Search for heavy neutral Higgs bosons produced in association with $b$-quarks and decaying into $b$-quarks at $\sqrt{s}=13\mathrm{TeV}$ with the ATLAS detector

A search for heavy neutral Higgs bosons produced in association with one or two $b$-quarks and decaying to $b$-quark pairs is presented using $27.8{\mathrm{fb}}^{-1}$ of $\sqrt{s}=13\mathrm{TeV}$ proton-proton collision data recorded by the ATLAS detector at the Large Hadron Collider during 2015 and 2016. No evidence of a signal is found. Upper limits on the heavy neutral Higgs boson production cross section times its branching ratio to $b\overline{b}$ are set, ranging from 4.0 to 0.6 pb at 95% confidence level over a Higgs boson mass range of 450 to 1400 GeV. Results are interpreted within the two-Higgs-doublet model and the minimal supersymmetric Standard Model.

Figure 1

Feynman diagrams for some of the leading-order processes for the production of a heavy neutral Higgs boson (denoted here by $\varphi$) in association with one or two $b$-quarks in the 5-flavor scheme. Diagrams (a) and (b) are unique to the 5-flavor scheme, while diagrams (c) and (d) appear in both the 4- and 5-flavor schemes.

Figure 2

Efficiency of the single $b$-tagged jet and double $b$-tagged jet triggers and their logical OR for signal events fulfilling the final selection of Sec. 4 as a function of the neutral Higgs boson mass for datasets collected in 2015 (top) and 2016 (bottom). The operating points for the single (double) $b$-tagged jet triggers correspond to $b$-quark identification efficiencies of 79% (72%) in 2015 and 60% (60%) in 2016.

Figure 3

Two-dimensional distributions of $p_{\mathrm{T}1}$ versus $m_{bb}$ (left column) and of $p_{\mathrm{T}2}$ versus $m_{bb}$ (right column) for events with the $bbb$ classification following preselection, summed over all three $n$-jet regions. Plots are shown for data (top row), the multijet MC sample (middle row), and the $m_{\varphi }=1200\mathrm{GeV}$ signal MC sample (bottom row). The multijet plots are normalized to an integrated luminosity of $27.8{\mathrm{fb}}^{-1}$ based on the cross sections provided by the event generators. The signal plots are normalized to $\sigma (pp\to b\overline{b}\varphi )\times \mathcal{B}(\varphi \to b\overline{b})=1\mathrm{pb}$.

Figure 4

Two-dimensional distributions of $p_{\mathrm{T}1}^{\prime }$ versus $m_{bb}^{\prime }$ (left column) and of $p_{\mathrm{T}2}^{\prime }$ versus $m_{bb}^{\prime }$ (right column) for events with the $bbb$ classification following preselection, summed over all three $n$-jet regions, using the PCA for $m_{\varphi }=1200\mathrm{GeV}$. Plots are shown for data (top row), the multijet MC sample (middle row), and the $m_{\varphi }=1200\mathrm{GeV}$ signal MC sample (bottom row). The multijet plots are normalized to an integrated luminosity of $27.8{\mathrm{fb}}^{-1}$ based on the cross sections provided by the event generators. The signal plots are normalized to $\sigma (pp\to b\overline{b}\varphi )\times \mathcal{B}(\varphi \to b\overline{b})=1\mathrm{pb}$. The minimum values of $p_{\mathrm{T}1}^{\prime }$ and $p_{\mathrm{T}2}^{\prime }$ in the final event selection are indicated by horizontal lines.

Figure 5

Distributions of $m_{bb}^{\prime }$ in simulated multijet events with the $bbb$ and $bbanti$ classifications following all selection requirements for the 3-jet (top row), 4-jet (middle row), and 5-jet (bottom row) regions. The definition of the PCA variable $m_{bb}^{\prime }$ depends on the mass hypothesis $m_{\varphi }$ and distributions are shown for $m_{\varphi }=600\mathrm{GeV}$ (left column) and $m_{\varphi }=1200\mathrm{GeV}$ (right column). For each case, the $bbanti$ distribution is normalized to an integrated luminosity of $27.8{\mathrm{fb}}^{-1}$ based on the cross section provided by the event generator, and the $bbb$ distribution is normalized to the same area as the $bbanti$ distribution. The ratios of $bbb$ to $bbanti$ are also shown.

Figure 6

Distributions of $m_{bb}^{\prime }$ in data with the $bbb$ and $bbanti$ classifications following all selection requirements for the 3-jet (top row), 4-jet (middle row), and 5-jet (bottom row) regions. The definition of the PCA variable $m_{bb}^{\prime }$ depends on the mass hypothesis $m_{\varphi }$ and distributions are shown for $m_{\varphi }=600\mathrm{GeV}$ (left column) and $m_{\varphi }=1200\mathrm{GeV}$ (right column). For each case, the $bbb$ distribution is normalized to the $bbanti$ distribution. The ratios of $bbb$ to $bbanti$ are also shown.

Figure 7

Postfit $bbb$ category distributions of $m_{bb}^{\prime }$ for the 600 GeV (left) and 1200 GeV (right) mass points in the 3-jet (top), 4-jet (middle), and 5-jet (bottom) categories. The prefit background shape is also shown in the top panels, and its ratio to the postfit shape is shown in the bottom panels (green dashed line). The signal shape (red dashed line) is overlaid for illustration.

Figure 8

Observed and expected upper limits on $\sigma (pp\to b\overline{b}\varphi )\times \mathcal{B}(\varphi \to b\overline{b})$ at 95% C.L. as a function of the Higgs boson mass in $27.8{\mathrm{fb}}^{-1}$ of $pp$ collision data at $\sqrt{s}=13\mathrm{TeV}$.

Figure 9

Observed and expected 95% C.L. exclusion limits for the hMSSM scenario as a function of $m_A$. The expected sensitivities for the $m_h^{\text{mod}+}$ and $m_h^{\text{mod}-}$ scenarios are also shown. The observed 95% C.L. limits for the $m_h^{\text{mod}+}$ and $m_h^{\text{mod}-}$ scenarios follow the same pattern with respect to their expected limits as the hMSSM observed limits. Limits are not shown for $\tan \beta >60$ since the Higgs boson coupling becomes nonperturbative for very large values of $\tan \beta$ in the considered models.

Figure 10

Observed and expected 95% C.L. exclusion limits for the flipped 2HDM scenario at $m_{\varphi }=450\mathrm{GeV}$ as a function of $\cos (\beta -\alpha )$. Limits are not shown for $\tan \beta >50$ since the Higgs boson coupling becomes nonperturbative for very large values of $\tan \beta$ in this model.

Figure 11

Observed and expected 95% C.L. exclusion limits for the flipped 2HDM scenario in the alignment limit as a function of $m_{\varphi }$. Limits are not shown for $\tan \beta >50$ since the Higgs boson coupling becomes nonperturbative for very large values of $\tan \beta$ in this model.