Radiation effects on pure silica and Ge-doped silica core optical fibers and fiber Bragg grating sensors

Since the Fukushima nuclear accident in 2011, nuclear facility safety and environmental impacts have received global attention. Distributed optical fiber sensing has become an essential tool for monitoring and safety assessment owing to its high sensitivity, immunity to electromagnetic interference, and strong multiplexing capabilities.^1–3 Temperature sensing is particularly valuable for monitoring containment temperature changes and detecting radioactive material release.

In recent years, the performance of fiber Bragg grating (FBG) temperature sensors in irradiation environments has been extensively studied, with significant differences in the radiation resistance of various types of optical fibers revealed. Environmental changes cause shifts in the Bragg center wavelength (λ_B).⁴ Irradiation alters the peak grating wavelength, spectral width, and amplitude, affecting its performance.^5,6 These changes result in increased transmission loss, signal attenuation, and temperature sensing errors.^7–9 Radiation-induced Bragg wavelength shift (RI-BWS) directly affects the measurement accuracy of FBGs, limiting their application in high-radiation environments.^10–12 Since the 1970s, when Hill et al.¹³ first discovered the photosensitivity of optical fibers and successfully fabricated the first Fiber Bragg Grating (FBG) using a standing wave method, FBGs have been widely used as an ideal monitoring tool across various fields, including energy, nuclear power, space, communications, defense, and radiotherapy. Based on silica-based optical fibers and nanomaterials, WaveFlex fiber biosensors exhibit high flexibility and multifunctionality, demonstrating great potential in key areas such as health monitoring and food safety, while also paving the way for future cross-disciplinary applications.^14–16 In 2003, Mihailov et al.¹⁷ employed femtosecond lasers combined with phase mask technology to inscribe FBGs in Ge-doped fibers, demonstrating excellent high-temperature stability. In 2020, Viveiros et al.¹⁸ used femtosecond laser direct writing to inscribe FBGs in single-mode fibers, achieving a temperature sensitivity of 12 pm/°C within the range of 23–300 °C. Additionally, Adriana Morana et al.⁴ found that radiation can induce defects in optical fibers through ionization or displacement damage, leading to a decline in optical transmission performance and a shift in Bragg wavelengths (RI-BWS). Using radiation-resistant fibers, such as pure silica core or fluorine-doped fibers, can significantly improve FBG stability in high-radiation environments. However, while hydrogen loading can enhance laser sensitivity in fibers, it may also increase their sensitivity to radiation. The underlying mechanisms of the post-irradiation Bragg wavelength drift and the microscopic processes at the atomic scale remain unclear.

Current research primarily focuses on experimentally evaluating the macroscopic performance of various optical fibers under irradiation, with limited exploration of the underlying microscopic mechanisms of irradiation effects. This study aims to investigate the microscopic mechanisms behind grating performance changes through systematic experiments and theoretical analysis. The radiation environments in space and nuclear power plants are highly complex, with radiation doses varying significantly depending on location. Space radiation doses range from 5 to 200 krad, while doses inside nuclear power plant containment structures range from 10³ to 10⁵ rad, and doses near the reactor core can reach 10⁵–10⁸ rad. Specifically, this study will develop optical fibers with excellent radiation resistance that can operate in these high-radiation environments while meeting the temperature requirements of nuclear power plants and spacecraft interiors (≤85 °C). Additionally, this research will focus on radiation-induced structural changes in materials, the formation and evolution of point defects, and their impact on the refractive index. The goal is to uncover the intrinsic reasons for the varying radiation tolerance of different optical fibers, providing a theoretical foundation and technical support for the future development of higher-performance radiation-resistant optical fibers.

A pure silica core fiber preform was fabricated using plasma chemical vapor deposition (PCVD) and drawn into fibers using a high-temperature drawing tower. The core material was pure silica, with F-doped silica cladding, and an acrylic coating. The cladding contains approximately 3 wt. % of fluorine and the numerical aperture (NA) was 0.126, and the core and cladding dimensions were 8 and 125 μm, respectively. A Ge-doped silica fiber preform was also fabricated, using modified chemical vapor deposition (MCVD) with solution doping, and drawn into fibers using a high-temperature drawing tower. The core was Ge-doped silica, with pure silica cladding, and a polyimide coating. The core is doped with approximately 3.4 wt. % germanium and the NA was 0.153, and the core and cladding dimensions were 8 and 125 μm, respectively.

Femtosecond laser pulses were used to induce localized nonlinear effects in the fibers, forming permanent refractive index changes and periodic grating structures.^19–23 The λ_B of the FBGs ranged from 1550 to 1560 nm. Herein, FBGs made from pure silica fibers are referred to as FBG-Si, whereas those made from Ge-doped fibers are referred to as FBG-Ge.

