Measurement of the $CP$ properties of Higgs boson interactions with $\tau$-leptons with the ATLAS detector

A study of the charge conjugation and parity ($CP$) properties of the interaction between the Higgs boson and $\tau$-leptons is presented. The study is based on a measurement of $CP$-sensitive angular observables defined by the visible decay products of $\tau$-lepton decays, where at least one hadronic decay is required. The analysis uses 139 fb$^{-1}$ of proton–proton collision data recorded at a centre-of-mass energy of $\sqrt{s}= 13$ TeV with the ATLAS detector at the Large Hadron Collider. Contributions from $CP$-violating interactions between the Higgs boson and $\tau$-leptons are described by a single mixing angle parameter $\phi_{\tau}$ in the generalised Yukawa interaction. Without assuming the Standard Model hypothesis for the $H\rightarrow\tau\tau$ signal strength, the mixing angle $\phi_{\tau}$ is measured to be $9^{\circ} \pm 16^{\circ}$, with an expected value of $0^{\circ} \pm 28^{\circ}$ at the 68% confidence level. The pure $CP$-odd hypothesis is disfavoured at a level of 3.4 standard deviations. The results are compatible with the predictions for the Higgs boson in the Standard Model.

12 December 2022

Contact: Higgs conveners internal

Eur. Phys. J. C 83 (2023) 563

e-print arXiv:2212.05833 - pdf from arXiv

Inspire record

Data points

Figures | Tables | Auxiliary Material

Figures

Figure 01a

Illustration of the τ-lepton decay planes for constructing the φ^*_CP observable in (a) H→τ⁺τ^-→π⁺π^-+2ν decay using the impact parameter method, (b) H→τ⁺τ^-→π⁺π⁰νπ^-π⁰ν using the ρ-decay plane method, and (c) H→τ⁺τ^-→π⁺π⁰νπ^-ν using the combined impact parameter and ρ-decay plane method. The decay planes are spanned by the spatial momentum vector of the charged decay particle of the τ-lepton (π^±) and either its impact parameter n^*± or the spatial momentum vector of the neutral decay particle of the τ-lepton (π⁰).

png (11kB) pdf (53kB)

Figure 01b

Figure 01c

Figure 02

Normalised φ^*_CP distributions in simulated H→τ⁺τ^-→π⁺π^-+2ν events at the generator level for different CP hypotheses. The predictions for a pure CP-even SM Higgs boson (scalar, red circle), a pure CP-odd hypothesis (pseudoscalar, green square), and CP-mix hypothesis (φ_τ = 45^°, blue triangle) are shown. The transverse momentum of the simulated τ leptons is required to be larger than 30 GeV (20 GeV) for the leading (sub-leading) τ lepton during the event generation.

png (60kB) pdf (16kB)

Figure 03a

Post-fit distributions of the π^±π⁰ invariant mass, m(π^±,π⁰), in the Z→ττ control regions in the (a) τ_lepτ_had and (b) τ_hadτ_had channels. The ℓ--1p1n (1p1n--1p1n) events are used in the τ_lepτ_had (τ_hadτ_had) channel in this control region. For the τ_hadτ_had channel only the m(π^±,π⁰) value of the leading τ_had is selected. 'Other backgrounds' include W, diboson, top, Z→ ℓℓ and H→ WW^*. The hatched uncertainty band includes all sources of uncertainty after the fit to data.

png (118kB) pdf (20kB)

Figure 03b

Figure 04a

Post-fit distributions of φ^*_CP in the signal regions (SRs), showing (a) τ_lepτ_had High SR, (b) τ_hadτ_had High SR, (c) τ_lepτ_had Medium SR, (d) τ_hadτ_had Medium SR, (e) τ_lepτ_had Low SR, and (f) τ_hadτ_had Low SR. The φ^*_CP bins are counted incrementally through all VBF and Boost categories and cover the range [0, 360]^° for each category. The best-fit H → ττ signal is shown in solid pink, while the red and green lines indicate the predictions for the pure CP-even (scalar, SM) and pure CP-odd (pseudoscalar) hypotheses, respectively, scaled to the predicted signal yield. 'Other backgrounds' include W, diboson, top, Z→ ℓℓ and H→ WW^*. The hatched uncertainty band includes all sources of uncertainty after the fit to data.

