Global fits and the 95 GeV diphoton excesses in the Supersymmetric Georgi-Machacek Model

TL;DR

This work tests whether the 95 GeV diphoton excesses can be explained within the supersymmetric Georgi-Machacek (SGM) model, which preserves the GM-like custodial Higgs sector but includes custodial higgsinos. A comprehensive global fit—incorporating the 125 GeV Higgs data, the 95 GeV diphoton signal via a light custodial singlet, Drell-Yan Higgs pair production, and ttH3±→τν constraints—shows compatibility only if the 95 GeV resonance is the lighter custodial singlet H, contributing about $5-7\%$ to electroweak symmetry breaking and with small VEV/mixing angles. The analysis predicts a tightly constrained spectrum: a fiveplet around $(185-195)\,\text{GeV}$ (SGM) and $(210-220)\,\text{GeV}$ (CGM), a triplet around $(133-140)\,\text{GeV}$ (SGM) versus $(145-150)\,\text{GeV}$ (CGM), and a custodial Higgsino sector with a lightest neutral state in the $(117-135)\,\text{GeV}$ range, forming a compressed spectrum that is challenging to probe experimentally. The Drell-Yan channel remains the dominant production mechanism for the 95 GeV state in the allowed region, distinguishing the SGM from the non-supersymmetric CGM, and providing concrete targets for future collider searches to test the custodial Higgsino–influenced GM-like sector.

Abstract

Recently the ATLAS and CMS experiments have reported modest excesses in the diphoton channel at around 95 GeV.~A number of recent studies have examined whether these could be due to an extended electroweak symmetry breaking (EWSB) sector, including the well known Georgi-Machacek (GM) model.~Here we examine whether the excesses can be explained by a light exotic Higgs boson in the \emph{Supersymmetric} GM (SGM) model which has the same scalar spectrum as the conventional GM model, but with a more constrained Higgs potential and the presence of custodial Higgsino fermions.~We perform a global fit of the SGM model including all relevant production and decay channels, some of which have been neglected in previous studies, which severely constrain the parameter space.~We find that the SGM model can fit the data if the LHC diphoton excesses at 95\,GeV are due to the lightest custodial singlet Higgs boson which contributes $(5-7)\%$ to EWSB, but \emph{cannot} accommodate the LEP $b\bar{b}$ excess, in contrast to other recent studies of the GM model.~Since the SGM model has a highly constrained Higgs potential, the rest of the mass spectrum is sharply predicted, allowing for targeted searches at the LHC or future colliders.~We also compare the SGM model with the non-supersymmetric GM model and identify how they can be distinguished at the LHC or future colliders.

Figures (8)

• Figure 1: Results of the (three dimensional) parameter scan over Higgs potential parameters $(\lambda_2,\,\lambda_4,\,M_1,\,M_2,\,v_\phi,\,v_\Delta)$ for the SGM (yellow) and CGM (green) models.
• Figure 2: Results of the scan for the $\mu$ terms of the superpotential in the SGM model, related to the $M_1$ and $M_2$ mass parameters as in Eq. \ref{['eq:muterms']}, in the $(\mu_\Delta$ vs. $\mu)$ plane.
• Figure 3: Left: Constraints on $(\kappa_V^h$ vs. $\kappa_f^h)$ coming from measurements of the 125 GeV Higgs boson at ATLAS (blue) and CMS (red). We also show the allowed parameter (as consistent with data as the SM) space in the SGM (yellow) and CGM (green) models. Right: Allowed parameter space for the SGM (yellow) and CGM (green) models in the electroweak triplet VEV ($s_H$) versus custodial singlet Higgs mixing angle ($s_\alpha$) plane with the mixing angles defined in Eq. \ref{['eq:hHmixing']}.
• Figure 4: Drell-Yan Higgs pair production of $H$ ($m_H = 95$ GeV) as a function of the custodial triplet mass $m_3$ for both the $W$ (blue) and $Z$ (orange) mediated channels. We also show the VBF + VH for $\kappa_V^H = 0.2$ (green) and gluon fusion for $\kappa_f^H = 0.1$ (red) production cross sections.
• Figure 5: Allowed parameter space for the SGM (yellow) and CGM (green) models in the planes of $(\kappa_f^H,\kappa_V^H)$ (upper left), ($\mathcal{B}_{bb}^H,\mathcal{B}_{\gamma\gamma}^H$) (upper right), and $(r_{\rm DY},r_V)$ (lower).