Pure and Ge-doped silica glasses were prepared using the sol–gel method and high-temperature sintering, which offers advantages over PCVD and MCVD, such as larger sizes, higher doping uniformity, and easier doping concentration control.²⁴ 

The irradiation experiments in this study were conducted at the Shanghai Institute of Applied Physics, Chinese Academy of Sciences, using ⁶⁰Co γ-ray as the radiation source. The radiation dose rate was 5 kGy/h and the total radiation doses were 100, 200, and 300 kGy. Three types of experimental samples were used: fiber preform optical fibers and FBG.

Figure 1 shows a schematic for the FBG temperature sensing setup. Light from a 1550 nm amplified spontaneous emission broadband source passes through a circulator into the FBG sensor, and the reflected light is detected using an optical spectrum analyzer (OSA). The FBG is heated and irradiated using a heating plate, while the OSA monitors changes in the reflected spectrum; λ_B shifts with temperature under various irradiation conditions were recorded. In this experiment, the heating plate controlled the temperature, heating the FBG from 25 to 80 °C, and the corresponding λ_B shifts were recorded to establish the relationship between temperature and λ_B.

Fiber composition test: An electron probe microanalyzer (EPMA, JXA-8230 JEOL, Japan) was used to perform a line scan analysis of Ge and F on the fiber core and cladding cross section. The uniformity and content of Ge and F were assessed based on the signal intensity fluctuations.

Fiber refractive index test: The refractive index distribution of the fiber end-face was measured using an IFA-100 analyzer (Interfiber Analysis, USA) with a spatial resolution of 0.5 μm and an accuracy of ±0.0001. Tests were conducted at a wavelength of 630 nm at room temperature.

Fiber loss test: Fiber loss represents the ability of a fiber to attenuate light energy at a specific wavelength or range of wavelengths. The cutback method was used for testing. The loss spectrum was measured over the wavelength range of 600–1750 nm.

Electron paramagnetic resonance (EPR) test: A Bruker E-580 EPR instrument was used to qualitatively and quantitatively detect unpaired electrons and analyze their local structural characteristics. The testing temperature was at room temperature.

Absorption spectrum test: The absorption spectrum of the fiber preform was measured using a Perkin-Elmer 950 UV/VIS/NIR spectrophotometer. The testing temperature was at room temperature.

Figure 2 shows the core composition and refractive index distribution of the pure and Ge-doped silica core fibers. Figures 2(a) and 2(b) depict the EPMA elemental distributions of Ge and F in the pure and Ge-doped silica core fibers, respectively. The core of the pure silica fiber contained almost no F, whereas the cladding was doped with approximately 3.0 wt. % F. The cladding of the Ge-doped fiber was pure silica with almost no Ge, whereas the core contained approximately 3.4 wt. % Ge.

Figures 2(c) and 2(d) show the refractive index distributions of the core in the pure and Ge-doped silica core fibers, respectively. Both fibers exhibited good structural uniformity. The F-doped cladding of the pure silica fiber had a significantly lower refractive index than the core, with a core–cladding refractive index difference (Δn) of approximately 5.5 × 10⁻³, and an NA of approximately 0.126. For the Ge-doped silica core fiber, the refractive index of the Ge-doped core was significantly higher than that of the pure silica cladding, with a core–cladding Δn of approximately 8.0 × 10⁻³, and an NA of approximately 0.153.

Figures 4(a) and 4(b) show the core loss spectra before and after irradiation of the pure and Ge-doped silica core fibers. At 1550 nm, the loss in the pure silica fiber increased from 0.417 to 5.57 dB/km after 100 kGy irradiation [Fig. 4(a)], whereas the Ge-doped silica fiber loss increased from 1.06 to 42.58 dB/km [Fig. 4(b)]. The increase in background loss for the Ge-doped fiber (∼41 dB/km) was approximately eight times higher than that of the pure silica fiber (∼5 dB/km) after 100 kGy irradiation, indicating that pure silica fiber is more radiation-resistant.

