Difference: LightQuarkEta (r45 vs. r44)

Studies of the light quark pseudorapidity distribution in t-channel single top

At the Tevatron, the t-channel single top quark signal has a very nice variable that can be exploited to separate the single top quark signal from the backgrounds. The light quark, also called spectator quark, that is scattered in the t-channel exchange of a W boson typically goes in the forward direction. The corresponding variable is called Q*Eta. In this variable, the pseudorapidity of the leading non-b-tagged jet is multiplied by Q, the charge of the lepton (i.e. the charge of the top quark). The distribution is peaked in the forward direction, at a pseudorapidity of about 1.5 to 2. The plot below shows this variable, this is from a study of Tevatron t-channel single top quark production at NLO done by us at MSU (PRD72,094027 (2005)). tevQEtaSmall.png

LHC, Atlas

At the LHC the situation is different. First, the initial beam is proton-proton rather than proton-antiproton. Thus, every distribution is symmetric in eta, and selecting events by charge of the lepton doesn't make any difference. Second, the initial beam energy is much higher, and top quarks are produced farther away from threshold and hence experience the top quark and the light quark jet will be more central.


The light quark pseudorapidity distribution isn't expected to change much when going from LO to NLO. Shown below is the output from the ONETOP program, comparing the light quark jet eta distribution between leading order and NLO. ONETOP_QEta ONETOP_QEta


Here we show the results of a quick study using MCFM. In this study, the top quark mass was set to 175GeV, and the same value was chosen for factorization and renormalization scales. The CTEQ6M PDF set was chosen. Both top and antitop quark production is included. With these settings, the total cross section at NLO for top quark production only is 150pb, and that for antitop production is 109pb. Note that MCFM uses massless b-quarks for s-channel and t-channel single top.

MCFM output, no cuts

  • t-channel light quark eta distribution from MCFM, before any cuts:

* t-channel light quark jet pT distribution from MCFM, before any cuts:

There is no clear peak at high pseudorapidities. The distribution is rather wide and almost flat out to an eta of about 2. For higher pseudorapidities it drops off quickly.

MCFM output, with Atlas-style cuts

The following cuts were made to mirror what has been used in recent Atlas analyses:

  • lepton pT>20GeV
  • MET>20GeV
  • At least two jets with jet pT>30GeV

The result is the following:

  • t-channel light quark eta distribution:
  • t-channel light quark jet pT distribution from MCFM, after Atlas-style cuts:

The distribution after cuts has changed and is peaked at the center (eta=0). The unique t-channel feature has disappeared. This is mostly the result of the jet pT cuts. With a looser cut of 20GeV on the pt of the jets, the distribution is more flat.

Leading Order Distributions

The light quark jet pseudorapidity distribution looks very different if only the leading order term is included in the calculation. Below are the two relevant distributions:

  • t-channel light quark eta distribution at LO, before any cuts:
  • t-channel light quark eta distribution at LO, after Atlas-style cuts:

Distributions separated by jet multiplicity

The light quark jet pseudorapidity distribution shows a peak in the forward direction at leading order. Here we check whether that peak is still visible if we use the NLO distribution, but restrict ourselves to 2-jet events. Shown below are is the eta distribution of the light quark jet at NLO, but separately for 2-jet events and 3-jet events. The 2-jet events are made up of LO events, events with virtual corrections, and events with real gluon emission for which the gluon (or b-quark) momentum was low or absorbed into one of the other quark jets. Conversely, 3-jet events are produced when the additional gluon (or jet) radiation is a high momentum so that it will be reconstructed and separated from the other objects in the event. Below are the two relevant distributions: * t-channel light quark eta distribution at NLO, but only for 2-jet events:

  • t-channel light quark eta distribution at NLO, but only for 3-jet events:

It is clear from these distributions that the real gluon emission diagrams are responsible for the flattening out of the eta distribution of the light quark jet.

-- ReinhardSchwienhorst - 06 Apr 2007 -- ReinhardSchwienhorst - 13 Apr 2007

This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback