We have studied prompt diphoton production in proton-antiproton collisions at a center-of-mass energy of 1.8 TeV by the Collider Detector at Fermilab CDF.
Diphoton production in hadron-hadron collisions is one of the clear probe for testing the Next-to-Leading Order (NLO) calculation of the Quantum Chromodynamics (QCD) because of the advantage of the detection of photons with small systematic uncertainty and good energy resolution. The measurement of the diphoton production cross section can be used to check the QCD calculation. By looking at the transverse momentum of the diphoton system, diphoton production is also a good probe for measuring the effects of multiple initial soft gluon radiations and the intrinsic transverse momentum of the initial state partons, which are beyond the NLO prediction.
The data were collected by the CDF detector placed at B\O \,area of the Fermilab Tevatron collider during 1992\,-\,1995 collider run. The total integrated luminosity was about 110\,$\hbox{pb}^{-1}$. The trigger system, including both the hardware and the software modules, required the collision events to have two large and isolated electromagnetic energy depositions in the Central Electromagnetic Calorimeter\,(\,CEM\,)\,.
At offline level, we selected EM clusters which had no tracks associated with them. Using the Central Electromagnetic Strip chamber (CES), which were embedded in the CEM near the shower maximum, we required that the EM shower profile should be consistent with that obtained from testbeam electrons. Candidate photons with large additional CES clusters were rejected. At $p_T\geq12$\,GeV/c, we obtained 652 events as the diphoton candidate events. The efficiencies of each selection criterion for the photons were calculated with various CDF physics datasets.
The backgrounds against prompt photons come from neutral mesons\, (\,$\pi^0$'s, $\eta$'s, and $K_s^0$'s\,)\,decaying into multiple photons. We cannot separate them on event-by-event basis, so we subtracted the backgrounds statistically by using the information from the CES shower profile and the conversion rate in the Central Preradiator detector (CPR) placed in front of the CEM.
The photon fraction in the diphoton candidates was evaluated as a function of the transverse momentum ($p_T$) of the photons. It showed a large photon purity at high $p_T$ region\,(0.6 at $p_T=25$\,GeV/c).
The differential cross section for diphoton production was measured as a function of the photon $p_T$, the invariant mass ($M$) of the diphotons, the azimuthal angle between diphotons ( $\Delta\phi$), the ratio of the transverse momenta of the two photons ($Z$) and the $p_T$ of the diphoton system. The results were compared with the NLO QCD prediction with CTEQ2M parton distribution functions. The results, as a function of the photon $p_T$, $M$ and $\Delta\phi$, were consistent with the NLO prediction, while the results on the $Z$ and the diphoton system $p_T$ showed slight differences between data and theory.