Experimental validation of Davis LUT module with photonic crystals for GATE simulations

Abstract Positron Emission Tomography (PET) image quality is partially determined by PET detector performance, which can be improved by increasing the optical photon harvest from scintillator to photodetector. Due to the steep index mismatch at this interface, the optical photon critical angle limits the harvest. Photonic Crystals (PhCs) are periodic nanostructures that can break through the critical angle limit, with a size comparable to the optical photon wavelength. Previous work including both simulation and experimental studies indicated that PhCs can enhance PET detectors’ optical photons harvest. However, computational models can be inaccurate due to the necessity of using wave optics modeling, which can be challenging when modeling fabricated PhCs due to physical defects (e.g. loss of periodicity). Our group has developed a new computational method based on the previous Look-Up-Table (LUT) Davis model implemented in the Monte Carlo simulation software GATE. The novel method models fabricated PhCs, including defects: first, we used laser characterization to analyze non-ideal PhCs and then used a reverse design approach to find the actual geometry of these ‘Effective PhCs’ LUT. We built the ‘Effective PhCs’ LUT and incorporated it into GATE. The new model was tested with 3×3×20 mm 3 lutetium yttrium oxyorthosilicate (LYSO). We compared the simulated light collection results (pulse height spectra) with experimental characterization of the LYSO-PhCs detector performance and showed the new model provided better prediction of the light collection than previous models, opening a path to model fabricated PhCs with imperfect geometries in Monte Carlo simulation tools such as GATE and Geant4. Although the light collection of LYSO-PhCs detectors was less than that of the LYSO-No PhCs detectors, the ‘Effective PhCs’ light prediction can advance the scintillator-PhCs detector design and guide the fabrication to reduce PhCs defects so that we can improve PET detector’s performance.