Constraints on effective field theories via quadruple-differential angular decay rates from $t$-channel single-top-quark production at $\sqrt{s}=13$ TeV with the ATLAS detector

TL;DR

This paper constrains dimension-six SMEFT Wilson coefficients in the top-quark sector using quadruple-differential angular decay rates in $t$-channel single-top production at $\sqrt{s}=13$ TeV with ATLAS and $140\ \mathrm{fb}^{-1}$. It introduces a Fourier-based $M$-function basis to project the angular observables $(\theta,\phi,\theta^{*},\phi^{*})$, and models EFT effects with a morphing technique across a six-parameter WC grid, while treating $C_{\varphi Q}$ via a cross-section scale. A joint likelihood combines 92 angular coefficients and 12 event yields across four channels, yielding constraints on seven WC combinations, with results compatible with the SM and competitive relative to prior ATLAS and CMS results. The method demonstrates a powerful, high-precision EFT probe leveraging top polarization and angular correlations, providing a framework for future EFT tests at the LHC.

Abstract

Events with $t$-channel single top quarks are used to probe effective field theory operators in $\sqrt{s}=13$ TeV proton-proton collision data corresponding to 140 fb$^{-1}$ recorded by the ATLAS detector at the Large Hadron Collider. An analysis method leveraging Fourier techniques applied to quadruple-differential decay rates based on observables containing angular information about the decays of the top quarks is used to achieve high sensitivity. The relevant effective field theory operators are those sensitive to top-quark decay and $t$-channel production vertices. Their Wilson coefficients are tightly constrained, with results compatible with Standard Model predictions.

Figures (12)

• Figure 1: Representative LO Feynman diagram for $t$-channel single-top-quark production and decay in the 4FS. Here $q$ represents a $\Pqu/\Pqc$ or $\Paqd/\Paqs$ quark, and $q^\prime$ represents a $\Pqd/\Pqs$ or $\Paqu/\Paqc$ quark. The initial typically arises from a gluon splitting into a pair.
• Figure 2: Illustration of the definition of the four angles $(\theta, \phi, \theta^*, \phi^*)$ in the $t\rightarrow Wb\rightarrow \Pl \nu b$ decay chain. The incoming protons, the spectator, top and quarks, the boson, and the lepton and neutrino are denoted by $p$, $q'$, , , , , and $\nu$, respectively. The angles $\theta$ and $\phi$ specify the direction of the -boson (or spectator-quark) momentum in the top-quark rest frame, while $\theta^*$ and $\phi^*$ specify the direction of the charged lepton momentum in the -boson rest frame.
• Figure 3: Overview of the analysis workflow. Events passing the preselection are separated into two CRs (CR-$W$, CR-) and an SR. In the SR, the four decay angles are reconstructed for each event and projected onto the orthonormal $M$-function basis, providing event-by-event coefficients that are then averaged as defined in Eq. (\ref{['eq:mfcomp']}). The dependence of these coefficients on the WC vector, $\vec{C}$, in both the $t$-channel single-top-quark and processes is modelled using morphing. Final observables consist of the $M$-function coefficients in the SR and the event yields in both the SR and CRs.
• Figure 4: Post-fit distributions of the four angles, namely (a) $\cos\theta$, (b) $\cos\theta^{*}$, (c) $\phi$, and (d) $\phi^{*}$ in the inclusive SR. The fitted predicted distributions are compared with the observed data, depicted as black points with statistical uncertainties. The uncertainty bands reflect the statistical and systematic uncertainties, taking into account the post-fit constraints from the profile likelihood fit. Correlations among all parameters obtained from the fit are accounted for in the calculation of the uncertainty bands. The lower plot for each angular distribution shows the ratio of the observed data to the fitted prediction in each bin, offering a quantitative measure of how well the two agree.
• Figure 5: The relative differences between the observed and SM predicted event counts in the SRs and CRs, for both the observed data and the fitted EFT model. The black points represent observed data, with the data statistical uncertainties as error bars. The grey dashed area indicates the total systematic uncertainties, while the red boxes are centred at positions determined by the best-fit values of the Wilson coefficients, with heights determined by systematic uncertainties constrained by the fit. The coloured scattered markers correspond to expected values when making various WC assumptions.