Characterization of Gamma Radiation Sensitivity in Cerium, Lutetium and Aluminum Doped Optical fiber with Long Period Gratings

1 Abstract—In this work, we present the first analysis of gamma radiation response in silica optical fibers with core doped with lutetium and aluminum (LuAG). By utilizing long period gratings (LPGs) inscribed within the fibers, we monitored the evolution of their properties in real time. These fibers were fabricated using modified chemical vapor deposition technique combined with nanocrystal doping, while the LPGs were inscribed using electric arc discharge method. The LPGs were subsequently exposed to gamma radiation with total dose up to 4.5 kGy at various dose rates (0.45, 0.9, and 1.8 kGy/h). A comprehensive investigation of the dose rate dependence on their response has been conducted and post-irradiation effects have also been evaluated. A remarkable shift in the LPG resonant wavelength, up to 50 nm at the highest dose, significantly surpassing the response of standard fibers, with minimal radiation-induced attenuation, has been observed. This LPG wavelength shift is directly related to gamma radiation induced refractive index changes (RIRIC) in the fiber, a phenomenon previously unexplored in this type of fiber. This This paragraph of the first footnote will contain the date on which you submitted your paper for review. This work was partially supported by the European Union - Next Generation EU under the Italian National Recovery and Resilience Plan (NRRP), Mission 4, Component 2, Investment 1.3, CUP E63C22002070006, partnership on “Telecommunications of the Future” (PE00000001 - program “RESTART”). Contract N° 23-05507S of the Czech Science Foundation and the project LasApp CZ.02.01.01/00/22_008/0004573 co-funded by the European Union and the Czech Republic. Romanian authors acknowledge the support of National Interest Infrastructure facility IOSIN-CETAL at INFLPR and national projects 119TE/2025 “LIFEFORMS” funded by the Romanian Ministry of Education and Research and FAIR-RO-004 “SPARC-RO” funded by the Institute of Atomic Physics. Research was also funded in Poland by POB Photonics of Warsaw University of Technology within the Excellence Initiative: Research University (IDUB) programme. (Corresponding authors: Andrei Stancalie, Jan Mrázek, Agostino Iadicicco) F. Esposito, A. Srivastava, S. Campopiano and A. Iadicicco are with Department of Engineering, University of Naples “Parthenope”, 80143 Naples, advancement paves the way for the understanding and exploitation of novel optical fiber materials in the field of ionizing radiation sensing and protection. Index Terms—dosimetry, fiber gratings, ionizing radiation, nanoparticle doped glass, aluminum garnets, multicompound optical fibers, optical fiber sensors. I. INTRODUCTION onizing radiation is commonly encountered in daily life, ranging from low doses emitted by medical and scientific devices to higher levels from nuclear reactors and waste facilities. Among the various types of radiation, gamma rays, characterized by very short wavelengths (0.001-0.1 nm) and photon energies above 100 keV, can penetrate most materials and are typically produced in radioactive decay, nuclear reactions, and cosmic events [1]. Advanced gamma radiation Italy (email: [email protected], anubhav.srivastava001@ studenti.uniparthenope.it, [email protected], [email protected]). A. Stancalie and R. Mihalcea are with Center for Advanced Laser Technologies (CETAL), National Institute for Laser, Plasma and Radiation Physics, Magurele RO-077125, Romania (e-mail: [email protected], [email protected]). I. Bartoň and J. Mrázek are with Institute of Photonics and Electronics, Czech Academy of Sciences, Prague, Czech Republic (e-mail: [email protected], mraze[email protected]) D. Negut is with “Horia Hulubei” National Institute for R&D in Physics and Nuclear Engineering, Magurele RO-077125, Romania (e-mail: [email protected]). M. Smietana is with Institute of Microelectronics and Optoelectronics, Warsaw University of Technology, 00-662 Warsaw, Poland, and Institute of Microelectronics and Photonics, Lukasiewicz Research Network, 02-668 Warsaw, Poland (email: [email protected]). © 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, incl. reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. This is the accepted version of the following article: Flavio Esposito et al.,"Characterization of gamma radiation sensitivity in cerium, lutetium and aluminum doped optical fiber with long period gratings", in Journal of Lightwave Technology, 2025. The final version is available at: https://doi.org/ 10.1109/JLT.2025.3622307. Characterization of gamma radiation sensitivity in cerium, lutetium and aluminum doped optical fiber with long period gratings Flavio Esposito, Andrei Stancalie, Anubhav Srivastava, Razvan Mihalcea, Ivo Bartoň, Daniel Negut, Stefania Campopiano, Mateusz Śmietana, Jan Mrázek, Agostino Iadicicco I 2 monitoring technologies aim to improve remote dose detection and minimize exposure, particularly in medical imaging and radiation therapy [2], [3]. Scintillators are central to these applications, converting ionizing radiation into UV or visible light that can be detected remotely using photodetectors or semiconductors [4], [5]. In this context, rare-earth doped materials such as yttrium- (YAG) and lutetium- (LuAG) aluminum garnets have proven effective in single-crystal and composite scintillators [6]–[9]. Optical fiber technology provides an alternative to bulky scintillators by serving as passive waveguides that transmit light from irradiated scintillators to protected detectors [10]. This setup reduces risk to electronics and allows integration in hazardous environments. Additionally, in radiation-exposed fibers, three optical phenomena have been identified: radiationinduced attenuation (RIA), refractive index change (RIRIC), and emission (RIE) [11]–[13]. These effects have enabled new fiber-based radiation sensing technologies [14]–[18]. Progress in fiber materials has broadened the field, introducing optical fibers with unconventional compositions and structures [19]–[22]. Dosimetry based on the RIA phenomenon has been investigated for various rare-earth doped fibers in [23], [24]. Moreover, scintillating fibers leveraging RIE and/or RIA offer high sensitivity and spatial resolution [25]–[27]. For instance, Ce-doped LuAG fibers have been optimized for proton therapy, with emphasis on energy, position, timing, and light collection [25]. However, challenges persist in calibration, temperature dependence, and fiber ageing [25], [26]. Moreover, new methods such as melt-in-tube [28] and nanoparticle doping [29] have produced Ce:YAG-doped fibers showing strong radioluminescence around 400 nm [30], [31], though high attenuation (>30 dB/m) limits long-range signal transmission. This paper builds on our preliminary proof-of-concept study [32] and provides a deep experimental investigation of gamma RIRIC in a silica fiber where the core is doped with lutetium and aluminum, as well as a small percentage of cerium. Fibers were fabricated using modified chemical vapor deposition (MCVD) and nanoparticle doping, achieving low optical losses. To address the hard challenges due to low power measurement linked to RIE, we focus on RIRIC using long period gratings (LPGs) inscribed via electric arc discharge (EAD). The LPGs were exposed to gamma radiation up to 4.5 kGy at dose rates of 0.45, 0.9, and 1.8 kGy/h. The resulting resonance wavelength shifts, significantly larger than those observed in Ge-doped fibers, represent the strongest RIRIC response recorded to date in such systems [16], with minimal interference from RIA in the operating range. The high sensitivity encourages further research and use as dosimeters for portable medical devices and for distributed monitoring of radiation leakage in nuclear facilities. II. MATERIALS AND METHODS A. LuAG optical fiber fabrication Cerium-doped lutetium aluminum garnet (LuAG) nanocrystals were prepared by common sol-gel synthesis followed by heat-treatment at 900 °C to induce the crystallization process [33]. The prepared powder was ground to a fine powder in a mortar. To form a stable colloidal solution, a total of 5 g of the LuAG powders was dispersed in 50 mL of ethyl alcohol (UV-grade, BC-Chemservis, Czech Republic) in ultrasonic bath for 30 min and then settled down for 3 h. The colloidal solution was decanted from the precipitate and used for nanoparticle doping. The fiber preform was made by modified chemical vapor deposition (MCVD) combined with nanocrystal-doping method, which was previously used in [29], [34] with nanocrystals of different nature. First, a porous silica soot layer was formed on the inside of the silica substrate tube. Then, the colloidal solution was soaked inside the silica soot layer and dried under a 10 sccm oxygen flow for 48 h. Finally, the doped silica soot layer was sintered at 1600 °C into a transparent glass layer and collapsed at 2100 °C into a preform. The preform was drawn into optical fiber with an outer diameter of 125 µm and coated with a UV-curable acrylate (DeSolite, 3471-3-14, UK). To prove the formation of LuAG nanocrystals after the preparation by sol-gel and heat treatment, X-ray diffraction (XRD) analysis was performed on a Bruker D8 Discover diffractometer with Bragg-Brentano geometry operating with Cu-Ka radiation (λ = 1.54056 Å, operating voltage 40 kV, current 40 mA, and integration time 200 ms). The nanocrystal morphology and fiber cross-section were observed by a scanning electron microscope (SEM) (Tescan Lyra 3, Czech Republic). The chemical composition of preforms was measured with Cameca SX100 electron microprobe (EMPA). Refractive index (RI) profiles of preform were measured with RI profiler (Photon Kinetics A2600, UK). The longitudinal homogeneity of the fiber and background losses at 850 nm were measured with an optical time-domain reflectometer (EXFO 715B, UK). B. LPG fabrication and measurements LPG is selected here for analyzing of the fiber material evolution, as its spectral features are highly sensitive to optical properties of fiber. An electric