**3. Radiation effects in photonic devices**

#### **3.1. Fiber Bragg gratings**

**•** annealing and photobleaching can contribute to partial recovery of the irradiation induced

Tests on photobleaching of two Erbium-doped fibers (EDFs) subjected to 1.3 and 2.24 kGy gamma irradiation were carried out at two laser wavelengths (532 and 976 nm). A higher efficiency was achieved with the shorter wavelength excitation. The annealing effect induced by the 976 nm laser is comparable to that present in the case of thermal annealing at 500 K [69].

In another experiment, an EDF was tested under electron irradiation (beam energy 1.2 MeV, total dose 10 kGy), and the output power, noise figure and the central wavelength were measured after the irradiation, with the pump at 980 nm and detection at 1550 nm. The central wavelength did not change after irradiation, while the output power and noise figure were deteriorated (i.e. the output power dropped from 10 to –60 dBm for the same pump level).

Single mode and multi-mode Yb-doped optical fibers, acting as amplifiers at 1064 nm, were subjected to gamma rays and mixed gamma-neutron irradiation, up to a total dose of 1 Gy(Si) [71]. It was noticed a linear decrease of the output power with the irradiation dose. A slight

In his PhD thesis, Fox [72] concluded that EDFs are more radiation sensitive than Yb3+-doped fibers, and the most radiation hardened to gamma ray are optical fibers of Er3+/Yb3+ co-doped type. In this investigation, the radiation induced degradation increases with the increase of the

Optical fiber preforms of Yb-free aluminosilicate core, Yb-doped Al-free silicate core, Ybdoped alumino-silicate cores and Yb-free germanosilicate core produced by MCVD and solution-doping techniques were irradiated by 45 kV X-rays, at ~2.5 Gy(SiO2)/ min, for doses up to 0.3–0.5 kGy [73]. The effects of the irradiation were studied by thermally stimulated

Investigations were carried out to evaluate the influence of H2-loading of EDFs, under pumping at 980 and 1480 nm, as the samples were subjected to gamma irradiation, having doses from 0.1 to 10 kGy. One such sample was a H2-loaded EDF, while the other was H2-free carbon-coated [74]. At 980 nm pumping a photobleaching effect was observed which increases the efficiency of the process especially in the H2-loaded optical fiber. No such effect was present with the pump radiation of 1480 nm. H2-loading and the use of a hermetical coating of the

One approach to enhance the radiation resistance of the optical amplifier consists in the use of Er-doped-nanoparticles optical fibers [76]. Another proposed method is based on the doping

Tests on a thulium doped optical fiber amplifier were performed under neutron irradiation and a 17.1 dB was observed for a dose of 720 Gy, which is close to the computed values [78].

The development of Bi co-doped silica optical fibers [79] having an extensive emission operation range (from 1100 to 1800 nm) called researchers attention on their investigation

optical fiber, which prevents H2 diffusion, produce a radiation resistant EDF [75].

of the optical fiber core with Ce, reducing in this way the radiation sensitivity [77].

Within 2 weeks, the two parameters partially recovered [70].

recovery was observed at room temperature, after 20 hours.

luminescence (TSL) and optical absorption measurements.

effects [65, 67, 68].

50 Radiation Effects in Materials

dose rate.

Fiber Bragg gratings were excessively studied under various type of ionizing radiation: gamma ray [12, 83, 84], neutron [85], mixed gamma-neutron [86–88] or 13.5 MeV protons [89]. In most cases, the changes of the FBG Bragg wavelength are relatively low. Tests were carried out using Type I, Type II, Type IA, Type IIA, chemical composition or fs laser engraved gratings, and were focused on the effect of fiber composition, dose rate, total dose, heating during the irradiation or possible photobleaching during the exposure to ionizing radiation [90]. For example, Type I gratings produced in H2-loaded standard communication optical fibers (SMF28TM) or B/Ge co-doped optical fibers present an increase of the central wavelength followed by a plateau, with an overall change between 10 pm to 34 pm, under 0.54 MGy exposure. In the case of Type II gratings written in B/Ge co-doped optical fibers exhibit a decrease of the central wavelength between -30 and -60 pm. Degradation (30–50%) of gratings reflectivity was produced after irradiation, while no recovery was observed in the studied sensors [91]. A quite significant impact of gamma radiation on Ge-doped core optical fibers was observed as such uncoated fibers are immersed in water during gamma exposure [83], for a total dose of up to 5 MGy. Generally, all the investigations carried out on radiation effects on FBGs focused on the possible use of such devices in radiation environments for temperature or humidity [12] monitoring. Under some design circumstances (optical fiber type, technology used to write the grating), these modifications are significant, making possible the use of FBGs as radiation detectors [92, 93].

