Where no-one has gone before

Guest post by Alexander Grohsjean

Born at the end of the ’70s, I was still in school when the heaviest of all quarks was discovered at the Tevatron: the top quark. Back then I had no idea what it was about. But reading an article in the newspaper I felt the excitement surrounding such a discovery. My interest for the smallest and most basic building blocks of the universe had been awakened. When I joined the CMS Collaboration in 2014, I had no doubt that the first measurement I would like to do was that of the production rates of top-quark pairs at the new energy regime of 13 TeV. Shortly after the restart of the LHC in summer this year, we began a journey where no-one has gone before.

img
A top-quark candidate in the CMS detector. Credit - CMS Collaboration

Since its discovery in 1995, the top quark has been measured with ever higher precision. By the end of its lifetime, the Tevatron provided a total of approximately 70,000 top-quark pairs. This is just a hundredth of what was produced at the LHC so far and there is a lot more to come. The large datasets from both accelerators have allowed us to pin down several properties of the top quark. So far, no deviation from the Standard Model has been found. The most precisely measured property of the top quark is its mass. With an uncertainty of only 0.64 GeV for a particle of around 173 GeV, the top quark’s mass is known to a level of better than 0.5%. And it is just this property that makes the top quark so special and the most puzzling particle among all quarks. Due to its high mass, the top quark has the strongest connection to the Higgs boson, suggesting that the top quark might play a special role in the mechanism that explains how particles acquire their mass, called “electroweak symmetry breaking”. Or, speaking more generally, the top quark is a very good candidate to find new physics beyond the Standard Model.

Top quarks are predominantly produced as quark-antiquark pairs. Only about 3% of tops decay into an electron-muon final state, in which one W boson from the top decay transforms into an electron (plus neutrino) and the other into a muon (plus neutrino). However, this is still one of the most attractive final states at the LHC as it provides an almost pure data sample of top-quark pairs: the contamination from background is only at the  5%-10% level. Thus it was this decay channel that we decided to explore when measuring the first 13TeV production rates for top quarks at CMS.

img
Example of a Feynman diagram of the studied top-antitop production process. Credit - Christian Schwanenberger and Ann Heinson

Preparations for the measurement commenced a long time before the actual data arrived. One of the first meetings took place in November 2014. Many groups joined that meeting and we discussed the best strategy for the measurement. At that time it was foreseen to collect about 1 fb–1 (inverse femtobarns), corresponding to around 800,000 top-quark events, providing sufficient statistics for a precision result. However, as any new start-up involves risks, we wanted to be on the safe side and decided to use two different approaches. On one hand a solid and robust cut-and-count approach where the number of data is simply counted; by measuring the efficiency of the detector to reconstruct top-quark events and the acceptance of the detector for these events, the production rates can be extracted using a text book formula. On the other hand we decided to also work on a much more sophisticated approach in which we would constrain the systematic uncertainties by using Monte Carlo simulations and data.

On the timescale of the summer conferences this year, only 42 pb–1 (inverse picobarns) of data were collected by CMS, i.e. just about 4% of the amount of data originally foreseen. Still, we wanted to take a first look since these were enough data to probably get a first glimpse of new physics that could have been revealed, for example, by the existence of a new heavy particle. Since we would anyway be limited in accuracy, we decided to use the simpler cut-and-count method. The analysis still required to manage and understand many ingredients such as the sophisticated Monte Carlo generators. Some of them were used for the first time and had to be carefully validated and tested. Furthermore we needed to measure and understand the trigger rate (triggers select the interesting physics events from the flood of other not-so-interesting ones), the efficiency to reconstruct and identify electrons and muons in the detector, as well as to measure jet energies and their resolution and many many things more.

In the end, after we had control over everything, we finally went where no-one has gone before and looked into the data. We had 220 selected events: of these we expected 28 to be background and the rest to come from pairs of top quarks.

Comparing our result to the predictions from theory, we found agreement. This means that on the one hand we did not find any hint for new physics, but on the other hand we could confirm that the predictions of the Standard Model are right, even for the new energy regime. Now we know over a very wide range – from 2 TeV at the Tevatron, over 7 and 8 TeV at the first run of the LHC, to 13 TeV at the current LHC Run-2, so over almost one order in magnitude – that top-quark pairs are produced at the rate predicted by the Standard Model. It also shows us once more, that the particle discovered 20 years ago is indeed the top quark.

Now, this was only the very first step in an unknown territory, but soon we will have more data to confirm this picture by analysing more production and decay channels, or, and even better, we will discover new physics. But up to now we can conclude that the Standard Model is still alive and actually healthier than ever before!

img
Measurement of top-quark pair-production rate compared to the Standard Model's predictions as a function of the center-of-mass energy. The new result of the CMS Collaboration at 13 TeV is displayed in red and is in agreement with the theory prediction (green band). Credit - CMS Collaboration

The views expressed in CMS blogs are personal views of the author and do not necessarily represent official views of the CMS collaboration.