arc discharge technique has been used for their fabrication, as detailed in [35]. Following fabrication, LPGs were exposed to temperature and strain changes using setups also described in [35]. Finally, they were transformed into a reflective configuration to ensure strain-free operation [36]. During the fabrication and measurement phases, the LPG spectra were captured in wide wavelength range (1100-1650 nm) using a broadband light source (SLED Exalos) and an optical spectrum analyzer (Yokogawa AQ6370B, Japan). C. Gamma irradiation The setup for gamma irradiation was outlined comprehensively in our prior publications [36], [37]. In summary, we utilized a compact irradiator (BRIT GC-5000, India) wherein the sensors were subjected to gamma radiation emitted by 60Co rods. A cumulative dose of approximately 4.5 kGy was administered for all samples. However, different dose rates were explored by means of calibrated lead shields, i.e. 1.8 kGy/h, 0.9 kGy/h, and 0.45 kGy/h, resulting in different exposure times to reach the same total dose. The irradiations 3 were performed at room temperature. The temperature rises nonlinearly at irradiation onset, with a rapid initial increase followed by saturation, independent of dose rate, and changes at most 6 °C in the worst case (longer exposure). To account for temperature variations, the chamber is equipped with a thermocouple positioned near the sensors, and readings are then used to remove the temperature contribution from the LPG response. Following the irradiation, the recovery effects were monitored for the subsequent 15 h, a duration sufficient to detect the most significant alterations. For real-time monitoring of LPGs, an optoelectronic acquisition system consisting of an interrogator (Micron Optics sm125, USA) was used operating within the wavelength range of 1510-1590 nm, with a full spectrum resolution of 5 pm and an acquisition time set to 10 s. To identify the LPG resonance wavelength, a centroid analysis was performed after truncating the raw spectrum data above 85% of the peak value. III. RESULTS AND DISCUSSION A. Characterization of LuAG optical fiber XRD was used to prove the structural properties of prepared LuAG nanocrystals, after sol-gel synthesis and heat treatment. The diffractogram is shown in Fig. 1(a) and it reveals a set of well-pronounced peaks arranged in a pattern fairly matching a standard pattern of pure LuAG registered in International Centre for Diffraction Data database, record number 73‑1368, proving a successful synthesis. Slight regular shift of peaks maxima can be attributed to the incorporation of Ce ions inside LuAG crystal lattice. No other peaks indicating formation of parasitic secondary crystal phase were observed. The mean primary nanocrystal size calculated using in-build operating software of the diffractometer reached 46 nm. To check the nanopowder morphology, we used SEM and Fig. 1(b) shows a representative image. It can be found that the LuAG nanopowder is composed of well-developed primary nanocrystals with almost spherical shapes with size within 3050 nm, which corresponds well to the calculated mean value of 46 nm. In some areas, the primary nanocrystals were aggregated in larger grains or clusters up to 120 nm in size. Fig. 1. (a) XRD pattern of LuAG nanocrystals with denoted (hkl) indices. (b) SEM visualization of LuAG powders prepared by thermal treatment. Fig. 2. LuAG doped fiber: (a) RI profile and concentration profile of the preform; (b) SEM micrograph of the cross-section of drawn fiber. The general properties of prepared preform are shown in Fig. 2(a). The concentration of Lu3+, Al3+ and Ce3+ were measured by EMPA. The concentration of Lu3+ and Al3+ ions reached 0.377 at.% (atomic percentage) and 0.654 at.%, respectively. The concentration of Ce3+ was by two orders lower (0.011 at.%). The concentration ratios of the elements showed that the general formula of implemented nanoparticles was Ce0.08Lu2.9Al5O11.94 according to the reference Lu3Al5O12 garnet chemical composition. The shape of the RI core-cladding difference closely matched the concentration profiles of the dopants reaching the maximum of Δn = 0.0038 RIU (RI unit). Next, the preform was drawn in an optical fiber with a core diameter of 10 ± 0.2 µm and a cladding diameter of 125 ± 1.5 µm, as reported in the SEM image of cross-section from Fig. 2(b). The fiber is single mode at wavelength of 1550 nm and can be seamlessly spliced to standard silica fibers using the automatic routines of fusion splicer. The optical fiber exhibited optical losses slightly below 1.3 dB/m at wavelengths of 850 nm and 1550 nm. Anyway, for the purpose of the work the fiber length per single LPG does not exceed 10 cm, what keeps the insertion loss of the device low. It should be noted that, after the fiber drawing stage, we could not verify the persistence of nanocrystals within the fiber core. This limitation is a well-recognized issue in silica fibers produced via the MCVD process and doped with oxide nanocrystals through the solution doping technique [38]. B. Properties