More recently, the use of FBGs developed in sapphire optical fiber was proposed for moni‐ toring very high temperature in nuclear reactors [94].

Through these studies, the paradigm was changed as [95]:


This research indicated a linear dependence of the Bragg wavelength shift with the irradiation dose for commercially available FBGs and grating produced in radiation hardened optical fibers, and this shift was monitored by measuring simultaneously with a thermocouple the temperature change in the irradiation plane. This aspect is important because no saturation or permanent modification of FBG wavelength was observed; hence these sensors can be used for on-line measurements. The principle of the charged particle beam diagnostics is illustrated in **Figure 8**, for the case of an electron beam having a diameter of about 100 mm. The spatial resolution is limited by the number of FBGs engraved in an optical fiber, their individual length and the distance between two adjacent FBGs. In the described proof-of-concept design, gratings having 12 mm and 4 mm length were employed. As compared to other solutions for charge particle beam monitoring (arrays of Faraday cups; ionization chambers; micro strip metal detector; pepper-pot device, slit-grid or rotating slits; scintillating screen or gas detector; flat panel detectors, arrays of p-i-n diodes; moving wire or vibrating wire scanners) there are several advantage of this approach: its immunity to electro-magnetic noise, remote monitoring capability and the possibility to multiplex the acquired signals by using few connecting lines.

The operation of the proposed instrument is similar to that employed for laser beam diagnos‐ tics. In the evaluation of a laser beam quality an image sensor is used to acquire the transversal distribution of the laser beam's intensity. Periodically, the electric charge generated inside the image sensors is removed upon reading and new acquisition starts. In this implementation the instrument operation is based on the proved reliability of the tested FBGs under electron beam exposure and on the linear shift of FBGs Bragg wavelength with the temperature in the detection plan, and hence with the deposed energy by the charge particle beam. Prior to the use, the FBGs were calibrated as it concerns their wavelength change vs. temperature.

When the charge particle beam (1) is propagating from the linear accelerator output (2) its diameter increases (4) in the detection plan (5) due to its divergence (**Figure 8a**). In the detecting plan (item 1 in **Figure 8b**) a mesh composed of FBGs (3) is placed, the sensors being embedded into a thermally insulating material to prevent the lateral dissipation of the heat. As the individual gratings are exposed to different beam energies their temperature increases according to the energy deposed on each detecting site. In this way, in time, a map of the transferred energy at each location is obtained.

a total dose of up to 5 MGy. Generally, all the investigations carried out on radiation effects on FBGs focused on the possible use of such devices in radiation environments for temperature or humidity [12] monitoring. Under some design circumstances (optical fiber type, technology used to write the grating), these modifications are significant, making possible the use of FBGs

More recently, the use of FBGs developed in sapphire optical fiber was proposed for moni‐

**•** FBGs were produced in radiation hardened optical fibers by two-beam interferometer and

**•** for the first time FBGs written in both standard commercial optical fibers and radiation hardener optical fibers were tested, exposed to the electron beam from a linear accelerator;

This research indicated a linear dependence of the Bragg wavelength shift with the irradiation dose for commercially available FBGs and grating produced in radiation hardened optical fibers, and this shift was monitored by measuring simultaneously with a thermocouple the temperature change in the irradiation plane. This aspect is important because no saturation or permanent modification of FBG wavelength was observed; hence these sensors can be used for on-line measurements. The principle of the charged particle beam diagnostics is illustrated in **Figure 8**, for the case of an electron beam having a diameter of about 100 mm. The spatial resolution is limited by the number of FBGs engraved in an optical fiber, their individual length and the distance between two adjacent FBGs. In the described proof-of-concept design, gratings having 12 mm and 4 mm length were employed. As compared to other solutions for charge particle beam monitoring (arrays of Faraday cups; ionization chambers; micro strip metal detector; pepper-pot device, slit-grid or rotating slits; scintillating screen or gas detector; flat panel detectors, arrays of p-i-n diodes; moving wire or vibrating wire scanners) there are several advantage of this approach: its immunity to electro-magnetic noise, remote monitoring capability and the possibility to multiplex the acquired signals by using few connecting lines. The operation of the proposed instrument is similar to that employed for laser beam diagnos‐ tics. In the evaluation of a laser beam quality an image sensor is used to acquire the transversal distribution of the laser beam's intensity. Periodically, the electric charge generated inside the image sensors is removed upon reading and new acquisition starts. In this implementation the instrument operation is based on the proved reliability of the tested FBGs under electron beam exposure and on the linear shift of FBGs Bragg wavelength with the temperature in the detection plan, and hence with the deposed energy by the charge particle beam. Prior to the use, the FBGs were calibrated as it concerns their wavelength change vs. temperature.

