Giorgio Bellettini, Giuseppe Latino (INFN-Pisa),
Vadim Rusu (Fermilab),
Marco Trovato, Caterina Vernieri (Scuola Normale Superiore - Pisa)


We present a study of the invariant mass spectra of jets in events with one identified lepton and missing transverse energy at the CDF~II experiment at the Fermilab Tevatron. This signature is sensitive to the possible productions of heavy particles produced in association with a W bosons. We use the full CDF dataset (fb-1). We perform a study of the invariant mass distribution of jet pairs in the final state. We find good agreement between data and standard model predictions. Therefore we set a 95% confidence level limit on the cross section of a possible resonance at 145 Gev/c2 produced in association with a W boson and decaying into a jet pair


At hadron colliders, the production of jet pairs in association with vector bosons offers measurements of fundamental standard model (SM) parameters and tests of theoretical predictions. In a previous publication, the CDF collaboration reported a disagreement between data and the SM prediction using a data sample corresponding to 4.3 fb-1 of data [1]. Assuming an excess of events over the background prediction appearing as a narrow Gaussian distribution, the statistical significance of the reported disagreement was 3.2 standard deviations. In this current document, we report an update the previous analysis using the full CDF Run~II data set, more than doubling the candidate event sample size. In addition to the larger data set, we present the investigation of a series of systematic uncertainties which were not conclusive using the prior, smaller, data sample, or were previously unconsidered. As a result of these studies, new calibrations of the detector response and instrumental backgrounds is performed, yielding far better agreement with the standard model prediction from Monte Carlo (MC) event generators. Using this improved model of the detector and the background processes, we perform measurements of diboson production cross sections in the semileptonic decay mode WW->lν+jj, WZ->lν+jj. Good agreement with the predictions of the SM via next-to-leading-order (NLO) predictions is found. We also present searches for the excess of events over the background prediction described in [1]. No significant excess is observed, and a detailed study of the affect of the new calibrations on the dijet mass spectra is presented in [2].

Event Selection

We select events with one and only one electron ET>20 GeV or muon with pT>20 GeV/c large transverse missing energy MET>25 GeV and exactly two jets with ET>30$ GeV and eta<2.4. In order to reject multijet backgrounds we impose the following cuts: trasnverse mass cut mT>30 GeV, azimuthal angle between the most energetic jet and MET, DeltaPhi(MET,j1)>0.4, difference in pseudo-rapidity between the two jets, DeltaEta(j1,j2)<2.5 and the transverse momentum of the dijet system pT(j1,j2)>40 GeV/c.

Event Selection

We use the standard jet energy scale from calorimeter level to hadron level jets. For MC processes we further correct the jet energy scale about 1% up for quark-jets and 8% down for gluon jets. Further details are described elsewhere [2]

Background modeling

We model the SM signal and background processes using a variety of Monte Carlo (MC) simulation programs. The diboson processes (WW, WZ and ZZ) are generated with Pythia. The top-quark pair production is generated with Pythia by assuming a top-quark mass of 172.5 GeV/c2. The single top-quark productions, both s- and t-channel, are modeled using MADGRAPH+Pythia. The productions of W/Z plus jets are simulated by Alpgen with showering and hadronization performed by Pythia. To model the multi-jets background we use a data driven method: for the muon sample we use non-isolated muons while for the electron sample we require objects to pass all the identification cuts except two non-kinematical one. We define those objects "non-electrons". Both non-isolated muons and non-electrons have to pass all the other selection cuts mentioned earlier. Non-isolated muon and non-electron samples are orthogonal to the electron and muon samples respectively. We observe two major problems in electron qcd model: trigger bias and wrong energy scale for the non-electrons. Details about those problems and the respective adjustments are reported in [2]. The QCD yields in the sample are estimated by performing a fit to the data in MET.

In the selected datasample we see the following background contamination

We check our modeling by comparing our predictions to the data


In order to test the hypothesis of the presence of other signals, fits must be performed, comparing the data with sums of signal and background predictions. These predictions have associated systematic uncertainties for both shapes and rates. Further info about the systematic uncertainties are available at [2]. The fit is performed in dijet invariant mass by maximizing a binned likelihood function. We want to mirror what was done in [2], therefore we use five templates for the fit: diboson (WW+WZ+ZZ), W/Z+jets, top (ttbar and single-top), QCD multi-jets background and a gaussian template with mean = 145 GeV/c2 and width=14.3 GeV/c2. Systematic uncertainties in both shapes and rates are treated as nuisance parameters in the fit. No excess with a mass of 145 GeV/c2 is observed. Therefore we set a 95% confidence level limit on the cross section of a possible resonance at 145 Gev/c2 produced in association with a W boson and decaying into a jet pair. Such a limit is 0.9 pb

The agreement between standard model predictions and data is satisfactory

[1], Invariant Mass Distribution of Jet Pairs Produced in Association with a W Boson in ppbar Collisions at 1.96  TeV Phys. Rev. Lett. 106, 171801 (2011)
[2], Invariant Mass Distribution of Jet Pairs Produced in Association with a W=Z boson in pp Collisions at CDF Public CDF Note

This public page is created by Marco Trovato.