of LPGs in LuAG fiber The practical implementation of LPG demands proper selection of the fabrication parameters for the specific fiber. Specifically, the following parameters were identified: arc power of 20 step (proprietary unit of Sumitomo Type-39 fusion splicer), arc time of 650 ms, electrode gap set to 0.8 mm, weight of 12.0 g, and length of the grating near 30 mm. The resulting modulation strength is slightly weaker if compared to that necessary to induce grating in SMF28 fiber [35], [39]. (a) (b) (211) (220) (321) (400) (420) (422) (431) (521) (440) (532) (631) (444) (640) (721) (642) (732) (800) (a) (b) 4 Fig. 3. (a) Spectra of LPGs written in LuAG fiber for different Λ shown with the offset. (b) Dispersion curves of λi vs Λ. Red area in (b) indicates the range of feasible Λ employed for gamma radiation experiments. In Fig. 3(a) are shown spectra of different LPGs inscribed in LuAG fiber with period Λ ranging from 445 µm to 520 µm. A narrow profile and maximum depth reaching approximately 30 dB of the attenuation bands can be observed. The labels λA, λB, and λC are used to identify the resonances associated to different cladding mode orders. With the aid of a numerical tool [40], it is expected that λA likely corresponds to a cladding mode of approximately sixth order. Additionally, in Fig. 3(b), corresponding phase-matching curves are depicted. It is evident that the result follows theoretical expectations [35], [41] and λ increase with Λ. Furthermore, the slope ∂λi/∂Λ changes from 1.3 nm/µm to 5.1 nm/µm considering λC (green line) and λA (light blue line), respectively. For the purpose of this work, we focused on LPGs with Λ = 495-500 μm (red area in Fig. 3(b)) as they exhibit an attenuation band related to a high order mode in the wavelength range 1510-1590 nm of interrogator used during gamma irradiation. An exemplary LPG with such properties is reported in blue in Fig. 3(a) having a Λ = 500 μm and λA = 1545 nm, which will be considered for further investigations. Moreover, regarding the physical mechanisms implicated in the applied grating formation, identifying the predominant mechanism remains a subject of ongoing investigation, heavily contingent upon the writing conditions and fiber type [41]. It is hypothesized that the primary effect of grating formation in this fiber arises from the RI alteration due to stress relaxation, as also suggested for standard and aluminosilicate fibers [42]. Next, the response of LPGs to temperature and strain was measured and compared to those of LPGs written in standard Ge-doped fiber (SMF28) using the same technique. We focused attention on λA at 1545 nm and on a band corresponding to a similar sixth order cladding mode for an LPG in SMF28. The investigation explored response to temperature within its range 20-60 °C, while the fiber was kept under constant tensile load using a 12 g weight applied through a pulley system to avoid undesired bending. Fig. 4. Sensing properties for LuAG LPG with period of 500 µm and λA at 1545 nm and comparison with a reference in SMF28, to (a) temperature and (b) force. As depicted in Fig. 4(a), the λA experiences a linear shift within the considered range reaching a sensitivity of ST = ∂λA/∂T = ‒4 pm/°C. The observed negative shift suggests that the presence of LuAG reduces the thermo-optic coefficient of the core relative to that of the silica cladding [43], [44]. The ST is significantly lower than the one obtained for SMF28, i.e. 98.8 pm/°C. Overall, it is at least 15-20 times lower than typically obtained for other fibers [35], [39], thus offering low crosssensitivity to temperature. Regarding the measurement of strain-induced effects, the Fig. 4(b) comparatively shows the wavelength shift of λA and corresponding SMF28 LPG versus applied longitudinal force up to 800 mN (at stable temperature). The shift can be considered as linear within the range investigated and reaches SF = ∂λA/∂F = 5.7 pm/mN. The λA shift results from both material properties (i.e., strain impact on effective RIs) and a waveguide effect (variation of grating period with strain), where the response tends to be negative for lower order modes and positive for higher order ones [45]. The value found in this case is higher than 1.7 pm/mN of SMF28based LPG [39]. In order to minimize cross-sensitivity to strain, the LPG was modified to reach a reflective configuration. C. Gamma irradiation effects For the gamma irradiation experiments, three similar LPGs written in the LuAG fiber, were investigated and exposed to different dose rates until reaching the same total accumulated dose of 4.5 kGy. They are named as LPG-I (irradiated at 1.8 kGy/h dose rate), LPG-II (at 0.9 kGy/h), and LPG-III (at 0.45 kGy/h). Moreover, the results have been compared to those of LPGs in SMF28 irradiated at 1.8 kGy/h dose rate. First, the attention is focused on the spectral response of LPG-I irradiated at the highest dose for 2.5 h, which well (a) (b) λA λB λC λA λA λA λB λB λB λC λC (a) (b) LuAG ST=‒4.0 pm/ C SMF ST=98.8 pm/ C LuAG SF=5.7 pm/mN SMF SF=1.7 pm/mN 5 illustrates the