Search

Processing high-definition images with single-photon sensitivity acquired above 500 frames per second (fps) will certainly find ground-breaking applications in scientific and industrial domains such as nano-photonics. However, current technologies for low light imaging suffer limitations above the standard 30 fps to keep providing both excellent spatial resolution and signal-over-noise. This paper presents the state of the art on a promising way to answer this challenge, the electron bombarded CMOS (ebCMOS) detector. A large-scale ultra fast single-photon tracker camera prototype produced with an industrial partner is described. The full characterization of the back-thinned CMOS sensor is presented and a method for Point Spread Function measurements is elaborated. Then the study of the ebCMOS performance is presented for two different multi-alkali cathodes, S20 and S25. Point Spread Function measurements carried out on an optical test bench are analysed to extract the PSF of the tube by deconvolution. The resolution of the tube is studied as a function of temperature, high voltage and incident wavelength. Results are discussed for both multi-alkali cathodes as well as a Maxwellian modelization of the radial initial energy of the photo-electrons.

International audience; This development is related to the design and the integration of a Monolithic Active Pixel Sensor (MAPS) into a photosensitive proximity focusing vacuum-based tube. This EBCMOS project is dedicated to the fluorescent and the bioluminescent high speed imaging. The results of the full characterization of the first prototype are presented. Comparative tests with different fluorescent dyes have been performed in biology laboratories. Preliminary conclusions on the ability of EBCMOS to perform fast single-molecule tracking will be given.

Scintillating crystal-based hybrid photon detectors have been demonstrated as viable single photon detectors since 1996 in the Lake Baikal neutrino telescope. Prior to this, the Philips XP2600 was developed under the DUMAND program, while more recently, developments at CERN have demonstrated the advantages of a true concentric geometry with a scintillator at the geometric centre of a spherical photocathode, giving almost 100% electrostatic collection efficiency over 3π solid angle coverage. We began the development of a new series of quasi-spherical crystal hybrid photon detectors (the Photonis XP2608 series) for use in the future KM3NeT cubic kilometer-scale deep sea neutrino telescope. The thrust of this R&D was to investigate the industrialization of the crystal hybrid photon detector to the point where it would represent a significant cost reduction per cubic kilometer of instrumented volume compared to conventional large photomultipliers, thereby allowing extremely large telescope target volumes. Such gains would arise through an all-glass envelope, ‘internal’ processing of the photocathode, and from the use of an inexpensive scintillating crystal or deposited phosphor viewed by a small photomultiplier. Details of these developments are presented.