Preamble

Optically Stimulated Luminescence and Thermoluminescence in Newly Developed LiMgPO₄:Gd

Kai-Yong Tang,¹,²,∗ Li Fu,¹ Si-Yuan Zhang,¹ Hai-Jun Fan,¹ Yan Zeng,¹,³ and Mo Zhou¹
¹State Key Laboratory of NBC Protection for Civilian, P.O. Box 1044, Ext. 203, Beijing 102205, China
²Solid Dosimetric Detector and Method Laboratory, P.O. Box 1044, Ext. 203, Beijing 102205, China
³School of Materials Science and Engineering, Shanghai University, Shanghai, 200444, China

Five samples of LiMgPO₄:Gd were prepared via five different production processes using a solid-state reaction method. The effects of the preparation process on optically stimulated luminescence (OSL) and thermoluminescence (TL) were investigated. Considering its high sensitivity, low fading, and minimum detectable dose (MDD), the LiMgPO₄:Gd phosphor heated to $900^{\circ}$C for 15 h is concluded to be optimal. The effects of annealing on the OSL sensitivity, relative residual OSL signals measured after 24 h of irradiation, and MDD of LiMgPO₄:Gd phosphors heated to $900^{\circ}$C for 15 h were also investigated. Considering its high sensitivity, low fading, and MDD, annealing at $350^{\circ}$C for 1 h is concluded to be optimal. The OSL signal of LiMgPO₄:Gd was derived from the principal TL glow peak. For a maximum integration time of 5 s, the OSL signal was stable, with no fading 30 days after irradiation. LiMgPO₄:Gd eliminated approximately 2.2% of the OSL signal at each readout for a readout time of 0.1 s, which is sufficient for fast and multiple OSL readout. The sensitivity of LiMgPO₄:Gd phosphor, annealed for 1 h at $350^{\circ}$C with a reading time of 0.1 s, was found to be approximately 98% of that observed for α-Al₂O₃:C(TLD-500k), which should be sufficient for low-dose measurements in personal, workplace, and environmental dosimetry.

Keywords: Fading; LiMgPO₄:Gd; Optically stimulated luminescence; Phosphors; Thermoluminescence.

Introduction

Optically stimulated luminescence (OSL) and thermoluminescence (TL) refer to phenomena whereby a material previously subjected to irradiation releases a luminescent signal in response to the application of an appropriate optical or thermal stimulus \cite{1-4}. TL is a well-established technology that is widely used in various dosimetric applications, including personal, workplace, and environmental dosimetry; medical dosimetry; imaging of ionizing radiation; archaeological and geologic dating; and assessment of radiation accidents \cite{1,5}. Currently, thermoluminescent dosimeters (TLDs) are predominantly used to assess individual radiation exposure \cite{6-8}, with LiF:Mg,Cu,P (GR-200A) being the most prominent example \cite{9-13}.

The basic principles of TL and OSL methods are the same, except that the former uses heat to read the signal, whereas the latter uses light stimulation \cite{14}. The distinctive stimulation patterns associated with the TL and OSL methods provide the latter with certain advantages \cite{2,15}. A major advantage of the OSL method over the TL method is the ability to read signals at room temperature without heating the detector, allowing dosimeters to be used in combination with plastic adhesives at room temperature. In the OSL mode, the intensity of the excitation light and the excitation time can be controlled to read out only a portion of the signal at a time \cite{14}, in contrast to the TL mode, in which all signals are read simultaneously.

It is possible to perform multiple readout sessions and achieve a rapid readout by reading only a portion of the signal simultaneously. In the InLight reader, each detector was stimulated for a duration of 1 s. To provide multiple readings, a percentage of the OSL signal was discharged with weak stimulation, with a value of only 0.07%. In contrast, a value of 0.25% was obtained with strong stimulation \cite{14}. Recently, several Individual Monitoring Service (IMS) laboratories have transitioned from using TLD to OSL dosimeters \cite{16}.

In view of the benefits offered by OSL technology, numerous research teams have addressed the development of new OSL materials; however, only two commercially available OSL dosimeters based on Al₂O₃:C and BeO have been reported, in contrast to the significantly higher number of TL dosimeters \cite{1}. Interestingly, α-Al₂O₃:C is an ideal OSL material with high stability and sensitivity. Many methods have been adopted to prepare α-Al₂O₃:C OSL dosimeters; however, only α-Al₂O₃:C OSL dosimeters prepared using crystal growth technology have the highest OSL sensitivity \cite{17-22}. The growth of carbon-doped alumina crystals is susceptible to fluctuations in the growth conditions, which can result in considerable variations in the properties of α-Al₂O₃:C detectors between samples and batches \cite{14}. To enhance the uniformity of OSL dosimeters, commercially available Luxel+ and InLight OSL dosimeters were manufactured by blending α-Al₂O₃:C powders with polyester and depositing them on a transparent polyester film \cite{14}. This resulted in the initial dose of the linear range of the dose response of the product being worse than that of the material itself. It has been reported that BeO exhibits a pronounced exponential decline in sensitivity upon initial use \cite{23}, accompanied by signal fading following radiation exposure \cite{24}.

Many countries are developing novel and enhanced OSL materials, and many materials with OSL properties have been reported \cite{25-29}. However, they have not been widely adopted in commercially available dosimeters because of different inherent limitations, one of which is high fading. In particular, the high fading exhibited by OSL materials used in personal dosimetry and environmental monitoring applications is unacceptable because it inevitably leads to substantial dosimetry inaccuracies. The minimal detectable dose of Li₂B₄O₇:Ag is as low as 15 µGy; however, its CW-OSL signal decreased by approximately 30% over a 24-h period after irradiation \cite{30}. Following one day of radiation exposure, the OSL signals of CaSO₄:Dy exhibited a decay of approximately 50%, whereas those of CaSO₄:Eu exhibited a decay of approximately 10% \cite{31,32}. The OSL signals from Mg₄SiO₄:Tb and LiAl₂O₄:Tb exhibited a decline in intensity between 30 and 40%, spanning approximately 20 h \cite{33}. In a study on NaMgF₃ doped with Ce, Tm, and Tb, the OSL response decreased by 20, 13, and 35%, respectively, after irradiation for 1 d, 60 h, and 5 d \cite{34,35}. The minimum detectable doses (MDDs) for MgO:Li,Ce, and Sm are found to be as low as 0.2 µGy; however, a 15% fading in the OSL signals was observed within the initial hour \cite{36}.

