The CMS experiment has conducted a comprehensive study of the rates of production of top quark pairs created along with a photon, both inclusively and as a function of several kinematic variables.
As the heaviest fundamental particle, the top quark plays a crucial role in particle physics. At the Large Hadron Collider (LHC), top quarks are abundantly produced in pairs via the strong interaction, allowing precision tests of the Standard Model (SM) of particle physics and potential discoveries beyond it.
While top quark pair production in general is well studied, some more distant corners of the process have not been explored as thoroughly and can give us complementary information. Among these is the photon-associated top quark production, “ttγ”, which probes the coupling between the top quark and the photon. This interaction is precisely predicted by quantum electrodynamics (QED), so even small deviations could hint at new physics phenomena.
A new CMS measurement of the ttγ process was presented at the La Thuile 2025 conference by Beatriz Ribeiro Lopes, a CERN postdoctoral researcher who has been leading the analysis effort. The study used proton-proton collision data collected by CMS from 2016 to 2018.
In this process, two top quarks are produced with a photon. Each top quark decays into a W boson and a bottom quark, t → Wb. The analysis focuses on events where W bosons decay into a lepton (electron or muon) and a neutrino, as these leave a very clean signature in the detector. Since neutrinos escape undetected, signal events are identified by a photon, two leptons, two b jets, and missing energy. A signal-like event is displayed below.
Above: An event recorded with the CMS detector at a center-of-mass energy of 13 TeV, compatible with the ttγ signature.
Which particle emits the photon? It could be a top quark—most interesting for our study—but also a quark from the colliding protons, or even a top quark decay product. Experimentally, it is challenging to distinguish between these cases.
A major difficulty in this analysis is simulating the complex signal process, so we can compare it to data. Existing simulation tools cannot accurately model ttγ at the required precision, as calculations fail to converge. Therefore, we use a trick: we simulate only a part of the signal at high precision, and the rest at lower precision, and then combine the two. It is the first time that this is done in the CMS experiment to achieve a better description of the ttγ process.
In order to extract the rate of production of ttγ, formally called the “cross section”, we look at the distribution of the angular distance between the photon and the closest lepton. This variable is very powerful to distinguish events where the photon originates from the top quark decay products—in which case it is typically close to the leptons—from those where the photon is emitted from the top quarks or the initial quarks—in which case it is more separated from the leptons. This is shown in the figure below, where the orange contribution corresponds to simulated events with a photon from the decay, and the red contribution corresponds to events with a photon from the initial part of the process. By fitting the data to the prediction, we measure a cross section of 134±8 fb for the ttγ process, which is consistent with the predictions from the standard model of particle physics.
Above: Distribution of the angular distance between the photon and the closest lepton (electron or muon). The expected number of events for the signals after fitting them to the data are represented in the orange and red histograms, while the other colors represent the backgrounds. The black points show the data events.
We also measured the cross section as a function of seven kinematic observables. Energy-related variables, like transverse momenta, show good agreement with the predictions. However, angular variables reveal a slight tension with respect to the predictions, suggesting challenges in modeling photon emission in top quark events. This information can help theorists refine models of this process.
A key result is the measurement of the ratio between the cross section of the ttγ process and the total cross section for production of top quark pairs, the first at the LHC. Ratios help reduce uncertainties, as shared uncertainties in the numerator and denominator cancel out. By analyzing top quark pair events with and without a reconstructed photon simultaneously, CMS determines a ratio of (1.25 ± 0.05)%. This means that about 1% of top quark pairs are produced with an energetic photon. This ratio was also measured differentially, as a function of two different observables, to study whether this percentage changes for events with different kinematic properties.
The measurement of this ratio is still limited by the size of the data set; larger datasets currently being recorded in Run 3 will allow us to significantly improve the result. This will help to further deepen our understanding of the top quark and its interactions with photons, and as a consequence, our understanding of fundamental particle physics.