Search for heavy Higgs bosons A/H decaying to a top-quark pair in pp collisions at √ s = 8 TeV with the ATLAS detector

ATLAS-CONF-2016-073

3 August 2016

These preliminary results are superseded by the following paper:

EXOT-2016-04
ATLAS recommends to use the results from the paper.

Content

Preview

Main document (CDS record) - internal

pdf from CDS

Figures Tables

Abstract

A search for a pseudoscalar (A) or scalar (H) heavy Higgs boson decaying into a top-quark pair ($t\bar{t}$) has been conducted in 20.3 fb$^{-1}$ of data collected by the ATLAS detector at the Large Hadron Collider in proton--proton collisions at a centre-of-mass energy of $\sqrt{s}=8$ TeV. The analysis relies on the invariant mass spectrum of the $t\bar{t}$ pair in final states with an electron or muon, large missing transverse momentum, and at least four jets. Interference effects between the signal process and $gg\rightarrow t\bar{t}$ production in the Standard Model, which heavily distort the signal shape from a single peak to a peak-dip structure, are taken into account. Exclusion limits are derived for two resonance masses (500 and 750 GeV), as a function of the parameter tanβ in Two-Higgs-Doublet Models.

Figures

Figure 01a:
The processes generated in MGMCatNLO for this analysis: (a,b) the SM tt background, and (c) the pseudoscalar or scalar signal gg→ A/H → tt.

png (26kB) eps (26kB) pdf (11kB)

Figure 01b:
The processes generated in MGMCatNLO for this analysis: (a,b) the SM tt background, and (c) the pseudoscalar or scalar signal gg→ A/H → tt.

png (19kB) eps (20kB) pdf (9kB)

Figure 01c:
The processes generated in MGMCatNLO for this analysis: (a,b) the SM tt background, and (c) the pseudoscalar or scalar signal gg→ A/H → tt.

png (26kB) eps (29kB) pdf (12kB)

Figure 02a:
Distributions of the invariant mass of the tt pair from the decay of a pseudoscalar resonance A at parton level before the emission of final-state radiation for the pure resonance signal S (blue) and signal+interference contribution S+I (red). Left column: m_A = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_A = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. Events from all tt decay modes are included and no selection requirements are imposed. All distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (31kB) pdf (18kB)

Figure 02b:
Distributions of the invariant mass of the tt pair from the decay of a pseudoscalar resonance A at parton level before the emission of final-state radiation for the pure resonance signal S (blue) and signal+interference contribution S+I (red). Left column: m_A = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_A = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. Events from all tt decay modes are included and no selection requirements are imposed. All distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (31kB) pdf (18kB)

Figure 02c:
Distributions of the invariant mass of the tt pair from the decay of a pseudoscalar resonance A at parton level before the emission of final-state radiation for the pure resonance signal S (blue) and signal+interference contribution S+I (red). Left column: m_A = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_A = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. Events from all tt decay modes are included and no selection requirements are imposed. All distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (32kB) pdf (18kB)

Figure 02d:
Distributions of the invariant mass of the tt pair from the decay of a pseudoscalar resonance A at parton level before the emission of final-state radiation for the pure resonance signal S (blue) and signal+interference contribution S+I (red). Left column: m_A = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_A = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. Events from all tt decay modes are included and no selection requirements are imposed. All distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (32kB) pdf (18kB)

Figure 02e:
Distributions of the invariant mass of the tt pair from the decay of a pseudoscalar resonance A at parton level before the emission of final-state radiation for the pure resonance signal S (blue) and signal+interference contribution S+I (red). Left column: m_A = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_A = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. Events from all tt decay modes are included and no selection requirements are imposed. All distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (31kB) pdf (16kB)

Figure 02f:
Distributions of the invariant mass of the tt pair from the decay of a pseudoscalar resonance A at parton level before the emission of final-state radiation for the pure resonance signal S (blue) and signal+interference contribution S+I (red). Left column: m_A = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_A = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. Events from all tt decay modes are included and no selection requirements are imposed. All distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (32kB) pdf (17kB)

