Search for associated production of a Higgs boson and a single top quark in proton-proton collisions at $\sqrt{s}=13\mathrm{TeV}$

A search is presented for the production of a Higgs boson in association with a single top quark, based on data collected in 2016 by the CMS experiment at the LHC at a center-of-mass energy of 13 TeV, which corresponds to an integrated luminosity of $35.9{\mathrm{fb}}^{-1}$. The production cross section for this process is highly sensitive to the absolute values of the top quark Yukawa coupling, $y_t$; the Higgs boson coupling to vector bosons, $g_{H\mathrm{VV}}$; and, uniquely, their relative sign. Analyses using multilepton signatures, targeting $H\to WW$, $H\to \tau \tau$, and $H\to ZZ$ decay modes, and signatures with a single lepton and a $b\overline{b}$ pair, targeting the $H\to b\overline{b}$ decay, are combined with a reinterpretation of a measurement in the $H\to \gamma \gamma$ channel to constrain $y_t$. For a standard model–like value of $g_{H\mathrm{VV}}$, the data favor positive values of $y_t$ and exclude values of $y_t$ below about $-0.9{y_t}^{\mathrm{SM}}$.

Figure 1

Leading-order Feynman diagrams for the associated production of a single top quark and a Higgs boson in the $t$ channel, where the Higgs boson couples to either the top quark (left) or the $W$ boson (center), and one representative diagram of $tHW$ production, where the Higgs boson couples to the top quark (right).

Figure 2

Distributions of discriminating observables for the same-sign ${\mu }^{\pm }{\mu }^{\pm }$ channel, normalized to $35.9{\mathrm{fb}}^{-1}$, before fitting the signal discriminant to the data. The grey band represents the unconstrained (prefit) statistical and systematic uncertainties. In the panel below each distribution, the ratio of the observed and predicted event yields is shown. The shape of the two $tH$ signals for ${\kappa }_t=-1.0$ is shown, normalized to their respective cross sections for ${\kappa }_t=-1.0$, ${\kappa }_{\mathrm{V}}=1.0$.

Figure 3

Distributions of discriminating observables for the same-sign $e^{\pm }{\mu }^{\pm }$ channel, normalized to $35.9{\mathrm{fb}}^{-1}$, before fitting the signal discriminant to the data. The grey band represents the unconstrained (prefit) statistical and systematic uncertainties. In the panel below each distribution, the ratio of the observed and predicted event yields is shown. The shape of the two $tH$ signals for ${\kappa }_t=-1.0$ is shown, normalized to their respective cross sections for ${\kappa }_t=-1.0$, ${\kappa }_{\mathrm{V}}=1.0$.

Figure 4

Distributions of discriminating observables for the three-lepton channel, normalized to $35.9{\mathrm{fb}}^{-1}$, before fitting the signal discriminant to the data. The grey band represents the unconstrained (prefit) statistical and systematic uncertainties. In the panel below each distribution, the ratio of the observed and predicted event yields is shown. The shape of the two $tH$ signals for ${\kappa }_t=-1.0$ is shown, normalized to their respective cross sections for ${\kappa }_t=-1.0$, ${\kappa }_{\mathrm{V}}=1.0$.

Figure 5

Prefit classifier outputs, for the ${\mu }^{\pm }{\mu }^{\pm }$ channel (left), $e^{\pm }{\mu }^{\pm }$ channel (center), and three-lepton channel (right), for training against $t\overline{t}V$ (top row) and against $t\overline{t}+\text{jets}$ (bottom row). In the box below each distribution, the ratio of the observed and predicted event yields is shown. The shape of the two $tH$ signals for ${\kappa }_t=-1.0$ is shown, normalized to their respective cross sections for ${\kappa }_t=-1.0$, ${\kappa }_{\mathrm{V}}=1.0$. The grey band represents the unconstrained (prefit) statistical and systematic uncertainties.

Figure 6

Postfit categorized classifier outputs as used in the maximum likelihood fit for the ${\mu }^{\pm }{\mu }^{\pm }$ channel (left), $e^{\pm }{\mu }^{\pm }$ channel (center), and three-lepton channel (right), for $35.9{\mathrm{fb}}^{-1}$. In the box below each distribution, the ratio of the observed and predicted event yields is shown. The shape of the $tH$ signal is indicated with ten times the SM.

Figure 7

Output values of the SC-BDT.

Figure 8

Response values of the FC-BDT. The background consists of $t\overline{t}+b\overline{b}$, $t\overline{t}+1\overline{b}$, and $t\overline{t}+2\overline{b}$ events.

Figure 9

Prefit classifier outputs of the signal classification BDT for the three-tag channel (left) and the four-tag channel (right), for $35.9{\mathrm{fb}}^{-1}$. In the box below each distribution, the ratio of the observed and predicted event yields is shown. The shape of the $tH$ signal is indicated with 800 times the SM.

Figure 10

Postfit classifier outputs of the signal classification BDT as used in the maximum likelihood fit for the three-tag channel (left) and the four-tag channel (right). In the box below each distribution, the ratio of the observed and predicted event yields is shown. The shape of the $tH$ signal is indicated with 800 times the SM.

Figure 11

Pre- (left) and postfit (right) classifier outputs of the flavor classification BDT for the dilepton selection. In the box below each distribution, the ratio of the observed and predicted event yields is shown.

Figure 12

Acceptance and selection efficiency for the $tHq$ (red) and $tHW$ (blue) signal processes as a function of ${\kappa }_t/{\kappa }_{\mathrm{V}}$, for the $t\overline{t}H$ leptonic (solid lines) and $t\overline{t}H$ hadronic categories (dashed lines) of the $H\to \gamma \gamma$ measurement.

Figure 13

Scan of $-2\mathrm{\Delta }\ln (\mathcal{L})$ vs ${\kappa }_t$ for the data (black line) and the individual channels (blue, red, and green), compared to Asimov data sets corresponding to the SM expectations (dashed lines).

Figure 14

Observed and expected 95% CL upper limit on the $tH$ cross section times combined $H\to W{W}^{*}+\tau \tau +Z{Z}^{*}+b\overline{b}+\gamma \gamma$ branching fraction for different values of the coupling ratio ${\kappa }_t$. The expected limit is calculated on a background-only data set, i.e., without $tH$ contribution, but including a ${\kappa }_t$-dependent contribution from $t\overline{t}H$. The $t\overline{t}H$ normalization is kept fixed in the fit, while the $tH$ signal strength is allowed to float.