When the charge particle beam (1) is propagating from the linear accelerator output (2) its diameter increases (4) in the detection plan (5) due to its divergence (**Figure 8a**). In the detecting plan (item 1 in **Figure 8b**) a mesh composed of FBGs (3) is placed, the sensors being embedded into a thermally insulating material to prevent the lateral dissipation of the heat. As the

**•** a mash of FBGs was used for beam diagnostics of charged particle beams, as a novelty.

as radiation detectors [92, 93].

52 Radiation Effects in Materials

deep ultraviolet fs laser radiation;

toring very high temperature in nuclear reactors [94].

Through these studies, the paradigm was changed as [95]:

**Figure 8.** The operation principle of the electron beam analyzer: (a) the moving shutter used to interrupt FBGs expo‐ sure to the electron beam: (1) the incident electron beam; (2) beam diameter at the exit of the linear accelerator focusing system; (3) moving shutter; (4) the electron beam diameter at its incidence on the detection plan; (5) detecting plan. (b) The position of the FBGs mash: (1) the detecting plan; (2) electron beam pattern on the detection plan; (3) FBGs; (4) connecting optical fibers; (5) optical fiber interrogator; (6) bars symbolizing the integral energy deposited on a particu‐ lar FBG at a specified moment. (c) Example of the data acquired for the 8 kGy dose [95]. (d) Top view of the shutter and cooling system.

The detector operates as an energy integrator (dose related measurements). Permanently, the central wavelength of all sensors is acquired by a Micron Optics optical fiber interrogator, which makes possible the real time mapping of the electron beam cross section energy distribution. In order to avoid saturation, periodically the FBGs signal has to be "reset" by blocking the electron beam and forced cooling the FBGs array. For this purpose a thick (20 mm) Al shutter restricts under the software control (**Figure 8a** and **8d**) the exposure of the FBGs by interrupting the electron beam. During the interval the beam is blocked, a cooler pushes air over the sensors' matrix. The Al shutter goes back and forth as the array has to be exposed or cooled. The flow chart for the instrument operation is presented in **Figure 9**.

**Figure 9.** The flow chart of the electron beam analyzer.

During the "Initialization" step, the shutter is closed, the FBGs are cooled, and the functioning parameters set by the operator are introduced in the system. If a STOP command was issued by the operator the system stops and displays in 3D format the integrated energy to which each sensor was subjected, measured as Bragg wavelength shift in response to the local temperature increase. If no STOP command was given the system periodically acquires on the central wavelength shift for each sensor and checks if the maximum set temperature was reached by any of the FBGs. If this temperature was reached the system stops and displays the results as a 3D representation. If no such signal was received the software commands during the "Data acquisition" step the closing of the shutter and the cooling of the FBGs array. The periodicity of the cooling cycles, and the upper limit of the temperature to which any of the sensors can be subjected are set by the operator. The Bragg wavelength shift with the dose increase is given in **Figure 10**. In this case, during the irradiation pause, the grating's cooling was obtained by convection in air and not by forced cooling. The proposed instrument can be used during the adjusting process of the charged particle accelerator or to check the stability in time of the output beam.

**Figure 10.** The change of the Bragg wavelength with the dose increase. Periodically, between two exposures the gra‐ ting was cooled by natural convection [95].

A novelty in the field can be considered the investigation of FBGs performances written in polymer fibers under fast neutrons irradiation. In the paper it is suggested that neutrons produce a degradation of the fiber structure, which in turn causes a shift of the Bragg wave‐ length up to 14 pm. The wavelength change with neutron dose can be exploited in radiation dosimetry [96].

