Generation of scale invariant magnetic fields in bouncing universes

TL;DR

This paper investigates the generation of scale-invariant primordial magnetic fields in bouncing universe models with non-minimal electromagnetic coupling, using numerical methods to analyze their evolution across the bounce.

Contribution

It introduces a new time variable for numerical analysis and demonstrates the potential for generating observable magnetic fields in bouncing cosmologies.

Findings

Scale-invariant magnetic spectra can form before the bounce for certain parameters.

Magnetic fields of observed strength can be produced in these models.

Backreaction effects may challenge the viability of the scenarios.

Abstract

We consider the generation of primordial magnetic fields in a class of bouncing models when the electromagnetic action is coupled non-minimally to a scalar field that, say, drives the background evolution. For scale factors that have the power law form at very early times and non-minimal couplings which are simple powers of the scale factor, one can easily show that scale invariant spectra for the magnetic field can arise before the bounce for certain values of the indices involved. It will be interesting to examine if these power spectra retain their shape after the bounce. However, analytical solutions for the Fourier modes of the electromagnetic vector potential across the bounce are difficult to obtain. In this work, with the help of a new time variable that we introduce, which we refer to as the ${\rm e}$-${\cal N}$-fold, we investigate these scenarios numerically. Imposing the initial conditions on the modes in the contracting phase, we numerically evolve the modes across the bounce and evaluate the spectra of the electric and magnetic fields at a suitable time after the bounce. As one could have intuitively expected, though the complete spectra depend on the details of the bounce, we find that, under the original conditions, scale invariant spectra of the magnetic fields do arise for wavenumbers much smaller than the scale associated with the bounce. We also show that magnetic fields which correspond to observed strengths today can be generated for specific values of the parameters. But, we find that, at the bounce, the backreaction due to the electromagnetic modes that have been generated can be significantly large calling into question the viability of the model. We briefly discuss the implications of our results.