Measurement of the top quark pair production cross section based on lepton+jets event kinematics

The full paper is available here.

What is the top quark ?
One of the earliest questions pondered by mankind is what our world is made of or, alternatively, what are the building blocks of the universe. This question has had many different answers as science explored matter to ever smaller scales searching for its fundamental constituents.

Present knowledge of the building blocks of matter and their interactions is described in the so called "Standard Model" (SM). Within this framework, matter is comprised of six leptons and six quarks, and the interactions among them are mediated by the exchange of other particles, called gauge bosons. Quarks and leptons are assigned to three "generations", according to their hierarchy in mass.

The sixth quark, called the "top" quark, was discovered only 10 years ago at the Fermi National Accelerator Laboratory (Fermilab) by the CDF and experiments, 20 years after being predicted by the SM. The top quark is extremely heavy for a fundamental particle: it is about as heavy as a gold nucleus, and it decays very rapidly before it can form any bound state. This feature, which is not shared by any other quark, provides a unique opportunity to study directly the decay of quarks. Also, it is thought that the large mass of the top quark is intimately connected to the mass-generating process within the SM, described via exchanges of the Higgs boson. As a result, measurements of the top quark mass provide predictions of the mass of the yet undiscovered Higgs particle. These properties make the top quark a unique test stand for the SM.

Top quark production and decay
In the SM, top quarks are predominantly generated in pairs via the strong interaction, that is, the top quark (t) is most often produced together with its antiparticle, the antitop quark (t¯). Both the top quark and the antitop quark decay into a lighter quark and a W boson (carrier of the weak force). Since W bosons can decay in different ways, the final signature of the top quark pair decay depends on the decay mode of the two produced W bosons. The possibilities are: Our analysis focuses on the second case, specifically, when one W boson decays into an electron (e) or muon (μ) and its corresponding neutrinoe or νμ). In addition to the leptons, four quarks are produced: two from the original top quark decays and two from the second W boson. The quarks form collimated particle showers that we refer to as "jets", hence giving the studied decay channel the name "lepton+jets".

Finding the top quark
At the Tevatron, top quark pairs can be produced in proton-antiproton collisions, and their decay products observed in the DØ detector. Each recorded proton-antiproton collision is called an event. Since the production of a top quark pair is a very rare process, the search for events containing top quark pairs among all the produced events is similar to looking for a needle in a haystack. Consequently, very elaborate data-analysis methods are required to separate the desired events from background.

In accordance with the above description of the lepton + jets signature of the top quark pair decay, we require one energetic electron or muon in our events, together with at least four jets. Since neutrinos interact exceedingly weakly, they traverse the detector undetected. This causes an apparent imbalance in momentum conservation in the event that we also require to observe. After this selection, there are two main sources of background that can mimic a top quark pair decay:

Multijet events can be suppressed by applying stringent quality criteria to the observed leptons. The remaining background can be distinguished from signal by exploiting specific kinematic properties of top quark pair events compared to background – for example, due to the large top quark mass, its observed decay particles are more energetic than the particles in a non-top event, and since the top quark pair is produced mainly at rest, its decay particles will be emitted more isotropically, whereas the particles in a typical background event tend to be emitted in a direction close to the initial proton-antiproton axis. Such differences between signal and background can be combined into a discriminating likelihood variable that will tend towards values of 1 for signal events and 0 for background events. By comparing the shape of distributions in this variable in the selected data with expectations for signal and background, obtained from simulations or special data samples, we can extract the amount of top quark, W boson and multijet events that maximise the probability to be compatible with the observed distribution. The result of this determination is shown in the figures below for events containing electrons (left/up) and muons (right/down).

The result
The goal of our measurement is to determine the rate at which top quark pairs are being produced and to compare our observation with the SM prediction. In analogy with a dart-game, where the rate of hits depends on the size of the dart board, we use a cross section in units of area that provides a measure of the production rate for a particular process: 1 barn (b) = 10-28 m2.

As already mentioned, the production cross section for top quark pairs is tiny. The total proton-antiproton cross section is 61 millibarns (1 mb = 10-3 b), whereas the SM expectation for top quark pair production is 6.8 picobarns (1 pb = 10-12 b), 10 orders of magnitude lower than the total proton-antiproton interaction or cross section. Taking into account the size of the total data sample, the inefficiencies (losses) due to our selection criteria, and the fraction of events in which a top quark pair leads to an electron+jets or muon+jets final state (17% each), we obtain the total top quark pair production cross section from the determined number of signal events of 6.7 pb ± 1.4 pb, where the 1.4 pb reflects statistical uncertainty of the small event sample. This is in very good agreement with SM expectations. In addition, the individual cross sections for the electron+jets and muon+jets data agree with each other and with predictions.

Using the selections for top quark events described above, properties of the top quark – such as its mass – can be also studied. Other decay channels of top quark pairs are also being checked for agreement with SM predictions. By refining our analysis methods, and using the continuously accumulating data (a data sample at least a factor of 20 that used in this analysis is expected to be available by 2009), we are moving towards precision measurements in top quark physics, probing the validity of the SM and looking for new discoveries.

If you have any questions about this analysis, please contact the primary authors C. Gerber, T. Golling, G. Otero, M.-A. Pleier, E. Shabalina and J.-R. Vlimant.