Lattice study of classical inflaton decay

TL;DR

This work numerically investigates non-perturbative inflaton decay into a second scalar via fully nonlinear lattice simulations, revealing that broad parametric resonance is often tamed by backreaction and, crucially, by scatterings when the decay product has non-negligible self-interactions. The authors show a two-stage decay: an initial fast resonant transfer followed by a slow, scattering-dominated phase, with the final outcome strongly dependent on the self-coupling $\lambda_{\chi}$ and on whether the inflaton is massive ($V\propto \phi^2$) or massless ($V\propto \phi^4$). In particular, $\lambda_{\chi}\gg g$ can suppress resonance, drastically prolonging the decay, while in Type II models the inflaton decays into its own fluctuations essentially as if decoupled from $\chi$. The results, extendable to expanding universes via conformal rescaling, have important implications for the occurrence and efficiency of preheating in realistic inflationary scenarios.

Abstract

We study numerically the decay of the inflaton by solving the full non-linear equations of motion on the lattice. We confirm that parametric resonance is effective in transferring energy from the inflaton to a scalar field as long as the self-interactions of the second field are very small. However, in the very broad resonance case (q>>1) the decay rate is limited by scatterings, which significantly slows down the decay. We also find that the inflaton cannot decay via parametric resonance into a scalar field with moderate self-interactions. This means that the preheating stage may be completely absent in many natural inflationary models.

Figures (7)

• Figure 1: The EV of the inflaton as a function of time ($m=10^{-6}\Phi_0$, $\lambda_\chi=\lambda_\phi=0$, $g=10^{-8}$.)
• Figure 2: The variances of the fields as a function of time ($m=10^{-6}\Phi_0$, $\lambda_\chi=\lambda_\phi=0$, $g=10^{-8}$).
• Figure 3: The energy densities of the fields as a function of time ($m=10^{-6}\Phi_0$, $\lambda_\chi=\lambda_\phi=0$, $g=10^{-8}$.)
• Figure 4: $\chi$-field occupation numbers vs. momentum at early times. Three resonances are visible ($t^*=\sqrt{g}\Phi_0 t$).
• Figure 5: $\chi$-field occupation numbers vs. momentum at later times, when scatterings dominate ($t^*=\sqrt{g}\Phi_0 t$).