The aim of this study was to improve the time stability of the OSL signal of LiMgPO₄ to achieve fast and multiple OSL readouts. Five samples of LiMgPO₄:Gd phosphor powder were prepared using five different preparation methods. The effects of the preparation process on the OSL decay curves, TL glow curves, OSL sensitivity, MDD, and fading characteristics of LiMgPO₄:Gd phosphors after 24 h of irradiation were investigated. The impact of annealing on the OSL sensitivity, relative residual OSL signals measured after 24 h of irradiation, and MDD of LiMgPO₄:Gd phosphors heated to $900^{\circ}$C for 15 h were also investigated. The fundamental characteristics of LiMgPO₄:Gd annealed at $350^{\circ}$C for one hour were examined.

LiMgPO₄ is one of the numerous new OSL materials with excellent dosimetric properties that have been investigated by research teams worldwide \cite{26,37-50}; however, its main drawback is high fading. A recent study summarized the OSL fading behavior of LiMgPO₄. Various conditions, including changes in the dopant species, impurity concentrations, preparation processes, and prereadouts to eliminate the effects of low-temperature peaks, have been used to minimize the fading of LiMgPO₄ phosphors. It has been observed that fading can be reduced by preheating or optical treatment; however, the application of preheating or optical treatment resulted in a significant reduction in OSL intensity, with a reduction of 60% to 70% \cite{51}. Similarly, the fading of LiMgPO₄:Tb,Sm,B was reduced from 30% without pretreatment to approximately 7% by preheating at $160^{\circ}$C for 30 s after 30 days of irradiation \cite{52}. One day after irradiation, the fading of LiMgPO₄ doped with B alone was reduced from 15% to 30% for LiMgPO₄:Tb,B \cite{53}. The low-temperature peak of LiMgPO₄:B is still very high, and the fading is still unable to meet the requirements of personal dose monitoring. By varying the preparation temperature, the fading of the LiMgPO₄:Tb,Sm, and B phosphors was significantly reduced, with the residual OSL signal as high as 92.3% after 24 h of irradiation \cite{47}.

The fading of the OSL signals in LiMgPO₄:Er after irradiation has been reported to be reduced by improving the preparation process \cite{41}. It was found that the fading of LiMgPO₄:Er samples heated to $900^{\circ}$C for 10 h was less than 5% when observed within 30 d after irradiation and when the integration time was greater than or equal to 20 s; however, the fading was greater than 5% when the integration time was less than 20 s. To gain insight into the features of energy transfer in doped LiMgPO₄, LiMgPO₄:RE (RE = Nd, Sm, Gd, Tb, Dy, Ho, Er, Tm) was synthesized \cite{39}. Both LiMgPO₄:Er and LiMgPO₄:Gd exhibit broad emission bands at 360 nm in their TL spectra, and the intensity of the latter is higher than that of the former \cite{39}. However, the OSL characteristics of LiMgPO₄:Gd were not studied.

Significant efforts have been made to develop new OSL materials with properties comparable to those of Al₂O₃:C, including a simple synthetic process, easily available raw materials, low cost, and suitability for mass production.

Materials and Methods

A. Material Synthesis

A simple solid-state reaction method was used to synthesize LiMgPO₄:Gd materials. The following reagents were weighed according to a stoichiometric ratio: Li₂CO₃ (lithium carbonate), 3MgCO₃·Mg(OH)₂·3H₂O (magnesium carbonate, 99.9%), and NH₄(H₂)PO₄ (ammonium dihydrogen phosphate, 99.99%), then thoroughly mixed. An appropriate quantity (0.25 mol%) \cite{39} of Gd₂O₃ was added to the mixture, which was then mixed thoroughly.

Five distinct processes were employed to prepare the LiMgPO₄:Gd phosphors. The synthesis was divided into three stages, with the first two stages being identical across all five processes and only the third stage differing.

In the first phase, the mixture was placed in an alumina crucible and positioned in the center of a tube furnace. It was heated from room temperature to $300^{\circ}$C over 30 min, then maintained at $300^{\circ}$C for 2 h. The crucible was then removed from the furnace and rapidly cooled to room temperature. The samples were manually pulverized and sieved. In the second step, the material obtained in the first step was heated from 300 to $650^{\circ}$C over 30 min, then maintained at $650^{\circ}$C for 2 h. For the third step, preparations 1-3 involved heating the material from the second step from 650 to $900^{\circ}$C over 30 min, followed by additional periods of 10, 15, or 20 h at $900^{\circ}$C, respectively. In the third step of preparations 4 and 5, the material from the second step was heated from 650 to $950^{\circ}$C and $1000^{\circ}$C, respectively, over 30 min, then maintained at $950^{\circ}$C and $1000^{\circ}$C for 2 h, respectively.

For comparison, an α-Al₂O₃:C detector (TLD-500K) with a 5 mm diameter and 0.9 mm thickness was used.

B. Characterizations

The powder structure was characterized by powder X-ray diffraction (XRD; Bruker D8 Advance). Scanning electron microscopy (SEM) and energy dispersive spectroscopy (EDS) were performed to confirm the presence of Gd in the structure using a Hitachi SU8600.

The Risø TL/OSL reader TL/OSL-DA-20 was used to measure the OSL decay curves and TL glow curves of the samples, as well as for β-source irradiation. Many studies have been published on this TL/OSL reader; therefore, it will not be described in detail here \cite{41,52}. A continuous-wave OSL (CW-OSL) method was used for OSL measurements, with stimulus power at 90% and duration of 88 s. TL glow curves were measured from room temperature to $450^{\circ}$C at a heating rate of $5^{\circ}$C s⁻¹.

To test the effect of the preparation process on the OSL decay curves, TL glow curves, OSL sensitivity, MDD, and fading characteristics of the LiMgPO₄:Gd phosphors after 24 h of irradiation, the five types of samples were fully bleached using the TL readout method, which involves pretreatment at a ramp rate of $5^{\circ}$C per second, followed by a readout temperature of $450^{\circ}$C; then, their OSL background signals were measured. A $^{90}$Sr/$^{90}$Y β source was used to irradiate the samples for 2 s at a nominal dose rate of 50 µGy s⁻¹ to measure the OSL decay curve, TL glow curve, and fading signal 24 h after irradiation.

