Issue 50

N. Martini et alii, Frattura ed Integrità Strutturale, 50 (2019) 471-480; DOI: 10.3221/IGF-ESIS.50.39 476 adequate to observe a possible low-frequency sharpness drop in scintillating screens [39]. The variation in the ESF curves that can be depicted from Figs.5(a,b), can be attributed to a statistical noise component mostly prominent at the edge surface area. Fig.3d shows results for the modulation transfer function of the CMOS sensor combined with the 118.9 μ m CaWO 4 screen, under the RQA-5 (70kVp) beam quality. MTF values were found high across the examined spatial frequency range. (a) (b) (c) (d) Figure 3 : (a) ESFs of a 118.9 μm CaWO 4 /RadEyeHR CMOS combination, following the IEC 2015 method, (b) averaged and Fermi fitted ESF, (c) LSF, (d) Modulation transfer function following the IEC 2015 protocol, under the RQA-5 beam quality. Figs.(4,5) show the variation of GTF and eGTF with spatial frequency for the CaWO 4 phosphor screen measured at 70 kVp. The difference between this curve and the corresponding MTF curve is due to the influence of X-ray absorption and optical emission on GTF, which are more apparent at lower frequencies. As frequency increases, the influence of MTF on GTF is more significant than the corresponding influence of detector quantum gain (DQG: 18.17 at 70 kVp [5]) causing a further decrease in the GTF (Fig.4). Figs.5(a-d) shows indicative effective GTF results of the CaWO 4 screen with various optical detectors. Results are shown up to 5 cycles/mm, since the measured MTF incorporates the MTF of the CMOS semiconductor which however has been reported to be higher than 0.984, thus the calculation error is minimum [8,50-52]. The best optical detector-screen com- bination was obtained for a charge-coupled device having broadband anti-reflection (AR) coating (Fig.5c) with an eGTF value of 17.52 at zero spatial frequency. This value reduces GTF only by 3.54% (spectral matching factor: 0.964 [5]). The eGTF values of the CCD, is followed by the Hamamatsu MPPC silicon photomultipliers S10985 (Fig.5b) (eGTF: 17.39 at 0 cycles/mm, matching factor: 0.957), the GaAs photocathode (eGTF: 17.36 at 0 cycles/mm, matching factor: 0.955) (Fig. 5a) and the non-passivated amorphous hydrogenated silicon photodiode (a-Si:H) (Fig.5d) (eGTF: 17.39 at 0 cycles/mm, matching factor: 0.948), employed in thin film transistors in active matrix flat panel detectors. CaWO 4 also shows very good eGTF values with Sensl’s silicon PMTs, with eGTF value 14.21 at 0 cycles/mm, for the MicroFM-10035 (matching factor: 0.782), with the MicroFB-30035-SMT (eGTF: 15.25 at 0 cycles/mm, matching factor: 0.839 and with the MicroFC-30035 (eGTF: 15.94 at 0 cycles/mm, matching factor: 0.877 (Fig.5b). Furthermore, it showed very good eGTF with Hamamatsu flat panel position sensitive photomultipliers, such as the H8500C-03 (eGTF: 14.60 at 0 cycles/mm, matching factor: 0.80) (Fig.5c). It is of importance to note that eGTF showed good values when CaWO 4 is combined with complementary metal-oxide semiconductors, used in digital radiography and mammography systems, showing maximum when coupled with a hybrid blue CMOS (eGTF: 17.39 at 0 cycles/mm, matching factor: 0.854) [53].