overall response induced by radiation. The results are depicted in Fig. 5(a), where the spectra highlighted in violet, light blue, and yellow represent the readouts taken before irradiation, during irradiation at the maximum dose, and after recovery post-irradiation, respectively. The primary variation observed is in the resonance wavelength, whereas no significant RIA could be observed even after such doses. Specifically, a red shift of about 50 nm was observed at the maximum irradiation dose of 4.5 kGy, with some recovery effects observed during 15 h following the exposure turn off. Finally, no grating erasure effects were observed even at the highest dose making EAD technology for grating fabrication one of the most suitable for these applications [16]. Regarding the physical mechanisms involved, the observed wavelength shift can be mainly associated with RIRIC occurring in the optical fiber core due to the presence of dopants [12], [16], [36]. To visualize the response during gamma irradiation, the realtime tracking of the wavelength shift is gathered in Fig. 5(b) with solid blue line for the LPG-I. The resonance wavelength experiences a red shift during the irradiation without reaching saturation within the investigated dose range. During the recovery phase, a slight blue shift in resonance wavelengths was observed, indicating that radiation effects persist after exposure. This behavior is due to radiation interacting with the fiber glass and dopants, forming defects and breaking chemical bonds. After irradiation stops, recovery begins through thermal annealing, dopant ion migration, and natural recombination of charge carriers [13]. A maximum wavelength shift Δλ = 49.8 nm was obtained when the maximum dose was attained as reported in Fig. 5(b) with blue line. Concerning the recovery stage, the total accumulated shift for LuAG LPG-I remained equal to 41.7 nm within 15 h after the exposure, i.e. the wavelength shift recovery is about 16%. When LPG in SMF28 fiber is irradiated at the same dose rate the shift is an order of magnitude lower. It is worth noting that for the curve related to SMF28 in Fig. 5(b), the irradiation duration is 20 h reaching a total dose of 36 kGy and further monitoring demonstrated a 23% recovery [36]. Fig. 5. (a) Spectra of LPG-I irradiated at dose rate of 1.8 kGy/h: before irradiation, at highest dose, and after recovery; (b) Resonance wavelength vs time during irradiation and recovery of LuAG LPGs compared with SMF28. The attention is subsequently focused on LPG-II and LPGIII for which the dose rate was decreased vs LPG-I by 50% and 75%, respectively. Despite different dose rates, the overall trends are similar in Fig. 5(b). The maximum wavelength shift for LPG-II reached 31.9 nm and after recovery it remained around 25 nm (green line), i.e. a recovery of 22%; whereas for LPG-III the shift was only up to 16.4 nm with a permanent effect of about 12 nm after recovery (red line), i.e., 27%. Therefore, the influence of the dose rate on response and recovery is evident, and although this could have been anticipated, it has never been experimentally documented in literature until now for any type of optical fiber. Fig. 6(a) illustrates the shift for the three samples vs total absorbed dose (hence the LPG responses correspond to “isodose rate” curves), where the maximum shift is related to the dose, but strongly dependent on the dose rate as well. Although no saturation is achieved, it is evident that overall behavior deviates from linearity when the dose is increased. Furthermore, the same figure demonstrates that the LuAG LPGs exhibit higher sensitivity than SMF28 even at the lowest dose rates studied. Specifically, a sensitivity of 4.4, 4.8, and 21.6 nm/kGy was obtained at lower doses (below 0.5 kGy) when the dose rate is 0.45, 0.9, and 1.8 kGy/h, respectively. Such values correspond to theoretical resolutions of 0.2, 0.2, and 0.05 Gy, respectively, for the three previously mentioned cases, assuming a spectral resolution of 1 pm and the sensitivity values obtained. It should be noted that these resolution values are purely theoretical, as they have not been experimentally demonstrated under actual measurement conditions. The same results can be plotted versus the dose rate for different total doses, creating “iso-dose” curves. Fig. 6(b) presents the iso-dose curves for various dose values (0.5, 1.5, (a) (b) LPG-I @ 1.8 kGy/h LPG-II @ 0.9 kGy/h LPG-III @ 0.45 kGy/h SMF @ 1.8 kGy/h 6 2.5, 3.5, and 4.5 kGy). The response depends on both dose rate and total dose in a non-linear manner. Notably, at lower doses (e.g., the 0.5 kGy iso-dose curve), the dependence on dose-rate is minimal for dose rates below 0.9 kGy/h, but increases with higher dose rates. At a dose of 2.5 kGy, the iso-dose curve (green line) appears nearly linear. Conversely, at the highest dose (4.5 kGy iso-dose curve), the dependence on dose rate decreases as the dose rate increases. Fig. 6. Resonance wavelength shift of LuAG LPGs and comparison SMF28: (a) iso-dose rate curves versus total dose; (b) iso-dose curves