Search for dijet mass resonances at CDF (1.1 fb-1 analysis)

Primary contacts: Kenichi Hatakeyama¹, Anwar Bhatti¹, Robert Harris²

Rockefeller University¹
Fermilab²

for the CDF Collaboration

Submitted to PRD: arXiv:0812.4036

			FIG. 1: (a) The measured dijet mass spectrum for both jets to have |y| < 1 compared to the NLO pQCD prediction obtained using the CTEQ6.1 PDFs. (b) The ratio of the data to the NLO pQCD prediction. The experimental systematic uncertainties, theoretical uncertainties from PDF, the ratio of MRST2004/CTEQ6.1, and the dependence on the choice of renormalization and factorization scales are also shown. An additional 6% uncertainty in the determination of the luminosity is not shown. [eps| gif]
			FIG. 2: (a) The measured dijet mass spectrum (points) fitted to Eq. (2) (dashed curve). The bin-by-bin unfolding corrections is not applied. Also shown are the predictions from the excited quark, q*, simulations for masses of 300, 500, 700, 900, and 1100 GeV/c2, respectively (solid curves). (b) The fractional difference between the measured dijet mass distri- bution and the fit (points) compared to the predictions for q* signals divided by the fit to the measured dijet mass spec- trum (curves). The inset shows the expanded view in which the vertical scale is restricted to +/-0.04. [eps| gif]
			FIG. 3: Dijet mass distributions for simulated signals of the q*, RS graviton, W', and Z' with the mass of 800 GeV/c2. [eps| gif]
			FIG. 4: Observed 95% C.L. upper limits on new particle production cross sections times the branching fraction to dijets obtained with the signal shapes from (a) W', (b) Z', (c) RS graviton, and (d) q* production. Also shown are the cross section predictions for the production of W', Z', RS graviton, #rhoT8, q*, axigluon, flavor-universal coloron, and E6 diquark for the set of parameters described in the text. The limits and theoretical predictions are for events in which both of the leading two jets have |y| < 1. [eps| gif]

Dijet mass distributions from QCD and expected signals
	Excited quark: Dijet mass spectrum expected from QCD and excited quark production with excited quark mass=300, 500, 700, 900, and 1100 GeV/c2. The excited quark decaying to a quark-gluon pait is simulated with Pythia with f=f'=fs=1. [eps| gif] W': The W' production is simulated with Pythia with SM couplings. The W' production cross section is multiplied by a factor of 1.3 to account for the k-factor. [eps| gif] Z': The Z' production is simulated with Pythia with SM couplings. The Z' production cross section is multiplied by a factor of 1.3 to account for the k-factor. [eps| gif] Randall-Sundrum graviton: The R-S graviton production is simulated with Pythia with k/MPL=0.1. The R-S graviton production cross section is multiplied by a factor of 1.3 to account for the k-factor. [eps| gif]
Dijet mass distribution from four models that produce dijet mass resonances
	Dijet mass distribution from four models that produce dijet mass resonances. [eps| gif]
Test of a parametrization form used for modeling QCD production with Pythia and Herwig dijet MC samples
	The dijet mass spectra from Pythia and Herwig Monte Carlo and fits to the parametrization form: Pythia [eps| gif] Herwig [eps| gif] (Note: these spetra are the "detector-level" dijet mass distributions, not corrected to the particle-level.)
Dijet mass distribution from data and fit to a parametrization form
	The measured dijet mass spectrum and results of a fit to the parametrization form: [eps| gif] No significant indication of resonant structure is observed. (Note: this measured spetrum is the "detector-level" dijet mass distribution, not corrected to the particle-level.)
Effect of systematic uncertainties on limits for new particle production decaying into dijets.
	The four parameters in the parametrization form (shown above) are not a priori determined. When forming likelihood distributions as function of signal cross sections at each new particle mass, the likelihood is maximized keeping these four parameters free (profiling). The other systematic uncertainties from (1) jet energy scale (JES), (2) jet energy resolution, (3) luminosity, (4) trigger prescale, and (5) trigger efficiency are folded into the likelihood by the Bayesian approach. The effects of these systematic uncertainties on the limits are shown in: top [eps| gif] bottom [eps| gif]
Limits on production of new particles decaying into dijets.
	[top] 95% C.L. limits on the Randall-Sundrum graviton and color-octet technirho producion (red), and limits on the excited quark, axigluon, flavor-universal coloron, and E6 diquark (black), compared with the theoretical predictions for these particle production. [eps| gif] [bottom] 95% C.L. limits on the W' and Z' production compared with their theoretical predictions. [eps| gif] The shown cross section predictions and limits are for the total particle production cross section times branching fraction to dijets times the kinematic acceptance that the leading two particle-level jets are within |y(jet1,2)|<1.

Mass exclusion:

Observed mass exclusion range	Model description
260-870 GeV/c2	Excited quark (f=f'=fs=1)
260-1100 GeV/c2	Color-octet technirho [top-color-assisted-technicolor (TC2) couplings, M'8=0, M(pi228)=5M(rho)/6, M(pi221)=M(pi228)/2, M8=5M(rho)/6]
260-1250 GeV/c2	Axigluon and flavor-universal coloron (mixing of two SU(3)'s, cot(theta)=1)
290-630 GeV/c2	E6 diquark
280-840 GeV/c2	W' (SM couplings)
320-740 GeV/c2	Z' (SM couplings)

• The measurement of the dijet mass differential distribution corrected to the particle-level and comparisons with the QCD predictions [link] (under the CDF QCD group page).