Collider Signatures of Singlet Extended Higgs Sectors

TL;DR

This paper analyzes collider signatures of Higgs sectors in singlet-extended MSSM frameworks (NMSSM, nMSSM, UMSSM, sMSSM). It demonstrates that singlet mixing can suppress traditional Higgs decay signals while opening invisible decay channels to neutralinos and cascade decays, making some Higgs states detectable only through indirect or invisible means at the LHC. The authors compute production cross sections, signal significances for CMS/ATLAS projections, and quantify invisible decay sensitivity with the observable $\xi_i^2 \equiv \mathrm{BF}(H_i\to inv.)\,\xi_{VVH_i}^2$, and also discuss Higgs coupling measurements at the LHC and ILC for model discrimination. They find that in many scenarios at least one Higgs remains discoverable in visible modes, but in the n/sMSSM a substantial fraction of parameter space yields no direct CP-even Higgs discovery, necessitating invisible or cascade channels. The work outlines a multi-channel program combining visible searches, invisible Higgs analyses, and coupling measurements to distinguish singlet-extended SUSY from the MSSM.

Abstract

We explore the collider signatures of the Higgs sectors in singlet-extended MSSM models. We find that even with reduced couplings due to singlet mixing, a significant portion of the parameter spaces have a discoverable Higgs via traditional decay modes or via invisible decays (directly to neutralinos or through cascade decays to neutralinos and neutrinos). For illustrative points in parameter space we give the likelihood of Higgs discovery. In cases where neither traditional nor invisible modes can discover the Higgs, the neutralino sector may provide evidence for the extended models.

Figures (8)

• Figure 1: Illustrative Higgs composition $(H_d, H_u, S)$ for the models in (a) a decoupled singlet scenario and (b) a strongly mixed singlet scenario. In the decoupled scenario, the extended model has a spectrum similar to that of the MSSM, but contains an additional singlet Higgs that is heavy in the NMSSM and UMSSM and light in the n/sMSSM. Parameters used for this illustration are $\tan \beta = 10$, $s=800$ GeV, $\mu_{\rm eff} = 130$ GeV, $M_2 = 250$ GeV, $A_{\lambda} = 1$ TeV and $\theta_{E_6} = 0.67$ for the UMSSM. The n/sMSSM parameter values are $\tan \beta = 5$, $s=400$ GeV, $\mu_{\rm eff} = 210$ GeV and $M_2 = -140$ GeV. For (a) $\kappa=0.7$ and $A_{\kappa}=-1$ TeV in the NMSSM and $t_F = -0.025$ TeV$^2$ and $t_S =-0.00125$ TeV$^3$ in the n/sMSSM; for (b) $\kappa=-0.11$ and $A_{\kappa}=100$ GeV in the NMSSM and $t_F = -0.0625$ TeV$^2$ and $t_S = -0.0125$ TeV$^3$ in the n/sMSSM and $s=550$ GeV in the UMSSM. $A_{\lambda}, A_{\kappa}$ and $t_S$ are respectively the soft parameters associated with $\lambda, \kappa$ and $t_F$.
• Figure 2: Production cross section of the lightest Higgs in the MSSM and extended MSSM models at the LHC via (a) gluon fusion, (b) Weak Boson Fusion (WBF), (c) Higgstrahlung of a $W$ boson, (d) Higgstrahlung from a $Z$ boson and (e) associated production with top pairs. Most of the MSSM $H_1$ points lie close to the SM curves because they are in the MSSM decoupling limit.
• Figure 3: Signal significance $S/\sqrt{B}$ at CMS ref:cmstdr with 30 fb$^{-1}$ of data in the (a) MSSM (b) NMSSM (c) n/sMSSM (d) UMSSM. The significance of the lightest Higgs is dominated by $gg\to H$ with $H\to \gamma \gamma$ and $H\to ZZ \to 4l$. The $H\to ZZ \to 4l$ mode provides the best signal for heavier Higgs masses, even though these Higgs states are typically singlet or MSSM-like and weakly couple to the $Z$ bosons. No K-factors have been applied as in the ATLAS and CMS analyses.
• Figure 4: Same as in Fig. \ref{['fig:cmssig']}, except for ATLAS ref:atltdr with 100 fb$^{-1}$ of data.
• Figure 5: Sensitivity of the ATLAS detector to the quantity $\xi_i^2$ of Eq. \ref{['eq:invfrac']} for invisibly decaying Higgs bosons. The solid maroon curve indicates the expected sensitivity with 10 fb$^{-1}$ of data while the dashed blue curve represents 30 fb$^{-1}$ of data. The points are predictions of the NMSSM, n/sMSSM and UMSSM models as labeled.