versus dose rate. It is important to note that despite literature [16], [46], [47], this is the first comprehensive experimental investigation of gamma radiation effect versus dose and dose rate in LPGs. The only previous study on dose rate dependence was in [48], where the authors found a change in measured wavelength shift of chiral LPGs in different fibers of about 10-20% when the dose rate was changed from 0.1 Gy/s to 0.9 Gy/s (for a 20 kGy total dose). As far as sensitivity is concerned, the response of LuAG LPGs is approximately one order of magnitude higher than any previously reported value, as summarized in Table I. To provide a broader comparison with literature, it is worth mentioning the performance of other, more mature fiber technologies. For example, P-doped fibers interrogated using Optical TimeDomain Reflectometry based on the RIA effect, as in [51], demonstrated an accuracy of ±6% under gamma rays, X-rays, and protons for space applications, with maximum doses up to 344 Gy. The RIE phenomenon is also extensively exploited for dosimetry, as in [52], where a Ce-doped silica fiber exhibited a maximum deviation of 4% when measuring a 14 MeV neutron flux in nuclear fusion-related facilities. Finally, for low-dose medical applications, [53] reports the use of a plastic fiber with scintillating material at the tip for real-time in-vivo dosimetry during brachytherapy at doses below 200 Gy. In this context, the high sensitivity achieved with LuAG LPGs, combined with further comprehensive modeling of dose and dose-rate dependence, encourages future research aimed at exploiting this configuration to develop innovative dosimeters and to advance their maturity level. Table I. Summary about radiation response of LPGs from literature. Fiber LPG mode order Radiation type Total dose Dose rate [kGy/h] Max. Δλ [nm] Max. sensitivity [nm/kGy] (range) Ref. B/Ge doped 3rd Gamma 52 kGy 2.6 9.8 1.1 (0-0.5 kGy) [36] B/Ge doped 10th Gamma 1.5 MGy 1.3 37.4 0.6 (0-6.5 kGy) [49] Ybdoped YAG 1st X-ray 1.6 MGy N/A 10.9 0.2 (0-40 kGy) [50] B/Ge doped 7th Proton 1.2 MGy 8.5 44 2.4 (0-0.5 kGy) [46] Cedoped LuAG 6th Gamma 4.5 kGy 0.45, 0.9, 1.8 49.8 21.6 (0-0.5 kGy) This work IV. CONCLUSIONS In this work, we have reported for the first time the suitability of LPGs in silica optical fibers doped with cerium, lutetium and aluminum to ionizing radiation, in particular to gamma, based on radiation-induced refractive index changes mechanism. The optical fiber was fabricated using the MCVD method, incorporating LuAG powders through nanocrystal doping. This process yielded a single mode optical fiber with low transmission losses, which can be easily spliced with standard fibers (e.g., SMF28). LPGs were then inscribed in the LuAG fiber using the electric arc method with high tunability, producing gratings comparable to those in standard fibers. To assess their performance, a broad experimental evaluation was performed where the LPGs were subjected to gamma radiation doses up to 4.5 kGy at different dose rates of 0.45, 0.9, and 1.8 kGy/h. Real-time monitoring revealed a significant wavelength shift i.e. an order of magnitude higher than for reference devices in standard Ge-doped fibers, with the shift for a given absorbed dose increasing with the dose rate, and insignificant radiation-induced attenuation. The key innovation deals with high sensitivity to gamma radiation never achieved in optical fiber LPG. Such a high sensitivity encourages further research in using as dosimeters. It is desirable that, with further tailoring of the fiber features, the sensitivity can be improved to meet the requirements of novel scenarios such as medical application bringing the advantages of optical fiber sensing technology to dosimeter applications. Moreover, the results may be of interest for a broader community including the ones related to lasers and nonlinear optics where nanocrystal doped fibers are promising alternatives [28]. REFERENCES [1] C. Zeng et al., “Development of Polymer Composites in Radiation Shielding Applications: A Review,” J. Inorg. Organomet. Polym. Mater., vol. 33, no. 8, pp. 2191–2239, Aug. 2023, doi: 10.1007/s10904-023-02725-6. [2] K. A. Pradeep Kumar, G. A. Shanmugha Sundaram, B. K. Sharma, (a) (b) LPG-I @ 1.8 kGy/h LPG-II @ 0.9 kGy/h LPG-III @ 0.45 kGy/h SMF @ 1.8 kGy/h 7 S. Venkatesh, and R. Thiruvengadathan, “Advances in gamma radiation detection systems for emergency radiation monitoring,” Nucl. Eng. Technol., vol. 52, no. 10, pp. 2151–2161, Oct. 2020, doi: 10.1016/j.net.2020.03.014. [3] A. M. Zeidell et al., “Organic Field‐Effect Transistors as Flexible, Tissue‐Equivalent Radiation Dosimeters in Medical Applications,” Adv. Sci., vol. 7, no. 18, Sep. 2020, doi: 10.1002/advs.202001522. [4] M. Koshimizu, “Recent progress of organic scintillators,” Jpn. J. Appl. Phys., vol. 62, no. 1, p. 010503, Jan. 2023, doi: 10.35848/1347-4065/ac94fe. [5] Z. Lin, S. Lv, Z. Yang, J. Qiu, and S. Zhou, “Structured Scintillators for Efficient Radiation Detection,” Adv. Sci., vol. 9, no. 2, Jan. 2022, doi: 10.1002/advs.202102439. [6] M. Nikl, “Scintillation