

The minimal allowed values of $\phi_{k_*}$4 (for low $\phi_{k_*}$5) decrease with $\phi_{k_*}$6, reaching as low as $\phi_{k_*}$7 for $\phi_{k_*}$8, further distinguishing these models from integer-power counterparts.

Comparison with Experimental Data

All considered P-models (with $\phi_{k_*}$9) can accommodate current combined data at the $$V(\phi,\alpha,n,\Lambda_{inf}) = \Lambda_{inf}^4 \frac{\phi^{2n}}{\phi^{2n} + (\sqrt{3\alpha/2})^{2n}},$$0 level. However, increasing precision—especially a decrease in the upper bound on $$V(\phi,\alpha,n,\Lambda_{inf}) = \Lambda_{inf}^4 \frac{\phi^{2n}}{\phi^{2n} + (\sqrt{3\alpha/2})^{2n}},$$1—intensifies theoretical constraints, particularly in the presence of non-perturbative reheating phenomena. The inclusion of fragmentation is essential for an accurate determination of theoretical predictions, with the impact varying critically with $$V(\phi,\alpha,n,\Lambda_{inf}) = \Lambda_{inf}^4 \frac{\phi^{2n}}{\phi^{2n} + (\sqrt{3\alpha/2})^{2n}},$$2.

Theoretical and Phenomenological Implications

Directly constraining inflationary parameters via reheating temperatures inferred from CMB observables negates the need for arbitrary assumptions about e-foldings. The explicit dependence on $$V(\phi,\alpha,n,\Lambda_{inf}) = \Lambda_{inf}^4 \frac{\phi^{2n}}{\phi^{2n} + (\sqrt{3\alpha/2})^{2n}},$$3 opens new parameter regimes, particularly relevant for physics beyond the Standard Model tied to the early universe, including baryogenesis and dark matter genesis. This framework allows for imposition of model-specific lower bounds on $$V(\phi,\alpha,n,\Lambda_{inf}) = \Lambda_{inf}^4 \frac{\phi^{2n}}{\phi^{2n} + (\sqrt{3\alpha/2})^{2n}},$$4 from CMB data, which may surpass traditional BBN constraints.

Future CMB B-mode searches (e.g., LiteBIRD, CMB-S4) approaching $$V(\phi,\alpha,n,\Lambda_{inf}) = \Lambda_{inf}^4 \frac{\phi^{2n}}{\phi^{2n} + (\sqrt{3\alpha/2})^{2n}},$$5 are expected to critically probe, and in some cases eliminate, large classes of P-models. The analytic and numerical tools demonstrated here represent an essential methodology for such theoretical-experimental interplay.

Conclusion

This analysis of polynomial $$V(\phi,\alpha,n,\Lambda_{inf}) = \Lambda_{inf}^4 \frac{\phi^{2n}}{\phi^{2n} + (\sqrt{3\alpha/2})^{2n}},$$6-attractor P-models establishes a highly constrained mapping between the reheating temperature and CMB observables. The approach incorporates both perturbative and non-perturbative reheating, with careful attention to inflaton fragmentation effects. The derived bounds on $$V(\phi,\alpha,n,\Lambda_{inf}) = \Lambda_{inf}^4 \frac{\phi^{2n}}{\phi^{2n} + (\sqrt{3\alpha/2})^{2n}},$$7 and $$V(\phi,\alpha,n,\Lambda_{inf}) = \Lambda_{inf}^4 \frac{\phi^{2n}}{\phi^{2n} + (\sqrt{3\alpha/2})^{2n}},$$8—sharpened by physical reheating bounds and fragmentation—are compared systematically with experimental data, indicating that while all considered models remain viable under present constraints, imminent experimental advances will impose strong discriminatory power. This formalism provides a template for future analyses of inflationary models in the high-precision cosmological era.