#### **3.2. Long period gratings**

The detector operates as an energy integrator (dose related measurements). Permanently, the central wavelength of all sensors is acquired by a Micron Optics optical fiber interrogator, which makes possible the real time mapping of the electron beam cross section energy distribution. In order to avoid saturation, periodically the FBGs signal has to be "reset" by blocking the electron beam and forced cooling the FBGs array. For this purpose a thick (20 mm) Al shutter restricts under the software control (**Figure 8a** and **8d**) the exposure of the FBGs by interrupting the electron beam. During the interval the beam is blocked, a cooler pushes air over the sensors' matrix. The Al shutter goes back and forth as the array has to be exposed or cooled. The flow chart for the instrument operation is presented in **Figure 9**.

During the "Initialization" step, the shutter is closed, the FBGs are cooled, and the functioning parameters set by the operator are introduced in the system. If a STOP command was issued by the operator the system stops and displays in 3D format the integrated energy to which each sensor was subjected, measured as Bragg wavelength shift in response to the local temperature increase. If no STOP command was given the system periodically acquires on the central wavelength shift for each sensor and checks if the maximum set temperature was reached by any of the FBGs. If this temperature was reached the system stops and displays the results as a 3D representation. If no such signal was received the software commands during the "Data acquisition" step the closing of the shutter and the cooling of the FBGs array. The periodicity of the cooling cycles, and the upper limit of the temperature to which any of the sensors can be subjected are set by the operator. The Bragg wavelength shift with the dose increase is given in **Figure 10**. In this case, during the irradiation pause, the grating's cooling

**Figure 9.** The flow chart of the electron beam analyzer.

54 Radiation Effects in Materials

Long period gratings (LPGs) were tested only under gamma ray exposure [90]. Most of the measurements were performed off-line [97–99], with some exceptions when the device behavior was monitored during the irradiation [100, 101]. The LPGs were obtained by different techniques: CO laser or UV engraving [99], electric arc-discharge (EAD) technique [101], CO2 laser point-by-point writing [97], as chiral gratings [100], having a turnaround point (TAP) design produced by a CO2 laser [98]. Various optical fibers were employed: N-doped and Gedoped optical fibers [99], pure-silica-core/ F-doped silica cladding fibers [101], SMF-28TM [97], core doped with P, Ge, F, rare earth elements [100], photosensitive fibers (B/Ge co-doped) [98]. The irradiation conditions are modified from 5 kGy [97] up to 1.47 MGy [99]. Depending on the optical fiber type, the writing technique, the dose rate/total dose used, the reported results led to different conclusions: (i) no effect was observed on the LPG transmission spectrum for doses lower than 5 kGy [97], lower than 500 kGy [101] or as high as 1.47 MGy [99], (ii) a 10 nm shift of the transmission deep was noticed for doses of 100 kGy [100], (iii) very high shift of the dual resonance deep (35 nm) is present for a dose of 6 kGy [98]. This variety of investigations outcomes suggest that, by an appropriate selection of the above mentioned parameters, LPGs can be produced either to be radiation hardened or with pronounced radiation sensitivity, appropriate to be used in radiation dosimetry. No change in the temperature sensitivity of LPGs upon irradiation was observed [101].

A new proposed approach in studying gamma irradiation effects on LGPs Includes several novelties [102, 103]:


The use of the OFDR improved drastically the detection S/N as compared to classical reading with an optical spectrum analyzer. During the irradiation the sensors were encapsulated into ceramic radiation transparent cases to avoid any strain induced changes in the LPG spectrum, and in the mean time, the gratings were placed into a thermally insulated box. The temperature was permanently monitored both inside this box and in the irradiation chamber and temper‐ ature related corrections were applied to the LPG characteristics. Based on the referred papers, **Figure 11** illustrates comparatively the behaviour of gamma irradiation on the LPG developed into a standard communication optical fiber (LPGsc) and one written in a radiation hardened optical fiber (LPGrh). The wavelength deep of the two samples move in opposite directions, the wavelength of the grating produced in the standard communication fiber increases with the dose increase, while for the other one the wavelength decreases with the dose increase. Besides that, it can be noticed the magnitude of the two changes, LPGsc being by far more

**Figure 11.** The change of the wavelength deep for two LPGs written in standard communication optical fibers LPGsc and in F-doped core optical fiber LPGrh.

sensitive. The fluctuations of the wavelength deep values during gamma exposure as well as the partial saturation trend are associated to the on-off operation of the irradiation facility. In this way, for the first time the recovery effect during the irradiation pauses was observed.

Because for the LPGrh the wavelength change is by far less significant than that in the case of LPGsc, it can be expected that the first one can be used in radiation environments as a sensor for temperature/humidity, while the second LPG can be employed within specific dose range as an optical fiber-based dosimeter.
