Erratum: Ricci reheating reloaded

We rectify an erroneous rescaling factor that affects the last coefficient of the parametric formulas in equations (2.28), (2.29) and (2.34). New figures are included in order to correct for the mismatch. For a smoother read, we refer the reader to the updated arXiv version of this work arXiv:2307.03774.


JCAP06(2024)E01
The original version of this work contained an erroneous rescaling factor in the conversion from lattice quantities in χ kin -units to plotted quantities in H kin -units.The shape of the parametric formulas (2.26), (2.27), (2.31) and (2.33) is unaltered ⟨χ 2 br (λ, ν)⟩ rms = 4H 2 kin exp (α 1 + α 2 ν + α 3 ln ν) , ρ χ br (λ, ν) = 16H 4 kin exp (β 1 + β 2 ν + β 3 ln ν) , z rad (λ, ν) = γ 1 + γ 2 ν , ρ χ rad (λ, ν) = 16H 4 kin exp (δ 1 + δ 2 ν + δ 3 ln ν) , while the coefficients in (2.28), (2.29), (2.32) and (2.34) are now with n = − log λ.Note that this modification influences some rough estimates present in the body of the paper.In particular, the typical values for the heating temperatures are now of the order T ht ∼ 10 5 − 10 12 GeV for H kin = 10 11 GeV (see eq. (3.4)) and the number of inflationary e-foldings in eq.(3.9) is now 58.0 < N < 63.8.In spite of affecting some figures in the article (included here for the sake of clarity), this discrepancy does not alter neither the methodology we have followed, nor its final accomplishments.We also noted a minor typographical error in the writing of eq.(3.2), which should read Deriving this expression from the definition of heating efficiency in (3.1) with the help of (2.27) would have given the correct final form.The same is true for (3.4) and (3.6), which do not contain the a kin factor: For an easier and smoother reading of the work, we refer the reader to the arXiv version of this article arXiv:2307.03774, which contains the updated formulas and figures.

Figure 1 .Figure 2 .Figure 3 .Figure 4 .
Figure1.Evolution of the volume-averaged kinetic (K), gradient (G), potential (V ) and interaction (I) contributions in (2.23)) to the total energy density ρ χ for exemplary values ν = 10 and λ = 10 −4 .Each energy component is multiplied by a 4 to highlight asymptotic radiation-like behaviours.In order to facilitate the comparison, we display the absolute values of the interaction and the total-energy terms, since they are not positive definite at all times.The timescales identifying the backreaction time z br and the virialisation time z rad for this specific case are indicated at the top of the image.

Figure 5 .
Figure 5. Ratio of the energy density stored in gradients and the total energy density of the spectator field at z rad , as a function of the model parameters ν and λ.

Figure 6 .
Figure 6.Parametric dependence of the heating efficiency Θ ht on the model parameters λ and ν for a fiducial energy scale of kination H kin = 10 11 GeV.Solid lines correspond to the values computed from the simulations scanning of the parameter space, while the dashed lines correspond to the fitting formula (3.2).

Figure 7 .
Figure 7. Predicted values of tensor-to-scalar ratio and spectral tilt in an α-attractor quintessentialinflation scenario involving only gravitational particle production.Lines of different colours correspond to different values of the H kin parameter, where the end points correspond to the largest (left) and smallest (right) heating efficiencies achieved in our simulations.The vertical grey lines delimit the region allowed by the numerically-obtained heating efficiency.The results are shown in comparison with the constraints from Planck + Bicep/Keck at one and two sigma.

Figure 8 .Figure 9 .Figure 10 .
Figure 8. Energy components of the model as defined in (2.23).Each component has been multiplied by a 4 in order to highlight the radiation-like behaviour.In this figure, the model parameters have been set to ν = 20, λ = 10 −3 , g = 10 −3 .