detectors for x-rays,” Meas. Sci. Technol., vol. 17, no. 4, pp. R37–R54, Apr. 2006, doi: 10.1088/09570233/17/4/R01. [7] W. Chewpraditkul et al., “Scintillation Properties of LuAG:Ce, YAG:Ce and LYSO:Ce Crystals for Gamma-Ray Detection,” IEEE Trans. Nucl. Sci., vol. 56, no. 6, pp. 3800–3805, Dec. 2009, doi: 10.1109/TNS.2009.2033994. [8] Y. Isokawa, D. Nakauchi, G. Okada, N. Kawaguchi, and T. Yanagida, “Radiation induced luminescence properties of Ce-doped Y2O3-Al2O3-SiO2 glass prepared using floating zone furnace,” J. Alloys Compd., vol. 782, pp. 859–864, Apr. 2019, doi: 10.1016/j.jallcom.2018.12.245. [9] M. Koshimizu, “Composite scintillators based on polymers and inorganic nanoparticles,” Funct. Mater. Lett., vol. 13, no. 06, p. 2030003, Aug. 2020, doi: 10.1142/S1793604720300030. [10] M. Jelinek, O. Cip, J. Lazar, and B. Mikel, “Design and Characterisation of an Optical Fibre Dosimeter Based on Silica Optical Fibre and Scintillation Crystal,” Sensors, vol. 22, no. 19, p. 7312, Sep. 2022, doi: 10.3390/s22197312. [11] S. Jetschke, S. Unger, A. Schwuchow, M. Leich, and M. Jäger, “Role of Ce in Yb/Al laser fibers: prevention of photodarkening and thermal effects,” Opt. Express, vol. 24, no. 12, p. 13009, Jun. 2016, doi: 10.1364/OE.24.013009. [12] S. Girard et al., “Overview of radiation induced point defects in silica-based optical fibers,” Rev. Phys., vol. 4, p. 100032, Nov. 2019, doi: 10.1016/j.revip.2019.100032. [13] S. Girard et al., “Overview of Radiation Effects on Silica-Based Optical Fibers and Fiber Sensors,” IEEE Trans. Nucl. Sci., vol. 72, no. 4, pp. 982–1020, Apr. 2025, doi: 10.1109/TNS.2024.3511455. [14] S. Girard et al., “Recent advances in radiation-hardened fiber-based technologies for space applications,” J. Opt., vol. 20, no. 9, p. 093001, Sep. 2018, doi: 10.1088/2040-8986/aad271. [15] A. Morana et al., “Radiation Effects on Fiber Bragg Gratings: Vulnerability and Hardening Studies,” Sensors, vol. 22, no. 21, p. 8175, Oct. 2022, doi: 10.3390/s22218175. [16] F. Esposito, A. Srivastava, S. Campopiano, and A. Iadicicco, “Radiation Effects on Long Period Fiber Gratings: A Review,” Sensors, vol. 20, no. 9, p. 2729, May 2020, doi: 10.3390/s20092729. [17] A. Stancalie et al., “Electron Radiation Impact on Long Period Gratings in Different Optical Fibers,” IEEE Sens. J., vol. 25, no. 2, pp. 2806–2813, Jan. 2025, doi: 10.1109/JSEN.2024.3508885. [18] G. Breglio et al., “Innovative Photonic Sensors for Safety and Security, Part III: Environment, Agriculture and Soil Monitoring,” Sensors, vol. 23, no. 6, p. 3187, Mar. 2023, doi: 10.3390/s23063187. [19] H. T. Zubair et al., “Real-time radiation dosimetry using P-doped silica optical fiber,” Measurement, vol. 146, pp. 119–124, Nov. 2019, doi: 10.1016/j.measurement.2019.06.010. [20] V. Prajzler, K. Masopoustova, and Z. Sarsounova, “Gamma radiation effects on plastic optical fibers,” Opt. Fiber Technol., vol. 72, p. 102995, Sep. 2022, doi: 10.1016/j.yofte.2022.102995. [21] S. Girard et al., “Radiation Effects on Ytterbiumand Ytterbium/Erbium-Doped Double-Clad Optical Fibers,” IEEE Trans. Nucl. Sci., vol. 56, no. 6, pp. 3293–3299, Dec. 2009, doi: 10.1109/TNS.2009.2033999. [22] I. Veronese et al., “Real-time dosimetry with Yb-doped silica optical fibres,” Phys. Med. Biol., vol. 62, no. 10, pp. 4218–4236, May 2017, doi: 10.1088/1361-6560/aa642f. [23] M. Lezius et al., “Radiation Induced Absorption in Rare Earth Doped Optical Fibers,” IEEE Trans. Nucl. Sci., vol. 59, no. 2, pp. 425–433, Apr. 2012, doi: 10.1109/TNS.2011.2178862. [24] A. Gallet et al., “Radiation-induced attenuation in aluminumand aluminum-thulium-doped optical fibers,” Appl. Opt., vol. 62, no. 16, p. E24, Jun. 2023, doi: 10.1364/AO.480576. [25] K. Rusiecka et al., “A systematic study of LYSO:Ce, LuAG:Ce and GAGG:Ce scintillating fibers properties,” J. Instrum., vol. 16, no. 11, p. P11006, Nov. 2021, doi: 10.1088/1748-0221/16/11/P11006. [26] X. Wang et al., “Light response uniformity of LuAG:Ce scintillating single crystal fiber grown by laser heated pedestal growth method,” Radiat. Meas., vol. 169, p. 107017, Dec. 2023, doi: 10.1016/j.radmeas.2023.107017. [27] A. Gierej et al., “Mass-manufacturable scintillation-based optical fiber dosimeters for brachytherapy,” Biosens. Bioelectron., vol. 255, p. 116237, Jul. 2024, doi: 10.1016/j.bios.2024.116237. [28] Y. Pang et al., “Recent Advances in Fabrication and Applications of Yttrium Aluminum Garnet-Based Optical Fiber: A Review,” Materials (Basel)., vol. 17, no. 14, p. 3426, Jul. 2024, doi: 10.3390/ma17143426. [29] I. Kasik et al., “Erbium and Al2O3 nanocrystals-doped silica optical fibers,” Bull. Polish Acad. Sci. Tech. Sci., vol. 62, no. 4, pp. 641– 646, Dec. 2014, doi: 10.2478/bpasts-2014-0070. [30] M. Jia et al., “Improved scintillating properties in Ce:YAG derived silica fiber with the reduction from Ce4+ to Ce3+ ions,” J. Lumin., vol. 221, p. 117063, May 2020, doi: 10.1016/j.jlumin.2020.117063. [31] R. E. Shaw et al., “Impact of rare earth doping on the luminescence of lanthanum aluminum silicate glasses for radiation