Search for pair-production of vector-like $T$ quarks decaying into a top quark and a spin-0 particle in the diphoton final state in proton proton collisions at $\sqrt{s}=13$ TeV with the ATLAS detector

Technical Summary: Search for Pair-Production of Vector-Like $T$ Quarks Decaying into a Top Quark and a Spin-0 Particle in the Diphoton Final State

Problem and Motivation
Following the 2012 discovery of the Higgs boson, high-energy physics has focused on determining whether the Higgs is fundamental or composite and addressing the hierarchy problem. Theories such as partial compositeness propose the existence of Vector-Like Quarks (VLQs), specifically top partners ($T$), which mix with Standard Model (SM) fermions. While previous searches have largely assumed VLQ decays into SM particles (e.g., $T \to tH, tZ, tW, bW$), models involving compositeness often predict additional exotic scalars or pseudoscalars ($S$). In such scenarios, a top partner could decay via $T \to St$, where $S$ is a new spin-0 particle. If $S$ decays into SM particles, these cascade decays represent a search topology not previously explored by the ATLAS or CMS collaborations. This analysis targets the specific process of pair-produced VLQs ($pp \to T\bar{T}$) where one $T$ decays via $T \to St \to t\gamma\gamma$ and the other via $T \to St \to tX$, with $X$ representing any allowed final state ($\gamma\gamma, Z\gamma, WW, ZZ, t\bar{t}$).

Methodology
The analysis utilizes the full Run 2 dataset collected by the ATLAS detector at $\sqrt{s}=13$ TeV, corresponding to an integrated luminosity of 140 fb$^{-1}$.

• Event Selection: Events are selected requiring at least two isolated photons with transverse energy $E_T > 38$ GeV and pseudorapidity $|\eta| < 2.47$ (excluding the barrel-endcap transition region). At least one $b$-tagged jet is required. The invariant mass of the diphoton system ($m_{\gamma\gamma}$) is required to be greater than 150 GeV to avoid the Higgs boson mass region.
• Kinematic Discriminator: The primary discriminant against the SM background is the scalar sum of the transverse momenta of all visible objects (jets, leptons, photons) plus the missing transverse momentum ($E_T^{\text{miss}}$), denoted as $p_T^{\text{vis}} + E_T^{\text{miss}}$. This variable is chosen because the decay products of heavy resonances are expected to have large transverse momenta, whereas SM processes populate the lower end of this spectrum.
• Signal and Background Modeling:
  • Signal: Simulated using MadGraph5_aMC@NLO with Pythia 8 for parton showering. Signal samples cover $T$ masses ($m_T$) from 1000 to 1800 GeV and $S$ masses ($m_S$) from 200 to 1200 GeV. The analysis assumes a 100% branching ratio for $T \to St$.
  • Background: The continuum diphoton background is estimated using a data-driven approach. Control Regions (CRs) are defined in lower $p_T^{\text{vis}} + E_T^{\text{miss}}$ ranges to provide background templates. The background shape in the Signal Regions (SRs) is parameterized by an exponential function, $f(m_{\gamma\gamma}) = N e^{-m_{\gamma\gamma}/\tau}$, with parameters fitted to data.
• Statistical Analysis: An unbinned simultaneous likelihood fit is performed across four Signal Regions defined by increasing $p_T^{\text{vis}} + E_T^{\text{miss}}$ thresholds. Systematic uncertainties, including those from background modeling (spurious signal), photon energy scale/resolution, and luminosity, are incorporated via nuisance parameters.

Key Contributions

• Novel Search Topology: This is the first ATLAS search for pair-produced VLQs decaying through an exotic spin-0 particle into a diphoton final state. It addresses a gap in previous VLQ searches which typically assumed decays directly to SM bosons.
• Model-Independent Results: While the search is motivated by specific cascade models, the results are presented in a fiducial volume defined by kinematic requirements on the photons. This allows the limits to be reinterpreted in a wide variety of phenomenological models beyond the specific $T \to St$ scenario.
• Comprehensive Parameter Space: The analysis covers a broad range of masses ($m_T \in [1000, 1800]$ GeV, $m_S \in [200, 1200]$ GeV), excluding only kinematically suppressed regions where $m_S > m_T - m_t$.

Results

• Observation: No significant excess over the expected Standard Model background is observed in any of the Signal Regions. The data are consistent with the background-only hypothesis, with deviations remaining within $2\sigma$.
• Exclusion Limits: Upper limits at the 95% confidence level (CL) are set on the fiducial production cross-section times the branching ratio, $\sigma_{\text{fid}}(pp \to T\bar{T}) \times \mathcal{B}(S \to \gamma\gamma) \times \mathcal{B}(S \to X)$, as a function of $m_T$ and $m_S$.
• Model Interpretation: The results are interpreted within the framework of a specific partial compositeness model (Ref. [61]). In this model, assuming degenerate masses for the VLQs and the spin-0 particles, the analysis excludes $T$ masses below $m_T = 1620$ GeV for the scenario where the spin-0 particles decay into non-photonic final states (the $2\gamma$ case, corresponding to one $S \to \gamma\gamma$ and the other to non-photonic modes). The exclusion limits are weaker for scenarios involving more photons (e.g., $3\gamma$ or $4\gamma$ final states) due to lower branching ratios or acceptance.

Significance
The paper claims significance primarily as a pioneering search in an unexplored decay channel for Vector-Like Quarks. By targeting the $T \to St$ cascade with $S \to \gamma\gamma$, the analysis extends the sensitivity of VLQ searches to scenarios involving exotic scalars, which are predicted by composite Higgs models but were previously unconstrained by ATLAS. The lack of an observed signal allows the collaboration to set the first constraints on this specific topology, thereby narrowing the viable parameter space for models predicting top partners that decay via new spin-0 mediators. The results are presented with a high degree of model independence in the fiducial volume, ensuring their utility for future reinterpretations.