RICH pickings in Cassis

The ring imaging Cherenkov (RICH) technique is used extensively in nuclear, high-energy and astroparticle physics experiments to identify charged particles via the measurement of their Cherenkov emission angles, over momenta ranging from a few hundred MeV/c to several hundred GeV/c. The technological feats of the single-photon RICH detection technique – manifested in the most extreme case via the 3D imaging of single photoelectrons that have been created by Cherenkov photons and then been allowed to drift a metre or so through gas at atmospheric pressure – remain unmatched in any other detection technology.

In 1993, Eugenio Nappi of INFN Bari and Tom Ypsilantis of Collège de France launched a series of international workshops as a forum for reviewing new developments and perspectives in this powerful technique. RICH2010, the latest in the series, took place on 3–7 May in the French Mediterranean port of Cassis. A broad programme of invited and contributed talks, as well as poster presentations, attracted 115 participants from 25 countries, reflecting the expanding application of Cherenkov imaging in accelerator-based particle and nuclear physics, astroparticle physics and neutrino astronomy. In addition to 10 invited talks, the programme included 42 contributed talks, which were selected from some 80 submissions to allow time for extensive discussions; the other 34 contributions were presented in poster sessions.

From the LHC to Lake Baikal

The workshop began appropriately with a comprehensive review of the fundamentals of Cherenkov-light imaging and recent developments by Jurgen Engelfried of the University of San Luis Potosi. The opening session on operating RICH detectors in nuclear- and particle-physics experiments then provided an opportunity to see the first calibration measurements with real LHC data from the ALICE and LHCb experiments.

The High Momentum Particle Identification Detector (HMPID) in ALICE employs a perfluoro-n-hexane (C₆F₁₄) liquid radiator with photon imaging via a reflective caesium iodide (CsI) photocathode operating in a multiwire proportional chamber (MWPC) filled with methane (CH₄) at atmospheric pressure. The detector has already demonstrated the expected π/K separation up to 3 GeV/c and proton identification up to 5 GeV/c; a future upgrade should extend the momentum range beyond this. LHCb has two RICH systems: RICH1, operating with aerogel and perfluoro-n-butane (C₄F₁₀) radiators; and RICH2, with a carbon tetrafluoride (CF₄) radiator, to provide a combined particle-identification range over 2–100 GeV/c. The data already taken clearly demonstrate the identification of hyperons and strange mesons.

Also at CERN, the COMPASS experiment has a RICH detector that has been operating since 2002 in beam rates as high as 10⁸ Hz with a C₄F₁₀ gas radiator for hadron identification over 3–60 GeV/c. The detector was subsequently upgraded with multianode photomultiplier tubes (MAPMTs) replacing the four central CsI MWPCs with pad read-out. This should allow an improvement from the present operation at 40 MHz to deadtimeless operation at 100 MHz in the central region. The NA62 experiment will use a RICH detector with a 17 m neon radiator and a focal plane of 2000 PMTs. Designed for electron–muon separation between 15 and 35 GeV/c, it should begin data-taking in 2012.

At Brookhaven, the hadron-blind RICH in the PHENIX experiment has demonstrated extremely high efficiency for photon detection in windowless operation, with CF₄ serving as radiator and for photoelectron detection in a gas-electron multiplier (GEM) device with a CsI photocathode. Ionization from passing hadrons is trapped on electrodes in the GEM, allowing for clean electron identification in gold–gold collisions at the Relativistic Heavy Ion Collider.

Following the success of the BaBar experiment at SLAC, techniques for the detection of internally reflected Cherenkov (DIRC) light produced in quartz bars continue to evolve. The barrel focusing-DIRC for the proposed Super-B facility would be more compact than its BaBar ancestor. It would also have quartz blocks instead of water in the focusing zone and MAPMTs with sub-200 ps time resolution or multianode microchannel plate PMTs to replace conventional tubes and provide time-of-propagation (TOP) measurements that could resolve the colour of individual Cherenkov photons. The more compact forward region would use an aerogel radiator.

SuperKEKB, the upgraded B facility planned at KEK, will operate at beam luminosities of more than 80 times the current value of 10³⁴ cm^–2s^–1. The Belle II experiment will employ a challenging TOP detector that relies on new multichannel hybrid devices based on avalanche photodiodes (APDs) that are under development at Hamamatsu Photonics. The PANDA experiment at the Facility for Antiproton and Ion Research (FAIR), Darmstadt, will include a barrel DIRC that follows similar principles to barrel designs for Super-B and Belle II, while for the forward directions a DIRC detector with innovative disc geometry is under study.

The expansion in the use of imaging Cherenkov detectors in astroparticle physics, witnessed in earlier RICH workshops, continues to accelerate. Imaging air Cherenkov telescope (IACT) arrays use the Earth’s atmosphere as radiator. The High Energy Stereoscopic System (HESS) array in Namibia has discovered numerous high-energy gamma-ray sources and a legacy survey of the galactic plane is almost complete. Sensitivity will be further increased in HESS-II with a fifth and larger (600 m²) dish added at the centre of the array. The future Cherenkov Telescope Array (CTA), with around 80 dishes of three diameters, is expected to be approved in 2013. The CTA will greatly increase sensitivity to astrophysical gamma-ray sources through the exploitation of ongoing advances in mirror coatings, photon-camera technology and control systems for telescope positioning. Some of these advances are already becoming apparent in the innovative photon cameras of the MAGIC-II IACT and in the First Avalanche-photodiode Camera Test (FACT) project for a novel camera using Geiger-mode APDs (G-APDs), both on La Palma.