Figure 5 shows the Bragg reflection spectra and Bragg wavelength shift (BWS) of the two FBGs before and after irradiation as a function of temperature. Figures 5(a) and 5(c) show a redshift in λ_B for both FBG-Si and FBG-Ge as the temperature increases before irradiation, whereas Figs. 5(b) and 5(d) indicate that this redshift persists after 300 kGy of γ-ray irradiation. The BWS at a given temperature is obtained by subtracting the λ_B at room temperature (25 °C) from the λ_B at that temperature. Figures 5(e) and 5(f) compare the BWS values of FBG-Si and FBG-Ge, respectively, before and after irradiation. The BWS increased linearly with increasing temperature. By fitting the data, the slope of the plot of BWS vs temperature, which represents the temperature sensitivity coefficient of the sensor, can be determined. The temperature sensitivity of FBG-Si slightly increases from 13.64 to 13.92 pm/°C after 300 kGy irradiation [Fig. 5(e)]; the temperature sensitivity coefficient of FBG-Ge significantly increases from 11.36 to 12.48 pm/°C after 300 kGy irradiation [Fig. 5(f)]. These results indicate that γ-ray irradiation has little effect on FBG-Si but significantly impacts FBG-Ge.

Table I and Fig. 6 show the temperature sensitivities of FBG-Si and FBG-Ge at different irradiation doses. The sensitivity coefficients of both increase with the radiation dose. The difference lies in that, after 300 KGy of radiation, the temperature sensitivity of FBG-Si increased by 2.05%, while that of FBG-Ge increased by 9.86%. Under radiation conditions, FBG-Si demonstrates superior stability compared to FBG-Ge. This suggests that in radiation environments, the FBG-Si sensors have smaller temperature measurement errors than the FBG-Ge sensors.

Figure 7 shows the changes in the Bragg reflection spectra of FBG-Si and FBG-Ge with irradiation dose. Figures 7(a) and 7(b) display the changes at 25 °C for FBG-Si and FBG-Ge, respectively, whereas Figs. 7(c) and 7(d) show the changes at 80 °C. At the same temperature, both FBGs exhibit a redshift in λ_B as the irradiation dose increases, with a more significant shift at 80 °C than at 25 °C. The BWS at a given radiation dose is calculated by subtracting the λ_B before irradiation (0 kGy) from the center wavelength at that dose. Figures 7(e) and 7(f) show the BWS values of the two fibers at 25 and 80 °C, respectively, as a function of irradiation dose. Table II lists the temperature error values for FBG-Si and FBG-Ge at different temperatures and doses. FBG-Si exhibited more stable BWS changes across temperatures, whereas FBG-Ge was more affected by irradiation. Under the same irradiation dose and temperature conditions, FBG-Si exhibited smaller temperature errors, indicating better stability, likely owing to intrinsic material differences. With a fixed amplified spontaneous emission source input, the output power of FBG-Si was higher than that of FBG-Ge before irradiation. As the irradiation dose increased, the output power of both FBGs decreased, but that of FBG-Si remained consistently higher than that of FBG-Ge at all doses.

Figure 8 shows the continuous-wave electron paramagnetic resonance (CW-EPR) and RIA spectra of pure and Ge-doped silica glass, along with the change in the defect absorption intensity with irradiation dose. Figures 8(a) and 8(b) show the CW-EPR results for the pure and Ge-doped silica glasses after irradiation, respectively. In the pure silica glass, Si-E′, NBOHC, and POR defects appeared after irradiation, and their concentrations increased with irradiation dose. Similarly, the Ge-doped silica glass exhibitied Ge-E′, Ge(1), Ge(2), and H(II) defects after irradiation, with defect concentrations increasing as the dose increased.

Similarly, absorption spectra tests were performed on the pure and Ge-doped silica glasses. Figure 8(c) shows that, after irradiation, the pure silica glass mainly developed Si-E′, ODC(II), NBOHC, and POR defects. Gaussian peak fitting was applied to the RIA spectra to track the changes in defect concentrations with increasing irradiation doses. The spectrum was deconvoluted into five Gaussian components: a 4.85 eV peak (FWHM = 1.12 eV) corresponding to NBOHC defects, 4.8 eV peak (FWHM = 0.8 eV) related to POR defects, 5.15 eV peak (FWHM = 0.3 eV) attributed to oxygen vacancy defects [ODC(II)], 5.75 eV peak (FWHM = 0.8 eV) caused by Si-E′ centers, and 6.56 eV peak (FWHM = 0.48 eV) corresponding to ODC(II) defects.

Figure 8(d) shows that after irradiation, the Ge-doped silica glass developed Ge-E′, Ge(1), and Ge(2) defects, while the germanium lone pair electron center (GLPC) defects decreased. Gaussian fitting revealed five peaks: 4.49 eV (FWHM = 1.05 eV) and 5.77 eV (FWHM = 1.25 eV) peaks corresponding to Ge(1) defects, a 5.12 eV peak (FWHM = 0.45 eV) representing GLPCs, with a negative RIA peak indicating a decrease after irradiation, 5.8 eV peak (FWHM = 0.8 eV) attributed to Ge(2) defects, and 6.8 eV peak (FWHM = 0.85 eV) associated with Ge-E′ centers.

