Sample preparation

Y_2-xSm_xMgTiO₆ (x=0, 0.001, 0.002, 0.005, 0.01, 0.02, 0.05, 0.1 and 0.2) phosphors were synthesized using a high-temperature solid-state method. Stoichiometric amounts of Y₂O₃ (99.99%), MgO (99.99%),TiO₂ (99.99%), Sm₂O₃ (99.99%) were weighed and placed in an agate mortar for 0.5 h until a homogenous mixture was achieved. This uniformly ground powder was placed in a corundum crucible in a muffle furnace and pre-sintered in the air at 800 ℃ for 3 h. Subsequently, the temperature was rapidly increased to 1300 ℃ at a faster heating rate (7 ℃ s^-1), and the samples were calcined at 1300 ℃ for 9 h. The resulting block-shaped sintered products were then crushed and ground again using an agate mortar to obtain the final phosphor powder.

Testing method

X-ray diffraction (XRD) patterns were recorded using a Rigaku Ultima IV X-ray diffractometer with a Cu-K_α radiation source, a scanning range of 10° to 80°, and a scanning rate of 5°/min. Photoluminescence (PL) spectra were measured using a HITACHI F-7000 fluorescence spectrometer with a Xe lamp excitation source and a spectral resolution of 0.2 nm. Thermoluminescence (TL) measurements were performed using a Risø TL/OSL-15-B/C thermoluminescence/optically stimulated luminescence measurement instrument equipped with a ⁹⁰Srβ radiation source (1,4 GBq activity, 0.1 Gy/s dose rate). The distance between ⁹⁰Srβ radiation source and the sample was 5 mm, the distance between the detector and the sample was 55 mm, and the heating rate was 5 K/s during measurement. The three-dimensional thermoluminescence spectroscopy of the samples was measured using an LTTL3DS thermoluminescence spectrometer (Guangzhou Ruidi Technology Co., Ltd.). The irradiation source was an X-ray tube, the working voltage of the X-ray tube was 50 kV, the current was 150 μA, the dose rate was about 0.1 Gy/s, the heating rate was 5 K/s during measurement, the heating range was 300–750 K, the spectral range was 300–1000 nm, and the spectral resolution was 1 nm.

XRD analysis

The XRD patterns of Y_2-xSm_xMgTiO₆ (x=0, 0.005, 0.01, 0.02, 0.05 and 0.1) series samples are shown in Fig. 1. No information on Y₂MgTiO₆ is available in the inorganic crystal information database. Shannon [35] used the Rietveld method to analyze data, and the results proved that Dy₂MgTiO₆ and Y₂MgTiO₆ have very similar structures. Therefore, the standard Dy₂MgTiO₆ (ICDD 04-021-1637) card was used as a reference. As shown in the figure, the number and positions of the diffraction peaks for all samples are consistent with the standard card. The diffraction angle at 33°, corresponding to the characteristic site of Y³⁺ shifts to a small angle [36, 37]. According to Bragg's equation [38, 39], this decrease in diffraction angle indicates an increase in lattice spacing. Given the ionic radii of rY³⁺=0.1019 nm, rMg²⁺ =0.0720 nm, rTi⁴⁺=0.0605 nm [40], and the ionic radius of Sm³⁺ is rSm³⁺=0.1132 nm. The shift in the diffraction angle of the characteristic site of Y³⁺ to a small angle confirms the successful substitution of Y³⁺ by the larger Sm³⁺ ions within the lattice. This substitution does not alter the overall lattice structure or charge configuration, and the samples retain the monoclinic P2₁/n [37].

Fig. 1Full-screen View

XRD Spectra of Y_2-xSm_xMgTiO₆ (x=0, 0.005, 0.01, 0.02, 0.05 and 0.1) Samples

Figure 2 shows the morphology and lattice fringes of Y₂MgTiO₆ matrix and Y₂MgTiO₆:Sm. The figure reveals that the phosphor particles have irregular morphologies, with sizes ranging from approximately 400 nm to 2 μm. After Sm doping, the lattice fringes become denser, indicating decreased crystal plane spacing. This observation further supports the successful incorporation of the larger Sm ions into the matrix lattice. Combined with the XRD results, this confirms the successful synthesis of the phosphor under the described experimental conditions.