The ANTARES neutrino telescope, by contrast, looks downwards, using seawater as the radiator and the Earth to filter out up-going charged cosmic rays. At a depth of 2500 m in the Mediterranean Sea, the telescope was completed in 2008 and results based on an analysis of around 1000 detected neutrinos were presented. The constructional and operational experience gained in ANTARES represents a major step towards the KM3NeT multi-cubic-kilometre neutrino telescope for the deep Mediterranean, which is being pursued by a consortium that includes members of the ANTARES, NEMO and NESTOR neutrino-detector projects.

The Lake Baikal neutrino telescope has been running while increasing in size since 1993. With the deployment of additional optical detectors it will soon reach a target mass of a gigatonne of water. Nearby, the complementary 1 km² TUNKA-133 extensive air-shower array is coming into operation. This will be sensitive to primary cosmic rays in the energy range of 10¹⁵–10¹⁸ eV.

Glimpses of the future

The first permanent RICH installation outside the Earth’s atmosphere will soon be achieved with the launch of the Alpha Magnetic Spectrometer (AMS-02) in the final NASA space shuttle mission (STS-134) to the International Space Station in February 2011. The RICH subdetector consists of sodium fluoride and aerogel radiators, with a conical mirror and MAPMT photon detection. Tests at CERN last year using cosmic rays and a test beam confirmed the expected performance of the RICH subdetector.

Rapid, ongoing developments in solid-state, vacuum-based and gaseous photon detectors were reviewed in invited talks by Samo Korpar of Maribor, Toru Iijima of Nagoya and Silvia Dalla Torre of INFN Trieste, respectively. Leszek Ropelewski of CERN completed the picture with a fascinating evening seminar on developments in micropattern gas detectors (MPGDs) within the RD51 collaboration.

Solid-state, single-photon detectors continue to mature. Progress on the design of G-APDs has led to commercialized silicon photomultipliers, offering single-photon sensitivity with high detection efficiency, high gain for bias voltages less than 100 V, excellent (tens of picoseconds) timing resolution and operation in high magnetic fields. Current disadvantages include a high (temperature-sensitive) dark count rate, which increases with radiation exposure. Nonetheless, G-APDs have already demonstrated their adaptability in various detectors. They are combined with light-collecting Winston cones in successful prototype IACT cameras in MAGIC and FACT. More than 60,000 G-APD modules are currently installed in the near detector of the Tokai-to-Kamioka long-baseline neutrino experiment; G-APDs have also been successfully tested with an aerogel radiator in studies for the RICH detector for Belle II.

Vacuum-based photon detectors continue to diverge from classical PMT forms to include: new multianode types; versions with photoelectron gain achieved in the pores of a microchannel plate (MCP), affording the time resolution required for future DIRC devices; and hybrid devices that combine a pixellated silicon pin-diode sensor or APD with a photocathode and photoelectron acceleration potential of several kilovolts. LHCb’s RICH detectors represent the first implementation of such semiconductor hybrid devices in an operating experiment, while 144-channel hybrid APDs are foreseen as the baseline for Belle II. Quantum efficiencies continue to rise from typical values of 23% for bialkali (BA) photocathodes operating in the visible range to routinely produced “super” and “ultra” BAs, which approach 35% and 45% respectively. Such improvements are important for future water-based neutrino detectors, including MEMPHYS, because they allow for a bigger detector spacing and target volume.

While gaseous photon detectors – today with solid reflecting photocathodes rather than photosensitive vapours – remain the only approach to affordable large surfaces, great efforts have been made to inhibit the positive-ion feedback that limits photocathode lifetime and reduces operating speed. MPGDs based on stacked, perforated electrostatic layers in the GEM configuration have been implemented in several tracking detectors, including the hadron-blind RICH detector in PHENIX. Gaseous photon detectors for visible wavelengths so far remain elusive but many applications await, if they can be made cheaply enough.

Making the most of RICH detectors requires exceptional performance in many challenging technical areas, as highlighted in the invited talk by Clara Matteuzzi of INFN Milano. These include the purity of the solid, liquid or gas media – which the Cherenkov radiator transparency depends upon – as well as the transparency of radiator windows and reflectivity of focusing mirrors, which often operate at ultraviolet wavelengths. Groups in Novosibirsk and Japan have attained new levels of performance from aerogel radiators, in particular in terms of improved transparency and the production of tiles with customized refractive index.

In the tradition of the previous workshops, a session was also devoted to talks on pattern recognition and data analysis, where sophisticated methods and algorithms were presented. Last, in the conference summary, Blair Ratcliff of SLAC selected highlights from the many contributions at RICH2010 and revealed a picture of high “V²” (variety and vitality) in Cherenkov-light imaging.

The first half of the conference suffered appalling and uncharacteristic weather, but nevertheless the participants enjoyed a social programme that included a (rescheduled) boat visit to the Calanques of Cassis and a banquet in the barrel hall of the Domaine Bunan vineyard, near Bandol. The accompanying concert featured keyboard improvisations by Jacques Diennet of Ubris Studio, Marseille, accompanied by passing cosmic rays, made audible in the “Cosmophone”, which was introduced by its inventor, Claude Vallée of the Centre de Physique des Particules de Marseille. The concert continued with contemporary pieces played on vintage Provençial instruments by Jean-Marc Montera (lute and guitar) and the Meditrio ensemble. Fortunately, the weather smiled on the final days of the conference and on the closing ceremony, where the RICH2010 conference flag was presented to Takayuki Sumiyoshi of Tokyo Metropolitan University, in anticipation of RICH2013, which will take place in Japan.

• RICH2010 was sponsored by French and European private companies and institutions including CERN, IN2P3, Commissariat à l’énergie Atomique, Université de la Méditerranée Aix-Marseille II, Conseil General des Bouches du Rhône, Conseil General de la Région Provence – Alpes – Cote d’Azur and the town of Cassis.