To ascertain the effect of annealing temperature on the OSL sensitivity, MDD, and fading of LiMgPO₄:Gd 24 h after irradiation, LiMgPO₄:Gd powders synthesized at $900^{\circ}$C for 15 h were subjected to annealing treatment at temperatures between $350^{\circ}$C and $500^{\circ}$C at intervals of $50^{\circ}$C for an hour.

The background signal was quantified immediately following the annealing of the samples, which were then irradiated at ambient temperature for 2 s at a nominal dose rate of 50 µGy s⁻¹ using a $^{90}$Sr/$^{90}$Y β source integrated into the Risø reader. Subsequently, OSL signals were recorded. The samples were fully bleached using the TL readout method, which entailed pretreatment with a temperature ramp rate of $5^{\circ}$C s⁻¹, followed by a readout to $450^{\circ}$C, then irradiated at ambient temperature using a $^{90}$Sr/$^{90}$Y β source integrated into the Risø reader at a nominal dose rate of 50 µGy s⁻¹ for 2 s. The OSL signal was then recorded 24 h after irradiation.

To experimentally assess long-term fading of LiMgPO₄:Gd, a sample synthesized at $900^{\circ}$C for 15 h was used. Before irradiation, the samples were annealed at $350^{\circ}$C for one hour. To investigate the kinetic order of LiMgPO₄:Gd, the samples were completely bleached, which involved pretreatment at a rate of $5^{\circ}$C per second, followed by a readout temperature of $450^{\circ}$C, irradiation with a dose of 0.1 Gy using a β-in-situ irradiator, and subsequent reading of the TL glow curves at $450^{\circ}$C. The heating rate of the TL readings was set to $5^{\circ}$C s⁻¹, and this was repeated six times at irradiation doses of 0.2, 0.5, 1, 2, 5, and 10 Gy.

To experimentally assess long-term fading of LiMgPO₄:Gd, the sample was irradiated with a dose of 50 µGy from a $^{60}$Co gamma source. The OSL signals were read at 0 days (approximately 40 min) and 1, 2, 5, 10, 20, and 30 days after irradiation.

To study the reusability of LiMgPO₄:Gd, five samples, each weighing 10 mg, were extracted and placed at five distinct positions within the TL/OSL reader. The samples were fully bleached using the TL readout method, which entailed pretreatment with a temperature ramp rate of $5^{\circ}$C per second, followed by a readout to $450^{\circ}$C, then irradiated at ambient temperature using a $^{90}$Sr/$^{90}$Y β source integrated into the Risø reader at a nominal dose rate of 50 µGy s⁻¹ for 2 s before reading out the OSL signal. Ten such cycles were performed.

Results

A. X-ray Diffraction (XRD)

Figure 1 [FIGURE:1] shows the XRD patterns of LiMgPO₄:Gd samples produced through five distinct preparation processes in conjunction with Inorganic Crystal Structure Database (ICSD) Card No. 201138. The measured diffraction patterns agreed with the reference pattern with respect to the positions of the majority of reflections; however, the differences in sensitivity may also be attributable to variations in sample preparation.

B. Elemental Analysis

Figure 2 [FIGURE:2] depicts the map sum spectra of LiMgPO₄:Gd samples produced using five distinct preparation processes, confirming the presence of gadolinium in the structure.

C. Effect of Preparation Processes

The effects of the five preparation processes on LiMgPO₄:Gd phosphors are shown in Fig. 3 [FIGURE:3]. The OSL decay curves, normalized by the maximum of the LiMgPO₄:Gd phosphors produced using the five preparation processes, are shown in Fig. 3a. The LiMgPO₄:Gd phosphor heated to $900^{\circ}$C for 15 h exhibited the slowest decay, while the phosphor heated to $1000^{\circ}$C for 2 h exhibited the fastest decay. The remaining three curves exhibit relatively similar decay profiles.

The normalized OSL signals of the five LiMgPO₄:Gd phosphors prepared by the five processes, compared to TLD-500k for various integration times, are shown in Fig. 3b. The OSL signals were averaged over five readings for each data point, with all standard deviations less than 10%. The OSL signals of the LiMgPO₄:Gd phosphors from the five preparation processes are lower than those of the TLD-500k phosphors for the different integration times, with the highest OSL sensitivity obtained for the LiMgPO₄:Gd phosphor heated to $900^{\circ}$C for 15 h.

The relative residual OSL signals of the five LiMgPO₄:Gd phosphors prepared using the five preparation methods with different integration times, measured after 24 h of irradiation compared to those measured immediately after irradiation, are shown in Fig. 3c. The OSL signals were averaged over five readings for each data point. For the sample heated to $1000^{\circ}$C for 2 h, the relative residual OSL signals measured after 24 h of irradiation were approximately 0.701–0.88, while the relative residual OSL signals measured after 24 h of irradiation for the other four samples were stable across different integration times.

The OSL MDDs of the five LiMgPO₄:Gd phosphors prepared using the five processes are presented in Fig. 3d. The preparation process has a significant impact on the MDD, with relatively low MDDs observed in the samples heated to $950^{\circ}$C for 15 h across different integration times.

Typical TL glow curves and TL integrated signals of the five LiMgPO₄:Gd phosphors prepared using the five processes are presented in Figs. 3e and 3f. For clarity, Fig. 3e shows the TL signal between $50^{\circ}$C and $300^{\circ}$C because of the minimum TL signal between $300^{\circ}$C and $450^{\circ}$C. The LiMgPO₄:Gd phosphor heated to $900^{\circ}$C for 15 h exhibited the highest TL sensitivity, while the phosphor heated to $1000^{\circ}$C for 2 h exhibited the lowest TL sensitivity.

Considering its high sensitivity, low fading, and MDD, the LiMgPO₄:Gd phosphor heated to $900^{\circ}$C for 15 h is concluded to be optimal. The following study was conducted only on this LiMgPO₄:Gd phosphor.

