Status of invisible Higgs decays

TL;DR

The paper investigates how much of the Higgs-like state’s decays could be invisible or undetected, given LHC and Tevatron data up to 2012. It uses global fits with parametrized Higgs couplings and potential invisible branching fractions across three scenarios: SM-like couplings with invisibles, SM-like couplings plus new loop contributions, and general free couplings for fermions and vectors. The results show that invisible decays are tightly constrained for SM-like couplings (about 23% at 95% CL) but can be substantially larger when couplings deviate or new particles alter loop effects (up to ~60%), with even tighter bounds under additional positivity conditions (around 0.36). The work also connects collider constraints to dark matter searches, illustrating how monojet/VBF probes and XENON100 data jointly limit the Higgs’ invisible width and its DM portal implications, and it outlines the experimental prospects required to further advance these limits.

Abstract

We analyze the extent to which the LHC and Tevatron results as of the end of 2012 constrain invisible (or undetected) decays of the Higgs boson-like state at ~ 125 GeV. To this end we perform global fits for several cases: 1) a Higgs boson with Standard Model (SM) couplings but additional invisible decay modes; 2) SM couplings to fermions and vector bosons, but allowing for additional new particles modifying the effective Higgs couplings to gluons and photons; 3) no new particles in the loops but tree-level Higgs couplings to the up-quarks, down-quarks and vector bosons, relative to the SM, treated as free parameters. We find that in the three cases invisible decay rates of 23%, 61%, 88%, respectively, are consistent with current data at 95% confidence level (CL). Limiting the coupling to vector bosons, CV, to CV < 1 in case 3) reduces the allowed invisible branching ratio to 56% at 95% CL. Requiring in addition that the Higgs couplings to quarks have the same sign as in the SM, an invisible rate of up to 36% is allowed at 95% CL. We also discuss direct probes of invisible Higgs decays, as well as the interplay with dark matter searches.

Figures (8)

• Figure 1: $\Delta\chi^2$ distributions for the branching ratio of invisible Higgs decays. The full, dashed, and dotted lines correspond, respectively, to the cases of 1) SM couplings, 2) arbitrary $\Delta C_g$ and $\Delta C_\gamma$, and 3) deviations of $C_U,C_D,C_V$ from unity. In addition, we show as dash-dotted line the variant of case 3) with $C_U,C_D>0$ and $C_V\le 1$.
• Figure 2: ${\cal B}(H\to {\rm invisible})$ contours obtained by allowing for additional loop contributions $\Delta C_g$ and $\Delta C_\gamma$ to the effective couplings of the Higgs to gluons and photons. The red, orange and yellow areas (black, grey and light grey areas in greyscale print) are the 68%, 95% and 99.7% CL regions, respectively. The black and grey ellipses in the top plot show the 68% and 95% CL contours when invisible decays are absent. We find $\Delta C_g=-0.01_{-0.16}^{+0.24}$, $\Delta C_\gamma= 0.45_{-0.17}^{+0.19}$ and ${\cal B}(H\to {\rm invisible})=0.13_{-0.13}^{+0.24}$ (where the errors correspond to 1d profiling). At the best fit point, marked as a white star, we find $\chi^2_{\rm min}=12.03$ (for 18 d.o.f.).
• Figure 3: Fit of ${\cal B}(H\to {\rm invisible})$ allowing for deviations of $C_U$, $C_D$, $C_V$ from 1, but without extra loop contributions, i.e.$\Delta C_g=\Delta C_\gamma=0$. Same color code as in Fig. \ref{['fig:delta-cg-cp']}. Allowing up to 100% deviations in $C_U$ and $C_V$, the $2\sigma$ ($3\sigma$) limit is ${\cal B}(H\to {\rm invisible})<0.88$ (0.92).
• Figure 4: Same as Fig. \ref{['fig:cu-cd-cv']} but restricting either $C_V\le1$ (left) or $C_U>0$ (right).
• Figure 5: ${\cal B}(H\to {\rm invisible})$ contours obtained by requiring $C_V\le1$ and $C_U,C_D>0$. Same color code as in previous figures.