Characterization of low-energy protons by fluorescent nuclear track detectors based on lithium fluoride films on silicon substrates

In the last years, Fluorescent Nuclear Track Detectors (FNTDs) based on the visible radiophotoluminescence (RPL) of aggregate F2 and F3+ color centers (CCs) in lithium fluoride (LiF) crystals have been demonstrated. On the other hand, optically transparent polycrystalline LiF thin films, grown by thermal evaporation on Si(100) substrates, have been successfully used for proton beam advanced diagnostics, mainly through Bragg curve permanent recording and analysis. In this paper, they were tested as FNTDs for low-energy, nearly monochromatic, collimated proton beams produced by the vertical low-energy extraction line of the TOP-IMPLART proton linear accelerator in operation at ENEA Frascati, Italy. Cleaved LiF films were irradiated with the film plane approximately parallel to the beam propagation direction and the film edge directly exposed to the incident beam. The irradiation caused the formation of CCs along the proton tracks within the film. The luminescent track images were visualized with a fluorescence microscope under blue LED excitation. At the lower energy of ∼1 MeV, it was possible to record single entire proton tracks at a fluence of approximately 108 protons/cm2. Their lengths are comparable with those expected in the LiF film. Increasing the proton energy to ∼6 MeV, the luminescent tracks were observed mainly close to the expected Bragg peak position, i.e., at the penetration depth where it would be found in Si rather than in LiF, due to multiple Coulomb scattering. At both energies, by raising the fluence by two orders of magnitude, the superposition of a very high number of tracks allowed recording the luminescent Bragg curves of the proton beams in the LiF films. They were analyzed using two different methods, considering also the type of substrate and the film characteristics, allowing to estimate the beam energy spectrum. At ∼1 MeV, the Bragg curve was best fitted using a random-optimization approach, while at ∼6 MeV it was reproduced using depth-dose curves simulated in FLUKA.