Towards testing the general bounce cosmology with the CMB B-mode auto-bispectrum

TL;DR

This paper investigates the potential detectability of the B-mode polarization bispectrum in bounce cosmology models with non-minimal coupling, showing that upcoming CMB experiments could test these models.

Contribution

It demonstrates that non-minimal derivative coupling in bounce models can produce a detectable B-mode bispectrum signal in the CMB, unlike minimal models.

Findings

Signal-to-noise ratio can reach unity for  with =100.

The bispectrum amplitude can be significantly larger than in minimal models.

Upcoming CMB experiments have the potential to test these bounce cosmology models.

Abstract

It has been shown that a three-point correlation function of tensor perturbations from a bounce model in general relativity with a minimally-coupled scalar field is highly suppressed, and the resultant three-point function of cosmic microwave background (CMB) B-mode polarizations is too small to be detected by CMB experiments. On the other hand, bounce models in a more general class with a non-minimal derivative coupling between a scalar field and gravity can predict the three-point correlation function of the tensor perturbations without any suppression, the amplitude of which is allowed to be much larger than that in general relativity. In this paper, we evaluate the three-point function of the B-mode polarizations from the general bounce cosmology with the non-minimal coupling and show that a signal-to-noise ratio of the B-mode auto-bispectrum in the general class can reach unity for $\ell_{\rm max}\geq9$ and increase up to $5.39$ for $\ell_{\rm max}=100$ in the full-sky case. We also discuss the possibility to test the general class of bounce models by upcoming CMB experiments.