A new type of ring-imaging Cherenkov detector is being used for hadronic particle identification in the BaBar experiment at the SLAC B Factory (PEP-II). This detector is called DIRC, an acronym for Detection of Internally Reflected Cherenkov (Light). This paper will discuss the construction, operation and performance of the BaBar DIRC in detail.

A fast RICH prototype with a CsI photocathode has been built. The quantum efficiency has been estimated by two independent techniques: a UV lamp measurement and a detection of Cherenkov light produced in a liquid radiator. The crucial influence of the photocathode substrate on the quantum yield is demonstrated in this paper. The results obtained with our prototype have been used to simulate the performances of a Fast RICH for a B meson factory.

The principles of the DiRC ring imaging Čherenkov technique are briefly explained and its choice for the BaBar detector particle identification system is motivated. A large scale prototype of the DIRC for the BaBar experiment is then described. Details of the design of this prototype and its test in a hadronic particle beam at the CERN-PS are presented, and results from various prototype and test configurations are given. For example, after correcting for geometrical acceptance and estimated collection effects, the number of photoelectrons was measured to be 146 ± 1.8 ± 9 cm−1, for a track angle of 20° at zero photon transmission distance. The effective attenuation loss was measured to be 4.1 ± 0.7% per meter of bar length, and the observed single photon resolution was 10.0 ± 0.2 mrad. This performance is consistent with what was expected from earlier tests and Monte Carlo simulations, and will be fully adequate for the physics demands of the BaBar experiment.

The Bugey 3 experiment, designed to measure oscillations of reactor neutrinos, has used 3 identical detection modules, each of 600 liters, filled with a new 6Li-loaded liquid scintillator. These modules were located in two shielding bunkers, respectively 15 and 40 m away from the reactor core. We describe here the mechanical characteristics of these modules, their shielding, the associated electronics, the trigger, the acquisition systems, the calibration and monitoring of these detectors, and the Monte Carlo simulations of their response to particles. We conclude on the overall performance of this new detection technique which has allowed the recording of 120000 neutrino interactions with good neutron efficiency (49%), low background (2.5 evts/hr) and good energy resolution (4% at 4.4 MeV).

We have measured the transmission of a 25 μm thick foil of teflon AF in the ultra violet region. The properties of this material allow one to use it for protecting UV sensitive materials against external contaminations.

CsI photocathodes were studied in order to evaluate their potential use as large photoconverters in RICH detectors for the PID system at ALICE (LHC in heavy ion collider mode). It has been demonstrated that a quantum efficiency comparable to the reference value obtained on small samples can be obtained on CsI layers evaporated on large pad electrodes operated in a MWPC at atmospheric pressure. We present a survey of the results obtained in the laboratory on small samples irradiated with UV-monochromatic beams and with RICH detectors of proximity-focusing geometry at a 3 GeV/c pion beam.

We report here on the results obtained by the CERN RD26 collaboration on the production and characterization of large area photocathodes, susceptible to equip fast UV-photon imaging devices. Such detectors are planned for some Ring Imaging Cherenkov (RICH) detector projects, in particular HADES at SIS Darmstadt, BABAR at the SLAC asymmetric B-factory, and ALICE at the LHC (CERN).

The quantum efficiency of CsI photocathodes operating in a proportional gas chamber is measured in the wavelength range 185–220 nm. The tests are carried out with methane gas at atmospheric pressure using a pulsed UV lamp. Two different techniques are used to produce the photocathodes. The CsI is either evaporated under high vacuum or under low pressure of argon. The quantum efficiency of the photocathodes produced by the first technique is measured to be 9% at 190 nm while it is enhanced by a factor of 1.3 and 2 at 190 nm and 200 nm respectively for the photocathodes prepared with the second method. No significative drop of the quantum efficiency is observed over a total period of one month.

We performed a systematic investigation of the quantum efficiency of some solid reflective photocathodes in the spectral range 140–240 nm. The measurements were made without gaseous amplification in vacuum and in methane. No significant difference was found among CsI photocathodes prepared by vacuum deposition at different institutes, either from powders or from crystals of different origins, and measured either in vacuum or in methane. Amorphous silicon photocathodes were prepared by the plasma enhanced chemical vapor deposition technique. We present the results for several doping conditions of amorphous silicon and for p-n junctions. Some organometallic photocathodes, containing iron or other transition metals (cerium), were evaporated and measured. Among them decamethylferrocene exhibits the highest quantum efficiency in the range 190–240 nm.

The absolute quantum efficiency of two Philips XP2020Q photomultipliers and one Hamamatsu R2059 photomultiplier are measured in the VUV range for three wavelength values, (193, 201, and 253 nm). We used a deuterium lamp for which the absolute radiance has been calibrated; the wavelength range is selected by means of optical filters.

The Cherenkov light produced by 2.8 GeVc protons in a solid NaF radiator has been detected with a CsI photocathode coupled to a multiwire proportional chamber. The measurement of the number of photoelectrons gives an estimation of the photocathode quantum efficiency.

We describe a method and an experimental layout allowing the monitoring of photomultiplier gain. We use artificial blue light (spark gap with filter: 436 ± 20 nm) and three reference detectors. Short term and long term measurements are presented. The results indicate a precision better than 0.5% for the short term and 1.4% for the long term determinations. This gain monitoring system has been developed for a new neutrino oscillation reactor experiment (600 photomultipliers) starting at the Bugey nuclear plant.