Towards a Singularity-Free Inflationary Universe?

TL;DR

The paper analyzes whether a singularity-free inflationary history can be achieved in string-inspired gravity by a branch change from a pre-big-bang phase to a standard FRW era. Through phase-space methods, it proves an exact no-go theorem showing that dilaton potentials and stringy fluids cannot catalyze a graceful exit; higher-genus corrections likewise fail to enable the required transition, though they can support inflation if the dilaton decouples in the presence of a cosmological constant, with de Sitter-like behavior emerging asymptotically. Some non-singular, string-frame solutions exist when the conformal coupling B(φ) becomes negative, but their Einstein-frame interpretation involves connected singular histories, limiting their physical viability. Overall, the results strongly constrain graceful-exit scenarios in string cosmology and point to the need for higher-order α' corrections to realize a truly singularity-free inflationary universe.

Abstract

We consider the problem of constructing a non-singular inflationary universe in stringy gravity via branch changing, from a previously superexponentially expanding phase to an FRW-like phase. Our approach is based on the phase space analysis of the dynamics, and we obtain a no-go theorem which rules out the efficient scenario of branch changing catalyzed by dilaton potential and stringy fluid sources. We furthermore consider the effects of string-loop corrections to the gravitational action in the form recently suggested by Damour and Polyakov. These corrections also fail to produce the desired branch change. However, focusing on the possibility that these corrections may decouple the dilaton, we deduce that they may lead to an inflationary expansion in the presence of a cosmological constant, which asymptotically approaches Einstein-deSitter solution.