Low-pileup data pin down top-quark production

The top quark – the heaviest known elementary particle – differs from the other quarks by its much larger mass and a lifetime that is shorter than the time needed to form hadronic bound states. Within the Standard Model (SM), the top quark decays almost exclusively into a W boson and a b quark, and the dominant production mechanism in proton–proton (pp) collisions is top-quark pair (tt) production.

Measurements of tt production at various pp centre-of-mass energies at the LHC probe different values of Bjorken-x, the fraction of the proton’s longitudinal momentum carried by the parton participating in the initial interaction. In particular, the fraction of tt events produced through quark–antiquark annihilation increases from 11% at 13 TeV to 25% at 5.02 TeV. A measurement of the tt production cross-section thus places additional constraints on the proton’s parton distribution functions (PDFs), which describe the probabilities of finding quarks and gluons at particular x values.

In November 2017, the ATLAS experiment recorded a week of pp-collision data at a centre-of-mass energy of 5.02 TeV. Although the main motivation of this 5.02 TeV dataset is to provide a proton reference sample for the ATLAS heavy-ion physics programme, it also provides a unique opportunity to study top-quark production at a previously unexplored energy in ATLAS. The majority of the data was recorded with a mean number of two inelastic pp collisions per bunch crossing compared to roughly 35 collisions during the 13 TeV runs. Due to much lower pileup conditions, the ATLAS calorimeter cluster noise thresholds were adjusted accordingly, and a dedicated jet-energy scale calibration was performed.

Now, the ATLAS collaboration has released its measurement of the tt production cross-section at 5.02 TeV in two final states. Events in the dilepton channel were selected by requiring opposite-charge pairs of leptons, resulting in a small, high-purity sample. Events in the single-lepton final states were separated into subsamples with different signal-to-background ratios, and a multivariate technique was used to further separate signal from background events. The two measurements were combined, taking the correlated systematic uncertainties into account.

The measured cross section in the dilepton channel (65.7 ± 4.9 pb) corresponds to a relative uncertainty of 7.5%, of which 6.8% is statistical. The single-lepton measurement (68.2 ± 3.1 pb), on the other hand, has a 4.5% uncertainty that is primarily systematic. This measurement is slightly more precise than the single-lepton measurement at 13 TeV, despite the much smaller (almost a factor of 500!) integrated luminosity. The combination of the two measurements gives 67.5 ± 2.6 pb, corresponding to an uncertainty of just 3.9%.

The new ATLAS result is consistent with the SM prediction and with a measurement by the CMS collaboration, though with a total uncertainty reduced by almost a factor of two. It thus improves our understanding of the top-quark production at different centre-of-mass energies and allows an important test of the compatibility with predictions from different PDF sets (see figure 1). The result also provides a new measurement of high-x proton structure and shows a 5% reduction in the gluon PDF uncertainty in the region around x = 0.1, which is relevant for Higgs-boson production. Moreover, the measurement paves the way for the study of top-quark production in collisions involving heavy ions.