D. Effect of Annealing on LiMgPO₄:Gd

The effect of annealing on LiMgPO₄:Gd phosphors is shown in Fig. 4 [FIGURE:4]. Figure 4a shows the OSL signals of four distinct types of LiMgPO₄:Gd phosphors, each subjected to annealing at temperatures of $350^{\circ}$C, $400^{\circ}$C, $450^{\circ}$C, and $500^{\circ}$C for 1 h, with a single unannealed LiMgPO₄:Gd phosphor normalized to TLD-500k at various integration times. Each data point corresponds to the mean value of the OSL signals for five samples, with standard deviations less than 10%. The OSL sensitivity of LiMgPO₄:Gd can be enhanced by annealing, with the phosphor annealed at $350^{\circ}$C for 1 h exhibiting the highest OSL sensitivity. As the annealing temperature exceeds $350^{\circ}$C, the OSL sensitivity decreases with rising temperature. The initial and total OSL signals were found to be essentially identical to those of TLD-500k at an annealing temperature of $350^{\circ}$C, with the initial OSL signal of the LiMgPO₄:Gd phosphor annealed for 1 h at $350^{\circ}$C relative to TLD-500k approximately 0.98.

Figure 4b depicts the relative residual OSL signals after 24 h of irradiation for four distinct types of LiMgPO₄:Gd phosphors annealed at $350^{\circ}$C, $400^{\circ}$C, $450^{\circ}$C, and $500^{\circ}$C for 1 h, and a single unannealed phosphor. The highest residual OSL signal was observed for the unannealed phosphor. For annealed phosphors, the OSL signal can be considered stable when the integration time is less than 10 s.

Figure 4c depicts the OSL MDD for four distinct types of LiMgPO₄:Gd phosphors annealed at $350^{\circ}$C, $400^{\circ}$C, $450^{\circ}$C, and $500^{\circ}$C for 1 h, and a single unannealed phosphor. The OSL MDD of LiMgPO₄:Gd was significantly reduced by annealing, with the MDD of phosphors annealed for one hour at $350^{\circ}$C as low as 2.2 µGy when the integration time was one second, and 16.2 µGy when the integration time was 0.1 s.

Considering the high sensitivity, low fading, and MDD, annealing the LiMgPO₄:Gd phosphor at $350^{\circ}$C for 1 h was optimal. The following study was conducted only on this LiMgPO₄:Gd phosphor.

E. Basic Properties of LiMgPO₄:Gd

1. TL Dose Response and Kinetic Order

The effects of dose, heating rate, and OSL readings on the TL of the LiMgPO₄:Gd phosphor and OSL decay curve are shown in Fig. 5 [FIGURE:5]. In Fig. 5a, the glow curves for the LiMgPO₄:Gd phosphors were recorded under irradiation conditions of 0.1–10 Gy; the phosphors were initially maintained at room temperature and subsequently heated to $450^{\circ}$C at a constant rate of $5^{\circ}$C s⁻¹. For clarity, Fig. 5a shows the TL signal between $50^{\circ}$C and $250^{\circ}$C because of the minimum TL signal between $250^{\circ}$C and $450^{\circ}$C. The peak temperature of LiMgPO₄:Gd was found to be consistently positioned at $131.4 \pm 1.1^{\circ}$C for varying irradiation doses. The peak shift method \cite{5} indicates that the peak of the TL glow curve in the LiMgPO₄:Gd phosphor represents a first-order phenomenon because the temperature at which the peak occurs remains constant. The dose responses of the LiMgPO₄:Gd phosphors are shown in Fig. 5b. Over the investigated dose range of 0.1–10 Gy, the LiMgPO₄:Gd phosphors exhibited an excellent linear response (Fig. 5b).

2. TL Kinetics Parameters

Figure 5c shows five glow curves for the LiMgPO₄:Gd phosphors evaluated and recorded at ramping rates of 0.2 to $5^{\circ}$C s⁻¹. The samples were irradiated with 0.1 Gy and subjected to an increase in temperature from room temperature to $450^{\circ}$C. For clarity, the TL signal between $50^{\circ}$C and $250^{\circ}$C is shown in Fig. 5c because of the minimum TL signal between $250^{\circ}$C and $450^{\circ}$C. As the heating rate increased from $0.5^{\circ}$C s⁻¹ to $5^{\circ}$C s⁻¹, the peak temperature gradually increased and the peak and total integral TL signals decreased.

The variable heating rate (VHR) method \cite{5} was used to obtain the kinetic parameters of the peaks in the glow curve. This method requires only peak temperature information and can be applied to peaks with well-defined peak temperatures and first-order kinetics. The kinetic parameters of LiMgPO₄:Gd can be obtained using the VHR method because the peak of the TL glow curve in this sample appears to be a first-order phenomenon.

The theory of TL kinetics proposes that in first-order kinetics, the following relationship exists between the peak temperature ($T_m$) and heating rate ($\beta$) \cite{54}:

$$\ln\left(\frac{T_m^2}{\beta}\right) = \frac{E}{kT_m} + \ln\left(\frac{E}{ks}\right)$$

where $E$ is the activation energy, $k$ is the Boltzmann constant, and $s$ is the pre-exponential factor. When plotting the natural logarithm of $T_m^2/\beta$ against $1/(T_m)$, a line with a slope of $E/k$ and an intercept of $\ln(E/ks)$ is obtained, enabling calculation of the $E$ and $s$ parameters. The representation of $\ln(T_m^2/\beta)$ versus $1/(T_m)$ for the LiMgPO₄:Gd samples is shown in Fig. 5d. The $E$ and $s$ values calculated using Equation (1) and Fig. 5d were 1.44 eV and $4.82 \times 10^{17}$ s⁻¹, respectively.

3. TL after OSL

Figure 5e shows a comparison between the measured TL glow curve immediately following radiation exposure and the TL glow curve following the OSL readout. The peak height of the TL glow curve of the LiMgPO₄:Gd sample measured immediately was approximately 14 times greater than that of the TL glow curve measured after OSL. The total area of the TL peak after the OSL measurements was 8.5% of the area recorded immediately after irradiation over a temperature range of $50-250^{\circ}$C. It can be observed that the OSL signal present in LiMgPO₄:Gd is derived from the principal TL glow peak, and the majority of traps that lead to this signal can be eliminated using OSL readings.

