Three’s no crowd: ATLAS measures tri-boson production

18 August 2023 | By

At the LHC, force-carrying particles can be produced by themselves, in pairs or even in trios. This is because the particles that carry the weak and electromagnetic forces – the W and Z bosons and the photon (γ) – are able to interact with each other directly. Indeed, the strengths of these interactions are predicted by the electroweak sector of the Standard Model, making their measurement an excellent test of the theory. Using the vast amount of data collected during Run 2 of the LHC (2015 to 2018) and new analysis techniques, the ATLAS Collaboration has reported three separate new measurements of tri-boson processes.

The ATLAS Collaboration has announced the first observation of two different tri-boson processes: the simultaneous production of a W boson with two photons (Wγγ) and the production of a W boson, Z boson and photon (WZγ). The production of a Z boson with two photons (Zγγ) was first observed in 2016 using data from LHC Run 1 (2010-2013). In a new publication, the ATLAS Collaboration expanded the scope of this initial observation using Run 2 data.

These processes are difficult to measure, as they are extremely rare and can be hard to disentangle from other physics processes that mimic their signatures in the detector. For these new analyses, physicists looked for “leptonic” decays of the W and Z bosons. Specifically, for Z bosons decaying to a pair of electrons or muons, and W bosons decaying to an electron or muon, accompanied by a neutrino. While electrons, muons and photons interact directly with the ATLAS detector and are well measured, neutrinos are invisible to the detector. They are reconstructed indirectly as missing transverse energy. So, for each process, events with various combinations of electrons, muons, photons and missing transverse energy were selected from the Run 2 dataset.

The ATLAS Collaboration has observed three different tri-boson processes: the production of a W boson with two photons (Wγγ), the production of a W boson, Z boson and photon (WZγ), and the production of a Z boson with two photons (Zγγ).

Proton Collisions,Event Displays,Physics,ATLAS
Figure 1: Display of a candidate W(→μν)γγ event. Orange lines indicate tracks in the inner detector with transverse momentum above 2 GeV. The longer light green boxes indicate calorimeter energy clusters associated to the two reconstructed photons while the red line indicates the track of the reconstructed muon. The dashed white line indicates the missing transverse momentum from a neutrino. (Image: ATLAS Collaboration/CERN)

The ATLAS Collaboration observed the Wγγ process with a significance of 5.6 standard deviations, above the 5 standard deviation threshold required to claim observation. This corresponds to a less than 0.00003% chance that the measurement was due to a statistical fluctuation. This result benefited from the increased 13 TeV centre-of-mass energy (thereby increasing the cross section of the process), and improved data-driven techniques used to estimate several backgrounds from misidentified objects in the detector. These included backgrounds from events in which one or both of the recorded photons were actually misidentified hadronic jets (originating from quarks or gluons), or misidentified electrons. Researchers measured the cross section with a precision of 17%. Example event displays for the electron channel and muon channel decays are shown in the header and Figure 1, respectively.

Proton Collisions,Event Displays,Physics,ATLAS
Figure 2: Event display of a W(→νμ)Z(→ee)y candidate. The muon (red line) is reconstructed from hits in the inner detector and muon spectrometer. The neutrino is inferred via the presence of missing transverse momentum (dashed grey line). The electrons are reconstructed from clusters of energy deposits in the electromagnetic calorimeter (green layer of the detector) which are matched to tracks in the inner detector. A third cluster in the electromagnetic calorimeter is not matched to any track and hence becomes a candidate for a photon (orange, right). (Image: ATLAS Collaboration/CERN)

The WZγ process was observed with a significance of 6.3 standard deviations, also well above the threshold to claim observation. Similar to the Wγγ analysis, the main challenges for this measurement were the data-driven estimates for the backgrounds from misidentified objects. The largest of these backgrounds is from hadronic jets being misidentified as either leptons or photons. Researchers also measured this cross section with a precision of 17%. An example event display for the muon channel is shown in Figure 2.

The Zγγ cross section was measured with a precision of 12%, making this the most precise measurement of this process to date. In their new analysis, physicists also measured the dependence of the cross-section on certain variables describing the Zγγ system – providing a more extensive test of Standard Model predictions. These measurements were used to constrain the magnitude of physics beyond the Standard Model that could arise through anomalous interactions between electroweak gauge bosons. The measured cross section for this process as a function of the transverse momentum of the two leptons from the Z boson is shown in Figure 3. Also shown in the figure is an example of a new physics contribution, if assumptions are made about the energy required to produce it. As good agreement was found between the data and prediction, physicists were able to confidently rule out such new physics effects at these energies.

The studies of tri-boson processes remain in an early stage but these recent results demonstrate the feasibility to do these kinds of measurements. The understanding of these processes is also crucial to measuring other physics processes – such as the production of W or Z boson alongside a Higgs boson which then decays to two photons – which will become evermore important in Run 3 of the LHC and beyond. With more data and higher energies, these processes will provide further insight into a relatively unexplored area of the Standard Model and could even reveal new physics beyond our current understanding.

Figure 3: The measured cross section of Zyy production as a function of the transverse momentum of the dilepton pair. The measurement, labelled data, is compared to predictions from the Standard Model given in blue and green. If beyond-Standard-Model physics becomes accessible at 1 TeV, the size of the effect predicted by an example model is shown in orange. (Image: ATLAS Collaboration/CERN)

About the event display in the header: Display of a candidate W(→eν)γγ event. Orange lines indicate tracks in the inner detector with transverse momentum above 2 GeV. The coloured boxes indicate energy deposits in calorimeter cells with transverse energies above 250 MeV in the liquid argon electromagnetic calorimeter (green), 250 MeV in the tile calorimeter (yellow), and 800 MeV in the hadronic endcap calorimeter (cyan). The longer light green boxes indicate calorimeter energy clusters which are either associated with a green track in the inner detector for the reconstructed electron, or not for the case of the two reconstructed photons. The dashed white line indicates the missing transverse momentum. No jet has been reconstructed with transverse momentum above 20 GeV. (Image: ATLAS Collaboration/CERN)

Learn more