sensing,” Opt. Mater. Express, vol. 12, no. 9, p. 3621, Sep. 2022, doi: 10.1364/OME.463119. [32] F. Esposito et al., “Outstanding sensitivity to ionizing radiation of custom optical fibers evaluated in real time by long period gratings,” in Proc. SPIE 13639, 29th International Conference on Optical Fiber Sensors, May 2025, vol. 13639, p. 1363996, doi: 10.1117/12.3061564. [33] A. Potdevin, G. Chadeyron, D. Boyer, and R. Mahiou, “Sol‐gel based YAG:Ce 3+ powders for applications in LED devices,” Phys. status solidi c, vol. 4, no. 1, pp. 65–69, Jan. 2007, doi: 10.1002/pssc.200673550. [34] I. Kasik, P. Peterka, J. Mrazek, and P. Honzatko, “Silica Optical Fibers Doped with Nanoparticles for Fiber Lasers and Broadband Sources,” Curr. Nanosci., vol. 12, no. 3, pp. 277–290, Apr. 2016, doi: 10.2174/1573413711666150624170638. [35] F. Esposito, R. Ranjan, S. Campopiano, and A. Iadicicco, “ArcInduced Long Period Gratings from Standard to PolarizationMaintaining and Photonic Crystal Fibers,” Sensors, vol. 18, no. 3, p. 918, 2018, doi: 10.3390/s18030918. [36] F. Esposito et al., “The Impact of Gamma Irradiation on Optical Fibers Identified Using Long Period Gratings,” J. Light. Technol., vol. 41, no. 13, pp. 4389–4396, Jul. 2023, doi: 10.1109/JLT.2022.3191163. [37] F. Esposito et al., “Optical properties of thin films monitored in real-time at high gamma radiation doses using long period fiber gratings,” Opt. Laser Technol., vol. 176, p. 111019, Sep. 2024, doi: 10.1016/j.optlastec.2024.111019. [38] S. Yoo et al., “Ytterbium-doped Y2O3 nanoparticle silica optical fibers for high power fiber lasers with suppressed photodarkening,” Opt. Commun., vol. 283, no. 18, pp. 3423–3427, Sep. 2010, doi: 10.1016/j.optcom.2010.04.093. [39] F. Esposito, A. Srivastava, S. Campopiano, and A. Iadicicco, “Sensing Features of Arc-induced Long Period Gratings,” Proceedings, vol. 15, no. 1, 2019, doi: 10.3390/proceedings2019015029. [40] F. Esposito, A. Stancalie, C.-D. Negut, S. Campopiano, D. Sporea, and A. Iadicicco, “Comparative Investigation of Gamma Radiation Effects on Long Period Gratings and Optical Power in Different Optical Fibers,” J. Light. Technol., vol. 37, no. 18, pp. 4560–4566, Sep. 2019, doi: 10.1109/JLT.2019.2910639. [41] G. Rego, P. Caldas, and O. V Ivanov, “Arc-Induced Long-Period Fiber Gratings at INESC TEC. Part I: Fabrication, Characterization and Mechanisms of Formation,” Sensors, vol. 21, no. 14, p. 4914, Jul. 2021, doi: 10.3390/s21144914. [42] G. Rego et al., “Arc-induced long-period gratings in aluminosilicate glass fibers,” Opt. Lett., vol. 30, no. 16, pp. 2065–2067, Aug. 2005, doi: 10.1364/OL.30.002065. [43] R. L. Aggarwal, D. J. Ripin, J. R. Ochoa, and T. Y. Fan, “Measurement of thermo-optic properties of Y3Al5O12, Lu3Al5O12, YAIO3, LiYF4, LiLuF4, BaY2F8, KGd(WO4)2, and KY(WO4)2 laser crystals in the 80–300K temperature range,” J. 8 Appl. Phys., vol. 98, no. 10, Nov. 2005, doi: 10.1063/1.2128696. [44] G. Rego, “Temperature Dependence of the Thermo-Optic Coefficient of SiO2 Glass,” Sensors, vol. 23, no. 13, p. 6023, Jun. 2023, doi: 10.3390/s23136023. [45] X. Shu, L. Zhang, and I. Bennion, “Sensitivity characteristics of long-period fiber gratings,” J. Light. Technol., vol. 20, no. 2, pp. 255–266, 2002, doi: 10.1109/50.983240. [46] G. M. Berruti et al., “Analysis of uncoated LPGs written in B-Ge doped fiber under proton irradiation for sensing applications at CERN,” Sci. Rep., vol. 10, no. 1, p. 1344, Dec. 2020, doi: 10.1038/s41598-020-58049-9. [47] P. Mesonero-Santos et al., “Effect of Low-Doses of Gamma Radiation on Electric Arc-Induced Long Period Fiber Gratings,” Sensors, vol. 21, no. 7, p. 2318, Mar. 2021, doi: 10.3390/s21072318. [48] H. Henschel, S. K. Hoeffgen, J. Kuhnhenn, and U. Weinand, “High Radiation Sensitivity of Chiral Long Period Gratings,” IEEE Trans. Nucl. Sci., vol. 57, no. 5, pp. 2915–2922, Oct. 2010, doi: 10.1109/TNS.2010.2059043. [49] S. Kher, S. Chaubey, S. M. Oak, and A. Gusarov, “Measurement of γ -Radiation Induced Refractive Index Changes in B/Ge Doped Fiber Using LPGs,” IEEE Photonics Technol. Lett., vol. 25, no. 21, pp. 2070–2073, 2013, doi: 10.1109/LPT.2013.2281200. [50] Q. Guo et al., “Effect of X-ray Radiation and High Temperature on Long Period Fiber Gratings in Crystal-Derived Silica Fiber,” J. Light. Technol., vol. 42, no. 23, pp. 8421–8427, Dec. 2024, doi: 10.1109/JLT.2024.3435838. [51] A. Meyer et al., “Characterization and Radiation Testing of an Embedded, Distributed Fiber Dosimetry System for Space Applications,” J. Light. Technol., vol. in press, 2025, doi: 10.1109/JLT.2025.3587288. [52] L. Weninger et al., “Potential of radioluminescent silica-based optical fibers for 14 MeV neutron beam monitoring,” Results Opt., vol. 19, p. 100807, May 2025, doi: 10.1016/j.rio.2025.100807. [53] P. Woulfe, F. J. Sullivan, L. Byrne, and S. O’Keeffe, “First Clinical Evaluation of Optical Fiber Sensor for Real-Time In-Vivo Dosimetry (RTIVD) During Seed Implantation for Prostate Low Dose Rate Brachytherapy,” IEEE Sensors Lett., vol. 9, no. 8, pp. 1– 4, Aug. 2025, doi: 10.1109/LSENS.2025.3583237.