png (145kB) pdf (27kB)

Figure 04b

Figure 04c

Figure 04d

Figure 04e

Figure 04f

Figure 05

One-dimensional likelihood scan of the CP-mixing angle φ_τ. The observed (expected) value of φ_τ is 9^°±16^° (0^°±28^°) at the 68% confidence level (CL), and ± 34^° (_-70^°^{+75^°}) at the 2σ level. The CP-odd hypothesis is rejected at the 3.4σ (2.1σ expected) level.

png (49kB) pdf (15kB)

Figure 06

A 2D likelihood scan of the observed signal strength μ_ττ versus the CP-mixing angle φ_τ. The 1σ and 2σ confidence regions are shown.

png (54kB) pdf (15kB)

Figure 07

Combined post-fit distribution of φ^*_CP from all signal regions in both the τ_lepτ_had and τ_hadτ_had channels. Events are weighted with ln(1+S/B) for the corresponding signal region. The background is subtracted from data. The best-fit H → ττ signal is shown in solid pink, while the red and green lines indicate the predictions for the pure CP-even (scalar, SM) and pure CP-odd (pseudoscalar) hypotheses, respectively, all scaled to the best-fit H → ττ signal yield. The hatched uncertainty band includes all sources of uncertainty after the fit to data, and represents the same uncertainty in the total signal and background predictions as in Figure 4.

png (113kB) pdf (17kB)

Tables

Table 01

Notation for the dominant leptonic and hadronic τ-lepton decay modes used and their branching fractions. The symbol 'ℓ^±' stands for e^± or μ^±, and 'h^±' includes π^± and K^±. The parentheses show the hadronic decays involving π^± and their corresponding branching fractions.

png (18kB) pdf (47kB)

Table 02

Decay mode combinations of the τ-lepton pair and the corresponding methods to construct the φ^*_CP observable used in this analysis. The fraction of events for each decay mode combination relative to the total from all di-τ decay combinations (last column) is calculated using the τ-lepton decay mode branching fractions in Table 1.

png (25kB) pdf (40kB)

Table 03

Summary of selection criteria for the VBF and Boost categories in this analysis. The criteria are common to the τ_lepτ_had and τ_hadτ_had decay channels.

png (31kB) pdf (61kB)

Table 04

Summary of additional selection criteria for the signal regions in the τ_lepτ_had channel.

png (30kB) pdf (50kB)

Table 05

Summary of additional selection criteria for the signal regions in the τ_hadτ_had channel.

png (32kB) pdf (50kB)

Table 06

Free-floating parameters in the measurement. Observed and expected values are shown for the CP-mixing angle (φ_τ), the signal strength (μ_ττ) and various background normalisations for Z → ττ (NF_{Z → ττ}) corresponding to different signal phase-space regions.

png (14kB) pdf (48kB)

Table 07

Impact of different sources of uncertainty on the φ_τ measurement.

png (47kB) pdf (54kB)

Auxiliary material

Figure 01

Normalised φ^*_CP distribution before detector effects in simulated H→τ⁺τ^-→π⁺π⁰νπ^-π⁰ν (1p1n-1p1n) events for different CP hypotheses.

png (60kB) pdf (16kB)

Figure 02

Normalised φ^*_CP distribution before detector effects in simulated H→τ⁺τ^-→π⁺νπ^-π⁰ν (1p0n-1p1n) events for different CP hypotheses.

png (61kB) pdf (16kB)

Figure 03

Normalised φ^*_CP distribution before detector effects in simulated H→τ⁺τ^-→ℓ⁺ννπ^-ν (ℓ-1p0n) events for different CP hypotheses.

png (60kB) pdf (16kB)

Figure 04

Normalised φ^*_CP distribution before detector effects in simulated H→τ⁺τ^-→ℓ⁺ννπ^-π⁰ν (ℓ-1p1n) events for different CP hypotheses.

png (60kB) pdf (16kB)