Figure 03a:
Distributions of the invariant mass of the tt pair from the decay of a scalar resonance H at parton level for the pure resonance signal S (blue) and signal+interference contribution S+I (red) before the event selection. Left column: m_H = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_H = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set unity in all cases. All tt decay modes are included, namely semileptonic, dileptonic and fully hadronic. All distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (32kB) pdf (18kB)

Figure 03b:
Distributions of the invariant mass of the tt pair from the decay of a scalar resonance H at parton level for the pure resonance signal S (blue) and signal+interference contribution S+I (red) before the event selection. Left column: m_H = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_H = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set unity in all cases. All tt decay modes are included, namely semileptonic, dileptonic and fully hadronic. All distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (34kB) pdf (18kB)

Figure 03c:
Distributions of the invariant mass of the tt pair from the decay of a scalar resonance H at parton level for the pure resonance signal S (blue) and signal+interference contribution S+I (red) before the event selection. Left column: m_H = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_H = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set unity in all cases. All tt decay modes are included, namely semileptonic, dileptonic and fully hadronic. All distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (31kB) pdf (17kB)

Figure 03d:
Distributions of the invariant mass of the tt pair from the decay of a scalar resonance H at parton level for the pure resonance signal S (blue) and signal+interference contribution S+I (red) before the event selection. Left column: m_H = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_H = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set unity in all cases. All tt decay modes are included, namely semileptonic, dileptonic and fully hadronic. All distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (32kB) pdf (17kB)

Figure 03e:
Distributions of the invariant mass of the tt pair from the decay of a scalar resonance H at parton level for the pure resonance signal S (blue) and signal+interference contribution S+I (red) before the event selection. Left column: m_H = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_H = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set unity in all cases. All tt decay modes are included, namely semileptonic, dileptonic and fully hadronic. All distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (32kB) pdf (16kB)

Figure 03f:
Distributions of the invariant mass of the tt pair from the decay of a scalar resonance H at parton level for the pure resonance signal S (blue) and signal+interference contribution S+I (red) before the event selection. Left column: m_H = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_H = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set unity in all cases. All tt decay modes are included, namely semileptonic, dileptonic and fully hadronic. All distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (31kB) pdf (17kB)

Figure 04a:
The selection efficiencies for pure resonant signal samples with resonance masses of (a) 500 GeV and (b) 750 GeV. The A and H distributions show a difference of less than 5% at 500 GeV and less than 10% at 750 GeV.

png (33kB) eps (11kB) pdf (5kB)

Figure 04b:
The selection efficiencies for pure resonant signal samples with resonance masses of (a) 500 GeV and (b) 750 GeV. The A and H distributions show a difference of less than 5% at 500 GeV and less than 10% at 750 GeV.

png (32kB) eps (10kB) pdf (5kB)

Figure 05a:
Distribution of the tt invariant mass, reconstructed with the χ² algorithm, for S+I events with a pseudoscalar resonance A that fulfill the e+jets (blue) and μ+jets (red) selections. Left column: m_A = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_A = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. The distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (32kB) pdf (15kB)

Figure 05b:
Distribution of the tt invariant mass, reconstructed with the χ² algorithm, for S+I events with a pseudoscalar resonance A that fulfill the e+jets (blue) and μ+jets (red) selections. Left column: m_A = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_A = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. The distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (34kB) pdf (15kB)

Figure 05c:
Distribution of the tt invariant mass, reconstructed with the χ² algorithm, for S+I events with a pseudoscalar resonance A that fulfill the e+jets (blue) and μ+jets (red) selections. Left column: m_A = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_A = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. The distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (33kB) pdf (15kB)

Figure 05d:
Distribution of the tt invariant mass, reconstructed with the χ² algorithm, for S+I events with a pseudoscalar resonance A that fulfill the e+jets (blue) and μ+jets (red) selections. Left column: m_A = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_A = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. The distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (32kB) pdf (15kB)

