Could the 650 GeV Excess be a Pseudoscalar of a 3-Higgs Doublet Model?

TL;DR

The paper investigates whether a CP-odd Higgs $A$ with $m_A \approx 650$ GeV from an I(1+2)HDM can explain the CMS 650 GeV di-photon plus $b\bar b$ excess via $A \to h_{125} Z$ with $h_{125} \to \gamma\gamma$ and $Z \to b\bar b$, while also addressing the 95 GeV hints through a lighter $h_{95}$ and loop-enhanced $h_{95} \to \gamma\gamma$ from the inert sector. The framework uses a three-Higgs-doublet-like structure with two active and one inert doublet, Type-I Yukawas, and a rich scalar spectrum including $h_{95}$, $h_{125}$, $A$, $H^{\pm}$, and inert states, constrained by theory, collider searches, EW precision data, and dark matter experiments. A detailed parameter scan shows viable regions where the DM relic density is compatible with Planck and direct-detection bounds, and where $\sigma_{\gamma\gamma b\bar{b}}$ matches the CMS excess for $m_A$ in the 600–700 GeV range, with $\tan\beta \lesssim 5$ and $0.27 \lesssim \sin(\beta-\alpha) \lesssim 0.47$; BR$(A \to h_{95} Z)$ can dominate (around 79%), with non-negligible BRs to $h_{125} Z$ and $t\bar t$, all while predicting collateral signatures such as $pp \to A \to h_{95} Z$ and $pp \to A \to t\bar t$ for future tests. This approach ties together the 95 GeV and 650 GeV anomalies with DM phenomenology, offering concrete experimental targets for Run 3 and HL-LHC.

Abstract

In this study, we propose the interpretation of a 650 GeV excess observed at the Large Hadron Collider (LHC) by the CMS Collaboration in terms of the production of a CP-odd (or pseudoscalar) Higgs boson A, with mass around 650 GeV, decaying into the Standard Model (SM)-like Higgs state $h_{125}$ (in turn decaying into $γγ$) and a Z boson (in turn decaying into $b\bar b$), within a 3-Higgs Doublet Model (3HDM) featuring two active and one inert doublet, known as the I(1+2)HDM. This theoretical structure features a spectrum with both the SM-like Higgs boson (with a 125 GeV mass) and a lighter CP-even (or scalar) Higgs state with mass around 95 GeV, $h_{95}$, which is present in this scenario for the purpose of simultaneously explaining anomalies seen in the $b\bar b$, $γγ$ and $τ^+τ^-$ final states in searches for additional light Higgs states at the Large Electron-Positron (LEP) collider and LHC itself. It should be noted that, in the I(1+2)HDM, the inert sector presents loop-induced enhancements to the $h_{95} \to γγ$ width via inert charged Higgs states, providing a viable mechanism to explain, in particular, the observed (and most significant) di-photon excess at 95 GeV. Taking into account both experimental and theoretical constraints, our results can not only explain the aforementioned anomalies (possibly, aside from the $τ^+τ^-$, which is the most marginal one) but also predict, as collateral signals, resonant production of the same CP-odd scalar A followed by the decays: (i) $A \to h_{95} \, Z$, leading to the same $γγb \bar{b}$ final state displaying the original 650 GeV anomaly and (ii) $A\to t\bar t$, leading to a well-known and studied signature.

Figures (4)

• Figure 1: The relic density $\Omega h^2$ (left) and the cross section $\sigma_{\gamma\gamma b \bar{b}}$ (right) shown as a scatter plot in the $(m_{\chi},\,\sigma_{SI})$ plane. The XENONnT XENON:2023cxc and LUX-ZEPLIN LZ:2022lsvLZ:2024zvo upper limits are shown with solid, dotted and dashed lines respectively.
• Figure 2: The $\sigma_{\gamma\gamma b\bar{b}}$ values as function of $\tan\beta$ (left), $\sin({\beta-\alpha})$ (middle) and $m_A$ (right). The orange horizontal band corresponds to the value of the $\gamma\gamma b\bar{b}$ cross section consistent with the experimental measurement within $[-2.5\sigma,\,0.5\sigma]$.
• Figure 3: The scatter plots of the production cross section $\sigma_{\gamma\gamma b \bar{b}}$, as a function of the signal strengths $\mu_{\gamma\gamma}(h_{95})$ (top) and $\mu_{b\bar{b}}(h_{95})$ (bottom). The colour coding indicates one of the following BRs: $\text{BR}(A \to h_{95})$ (left), $\text{BR}(A \to h_{125} Z)$ (middle) and $\text{BR}(A \to t\bar{t})$ (rigtht). Furthermore, in orange(cyan) horizontal(vertical) band represents the measured value of $\sigma_{\gamma\gamma b \bar{b}}$ (signal strengths $\mu_{\gamma\gamma}(h_{95})$ and $\mu_{b\bar{b}}(h)$).
• Figure 4: The $\sigma({pp\to X\to HY\to \gamma\gamma \tau^+\tau^-})$ values as function of $m_X$ as obtained for the I(1+2)HDM points explaining the 650 and 95 GeV anomalies. Notice that, in such a scenario, one has $X=A$, $Y=Z$ and $H=h_{125}$.