A string-inspired model for the low-ℓ CMB

We present a semi--analytic exploration of some low--$ell$ angular power spectra inspired by "Brane Supersymmetry Breaking". This mechanism splits Bose and Fermi excitations in String Theory, leaving behind an exponential potential that is just too steep for the inflaton to emerge from the initial singularity while descending it. As a result, the scalar generically bounces against the exponential wall, which typically introduces an infrared depression and a pre--inflationary peak in the power spectrum of scalar perturbations. We elaborate on a possible link between this phenomenon and the low--$ell$ CMB. For the first 32 multipoles, combining the hard exponential with a milder one leading to $n_ssimeq 0.96$ and with a small gaussian bump we have attained a reduction of $chi^,2$ to about 46% of the standard $Lambda$CDM setting, with both WMAP9 and PLANCK 2013 data. This result corresponds to a $chi^,2/DOF$ of about 0.45, to be compared with a $Lambda$CDM value of about 0.85. The preferred choices combine naturally quadrupole depression, a first peak around $ell=5$ and a wide minimum around $ell=20$. We have also gathered some evidence that similar spectra emerge if the hard exponential is combined with more realistic models of inflation. A problem of the preferred examples is their slow convergence to an almost scale--invariant profile.

We present a semi-analytic exploration of some low-ℓ angular power spectra inspired by "brane supersymmetry breaking (BSB)". This mechanism splits Bose and Fermi excitations in string theory, leaving behind an exponential potential that is just too steep for the inflaton to emerge from the initial singularity while descending it. As a result, the scalar generically bounces against the exponential wall, which typically introduces an infrared depression and a pre-inflationary peak in the power spectrum of scalar perturbations. We elaborate on a possible link between this phenomenon and the low-ℓ CMB. For the first 32 multipoles, combining the hard exponential with a milder one leading to ns 0.96 and with a small Gaussian bump, we have attained a reduction of ξ2 to about 46% of the standard ACDM setting, with both WMAP9 and PLANCK 2013 data. This result corresponds to a ξ2/DOF of about 0.45, to be compared with a CDM value of about 0.85. The preferred choices combine naturally quadrupole depression, a first peak around ℓ = 5 and a wide minimum around ℓ = 20. We have also gathered some evidence that similar spectra emerge if the hard exponential is combined with more realistic models of inflation. A problem of the preferred examples is their slow convergence to an almost scale-invariant profile.