4. OSL Decay Curve

The decay curves per unit mass for LiMgPO₄:Gd and TLD-500k are compared in Fig. 5f. As the measurement time increased, LiMgPO₄:Gd began to decrease faster, then TLD-500k decreased faster. Therefore, the sensitivity of LiMgPO₄:Gd relative to TLD-500k decreased and then increased as the measurement time increased. During this investigation, the elimination of approximately 2.2% and 2.9% of the OSL signals was observed for LiMgPO₄:Gd and TLD-500k, respectively, following each readout at a readout time of 0.1 s.

5. OSL Long-Term Fading

Table 1 [TABLE:1] presents the relative residual OSL signals of the LiMgPO₄:Gd phosphors as functions of time after irradiation and integration time. For a maximum integration time of 5 s, the OSL signal was stable, with no fading 30 days after radiation exposure. The data points represent the mean values of the OSL signals obtained from five aliquots.

6. Reusability

Table 2 [TABLE:2] presents the repeatability of the normalized OSL signals for the LiMgPO₄:Gd phosphors with different integration times. The coefficient of variation for the repeated cycles of the LiMgPO₄:Gd samples at different integration times was less than 0.8%, demonstrating the excellent reproducibility of the powders.

Discussion

A. Sensitivity

The sensitivity of OSL materials to radiation is one of their most important properties. Low doses must be measured in personal, workplace, and environmental dosimetry, requiring OSL materials with high sensitivity. The sensitivity and MDD of integrated passive detectors for personal, workplace, environmental photon, and beta radiation monitoring are not specified in the IEC 62387 standard. Instead, they only correspond to the dose linearity floor of the entire dosimetry system, which is mandatory at 0.1 mSv. However, to assess the sensitivity of a novel material, it is common practice to compare the response signal of such a material with that of a well-known material used commercially in dosimetry systems \cite{1}, measured under identical conditions.

The two principal advantages of OSL dosimetry over TL dosimetry are the speed of readout and the capacity for multiple readouts. The former enhances the processing capacity of many dosimeters, whereas the latter safeguards against data loss and ensures the preservation of evidence. To satisfy the demand for rapid and multiple readouts, it is necessary to read only a minimal quantity of the OSL signal at any given time. To provide multiple readings, the InLight OSL Reader removes a minimal amount of the OSL signal per reading, with a weak stimulus resulting in a 0.07% reduction and a strong stimulus resulting in a 0.25% reduction \cite{14}.

In this study, LiMgPO₄:Gd eliminated approximately 2.2% of the OSL signal at each readout with a readout time of 0.1 s. This method can also be used for multiple readouts. For LiMgPO₄:Gd in this study, to eliminate the OSL signal by 0.07% per readout, the stimulus had to be reduced. Relative to TLD-500k, the initial OSL signal (i.e., the sensitivity of the LiMgPO₄:Gd phosphor annealed for 1 h at $350^{\circ}$C with a readout time of 0.1 s) was approximately 0.98, which is sufficient for low-dose measurements in personal, workplace, and environmental dosimetry.

B. Fading

The change in the OSL response of an irradiated material over time is referred to as OSL fading. A variety of new OSL materials have been investigated; however, most exhibit severe fading and are not routinely used for dosimetry. OSL signal fading is more difficult to solve than TL signal fading because OSL signals originate from different traps with different fading characteristics. OSL signal fading cannot be solved by splitting the spectrum into different peaks and removing unstable peaks as in TL signal fading. The problem of OSL signal fading can also be addressed by preheating, similar to TL signal fading; however, this approach appears to contradict the premise that the OSL method is suitable for timely processing of a large number of dosimeters. Development of an OSL material that stabilizes the signal over a long period after irradiation is the best approach to solve the problem of OSL signal fading.

The OSL signal fading can be significantly improved by modifying the preparation process. In the case of MgB₄O₇:Ce,Li phosphor, the observed fading of OSL signals following two preparation processes differed significantly. One study reported that the OSL signals of MgB₄O₇:Ce,Li phosphors exhibited fading of approximately 10–15% over the course of 72 h post-irradiation \cite{55}, while another observed that the OSL signals of MgB₄O₇:Ce,Li phosphors synthesized through distinct routes exhibited fading of less than 1% over a period of 40 d \cite{56}. In the case of LiMgPO₄:Tb,B, the fading of powder samples was reported to be approximately 20–25% after a period of three weeks following irradiation, although the crystalline samples exhibited a greater degree of fading, exceeding 50% within 24 h following irradiation \cite{42}. For LiMgPO₄:Er, the loss of OSL signal intensity after irradiation for 24 h was approximately 40–85% for samples heated to $1000^{\circ}$C for 2 h, whereas the loss of signal after irradiation for the same duration was below 5% for samples heated to $900^{\circ}$C for 10 h when the integration time was above 0.5 s \cite{41}.

For LiMgPO₄:Gd in this study, the relative residual OSL signals measured after 24 h of irradiation ranged from 0.701 to 0.880 for the sample heated for 2 h at $1000^{\circ}$C. In contrast, the relative residual OSL signals measured after 24 h of irradiation for the other four samples were stable at different integration times. The dependence of fading on the preparation process was less pronounced for LiMgPO₄:Gd than for LiMgPO₄:Er.

Fewer studies have addressed the relationship between OSL fading and integration time \cite{41}. Signal fading in the LiMgPO₄:Er samples was significantly dependent on integration time. The signal loss in LiMgPO₄:Er over 30 d of radiation exposure was reported to be less than 5% when the integration time was greater than 20 s, whereas the signal loss on day 30 was approximately 12% when the integration time was 0.1 s \cite{41}. No fading of the OSL signal of LiMgPO₄:Gd in this study was observed within 30 days of irradiation when integration times were less than five seconds; however, on the 30th day, a total OSL signal fading of approximately 18% was observed. To prevent fading, the integration time (i.e., readout time) of LiMgPO₄:Er must be at least 20 s \cite{41}; however, a readout time of 20 s is considerably longer than the typical requirement for a fast OSL readout and is not optimal for multiple readouts. The integration time (i.e., readout time) of LiMgPO₄:Gd in this study was as low as 0.1 s, which is sufficient for fast and multiple OSL readouts.

C. Superiority of LiMgPO₄:Gd

The properties and economics of α-Al₂O₃:C (TLD-500K) and LiMgPO₄:Gd are compared in Table 3 [TABLE:3]. Based on the results of this study (Figs. 3b, 4a, and 5f), LiMgPO₄:Gd presents significant advantages over the commercial dosimeter material α-Al₂O₃:C (TLD-500K). With comparable OSL sensitivity, effective atomic number (Table 3), and fading characteristics to satisfy fast and multiple measurements, LiMgPO₄:Gd offers the additional advantages of easy preparation and low cost.