Figure 05

Number of events for each hadronic τ lepton decay mode reconstructed in the τ_lepτ_had channel used in this analysis. The events include all signal region categories (Boost_1, Boost_0, VBF_1 and VBF_0) in the τ_lepτ_had channel within the Higgs boson mass window (110 GeV < m_ττ^MMC < 150 GeV). 'Other backgrounds' include W, diboson, top, Z→ ℓℓ and H→ WW^*. The hatched band represents the statistical uncertainty on the prediction.

png (117kB) pdf (17kB)

Figure 06

Number of events for each τ lepton pair decay mode combination reconstructed in the τ_hadτ_had channel used in this analysis. The events include all signal region categories (Boost_1, Boost_0, VBF_1 and VBF_0) in the τ_hadτ_had channel within the Higgs boson mass window (110 GeV < m_ττ^MMC < 150 GeV). 'Other backgrounds' include W, diboson, top, Z→ ℓℓ and H→ WW^*. The hatched band represents the statistical uncertainty on the prediction.

png (127kB) pdf (18kB)

Figure 07a

(a) Efficiency and (b) purity matrices of the POWHEG+PYTHIA VBF H → ττ signal sample used in the analysis. Each element in the efficiency matrix is normalised by the total number of events in its corresponding generated decay mode (column), while an element in the purity matrix is normalised by the total events in the corresponding reconstructed decay mode (row).

png (120kB) pdf (13kB)

Figure 07b

Figure 08

A schematic summary of the regions and fit parameters used in the fit. All signal regions (12) and control regions (5) in the diagram are defined in each of the τ_lepτ_had and τ_hadτ_had channels, giving a total of 24 (10) signal (control) regions in the likelihood fit. The CP-mixing angle (φ_τ) is the parameter-of-interest (POI). The H → ττ signal strength (μ_ττ), and four normalisation factors (NF) for the Z → ττ process are left free-floating in the fit. Each NF is shared between a control region and the respective signal regions in the same topology category, and across the τ_lepτ_had and τ_hadτ_had channels. The ℓ--1p1n (1p1n--1p1n) control region is defined in the τ_lepτ_had (τ_hadτ_had) channel to control π⁰-related uncertainties from data, as described in Section 6.

png (64kB) pdf (98kB)

Figure 09a

Post-fit event yields in the Z→ττ control regions in the (a) τ_lepτ_had and (b) τ_hadτ_had channels. The control regions are represented by the labels of the bins. 'Other backgrounds' include W, diboson, top, Z→ ℓℓ and H→ WW^*. The hatched uncertainty band includes all sources of uncertainty after the fit to data.

png (59kB) pdf (17kB)

Figure 09b

Figure 10

A 2D likelihood scan of the expected signal strength μ_ττ versus the CP-mixing angle φ_τ. The 1σ and 2σ confidence regions are shown.

png (54kB) pdf (16kB)

Figure 11

Candidate event display of a Higgs boson, produced in association with two additional jets (yellow cones), with the Higgs boson decaying to two τ leptons (blue cones) . Both τ leptons decay subsequently to a charged pion, characterised by a trajectory in the inner detector (blue line). Energy deposits in the electromagnetic calorimeter (green) not connected to a charged particle trajectory can be caused by additional neutral pions in the τ lepton decay. The white dashed line represents the missing transverse energy due to the undetected neutrinos from the τ lepton decays. The bottom left view shows the same objects projected onto the r-φ plane.

png (1MB) pdf (782kB)

Table 01

Summary of event selection requirements in the τ_lepτ_had channel (preselection). The transverse mass m_T is calculated from the momentum of the light lepton and E_T^miss. The quantities x₁ and x₂ are the momentum fraction carried by the visible decay products of the leading and sub-leading reconstructed τ leptons respectively, with the E_T^miss components decomposed in the collinear approximation.

png (27kB) pdf (59kB)

Table 02

Summary of event selection requirements in the τ_hadτ_had channel (preselection). The quantities x₁ and x₂ are the momentum fraction carried by the visible decay products of the leading and sub-leading reconstructed τ leptons respectively, with the E_T^miss components decomposed in the collinear approximation.

png (25kB) pdf (59kB)

Table 03

Expected sensitivities in excluding pure CP-odd Hττ coupling in different groups of signal regions.

png (37kB) pdf (41kB)