The exchange of a quark or gluon between interacting partons in hadronic collisions typically results in final-state particle production over several units of rapidity. In contrast, the exchange of a color-singlet is expected to yield a rapidity gap, defined as the absence of particles in a region of rapidity (or pseudorapidity) The observation of rapidity gaps between jets at both Fermilab in proton-antiproton collsions and DESY in proton-electron collisions implies the exchange of a hard color-singlet. The measured fraction of dijet events arising from color-singlet exchange is roughly 1% in proton-antiproton collisions and 10% in positron-proton collisions. These rates are too large to be explained by electroweak boson exchange and indicate a strong-interaction process.

We present results on dijet production via hard color-singlet exchange in proton-antiproton collisions at center-of-mass energies of 630 GeV and 1800 GeV using the DØ detector. The fraction of dijet events produced via color-singlet exchange is measured as a function of jet transverse energy, separation in pseudorapidity between the two highest transverse energy jets, and proton-antiproton center-of-mass energy. The results are consistent with a color-singlet fraction that increases with an increasing fraction of quark-initiated processes and inconsistent with two-gluon models for the hard color-singlet.

Fig. 1: (top left) Jet characteristics of the 630 GeV opposite-side jet data sample (solid line) and 1800 GeV opposite-side sample (dashed line). The normalized distributions are shown for (a) the average E_T, (b) Dh, and (c) average x of the two leading jets. (eps file)
Fig. 2: (top middle) Multiplicity in the region |h|<1 between the two leading jets for the high-E_T sample: (a) two-dimensional multiplicity: number of calorimeter towers (n_cal) vs. number of tracks (n_trk) (b) n_calonly with Negative Binomial fit. (eps file)
Fig. 3: (top right) Two-dimensional multiplicity n_cal vs. n_trk in the region |h|<1 for the (a) 1800-OS and (b) 630-OS samples. (eps file)
Fig. 4: (bottom left) The color-singlet fraction: (a) as a function of the second leading jet E_T; as a function of Dh between the two leading jets for (b) the low-E_T sample and (c) the high-E_T sample; (d) as a function of average x for each Dh bin in (b) and (c). Statistical error bars and relative normalization uncertainties for each sample (hatched bands) are shown. (eps file)
Fig. 5: (bottom middle) Fits of Monte Carlo models to the color-singlet fraction (a) as a function of E_T and (b)--(c) Dh for the low-E_T sample and the high-E_T sample, respectively. Shown are comparisons to BFKL jet level (solid line), BFKL parton level (dashed line), photon (dot-dashed line), and U(1) (dotted line) models. (eps file)
Fig. 6: (bottom right Fits of Monte Carlo models to the color-singlet fraction (a) as a function of E_T and (b)--(c) Dh for the low-E_T sample and the high-E_T sample, respectively. Shown are comparison to free-factor (solid line), soft-color (dashed line), single-gluon (dotted line), and simple two-gluon (dot-dashed line) models. (eps file)

DZero Collaboration

• Jet Production via Strongly-Interacting Color Singlet-Exchanges in pbarp Collisions
  S. Abachi et al, D0 Collaboration, PRL {76} 734 (1996), FERMILAB PUB-95/302-E.
• Rapidity Gaps Between Jets in pbarp Collisions at root-s=1.8 TeV
  S. Abachi et al, D0 Collaboration, PRL {72} 2332 (1994), FERMILAB PUB-94/005-E.

CDF Collaboration

• Events with a Rapidity Gap Between Jets in pbarp Collisions at sqrt(s)=630 GeV
  F.Abe et al, CDF Collaboration, Submitted to PRL, FERMILAB PUB-98/334-E.
• Dijet Production by Color-Singlet Exchange at the Fermilab Tevatron
  F.Abe et al, CDF Collaboration, PRL {80} 1156 (1998), FERMILAB PUB-97/283-E.
• Observation of Rapidity Gaps in pbarp Collisions at 1.8 TeV
  F.Abe et al, CDF Collaboration, PRL {74} 855 (1995).

ZEUS Collaboration

• Rapidity Gaps between Jets in Photoproduction at Hera
  M.Derrick et al, ZEUS Collaboration, PLB {369} 55 (1996).