LiMgPO₄:Gd is significantly easier to prepare than α-Al₂O₃:C. One disadvantage of α-Al₂O₃:C is its high melting point ($2050^{\circ}$C), which presents a challenge for its synthesis, and the harsh growth conditions caused by carbon doping, which can limit its optimization scope. The optimal conditions for the preparation of LiMgPO₄:Gd are heating to $900^{\circ}$C for 15 h.

The growth of carbon-doped alumina crystals is susceptible to fluctuations in growth conditions, resulting in considerable variations in the properties of different parts of the same crystal, as well as between different crystals. To improve the homogeneity of OSL dosimeters, commercially available Luxel+ and InLight OSL dosimeters were produced by blending α-Al₂O₃:C powders with polyester and depositing them on a transparent polyester film. This results in an initial dose situated within the linear range of the dose-response curve, exhibiting a considerably inferior response to that of the material itself, partly due to the fact that the OSL sensitivity of α-Al₂O₃:C powders is much lower than that of α-Al₂O₃:C crystals. The use of LiMgPO₄:Gd for the preparation of dosimeters, such as those of InLight OSL dosimeters, may confer certain advantages in terms of sensitivity. LiMgPO₄:Gd exhibits properties similar to those of Al₂O₃:C: it is easy to synthesize, readily available, inexpensive, and suitable for large-scale production.

Conclusions

In this study, five samples of LiMgPO₄:Gd phosphor powder were prepared via five different production processes using a solid-state reaction method. X-ray diffraction was used to determine the structures of the five prepared samples. The effects of the preparation process on the OSL decay curves, TL glow curves, OSL sensitivity, MDD, and fading characteristics of the LiMgPO₄:Gd phosphors after 24 h of irradiation were investigated. The highest OSL sensitivity was obtained for the LiMgPO₄:Gd phosphor heated to $900^{\circ}$C for 15 h. For the sample heated to $1000^{\circ}$C for 2 h, the relative residual OSL signals measured after 24 h of irradiation were approximately 0.701–0.88, while the relative residual OSL signals measured after 24 h of irradiation for the other four samples were stable across different integration times. The dependence of fading on the preparation process was found to be less pronounced for LiMgPO₄:Gd than for LiMgPO₄:Er. Considering its high sensitivity, low fading, and MDD, the LiMgPO₄:Gd phosphor heated to $900^{\circ}$C for 15 h is optimal.

The effect of annealing on OSL sensitivity, relative residual OSL signals measured after 24 h of irradiation, and MDD of LiMgPO₄:Gd phosphors heated to $900^{\circ}$C for 15 h were also investigated. The OSL sensitivity of LiMgPO₄:Gd was enhanced by annealing, with the phosphor annealed at $350^{\circ}$C for 1 h exhibiting the highest OSL sensitivity. For the annealed samples, the OSL signal can be considered stable when the integration time is less than 10 s. The OSL MDD of LiMgPO₄:Gd is significantly reduced by annealing. Considering its high sensitivity, low fading, and MDD, annealing at $350^{\circ}$C for 1 h is optimal.

The basic properties of LiMgPO₄:Gd annealed at $350^{\circ}$C for 1 h were investigated. Over the investigated dose range from 0.1 to 10 Gy, LiMgPO₄:Gd phosphors exhibited an excellent linear response. The peak in the TL glow curve of the LiMgPO₄:Gd sample represents a first-order phenomenon. $E$ and $s$ were 1.44 eV and $4.82 \times 10^{17}$ s⁻¹, respectively. The OSL signal of LiMgPO₄:Gd was derived from the principal TL glow peak. At a maximum integration time of 5 s, the OSL signal was stable, with no fading 30 d after radiation exposure. LiMgPO₄:Gd eliminated approximately 2.2% of the OSL signal at each readout at a readout time of 0.1 s, which is sufficient for fast and multiple OSL readout. The sensitivity of the LiMgPO₄:Gd phosphor annealed for 1 h at $350^{\circ}$C with a reading time of 0.1 s was found to be approximately 98% of that observed for TLD-500k, which is sufficient for low-dose measurements in personal, workplace, and environmental dosimetry.

References

[1] E.G. Yukihara, A.J.J. Bos, P. Bilski, et al., The quest for new optically stimulated luminescence materials: Needs, strategies and pitfalls. Radiat. Meas. 158, 106846 (2022). https://doi.org/10.1016/j.radmeas.2022.106846

[2] R. Majgier, G. Okada, Optically stimulated luminescence characteristics and dosimetric properties of copper-doped potassium sulphate. J. Lumin. 257, 119702 (2023). https://doi.org/10.1016/j.jlumin.2023.119702

[3] J. Xu, Z. Chen, M. Gai, et al., Fabrication and OSL properties of Eu³⁺-doped NaCaPO₄ glass-ceramics. Mater. Lett. 261, 126973 (2020). https://doi.org/10.1016/j.matlet.2019.126973

[4] J. Zhang, Y. Xiang, C. Wang, et al., Recent advances in optical fiber enabled radiation sensors. Sensors 22, 1126 (2022). https://doi.org/10.3390/s22031126

[5] W.S. McKeever, A Course in Luminescence Measurements and Analyses for Radiation Dosimetry (John Wiley & Sons Ltd, West Sussex, 2022), pp. 1–381.