Figure 05e:
Distribution of the tt invariant mass, reconstructed with the χ² algorithm, for S+I events with a pseudoscalar resonance A that fulfill the e+jets (blue) and μ+jets (red) selections. Left column: m_A = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_A = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. The distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (34kB) pdf (15kB)

Figure 05f:
Distribution of the tt invariant mass, reconstructed with the χ² algorithm, for S+I events with a pseudoscalar resonance A that fulfill the e+jets (blue) and μ+jets (red) selections. Left column: m_A = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_A = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. The distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (33kB) pdf (15kB)

Figure 06a:
Distribution of the tt invariant mass, reconstructed with the χ² algorithm, for S+I events with a scalar resonance H that fulfill the e+jets (blue) and μ+jets (red) selections. Left column: m_H = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_H = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. The distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (33kB) pdf (15kB)

Figure 06b:
Distribution of the tt invariant mass, reconstructed with the χ² algorithm, for S+I events with a scalar resonance H that fulfill the e+jets (blue) and μ+jets (red) selections. Left column: m_H = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_H = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. The distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (35kB) pdf (15kB)

Figure 06c:
Distribution of the tt invariant mass, reconstructed with the χ² algorithm, for S+I events with a scalar resonance H that fulfill the e+jets (blue) and μ+jets (red) selections. Left column: m_H = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_H = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. The distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (32kB) pdf (15kB)

Figure 06d:
Distribution of the tt invariant mass, reconstructed with the χ² algorithm, for S+I events with a scalar resonance H that fulfill the e+jets (blue) and μ+jets (red) selections. Left column: m_H = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_H = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. The distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (33kB) pdf (15kB)

Figure 06e:
Distribution of the tt invariant mass, reconstructed with the χ² algorithm, for S+I events with a scalar resonance H that fulfill the e+jets (blue) and μ+jets (red) selections. Left column: m_H = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_H = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. The distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (31kB) pdf (15kB)

Figure 06f:
Distribution of the tt invariant mass, reconstructed with the χ² algorithm, for S+I events with a scalar resonance H that fulfill the e+jets (blue) and μ+jets (red) selections. Left column: m_H = 500 GeV for tanβ values of (a) 0.4 (c) 0.68 (e) 9.0. Right column: m_H = 750 GeV for tanβ values of (b) 0.4 (d) 0.7 (f) 2.0. The parameter sin(β-α) is set to unity in all cases. The distributions are normalised to an integrated luminosity of 20.3 fb^-1.

png (33kB) pdf (15kB)

Figure 07a:
The m_tt^reco distribution of the events in the μ+jets channel signal regions after the profile-likelihood fit. Figure (a) shows the category 1, (b) the category 2, and (c) the category 3. The insert at the bottom of each plot shows the ratio of the data to the total background. The total background before the fit is indicated by a dashed line. The solid red line shows the expected distribution (scaled by a factor of 7 for better visibility) for a hypothetical pseudoscalar A with a mass of m_A=750 GeV.

png (95kB) eps (71kB) pdf (11kB)

Figure 07b:
The m_tt^reco distribution of the events in the μ+jets channel signal regions after the profile-likelihood fit. Figure (a) shows the category 1, (b) the category 2, and (c) the category 3. The insert at the bottom of each plot shows the ratio of the data to the total background. The total background before the fit is indicated by a dashed line. The solid red line shows the expected distribution (scaled by a factor of 7 for better visibility) for a hypothetical pseudoscalar A with a mass of m_A=750 GeV.

png (95kB) eps (70kB) pdf (11kB)

Figure 07c:
The m_tt^reco distribution of the events in the μ+jets channel signal regions after the profile-likelihood fit. Figure (a) shows the category 1, (b) the category 2, and (c) the category 3. The insert at the bottom of each plot shows the ratio of the data to the total background. The total background before the fit is indicated by a dashed line. The solid red line shows the expected distribution (scaled by a factor of 7 for better visibility) for a hypothetical pseudoscalar A with a mass of m_A=750 GeV.

