Event Displays,Physics,ATLAS
Figure 1: A 2018 ATLAS event display consistent with the production of a pair of W bosons from two photons, where the W bosons decay into a muon and an electron (visible in the detector) and neutrinos (not detected). The muon path (red line) and electron path (yellow line) are shown. The electron deposits its energy in the electromagnetic calorimeter (yellow blocks). The many particles reconstructed in the Inner Detector are shown in orange. Top left corner shows that these particles do not originate from the same interaction and are thus attributed to additional proton–proton interactions. (Image: ATLAS Collaboration/CERN)

The ATLAS Collaboration announces the first observation of two W bosons produced from the scattering of two photons — particles of light — at the International Conference on High-Energy Physics (ICHEP 2020).

Plots or Distributions,Physics,ATLAS
Figure 2: Feynman diagram depicting the production of a pair of W bosons from two photons in a four force-carrier interaction. The photons are scattered off of two protons which in the process lose energy but remain intact. (Image: ATLAS Collaboration/CERN)

In everyday life, two crossing light beams follow the rules of classical electrodynamics and do not deflect, absorb or disrupt one another. However, at the high energies seen in LHC collisions, effects of quantum electrodynamics become important. For a short moment, photons radiated off the incoming proton beams can scatter and transform into a particle–antiparticle pair which appears as light-by-light interactions in the detector. This process was first observed by the ATLAS Collaboration in 2019. Indeed, the Standard Model describes quantum electrodynamics as part of electroweak theory, which not only predicts that force-carrying particles – the W bosons, Z boson and photon – interact with ordinary matter, but also among themselves. 

The newly observed process proceeds via a very rare type of phenomenon where two photons collide to directly produce two W bosons of opposite electric charge via a four force-carrier interaction, among others (see Figure 2). Although the ATLAS and CMS Collaborations saw first evidence of this process in data recorded during Run 1 of the LHC (2011–2012), its observation required the substantially larger dataset taken during Run 2 (2015–2018). 

This rare process occurs as bunches of high-energy protons skim past each other in “ultra-peripheral collisions”, if only their surrounding electromagnetic fields interact. Quasi-real photons from these fields scatter off one another to produce a pair of W bosons and leave a distinct signature in the ATLAS experiment. As the skimming protons stay intact, the only detectable particles produced in the interaction are the visible decay products of the W bosons – namely, for this measurement, an electron and a muon with opposite electric charge. 

The ATLAS Collaboration has observed the rare interaction of two W bosons with two photons with statistical significance of 8.4 standard deviations.

Plots or Distributions,Physics,ATLAS
Figure 3: The distribution of the number of particles reconstructed in the ATLAS inner detector, in addition to the electron and the muon. The background process with the largest contribution is the W-boson-pair production from proton constituents; its simulation (blue) describes the observed data (black points) very well. The photon-induced W-boson pair production accumulates at low particle multiplicities (white area). (Image: ATLAS Collaboration/CERN)

ATLAS physicists had to overcome several unique challenges to observe this process, starting with separating the signal from background. LHC protons can break up into their constituents and fragment into several detectable particles at very low energy. In particular, W-boson pairs can be produced from the proton’s constituents. This background process is hundreds of times more likely to occur than the photon–photon production of W-boson pairs, and can mimic its signature. To enhance the signal over such a background, physicists only selected collisions where no other charged particles are measured in the vicinity of the electron and the muon, as reconstructed in the ATLAS Inner Detector.

Further, a typical collision event contains particles from 20 to 60 additional proton–proton interactions occurring simultaneously as bunches of proton cross in ATLAS. These additional particles can prevent the identification of signal events if they are produced in close proximity to the photon–photon interaction (see Figure 1). 

Physicists developed novel experimental techniques to precisely determine the contributions of these effects. Simulated events can be used to estimate the expected backgrounds, but detailed tuning on the data is needed to ensure that they provide a faithful description. Physicists performed auxiliary measurements using data consistent with resonant Z boson production, a well-understood process produced with high frequency and purity at the LHC. This dataset was used to count particles from both the additional proton-proton interactions and the proton fragmentation, and the findings allowed ATLAS physicists to tune the simulation of such events. The accurate description of these background processes made the observation of this rare phenomenon possible (see Figure 3, where the photon–photon signal interactions accumulate at low particle multiplicity).

A total of 307 events matching the selection requirements were found in the analysed dataset, of which 174 were attributed to be from the photon–photon production of W-boson pairs and the remaining events to various background processes. Such a yield corresponds to a statistical significance of 8.4 standard deviations, which is well above the established 5 standard deviations criterion for the unambiguous observation of a process. The cross section is measured to be 3.13 ± 0.42 fb. This means that only one or two such interactions occurred in the 30 trillion proton–proton interactions of a typical day of data-taking in 2018.

The four force-carrier interaction is an integral part of electroweak theory and, at the same time, can be sensitive to modifications of the Standard Model by unaccounted-for new physics. The experimental techniques presented by the ATLAS Collaboration will enable future measurements that can probe such modifications and test the electroweak theory in a novel way.