[6] H.H. Xiao, L.L. Liu, W.Y. Li, et al., TLD calibration and absorbed dose measurement in a radiation-induced liver injury model under a linear accelerator. Nucl. Sci. Tech. 34, 53 (2023). https://doi.org/10.1007/s41365-023-01211-5

[7] Q. Yang, H. Qin, L. Chen, et al., Metal-organic framework based thermoluminescence dosimetry. ACS Mater. Lett. (2023). https://doi.org/10.1021/acsmaterialslett.3c00326

[8] G. Zhang, J. Shen, P. Bao, et al., Assessment of occupational exposure of radiation workers at a tertiary hospital in Anhui Province, China, during 2013–18. Radiat. Prot. Dosim. 190, 237–242 (2020). https://doi.org/10.1093/rpd/ncaa098

[9] K. Tang, H. Cui, H. Zhu, et al., Study of a new LiF:Mg,Cu,P formulation with enhanced thermal stability and a lower residual TL signal. Radiat. Meas. 42, 24–28 (2007). https://doi.org/10.1016/j.radmeas.2006.07.001

[10] K. Tang, H. Cui, H. Zhu, et al., Influence of sintering temperatures on LiF:Mg,Cu,P with various magnesium concentrations. Radiat. Meas. 69, 7–11 (2014). https://doi.org/10.1016/j.radmeas.2014.07.016

[11] S. Zhang, K. Tang, H. Fan, et al., A competitive radioluminescence material for real-time dosimetry. Radiat. Meas. 151, 106719 (2022). https://doi.org/10.1016/j.radmeas.2022.106719

[12] K. Tang, S. Zhang, Real-time dosimeter based on LiF:Mg,Cu,P and SiPM. Radiat. Meas. 145, 106607 (2021). https://doi.org/10.1016/j.radmeas.2021.106607

[13] X.Y. Zhang, L. Chen, P. Luo, et al., TL response of LiF:Mg,Cu,P (GR200A and GR207A) exposed to high-energy ¹²C ions. Radiat. Meas. 78, 23–27 (2015). https://doi.org/10.1016/j.radmeas.2014.10.004

[14] E.G. Yukihara, S.W.S. McKeever, Optically Stimulated Luminescence: Fundamentals and Applications (Elsevier, New York, 2011), pp. 1–362.

[15] E.G. Yukihara, TL and OSL as research tools in luminescence: Possibilities and limitations. Ceram. Int. 49, 24356–24369 (2023). https://doi.org/10.1016/j.ceramint.2022.10.199

[16] C. Termsuk, L. Mitrayon, P. Rindhatayathon, et al., Results from 2023 interlaboratory comparison in Southeast and East Asia on OSL personal dosimeter performance in photon fields. Radiat. Meas. 175, 107181 (2024). https://doi.org/10.1016/j.radmeas.2024.107181

[17] X.B. Yang, H.J. Li, Q.Y. Bi, et al., Influence of carbon on the thermoluminescence and optically stimulated luminescence of α-Al₂O₃:C crystals. J. Appl. Phys. 104, 123112 (2008). https://doi.org/10.1063/1.3050344

[18] T. Sun, K. Tang, H. Cui, et al., Optically stimulated luminescence of α-Al₂O₃:C by the vertical gradient freezing (VGF) method. J. Lumin. 205, 568–571 (2019). https://doi.org/10.1016/j.jlumin.2018.10.014

[19] T. Sun, K. Tang, H. Cui, et al., Thermoluminescence of newly developed highly sensitive α-Al₂O₃:C by the vertical gradient freezing method. Radiat. Prot. Dosim. 184, 174–178 (2019). https://doi.org/10.1093/rpd/ncy196

[20] X. Yang, H. Li, Q. Bi, et al., Growth of α-Al₂O₃:C crystal with highly optically stimulated luminescence. J. Lumin. (2009). https://doi.org/10.1016/j.jlumin.2008.12.015

[21] X. Yang, J. Xu, H. Li, et al., Influence of pulling rates on the thermoluminescence and optically stimulated luminescence of α-Al₂O₃:C crystals grown by the edge-defined, film-fed growth technique. J. Am. Ceram. Soc. 92(10), 2265–2269 (2009). https://doi.org/10.1111/j.1551-2916.2009.03198.x

[22] C.C.P. Fontainha, N. Alves, W.B. Ferraz, et al., Very sensitive α-Al₂O₃:C polycrystals for thermoluminescent dosimetry. Appl. Radiat. Isot. 141, 234–240 (2018). https://doi.org/10.1016/j.apradiso.2018.05.004

[23] B. Broadhead, C. Noble, P. Ramachandran, A direct comparison of the optically stimulated luminescent properties of BeO and Al₂O₃ for clinical in-vivo dosimetry. Phys. Eng. Sci. Med. 45, 859–866 (2022). https://doi.org/10.1007/s13246-022-01155-x

[24] W. Abusaid, V. Altunal, Y. Akdeniz, et al., Studies of OSL properties of alkali- and rare earth-doped BeO based novel dosimeters for applications in external beam radiotherapy. Radiat. Phys. Chem. 212, 111136 (2023). https://doi.org/10.1016/j.radphyschem.2023.111136

[25] B. Marczewska, A. Sas-Bieniarz, OSL and RL of LiMgPO₄ doped with rare earth elements. Radiat. Meas. (2019). https://doi.org/10.1016/j.radmeas.2019.106205

[26] B. Dhabekar, S.N. Menon, E. Alagu Raja, LiMgPO₄:Tb,B - A new sensitive OSL phosphor for dosimetry. Nucl. Instrum. Meth. B 269, 1844–1848 (2011). https://doi.org/10.1016/j.nimb.2011.05.001

[27] C. Dotzler, G.V.M. Williams, U. Rieser, et al., Optically stimulated luminescence in NaMgF₃:Eu²⁺. Appl. Phys. Lett. 91(12), 121910 (2007). https://doi.org/10.1063/1.2786599

[28] J. Xu, Z. Chen, M. Gai, et al., Optically stimulated luminescence of Dy³⁺-doped NaCaPO₄ glass-ceramics. J. Rare Earth 38, 927–932 (2020). https://doi.org/10.1016/j.jre.2019.05.003

[29] Y.H. Wang, H. Chen, F. Chen, et al., Radiation dose detection using a high-power portable optically stimulated luminescence real-time reading system. Nucl. Sci. Tech. 29, 149 (2018). https://doi.org/10.1007/s41365-018-0484-z

[30] R. Hemam, L.R. Singh, A.I. Prasad, et al., Critical view on TL/OSL properties of Li₂B₄O₇ nanoparticles doped with Cu, Ag and co-doping Cu, Ag: Dose response study. Radiat. Meas. 95, 44–54 (2016). https://doi.org/10.1016/j.radmeas.2016.11.003

[31] P.B.R. Gasparian, A.L.M.C. Malthez, E.M. Yoshimura, et al., Study of the dosimetric properties of CaSO₄:Dy using OSL technique. 17, P04022 (2022). https://doi.org/10.1088/1748-0221/17/04/P04022

