New Particles Decaying to di-Jets

January 3, 2009 at 12:55 pm 3 comments

The CDF Collaboration recently published a very nice paper: Search for New Particles Decaying to di-Jets. The authors have taken the most basic kind of event available in a hadron collider, events with two energetic jets, and made a precise measurement of the di-jet mass spectrum, looking for a deviation that might be a signal for physics beyond the standard model. A brief summary is available from the corresponding CDF public web page.

No evidence for new physics was found.

Nonetheless, the paper is very interesting for several reasons. First, the data set is not small, corresponding to more than 1.1 fb-1, and the capabilities of the CDF detector have been pushed to their ultimate level, as far as jet physics is concerned. The previous run of the Tevatron, called Run I, collected only 10% of this luminosity, yet was able to find the top quark. Both CDF and D0 have about 5 fb-1 on tape, ready to be analyzed, and some analyses have been shown which use a large fraction of that amount. Some years ago, pessimists thought the Tevatron would not deliver more than 4 fb-1 before being shut down – we will certainly have 6 fb-1 and hopefully more like 8 fb-1 per experiment, at which point Higgs searches will become extremely fruitful.

Collider detectors like CDF measure the kinematics of individual tracks well, and electrons, photons and muons are also very clear and well measured. Jets, however, are bundles of hadrons which cannot be fully reconstructed as tracks (about a third of them are neutral) and which do not leave the kinds of showers in calorimeters that allow precise energy measurements. For various technical reasons, the calorimeters themselves do not deliver a nice, linear signal, so a whole tier of corrections is required before measurements of jet energies – and di-jet masses – are accurate. Many scientists at CDF contributed to the testing and refinement of these corrections, including some of the main authors of this paper. The control of the jet energy scale and the removal of measurement bias is central to this measurement, and the success of this analysis is proof of the mastery of these corrections by the relevant people within the CDF Collaboration. Let me point out that the di-jet mass spectrum presented in this paper extends over 1 TeV in energy, and the rate falls by eight orders of magnitude.

CDF di-jet mass spectrum

CDF di-jet mass spectrum

Second, the methodology of this kind of analysis – including the search for “bumps” and “shoulders” in the di-jet mass spectrum – is improved incrementally in this paper. In that sense, this paper and others like it serve as a “how-to” example for searches to be carried out in the future.

Third, the measurements are so accurate that the experimental errors (mainly coming from those jet energy corrections) are not large compared to the uncertainties coming from the parton distribution functions (PDF’s for short). The distribution of quark and gluon energies in the incoming proton and anti-proton beams is needed for predicting the di-jet mass distribution, yet we cannot calculate those parton (quark and gluon) energies without reference to many other empirical measurements from the past. The PDF’s amount to an intelligent parametrization of the momentum fraction of the partons based on these earlier measurements, so the uncertainties from the measurements as well as the arbitrariness of the parametrization unavoidably lead to an uncertainty in those parton energy distributions. One hopes that this measurement, and other similar measurements from both Tevatron experiments, can be folded back into the PDF’s to reduce these kinds of uncertainties.

ratio of measured spectrum to predicted spectrum.  Shows sensitivity to PDF's.

ratio of measured spectrum to predicted spectrum. Shows sensitivity to PDF's.

Finally, the CMS and ATLAS collaborations will make this same measurement at the LHC as soon as they get good collision data. Since the center-of-mass energy will be much higher than at the Tevatron (the Tevatron is slightly less than 2 TeV, and the LHC will run, hopefully, at 10 TeV this year, and 14 TeV in 2010), high di-jet masses can be reached with a lot less luminosity – just a small fraction of 1 fb-1. The control of the jet energy measurement, however, will take a lot of work and effort, and cannot be expected in the first weeks of LHC collision data. That said, both collaborations have detailed and tested procedures in place which will deliver the whole set of necessary corrections. A public document explains the CMS plan for jet energy corrections, for example. The craft involved in this kind of search will be inherited from the Tevatron experiments – indeed, some of the authors of the CDF paper are members of CMS…

Entry filed under: Particle Physics.

Antidote for “No Sympathy for Physics Departments” Update – Jet Measurements to be used in PDF fits

3 Comments Add your own

  • 1. dorigo  |  January 3, 2009 at 1:39 pm

    Hi Michael,

    great post, and I cannot but chuckle at this stupid “automatic generator of related posts” which picks my scribblings…

    About the measurement: I remember the Run I measurement of the dijet mass xs, of which I was a godparent. It was the first time I ran into that funny statistical problem of correlated uncertainties that make a comparison of data with different PDF sets so hard to perform by eye. The very same kind of plot you post above, in Run I was comparing data with CTEQ and MRST (or another set), and one of the two sets was off by two sigmas in all points, but had a shape ever so slightly more alike the experimental data, while the other was passing through all the points but had a slight bend in the shape. A chi2 comparison way preferred the former over the latter!


  • 2. Michael Schmitt  |  January 3, 2009 at 1:49 pm

    Thanks, Tommaso, I appreciate your comment very much! (But you are too modest about your own posts…)

    I also remember the Run I paper, which generated a certain amount of debate at the time. I was a little surprised that the authors of this latest paper did not compare to MRST as well as CTEQ. But I’ll bet the differences in the predictions are now very small…

    Being a godparent to a good paper is very rewarding since one has a chance to gain a lot of insight, more than one gains simply by reading the paper. 🙂

  • […] I wrote about the CDF di-jet mass spectrum and suggested that the measurement could be used to improve PDF’s. Ken Hatakeyama […]


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Trackback this post  |  Subscribe to the comments via RSS Feed

January 2009

%d bloggers like this: