Constraints on New Physics from Dijet Angular Distribution

We use the measurement of the dijet angular distribution, discussed at dijet_angle_1ab.html, to constrain the possibility of new physics beyond the standard model. All plots and numbers are PRELIMINARY. We use the variable R_chi: the ratio of the number of events with chi<2.5 to the number of events with 2.5< chi <5. Here chi=|exp(eta1-eta2)|=(1+|cosine theta*|)/(1-|cosine theta*|. Our measured value of R_chi versus mass is shown in Fig.1, with both the statistical and correlated systematic uncertainty. We form a chisquared between data and theory that includes both the uncorrelated statistical unceratainties and the correlated systematic uncertainties. In Fig. 2 the chisquared is plotted as a function of 1/Lambda**4, where Lambda is the scale of a left handed contact interaction among quarks (Eichten and Lane model, recently corrected in hep-ph/9605257 ). This is done for four different kinds of contact interactions: including u and d quarks only, or including all quarks, both with either positive or negative sign of the contact term. As in Fig.1, the theory has been normalized so that it gives NLO QCD with renormalization scale mu=Pt when the compositeness scale is Lambda=infinity. The 95% confidence level lower bound on Lambda (for composite u and d quarks only) is Lambda_ud(+)>1.6 TeV and Lambda_ud(-)>1.4 TeV. The former excludes our best fit from the jet Et distribution of approximately Lambda_ud(+)=1.6 TeV, while the latter isn't sufficiently stringent to exclude a similar fit to the jet Et distribution of Lambda_ud(-)=1.8 TeV. The dijet angular distribution is more sensitive to Lambda with a positive sign. In Fig. 3 we show our limits compared to the measured angular ratio; most of the difference in the height of the Lambda_ud(+) and Lambda_ud(-) limit curves is due to round-off of the limit to two significant figures, and the remainder is due to real differences in + and - sign angular distributions as a function of mass and there affect on the chisquared. Also derived in Fig. 2 is the limits on a flavor symmetric contact interaction: Lambda(+)>1.8 TeV and Lambda(-)>1.6 TeV. This kind of contact is theoretically preferred to avoid flavor changing neutral currents, and is therefore more appropriate for comparison with limits from Drell-Yan. In Fig.4 we show our new limits compared to the measured angular ratio; the same comments as for Fig. 3 apply here. In Fig.5 we compare the mass and angular distribution to QCD predictions using the same mass scale. In Fig. 6 we overlay compositeness curves (u and d only) in both mass and and angular distributions, and in Fig. 7 we do the same for a flavor symmetric contact interaction. In Fig. 8 we do a similar overlay for the coloron model of a new massive gluon (Chivukula, Cohen and Simmons). The angular distribution is currently not sensitive enough to exclude the value Lambda_C(-)=2 TeV proposed in hep-ph/9603311 to account for the jet Et excess. In Fig. 9 we do a similar overlay for the Hadrophyllic Z' model (Altarelli et al.). Again the angular distribution is not sufficiently sensitive to exclude a hadrophyllic Z' of mass 1.0 TeV proposed in hep-ph/9601324 to account for the jet Et excess.

• fig. 1: dijet_angle_ratio_comp_ud.ps--- Angular Ratio and Compositeness (only u and d composite)
• fig. 2: dijet_chisquared.ps--- Chisquared Distributions and Limits
• fig. 3: dijet_comp_ud_limit.ps--- Compositeness Limit (u and d only)
• fig. 4: dijet_comp_limit.ps--- Compositeness Limit (flavor symmetric)
• fig. 5: dijet_mass_angle.ps--- Mass and Angular Ratio
• fig. 6: dijet_mass_angle_comp_ud.ps--- Mass and Angular Ratio and compositeness (u and d only)
• fig. 7: dijet_mass_angle_comp.ps--- Mass and Angular Ratio and compositeness (flavor symmetric)
• fig. 8: dijet_mass_angle_coloron.ps--- Mass and Angular Ratio for coloron
• fig. 9: dijet_mass_angle_zprime.ps--- Mass and Angular Ratio for hadrophyllic Z'

Last updated May 13, 1996