Figures 8(e) and 8(f) show that as the irradiation dose increased, the radiation-induced defect concentration increased. In pure silica glass, Si-E′, ODC(II), NBOHC, and POR defects increased after irradiation, whereas in Ge-doped silica glass, Ge(1), Ge(2), and Ge-E′ defects increased, and GLPC defects decreased. The defect concentration in Ge-doped silica glass was significantly higher than that in pure silica glass after irradiation. Table III summarizes the structural models and spectral characteristics of radiation-induced defects from the absorption spectra.

Figure 10 shows the mechanism behind the FBG performance changes before and after irradiation. After irradiation, the formation of color centers increases the RIA, leading to higher transmission loss and weaker signal light. The Kramers–Kronig (K–K) formula indicates that in a radiation environment, the formation of radiation-induced color centers increases the absorption peak of the material. Changes in the absorption coefficient caused by these color centers (color center formation) lead to refractive index variations (Δn) through an integral relationship, causing BWS and increasing temperature measurement errors. Color centers change the optical properties of glass by altering its electronic structure. This increases the local electron density and refractive index. Although the fiber density remains nearly unchanged, the increased refractive index causes a redshift in the center wavelength, resulting in temperature measurement errors.

In this study, two high-temperature-sensitive fiber Bragg grating (FBG-Si and FBG-Ge) sensors were successfully developed based on self-fabricated pure and Ge-doped silica core fibers. The temperature response and radiation resistance of these FBGs were studied using spectroscopy and EPR, and the effects of irradiation on atomic-scale defects were explored. The results showed that both FBGs exhibited a linear redshift in λ_B with increasing temperature before and after irradiation. As the irradiation dose increased, the temperature sensitivity increased, but the rate of increase slowed, with FBG-Si exhibiting smaller changes and improved stability. At the same temperature, both FBGs exhibited redshifts and decreased output power at higher doses.

Gamma radiation, as high-energy electromagnetic waves, interacts with optical fiber materials through mechanisms such as the photoelectric effect, Compton scattering, and pair production. These interactions disrupt the local atomic structure, generating radiation-induced defects and color centers within the fiber. This process directly increases Radiation-Induced Absorption (RIA), leading to higher transmission loss and signal attenuation. The formation of radiation-induced color centers also alters the refractive index of the fiber, which significantly impacts FBG sensors by causing λ_B shifts, sensitivity changes, temperature measurement errors, and signal loss. Ultimately, these effects result in λ_B drift and inaccuracies in temperature measurements.

This study qualitatively explains the radiation effects on Fiber Bragg Gratings (FBGs), providing insights into their behavior in radiation environments. Enhancing the radiation resistance of FBGs can be achieved through various methods, including material doping (e.g., with F or Ce), hydrogen or deuterium loading, thermal annealing, pre-radiation treatment, and optimizing the fiber drawing process. Additionally, external coatings, such as carbon, polymer, or metal layers, can mitigate radiation effects. A combination of these approaches can significantly improve the radiation resistance of FBGs, enabling more accurate measurements in high-radiation environments. For example, incorporating metal coatings during fiber preparation can further expand the operational range of FBGs. This study's insights into radiation effects on FBGs mark a significant step forward, opening new possibilities for their application in radiation-prone environments.

This study was supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (No. XDB0650000) and the Lingchuang Research Project of the China National Nuclear Corporation (No. CNNC-LCKY-202281).

The authors have no conflicts to disclose.

Wen Hu: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Investigation (equal); Methodology (equal); Resources (equal); Software (equal); Supervision (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Chongyun Shao: Conceptualization (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Writing – review & editing (equal). Chunlei Yu: Funding acquisition (equal); Resources (equal). Lu Deng: Funding acquisition (equal); Resources (equal). Yuzhou Ming: Funding acquisition (equal); Resources (equal). Qing Ye: Funding acquisition (equal); Resources (equal). Xin Li: Resources (equal); Software (equal). Yinpeng Liu: Resources (equal). Mengda Wei: Resources (equal). Dongyu He: Resources (equal). Lili Hu: Conceptualization (equal); Formal analysis (equal); Methodology (equal); Project administration (equal); Resources (equal). Si-Yu Li: Funding acquisition (equal); Resources (equal); Supervision (equal); Validation (equal). Anlian Pan: Funding acquisition (equal); Methodology (equal); Project administration (equal); Resources (equal); Supervision (equal); Validation (equal). Meisong Liao: Conceptualization (equal); Funding acquisition (equal); Project administration (equal); Supervision (equal); Validation (equal).

The data that support the findings of this study are available within the article.