jump to navigation

Probing the top pair production mechanism March 9, 2007

Posted by dorigo in news, physics, science.

CDF just released a nice analysis which, although still seriously limited by its statistical power, starts to test in a new way our understanding of the mechanism at work when a top-antitop pair is produced by the 2-TeV collisions produced by the Tevatron’s proton and antiproton beams.

QCD, the theory of the strong force keeping together hadrons (such as protons and neutrons, but also more exotic, short-lived particles) is the basis of our description of the interactions between quarks and gluons – the constituents of hadrons.

When we collide a proton with an antiproton at high enough energy, what may happen is that a quark from a proton actually withstands a close encounter with an antiquark from the antiproton. Their annihilation creates energy, which may be released by the creation of a new quark-antiquark pair – and when we are lucky (once in ten billion collisions), a rare top-antitop pair!

Another mechanism that can create a heavy quark-antiquark pair is the collision of two gluons. These are also among the projectiles’ constituents, but they usually carry too little energy to be able to create such a powerful collision as those that may end up in the creation of top quarks.

Taking in account the distribution of probabilities to find quarks and gluons of a given energy in the colliding protons and antiprotons, QCD allows us to compute with good accuracy what fraction of the top-antitop pairs were actually created by gluon collisions: about 15%.

A brief parenthesis: at the Large Hadron Collider (LHC) in construction at the CERN laboratories in Geneva, the top quark pairs will be produced with opposite frequency by gluon (85%) and quark (15%) collisions: that is because the higher energy of the projectiles favors the gluons, and the fact that both beams of the LHC are made of protons makes it harder to find a high-energy antiquark (yes, there are antiquarks in the proton! But they sit silent and get little share of dad’s energy). 

What Ricardo Eusebi, Eva Halkiadakis, Sunil Somalwar, and Jared Yamaoka measured was precisely the fraction of gluon collisions yielding top pairs. They did so by considering several measurements  of the kinematics of top quarks in the final state, many of them boiling down to the same detail: the angle made by the emitted quarks from the colliding beams. Indeed, a quark-quark annihilation usually yields more central top quarks, so by measuring an angle called “theta-star” one can separate the two production mechanisms. Another detail is the spin correlation of the quarks, but I won’t discuss it here.

In the picture on the left you see what kind of separation we are talking about: indeed, there is only a little difference in the cosine of theta-star to play with, and if the statistics of the top quarks data sample used is not large enough, the measurement fails to produce a significant result. This is not the case, though: with 695 inverse picobarns of collisions, CDF puts a limit: the gluon-gluon fraction is smaller than 51%, at 95% confidence-level.

What that means is that CDF excludes that gluon-gluon collisions are the largest source of top pairs. This is of interest not only to test QCD, but also to test for potential new physics mechanisms of production of top quark pairs. If, for instance, the top pairs were produced by a high-mass resonance, the experimental result would be bound to disagree with Standard Model predictions (well, QCD). 

I think this is cool, and I look forward to new precise measurements of this quantity!

As for the past, I need to mention that there has already been another, largely independent measurement of the gluon-gluon fraction in top production by CDF. That analysis looked at the number of charged tracks produced by the collisions yielding top quarks, in the knowledge that gluon collisions produce more tracks, because gluons have a higher color charge! The group, led by Pekka Sinervo from Toronto University, was able to measure a gg fraction of 0.24+-0.26, in good agreement with both theory and with the newer determination. Testing the same quantity with different methods is a very useful thing to do, because when discrepancies are found, one learns a lot… Or things agree, which is also neat.


1. Kea - March 9, 2007

Phew! Did you run for the computer even before you’d had a shower? Anyway, cool post. We hope to have our version of ‘QCD’ up and running before 2009.

2. dorigo - March 9, 2007

Hi Kea,
😉 indeed! Mundane activities such as changing clothing or showering can wait if a stronger force drives me on the other side of the house… I do have wireless, but I have so far been unable to fit my laptop in a transparent plastic envelope to shower and blog at the same time…

Wow, QCD explained! That will be a nobel prize winner! And I will definitely nominate you!

3. Jester - March 9, 2007

Welcome back🙂 You even managed to drop by my place between the shower and the hairdryer😉

Question: how much is this sensitive to new physics? How light should be the resonance you mention in order to show up in this analysis?

4. dorigo - March 9, 2007

Hi Jester,

I think a ttbar resonance with a mass close to threshold – say 400 or 500 GeV – could make itself seen in the decay kinematics, if it made up for a sizable fraction of the top pairs.
That, however, is not the case, and in fact it would be like finding an elephant by the amount of nitrogen in its droppings rather than looking at it between the eyes. A ttbar resonance would be easy to spot in Tevatron data, and in fact CDF and D0 did set limits to its existence, by just reconstructing the total invariant mass of the system.

Despite that, I believe that a probing of the production mechanism retains a lot of interest.


Sorry comments are closed for this entry

%d bloggers like this: