The IR obstruction to UV completion for Dante's Inferno model with higher-dimensional gauge theory origin

Kazuyuki Furuuchi,Yoji Koyama

doi:10.1088/1475-7516/2016/06/037

Abstract

We continue our investigation of large field inflation models obtained from higher-dimensional gauge theories, initiated in our previous study [1]. We focus on Dante's Inferno model which was the most preferred model in our previous analysis. We point out the relevance of the IR obstruction to UV completion, which constrains the form of the potential of the massive vector field, under the current observational upper bound on the tensor to scalar ratio. We also show that in simple examples of the potential arising from DBI action of a D5-brane and that of an NS5-brane that the inflation takes place in the field range which is within the convergence radius of the Taylor expansion. This is in contrast to the well known examples of axion monodromy inflation where inflaton takes place outside the convergence radius of the Taylor expansion. This difference arises from the very essence of Dante's Inferno model that the effective inflaton potential is stretched in the inflaton field direction compared with the potential for the original field.

Highlights

Effective field theories1 allow us to make predictions with desired accuracy without knowing the full details of the underlying UV theory
Using the parameter values allowed by the CMB data obtained in section 2, we show in simple examples that the inflation takes place in the field range which is within the convergence radius of the Taylor expansion of the DBI action
In Dante’s Inferno model, which is most promising in this class of models, we have shown that the constraint on the sign of the quartic term in the potential of the massive vector field is in favor of the current observational upper bound on tensor-to-scalar ratio

Summary

Introduction

Effective field theories allow us to make predictions with desired accuracy without knowing the full details of the underlying UV theory. In [24] it has been shown that massive vector field theories which can be embedded to a UV theory whose S-matrix satisfies unitarity and canonical analyticity constraints do not have a Lorentz-symmetrybreaking vacuum In such theories, before the potential starts to go down, the contribution from higher order terms in the potential should come in to prevent Lorentz-symmetrybreaking local minimum, assuming that the potential is bounded from below. This figure should be looked together with the condition (2.16), 2πL5 1 × 102, which we have imposed to justify neglecting the contribution from the gauge field Vg(A) to the one-loop effective potential. (2.35) gives only a very mild constraint gA ≫ O(10−5), which is weaker than the bound given from Fig. 5 and (2.46)

Dante’s Inferno model with DBI action of a 5-brane

Summary and discussions

Calculation of the one-loop effective potential