[32] V. Guckan, V. Altunal, N. Nur, et al., Studying CaSO₄:Eu as an OSL phosphor. Nucl. Instrum. Meth. B 407, 145–154 (2017). https://doi.org/10.1016/j.nimb.2017.06.010

[33] J.C. Mittani, M. Prokić, E.G. Yukihara, Optically stimulated luminescence and thermoluminescence of terbium-activated silicates and aluminates. Radiat. Meas. 43, 323–326 (2008). https://doi.org/10.1016/j.radmeas.2007.10.004

[34] A. Jain, A. Tripathi, S. Aggarwal, Thermoluminescence studies of NaMgF₃:Tb dosimetry applications. Nucl. Instrum. Meth. B 543, 165096 (2023). https://doi.org/10.1016/j.nimb.2023.165096

[35] V. Guckan, S.W. Bokhari, V. Altunal, et al., Thermoluminescence and optically stimulated luminescence properties of NaMgF₃:Dy,Eu synthesized by hydrothermal method and DFT calculations for the bandgap. Mater. Res. Bull 167, 112373 (2023). https://doi.org/10.1016/j.materresbull.2023.112373

[36] L.C. Oliveira, E. Yukihara, O. Baffa, MgO:Li,Ce,Sm as a high-sensitivity material for Optically Stimulated Luminescence dosimetry. Sci. Rep-uk. 6, 24348 (2016). https://doi.org/10.1038/srep24348

[37] J. Guo, Q. Tang, C. Zhang, et al., Optically stimulated luminescence (OSL) of LiMgPO₄:Tm,Tb phosphor. J. Rare Earth 35, 525–529 (2017). https://doi.org/10.1016/S1002-0721(17)60943-8

[38] D.G. Kellerman, M.O. Kalinkin, R.M. Abashev, et al., Unusual intrinsic thermoluminescence in LiMgPO₄:Er. Chem. Phys. (2020). https://doi.org/10.1039/d0cp05185c

[39] D.G. Kellerman, M.O. Kalinkin, D.A. Akulov, et al., On the energy transfer in LiMgPO₄ doped with rare-earth elements. J. Mater. Chem. C. 9, 11272–11283 (2021). https://doi.org/10.1039/d1tc02211c

[40] M.Q. Gai, Z.Y. Chen, Y.W. Fan, et al., Synthesis of LiMgPO₄:Eu,Sm,B phosphors and investigation of their optically stimulated luminescence properties. Radiat. Meas. 78, 48–52 (2015). https://doi.org/10.1016/j.radmeas.2014.09.009

[41] L. Fu, K. Tang, S. Zhang, et al., Short-term fading studies of LiMgPO₄:Er optically stimulated luminescence. Radiat. Phys. Chem. 218, 111636 (2024). https://doi.org/10.1016/j.radphyschem.2024.111636

[42] D. Kulig, W. Gieszczyk, B. Marczewska, et al., Comparative studies on OSL properties of LiMgPO₄:Tb,B powders and crystals. Radiat. Meas. 106, 94–99 (2017). https://doi.org/10.1016/j.radmeas.2017.04.004

[43] Z. Yin, H. Chen, G. Feng, et al., Study on the thermoluminescence and optically stimulated luminescence of LiMgPO₄:Dy phosphors synthesized by different methods. Appl. Radiat. Isot. 201, 110990 (2023). https://doi.org/10.1016/j.apradiso.2023.110990

[44] M. Gai, Z. Chen, Y. Fan, et al., Synthesis and luminescence in LiMgPO₄:Tb,Sm,B phosphors with possible applications in real-time dosimetry. J. Rare Earth 31, 551–554 (2013). https://doi.org/10.1016/S1002-0721(12)60318-4

[45] L. Fu, K. Tang, H. Cui, et al., The effect of boron co-doping on thermoluminescence and optically stimulated luminescence properties of LiMgPO₄:Tb,Sm. Radiat. Meas. 167, 106986 (2023). https://doi.org/10.1016/j.radmeas.2023.106986

[46] H. Tang, L. Lin, C. Zhang, et al., High-Sensitivity and Wide-Linear-Range Thermoluminescence Dosimetry for High-Dose Radiation. LiMgPO₄:Tm,Tb,B Inorg. Chem. (2019). https://doi.org/10.1021/acs.inorgchem.9b00597

[47] L. Fu, H. Cui, H.J. Fan, et al., The dependence of thermoluminescence and optically stimulated luminescence properties of LiMgPO₄:Tb,Sm,B on maximum heating temperatures. Radiat. Eff. Defect S. 179, 2294053 (2023). https://doi.org/10.1080/10420150.2023.2294053

[48] J. Guo, C. Jian, C. Zeng, et al., Dosimetric and spectroscopic study of LiMgPO₄ doped with Tm³⁺ and Er³⁺. RSC Adv. 13, 4949 (2023). https://doi.org/10.1039/d2ra07109f

[49] X. Kong, Z. Fu, H. Que, et al., Effects of recording time and residue on dose-response by LiMgPO₄:Tb,B synthesized via improved sintering process. Nucl. Instrum. Meth. B 422, (2018). https://doi.org/10.1016/j.nimb.2018.02.032

[50] J. Guo, L. Zhao, Q. Tang, et al., Spectral study on energy transfer in the LiMgPO₄ phosphor doped with Tm³⁺ and Tb³⁺. J. Lumin. 228, 117613 (2020). https://doi.org/10.1016/j.jlumin.2020.117613

[51] A.L.M.C. Malthez, B. Marczewska, D. Kulig, et al., Optical and thermal pre-readout treatments to reduce the influence of fading on LiMgPO₄ OSL measurements. Appl. Radiat. Isot. 136, 118–120 (2018). https://doi.org/10.1016/j.apradiso.2018.02.022

[52] L. Fu, K. Tang, S. Zhang, et al., Fading performance of LiMgPO₄:Tb,Sm,B optically stimulated luminescence. Radiat. Meas. 175, 107165 (2024). https://doi.org/10.1016/j.radmeas.2024.107165

[53] S.N. Menon, B.S. Dhabekar, S. Kadam, et al., Fading studies in LiMgPO₄:Tb,B and synthesis of new LiMgPO₄ based phosphor with better fading characteristics. Nucl. Instrum. Meth. B 436, 45–50 (2018).