Vacuum Instabilities with a Wrong-Sign Higgs-Gluon-Gluon Amplitude

TL;DR

This paper investigates the theoretical and phenomenological implications of a wrong-sign Higgs-gluon-gluon amplitude, highlighting issues with vacuum stability and experimental constraints that challenge such scenarios.

Contribution

It provides a comprehensive analysis of vacuum stability and experimental constraints on models with a flipped sign in the Higgs-gluon-gluon coupling.

Findings

Stop loop scenarios are ruled out by vacuum decay and LEP constraints.

Color octet scalars and new fermions with flipped signs are highly constrained.

Most explanations for the sign flip face severe fine-tuning and stability problems.

Abstract

The recently discovered 125 GeV boson appears very similar to a Standard Model Higgs, but with data favoring an enhanced h to gamma gamma rate. A number of groups have found that fits would allow (or, less so after the latest updates, prefer) that the h-t-tbar coupling have the opposite sign. This can be given meaning in the context of an electroweak chiral Lagrangian, but it might also be interpreted to mean that a new colored and charged particle runs in loops and produces the opposite-sign hGG amplitude to that generated by integrating out the top, as well as a contribution reinforcing the W-loop contribution to hFF. In order to not suppress the rate of h to WW and h to ZZ, which appear to be approximately Standard Model-like, one would need the loop to "overshoot," not only canceling the top contribution but producing an opposite-sign hGG vertex of about the same magnitude as that in the SM. We argue that most such explanations have severe problems with fine-tuning and, more importantly, vacuum stability. In particular, the case of stop loops producing an opposite-sign hGG vertex of the same size as the Standard Model one is ruled out by a combination of vacuum decay bounds and LEP constraints. We also show that scenarios with a sign flip from loops of color octet charged scalars or new fermionic states are highly constrained.