Investigating the 95 GeV Higgs Boson Excesses within the I(1+2)HDM

In this work, we explore how the 2-Higgs Doublet Model (2HDM) Type-I, extended by an inert doublet, can provide an explanation for the recently observed excesses at the Large Hadron Collider (LHC) in the $γγ$ and $τ^+ τ^- $ final states. Hence, by imposing theoretical constraints and experimental bounds on the model parameter space, our findings show that a light CP-even Higgs boson, $h$, with a mass around 95 GeV, can account for these anomalies. This result aligns with the excess in $b\bar b$ signatures reported in earlier data from the Large Electron-Positron (LEP) collider.

💡 Research Summary

In this paper the authors investigate whether the excesses observed around a mass of 95 GeV in several LHC and LEP channels can be simultaneously explained within a three‑Higgs‑doublet framework called the I(1+2)HDM. The model consists of two active SU(2) doublets (Φ₁, Φ₂) and one inert doublet (η) that is odd under a Z₂ symmetry, so that η does not acquire a vacuum expectation value and its scalar components (χ, χₐ, χ⁺) can serve as dark‑matter candidates. The active sector follows a Type‑I Yukawa structure, meaning that all fermions couple only to Φ₂, and the CP‑conserving scalar potential contains twelve independent parameters after imposing a “dark democracy” reduction of the inert quartic couplings.

The authors impose a comprehensive set of theoretical constraints: perturbativity of all quartic couplings (|ρ_i| ≤ 4π), tree‑level unitarity (eigenvalues of 2→2 scalar scattering matrices ≤ 8π), and vacuum stability (positivity conditions on ρ₁, ρ₂, ρ_η and mixed inequalities). Experimental constraints include electroweak precision observables (S, T parameters with ΔS = −0.05 ± 0.07, ΔT = 0.00 ± 0.06, U = 0), the measured signal strengths of the 125 GeV Higgs boson (using HiggsSignals‑3, requiring χ² within 95 % C.L.), non‑SM Higgs searches (HiggsBounds‑6, keeping points with observed limit ratios below 0.95), and flavour observables (B → X_sγ, B_s → μ⁺μ⁻, B → τν, etc., via SuperIso).

Signal strengths for the putative 95 GeV scalar ϕ are expressed in the narrow‑width approximation as μ_{b b̄}=c_{ϕZZ}²·BR(ϕ→b b̄)/BR_SM, μ_{γγ}=c_{ϕtt}²·BR(ϕ→γγ)/BR_SM, and μ_{ττ}=c_{ϕtt}²·BR(ϕ→ττ)/BR_SM, where c_{ϕZZ} and c_{ϕtt} are the reduced couplings to Z bosons and top quarks, respectively. The experimental values to be reproduced are μ_LEP^{b b̄}=0.117 ± 0.057, μ_CMS^{ττ}=1.2 ± 0.5, and a combined μ_{γγ}=0.24 ± 0.09 from ATLAS and CMS.

A dedicated numerical scan is performed with a Fortran implementation of the I(1+2)HDM Type‑I. The scan ranges (Table 2) allow m_h∈