[en] In the framework of the Medipix2 Collaboration, a new photon-counting chip is being developed made of a 256x256 array of 55 μm-side square pixels. Although the chip was primarily developed for semiconductor X-ray imagers, we think that this type of device could be used in applications such as decommissioning of nuclear facilities where typical sources have γ-ray energies in the range of a few hundred keV. In order to enhance the detection efficiency in this energy range, we envisage connecting the Medipix2 chip to a CdTe or CdZnTe substrate (at least 1 mm thick). The small pixel size, the thickness of the Cd(Zn)Te substrate and the high photon energy motivate us to estimate first the spatial energy spreading following a photon interaction inside the detector. Estimations were made using the MCNP Monte Carlo package by simulating the individual energy distribution for each primary photon interaction. As an illustration of our results, simulating a 660 keV γ source, we found that there are two pixels on average, for each primary interaction, on which the deposited energy exceeds 50 keV. We have also made more accurate simulations using sub-pixels of side 11 μm, in order to evaluate the distance between the barycentre of the deposited energy and the photon impact point. As an example, with a 660 keV γ source, we found that the average of this distance reaches 67 μm when restricted to the events depositing more than 400 keV. If all events are taken into account, the mean distance is 26 μm, even though there is a small proportion of interactions where the scattered photon interacts somewhere else in the detector. Results are presented for different photon energies (60 keV, 660 keV, 1.25 MeV) and different materials (CdTe, GaAs)

[en] During the 1990's, the CEA has developed compact gamma-imaging systems, or gamma cameras, designed to localize irradiating sources quickly. Industrial products have become available, based on prototypes proven by laboratory tests and numerous on-site qualifications: experimental or power reactors, reprocessing facilities, glove boxes, fuel fabrication plants, waste storage area, etc. Gamma imaging was proven to be of significant interest in the following applications: - radiological characterisation of plants before or during decommissioning; - waste sorting; - nuclear material localisation in glove boxes; - radiation protection purposes. The paper will briefly describe the prototypes as well as the main outcomes of on-site operations. The basic performances will be reported as well, namely the detection limit and the operating range. The ability of dose-rate quantification will be also discussed. The second part will focus on new developments related to the extension of the operating range as well as an increased ease of use. The following items will be assessed: - The association of a dense shutter, which leads to a simple and efficient background reduction. - The combination with gamma spectrometry measurements, so as to identify the main radioisotopes of hot spots located by gamma imaging. For this purpose, compact and low cost CdZnTe probes fit with our need, as they provide good quality spectrum while being suitable for measurements under high dose rates. - The ability of replacing the pinhole collimator by a coded-mask aperture. Recent investigations showed that coded masks resulted in the enhancement of both of sensitivity and resolution. The basic principle and performances will be described. Thirdly, medium to long term evolutions will be discussed: one of the most significant expected breakthroughs might be the use of a pixelated CdTe detector, instead of the existing imaging detector (scintillator - MCP image intensifier - CCD). This new configuration should greatly reduce the size of gamma cameras. Developments and new results in this area will be highlighted. Another important issue deals with advanced processing of CdZnTe spectra, which could address one of the main drawback related with these detectors: the photo peaks can be strongly dissymmetrical, which can strongly affects the results. New deconvolution techniques, initially developed for actinides High-Resolution Gamma Spectrometry, might help to overcome this problem. (authors)

[en] The objective of the paper is to present characterization tools for specific metallic foams objects. As a first step, the objective is to obtain the geometric model of the foam as an input for computed fluid dynamics code. In a second stage, the water behavior in foam samples is studied in free conditions. To address these objectives, an X-ray microtomograph system based on a microfocus X-ray generator and the Medipix2 detector has been implemented. Such X-ray generator has a focal spot size in the magnitude of 1 μm that allows a geometric magnification of the sample without blurring images. Experimental constraints are details to observe in the foams in the magnitude of a few tens of micrometers or less, which implies very low contrast. For these reasons the Medipix2 detector is well suited as a very low noise detector. Two-dimensional images are acquired for 360 projections of the foam. The three-dimensional image is obtained by using an algebraic algorithm Ordered Subset Expectation Maximization (OSEM). Two spatial scales are studied in dry and wet conditions for the foams. Images acquired on a 10 mm² sample of foam with pixel size equal to 50 μm are used to make the analysis at a macroscopic scale of geometry and water kinetics. Images acquired on a 1 mm² sample of foam with pixel sizes equal to 5 μm are used for a thin characterization of the geometry and to reconstruct the water distribution along the struts of the foam

[en] Since the beginning of the 1990s, the CEA is being involved in the development of compact gamma cameras based on a pinhole collimator, a scintillator and an intensified CCD. In collaboration with the Kurchatov Institute, the CEA has recently developed a miniature coded mask, which significantly improves sensitivity and angular resolution. Laboratory tests have demonstrated excellent performance. They are summarized, accompanied with on-site test results. We also present preliminary results of Medipix2/CdTe, a semiconductor pixel detector