png (97kB) eps (79kB) pdf (12kB)

Figure 08a:
The m_tt^reco distribution of the events in the e+jets channel signal regions after the profile-likelihood fit. Figure (a) shows the category 1, (b) the category 2, and (c) the category 3. The insert at the bottom of each plot shows the ratio of the data to the total background. The total background before the fit is indicated by a dashed line. The solid red line shows the expected distribution (scaled by a factor of 7 for better visibility) for a hypothetical pseudoscalar A with a mass of m_A=750 GeV.

png (95kB) eps (71kB) pdf (11kB)

Figure 08b:
The m_tt^reco distribution of the events in the e+jets channel signal regions after the profile-likelihood fit. Figure (a) shows the category 1, (b) the category 2, and (c) the category 3. The insert at the bottom of each plot shows the ratio of the data to the total background. The total background before the fit is indicated by a dashed line. The solid red line shows the expected distribution (scaled by a factor of 7 for better visibility) for a hypothetical pseudoscalar A with a mass of m_A=750 GeV.

png (95kB) eps (71kB) pdf (11kB)

Figure 08c:
The m_tt^reco distribution of the events in the e+jets channel signal regions after the profile-likelihood fit. Figure (a) shows the category 1, (b) the category 2, and (c) the category 3. The insert at the bottom of each plot shows the ratio of the data to the total background. The total background before the fit is indicated by a dashed line. The solid red line shows the expected distribution (scaled by a factor of 7 for better visibility) for a hypothetical pseudoscalar A with a mass of m_A=750 GeV.

png (97kB) eps (80kB) pdf (12kB)

Figure 09a:
Observed (solid line) and expected (dashed line) upper limits on the signal strength parameter μ as a function of the parameter tanβ for a neutral pseudoscalar A with mass (a) m_A=500 GeV and (b) m_A=750 GeV. The blue line at μ = 1 corresponds to the signal strength in the type-II 2HDM.

png (95kB) eps (12kB) pdf (5kB)

Figure 09b:
Observed (solid line) and expected (dashed line) upper limits on the signal strength parameter μ as a function of the parameter tanβ for a neutral pseudoscalar A with mass (a) m_A=500 GeV and (b) m_A=750 GeV. The blue line at μ = 1 corresponds to the signal strength in the type-II 2HDM.

png (43kB) eps (12kB) pdf (5kB)

Figure 10a:
Observed (solid line) and expected (dashed line) upper limits on the signal strength parameter μ as a function of the parameter tanβ for a neutral scalar H with mass (a) m_H=500 GeV and (b) m_H=750 GeV. The blue line at μ = 1 corresponds to the signal strength in the type-II 2HDM.

png (91kB) eps (12kB) pdf (5kB)

Figure 10b:
Observed (solid line) and expected (dashed line) upper limits on the signal strength parameter μ as a function of the parameter tanβ for a neutral scalar H with mass (a) m_H=500 GeV and (b) m_H=750 GeV. The blue line at μ = 1 corresponds to the signal strength in the type-II 2HDM.

png (42kB) eps (12kB) pdf (5kB)

Tables

Table 01:
List of the MC generators, the order of cross-section calculation in QCD, and the PDF sets used in the generation of signal and background processes. All generator versions are the same as those in Ref. [1]. The background from multijet events was estimated in a fully data-driven way (Section 8).

png (40kB) pdf (52kB)

Table 02:
Data and expected background event yields after the resolved-topology selection. The uncertainty on the expected background yields is derived by summing all systematic uncertainties and the MC statistical uncertainty in quadrature. The expected yields and uncertainties are shown before the profile likelihood fit (described in the text) to the full dataset.

png (25kB) pdf (33kB)

© 2016 CERN for the benefit of the ATLAS Collaboration. Reproduction of the article, figures and tables on this page is allowed as specified in the CC-BY-4.0 license.
2024-05-19 01:23:00