Semi-visible higgs decay as a probe for new invisible particles

TL;DR

This work assesses HL-LHC sensitivity to new invisible particles produced in semi-visible Higgs decays within a dark-SMEFT framework up to dimension six and a Z2 symmetry. By focusing on associated Higgs production with ZH and H decays to leptons or jets plus missing energy, the authors derive perturbative unitarity bounds, impose Z invisible width constraints, and implement both cut-based and boosted decision tree analyses to distinguish operator structures. They find that derivative current operators are often constrained by the Z width at low DM masses, while Yukawa-type operators are most effectively probed by semi-visible Higgs decays, with multivariate techniques significantly enhancing sensitivity. The results show that certain semi-visible Higgs channels can reach branching fractions below the Higgs-neutrino floor and provide complementary information to direct and indirect dark matter searches, particularly for asymmetric or co-annihilating DM scenarios.

Abstract

We discuss the HL-LHC sensitivity to probe new invisible particles including scalars and fermions using semi-visible Higgs decays. The kinematics of these decays allow new particle masses below $m\lesssim 50$ GeV. We carry out our analysis within the framework of a dark-SMEFT effective theory with operators up to dimension six and a discrete $\mathbb{Z}_2$ symmetry under which the new particles are odd and the SM particles are even. We compare our results to those obtained from considering the invisible $Z$-width, as well as simple perturbative unitarity arguments. Finally, we outline kinematic strategies at the LHC to distinguish different operator structures of the postulated invisible particles.

Figures (10)

• Figure 1: Signal final states explored in this study for the semi-visible leptonic (left) and hadronic (right) Higgs decays.
• Figure 2: Representative Feynman diagrams illustrating semi-visible Higgs decays, independent of the production mode. The top panel shows the SM process, while the bottom panel presents possible EFT-induced contributions. The cross-hatched circles represent insertions of DSMEFT operators.
• Figure 3: Reconstructed invariant mass of the two leptons (top left), transverse mass of the two leptons and $\rm E{\!\!\!/}_T$ (top right), parton-level invariant mass of the invisible part: $\nu{\overline\nu}$ for SM, $\phi\phi$ for DSMEFT (center left), $\rm \rm E{\!\!\!/}_T$-distribution (center right), invariant mass of the reconstructed jets (bottom left), and transverse mass of the two jets and $\rm E{\!\!\!/}_T$ (bottom right). All histograms are normalized to unit area and thus show shape only. We use $C=1$, $\Lambda=1$ TeV, and $\sqrt{s}=14~{\textrm{TeV}}$ in panels.
• Figure 4: Transverse momentum of the leading (left) and sub-leading lepton (right) for SM and different DSMEFT operators. $C=1$, $\Lambda=1$ TeV, and $\sqrt{s}=14~{\textrm{TeV}}$ in the plots. All histograms are normalized to unit area and thus show shape only.
• Figure 5: Distributions of the invariant mass of the two leptons (top left) and the transverse mass of the two leptons and $\rm E{\!\!\!/}_T$ (top right) and $\rm E{\!\!\!/}_T$ (bottom). $C=1$, $\Lambda=1$ TeV, and $\sqrt{s}=14$ TeV. All histograms are normalized to unit area and thus show shape only.