Top Quark Asymmetry - Something Else To Watch
    By Tommaso Dorigo | August 10th 2010 10:13 AM | 8 comments | Print | E-mail | Track Comments
    About Tommaso

    I am an experimental particle physicist working with the CMS experiment at CERN. In my spare time I play chess, abuse the piano, and aim my dobson...

    View Tommaso's Profile
    CDF and DZERO, the two experiments at the Fermillab Tevatron collider, have studied top quark production since their own discovery of the heavy particle in 1995 (see here, here, and here for a three-post history of the top quark quest). As the number of produced proton-antiproton collisions kept increasing since 2002, the number and depth of studies performed in top-quark production events have similarly furthered our understanding of the fascinating physics of the heaviest known elementary particle.

    Top quarks are produced by the energetic interaction of a quark and an antiquark, or a pair of gluons. The high energy required for the creation of two top quarks guarantees that we can calculate with precision the rate and the features of their production, using the knowledge of quantum chromodynamics (QCD), the theory of the strong force that keeps nuclei together and that governs the interaction of their constituents: quarks and gluons.

    By measuring the observed characteristics of events featuring top quark decay signals, we can verify very cleanly the detailed theoretical predictions. This is what CDF and DZERO have been doing for 15 years: first by measuring the production rate; then by determining the decay modes of top quarks; then by looking for resonant production of top pairs, by searching for single top production, by measuring directly the top quark spin, the charge, etcetera.

    One thing that has been accessed with precision only with the recent wealth of analyzable data is the production asymmetry. This is basically a measure of how frequently the (anti-)top quark prefers to be emitted in the direction of the colliding (anti-)proton. Theory predicts that there should indeed be a small asymmetry in the production of top pairs by proton-antiproton collisions, but this asymmetry is entirely ascribed to second-order effects: at leading order the strong force should behave perfectly oblivious of the electric charge of the quarks. You might think of it as due to the fact that as far as colour charge is concerned, top and antitop quarks are the same particle.

    Previous measurements of the top asymmetry had shown some disagreement with predictions, but error bars were still large (I discussed the matter briefly here). Now, however, CDF made public a more precise analysis which utilizes over five inverse femtobarns of collisions, resulting in a dataset of over a thousand well-reconstructed top pair production events. And recently, DZERO also analyzed a similarly sized dataset, in search of their own precise determination of the top asymmetry. And what have the two experiments found ?

    The results are quite interesting: both experiments find a significant asymmetry, larger than what is expected by theoretical calculations. CDF finds a laboratory-frame parton-level asymmetry of 0.150+-0.055, when theory predicts 0.038+-0.006. DZERO instead finds a reconstructed asymmetry of 8+-4%, while 1+-2% would be expected by simulations.

    A technical note is needed here: the two results cannot be directly compared. That is  because CDF reduces the observation to a parton-level asymmetry to compare to a direct theoretical calculation, while DZERO does the opposite, passing the theory through a simulation of the reconstruction. Still, the two results agree in the essence: the production of top quarks is more asymmetric than what our present-day theoretical tools allow us to predict. This is evident if you compare the CDF and DZERO distributions of the top quark rapidity difference, below.

    Above, the CDF result with 5.3 inverse femtobarns of collisions. The black points are experimental data, and the two histograms stacked one atop the other are top production (in green) and background events (in blue). The black points "lean" to the right, which is the observed asymmetrical behaviour.

    And below, look at the DZERO distribution of 4.3 inverse femtobarns of top pair candidates. Here top is in red, and two different backgrounds are shown below it.  Again, the black points lean to the right: positively-charged top quarks tend to be emitted preferentially in the proton direction, if ever so slightly.

    The discrepancy is at the level of two standard deviations (a one-in-twenty chance) in both cases, so it is not significant enough to get particle physicists excited about new physics lurking in the production of top quarks: however, two plus two is almost three...

    ...Almost. Because if some systematic effect is causing one experiment to measure an asymmetry in top production, it is quite possible that the same effect affects also the other experiment lying a third of the ring away! These days I tend to enrich my posts with such sobering remarks, given the fact that I have -unjustly, I feel- fingered as a over-hyper of insignificant deviations of high-energy physics results from expectations. I think it is useful to keep the excitement high for the physics that the Tevatron collider is producing, but I still have to make my beliefs clear: I do not think these results are an indication of new physics! And further, let me also add my disclaimer of liability, to explain that I am speaking on behalf of nobody but myself only in this article:

    I do disclaim thee, o Liability!
    Ok - today's form of my disclaimer would not go very far in a libel cause, but it is a way to remember a friend, so allow me an eccentric twist.


    Go on T., I dare you. Claim that tentative evidence of Z' has been found.
    As we like to say, you might as well be hung for a sheep as for a lamb!

    Or a cow. I prefer to keep my head away from loops these days DB!

    As much I can say: it would be such a beautiful twist if we found a Z', and then a fourth generation of matter fields by inspecting multijet mass spectra at the Z' mass (see Andrea Giammanco's important point in the comments thread of my previous posting on 4th generation 2-sigma effects!).


    This finding is indeed intriguing. In fact, it may go hand in hand with the alleged breaking of CPT symmetry in neutrino physics as claimed by recent MINOS experiments. If confirmed as a first-order effect (and this remains to be seen), breaking of color symmetry between top and anti-top and violation of the mass equivalence of neutrinos and anti-neutrinos may very well be a portal to new physics. And this even before LHC sets new exclusion limits on SUSY!



    Arrrgghhh! Again, arXiv "moderated" a submission of mine. It's on the information content of the CKM and MNS matrices so it's down your alley, perhaps you can tell me what's seriously wrong with it: Quantum Information of Unitary Mixing Matrices. They rejected this even for gen-ph, the place where they put articles of doubtful use. Very annoying. I waited to submit it until Foundations of Physics accepted it as the sort of thing they publish and sent it out for review.

    Meanwhile, I'm studying for the physics GRE in October. 25 years ago I got a 990, I'm hoping to repeat.

    Well, the title sounds cranky to me, It sounds too much like "quantum information" was inserted there just for the effect as the whole makes little sense to me, after all the mixing matrices are sets of numbers not qubits so how can they contain quantum information?

    looks like either analysis could add some cuts (harder Ht, double tag, etc) to get a smaller dataset but with higher purity. and then have a smaller asymmetric background to subtract. did they do this and see that the asymmetry remained?
    -- jlm

    If this pair is the same particle, how can this assymmetry then be explained? Chirality alone would not do that? A partial wave-particle duality?

    Hi Ulla, the pair is a particle-antiparticle pair. The asymmetry is due to sub-leading effects, like Coulomb interactions etcetera.