Speaker
Description
The increased flux of coherent high-energy (> 20 keV) photons at fourth generation light sources enables many new experimental possibilities. At such energies, the absorption efficiency of Si (Z = 14) sensors is < 50 %. As the photoelectric absorption cross-section increases with the average atomic number (Z) of a sensor, materials such as GaAs:Cr (Z$_{\textrm{Av}}$ = 32) and CdZnTe (Z$_{\textrm{Av}}$ = 50) offer a solution, making it possible to exploit the advantages of hybrid pixel detectors (e.g. kHz framerates, large areas, low noise and high count-rates/large dynamic range) at higher photon energies.
However, the attenuation lengths of As and Ga K$_{\textrm{α}}$ fluorescence photons in GaAs:Cr (15.6/40.6 µm) as well as those of Te and Cd K$_{\textrm{α}}$ photons in CdZnTe (60.1/119.7 µm) are much greater than that of Si K$_{\textrm{α}}$ photons in Si itself (11.9 µm). This leads to the problem of auto-fluorescence in high-Z sensors, whereby a photon characteristic of one of the elements in the sensor deposits their energy some distance from the primary interaction site, causing a deterioration in spatial and energy resolution [1].
To develop ways of correcting for auto-fluorescence, we have measured the imaging performance (as quantified by the modulation transfer function) and spectral response of GaAs:Cr and CdZnTe sensors, supplied by DECTRIS and Redlen respectively, bonded to the JUNGFRAU 75 µm pitch charge-integrating ASIC [2]. This has been done at specific energies above and below the K-series fluorescence energies of the elements present in each sensor using monochromatic energies at the SYRMEP beamline of Elettra (Trieste, Italy).
We will present the results of these measurements and of complementary simulations, which enable elucidation of the extent to which auto-fluorescence causes a deterioration in the energy and spatial resolution of the tested devices. Furthermore, we will outline our planned work developing neural networks using these experimental and simulated datasets to identify fluorescence events in the sensor and their corresponding parent interaction. This is with the aim of reconstructing the primary event, thereby improving the spatial and energy resolution of GaAs:Cr and CdZnTe sensors.
[1] S. Chiriotti et al., JInst, 17, 2022
[2] A. Mozzanica et al., JInst, 11, 2016
K.A. Paton gratefully acknowledges funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 884104 (PSI-FELLOW-III-3i).
| Workshop topics | Sensor materials, device processing & technologies |
|---|
