Search for the production of a Higgs boson in association with a single top quark in $pp$ collisions at $\sqrt{s} = 13$ TeV with the ATLAS detector

TL;DR

This ATLAS study searches for Higgs boson production in association with a single top quark in $pp$ collisions at $\sqrt{s}=13$ TeV using $140\ \mathrm{fb}^{-1}$. It employs three orthogonal channels with multivariate discriminants to extract the signal strength $\mu_{tH}$ and constrain dominant backgrounds via a global profile likelihood fit, reporting a measured value of $\mu_{tH}=8.1\pm2.6\ \mathrm{stat}\pm2.0\ \mathrm{syst}$ and a significance of $2.8$ standard deviations for the SM hypothesis. The analysis also interprets an inverted top-quark Yukawa coupling sign, obtaining $\mu_{tH}(\kappa_t=-1)=1.2\pm0.4\ \mathrm{stat}\pm0.5\ \mathrm{syst}$ with a 95\% CL upper limit of 2.4, indicating compatibility with SM expectations but some data preference for the inverted-sign scenario. The study demonstrates the power of a multi-channel, MVA-driven approach to probe the Higgs–top coupling and its sign in the $tH$ production mode, with implications for precision tests of the SM and potential new physics in the Higgs sector.

Abstract

A search for the production of a Higgs boson in association with a single top quark, $tH$, is presented. The analysis uses proton-proton collision data corresponding to an integrated luminosity of $140\mathrm{fb}^{-1}$ at a centre-of-mass energy of $13$ TeV, collected by the ATLAS detector at the LHC. The search targets Higgs-boson decays into $b\bar{b}$, $WW^{*}$, $ZZ^{*}$, and $ττ$, accompanied by an isolated lepton (electron or muon) from the top-quark decay. Multivariate techniques are employed to enhance the separation between signal and background processes. The observed signal strength, $μ_{tH}$, defined as the ratio between the measured cross-section and the predicted Standard Model value, is $μ_{tH}~=~8.1~\pm~2.6~\mathrm{(stat.)}~\pm~2.0~\mathrm{(syst.)}$. The significance of the observed (expected) signal above the background-only expectation is $2.8$ ($0.4$) standard deviations. The corresponding observed (expected) upper limit at the 95% confidence level on the $tH$ cross-section is found to be $13.9$ ($6.1$) times the value predicted by the Standard Model. An interpretation with an inverted sign of the top-quark Yukawa coupling is performed, and the signal strength and corresponding limit are reported.

Figures (8)

• Figure 1: Representative dominant Feynman diagrams for (a) and (b) production at LO, in the 4FS. The event topology (a) is characterised by two heavy objects, the top quark and the Higgs boson plus an energetic light-flavour quark (dominantly a $u$-quark), referred to as the spectator quark, and an extra soft from gluon splitting. The event topology (b) is characterised by three heavy objects, the top quark, the Higgs boson and a boson and an extra soft from gluon splitting.
• Figure 2: The data and estimated signal-plus-background yields in the SR and CRs of the (a) $H\rightarrow b\bar{b}$ channel and (b) $3\Pl$ and $2\Pl\,\text{SS}$ channels. The term $e/\mu_\text{Fakes}$ refers to backgrounds from non-prompt or misidentified leptons, namely contributions from heavy-flavour decays and photon conversions. Both signal and background events are estimated with a likelihood fit to data ("Post-Fit"), as described in \ref{['sec:back', 'sec:result']}. The uncertainty band includes statistical and systematic contributions.
• Figure 3: The (a) fitted values of the $\mu_{\Pqt{}\PH}$ and (b) 95% CL upper limits in the individual channels and in the combined measurement. The values reported for the individual channels are obtained by fitting $\mu_{\Pqt{}\PH}$ separately for each channel, whereas the combined measurement is derived by simultaneously fitting all channels together. The observed limits are shown (solid black lines), together with the expected limits under the SM hypothesis (dotted black lines). A grey line is added in correspondence of the $\mu_{\Pqt{}\PH}$ equal to one. In the case of the expected limits, the one- and two-standard-deviation uncertainty bands are also shown.
• Figure 4: The impact of the top 20 most important systematic uncertainties on the fitted $\mu_{\Pqt{}\PH}$ in the SM scenario. The uncertainties are listed in decreasing order of their impact on $\mu_{\Pqt{}\PH}$. The filled boxes show the relative impact on $\mu_{\Pqt{}\PH}$, referring to the top $x$-axis. The points, referring to the bottom $x$-axis, show the deviations on the fitted NPs $\boldsymbol{\theta}_{0}$ from their nominal values, expressed in terms of standard deviations with respect to their nominal uncertainties. The associated black uncertainty bars show the fitted uncertainties of the NPs, relative to their nominal uncertainties. Filled markers refer to systematic uncertainties, while empty black markers refer to statistical uncertainties in the MC simulation in a given bin of an analysis region and they are labelled as $\upgamma$. The results shown are extracted from the fit including all channels.
• Figure 5: Comparison between data and the signal-plus-background prediction for the event yields in the (a) $H\rightarrow b\bar{b}$, (b) $2\Pl\,\text{SS}$ and (c) $3\Pl$ SRs. The term $e/\mu_\text{Fakes}$ refers to backgrounds from non-prompt or misidentified leptons, namely contributions from heavy-flavour decays and photon conversions. The SM signal and the background contributions after the likelihood fit to data ("Post-Fit") under the signal-plus-background hypothesis are shown as filled histograms. A dashed red line is added to the plot to show the signal contribution normalised to the measured value multiplied by 50 for the $H\rightarrow b\bar{b}$ channel and by 3 for the $2\Pl\,\text{SS}$ and $3\Pl$ channels. The ratio of the data to the total post-fit prediction ("Pred.") is shown in the lower panel. The combined statistical and systematic uncertainty in the prediction is indicated by the grey hatched band.