Fig. 2Full-screen View

TEM of Y_2-xSm_xMgTiO₆

Photoluminescence analysis

Figure 3 shows the PL of Y_2-xSm_xMgTiO₆ (x=0.001, 0.002, 0.005, 0.01, 0.02, 0.05 and 0.1) series phosphors. With a monitoring wavelength of 605 nm, the excitation spectrum spanned from 340 to 500 nm. The strongest absorption peak is at 409 nm, which belongs to the characteristic transition of Sm^{3+ 6}H_5/2→6P_3/2. The absorption peaks at 349 nm, 366 nm, 379 nm, 422 nm, 443 nm, and 474 nm correspond to the characteristic transitions of Sm^{3+ 6}H_5/2→4H_9/2, ⁴D_3/2, ⁶P_7/2, ⁶P_5/2, ⁴F_5/2, ⁴I_13/2, respectively[41]. When excited by 409 nm purple light, four prominent emission peaks are observed at 568 nm, 605 nm, 652 nm and 715 nm, corresponding to the characteristic transitions of Sm^{3+ 4}G_5/2→⁶H_5/2, ⁶H_7/2, ⁶H_9/2, ⁶H_11/2, respectively[42, 43]. With increasing Sm³⁺ doping concentration, the positions and shapes of the emission peaks remained unchanged, while the intensity initially increased and subsequently decreased. The optimal doping concentration was found to be x=0.05, where the photoluminescence emission intensity reached a maximum. Further increases in the Sm³⁺ doping concentration significantly decreased emission intensity, indicating concentration quenching.

Fig. 3Full-screen View

(Color online) Photoluminescence spectra of Y_2-xSm_xMgTiO₆ (x=0.001, 0.002, 0.005, 0.01, 0.02, 0.05 and 0.1) samples

Doping concentration optimization

The thermoluminescence response was optimized by varying the doping concentration of Sm³⁺ to determine the phosphor with the best TL yield. Samples with different doping concentrations (30 mg each) underwent TL testing using the following procedure: (1) preheating to 773 K for 10 s; (2) cooling to room temperature; (3) irradiation with a ⁹⁰Srβ radiation source at a dose of 100 Gy; 4) TL measurement at a heating rate of 5 K/s. The measurement results are presented in Fig. 4. As can be seen from the figure, with an increase in the Sm³⁺ doping concentration, the integrated TL intensity also increased, reaching a maximum at x=0.02, which was then selected for subsequent experiments. Further increases inSm³⁺ concentration led to concentration quenching and decreased integrated thermoluminescence intensity. It can be seen from the figure that the Y₂MgTiO₆ matrix has a weak thermoluminescence peak, and the thermoluminescence range is between 460 K and 630 K. After introducing Sm³⁺, the thermoluminescence peak of the phosphor extended to the high-temperature zone. There was evident thermoluminescence at 675 K. In addition, the shape of the thermoluminescence curve changes with the Sm³⁺ doping concentration. As shown in the inset of Fig. 3, when the Sm³⁺ doping concentration was x<0.02, the thermoluminescence peak around 610 K was more intense than the peak at approximately 530 K. Conversely, for Sm³⁺ doping concentrations x>0.02, the thermoluminescence peak at approximately 530 K exhibited a higher intensity than the peak at approximately 610 K. At a Sm³⁺ doping concentration of x=0.02, a balanced state is observed between the intensities of the two prominent TL peaks. However, when concentration quenching occurs at higher doping levels, the peak intensity at approximately 610 K decreases more rapidly than at approximately 530 K.

Fig. 4Full-screen View

(Color online) The thermoluminescence spectra of Y_2-xSm_xMgTiO₆ (x=0, 0.001, 0.002, 0.005, 0.01, 0.02, 0.05 and 0.1) samples by irradiate with ⁹⁰Srβ radiation source for 100 Gy. The variation trend of thermoluminescence intensity with Sm3+ doping concentration at 530 K and 610 K is illustrated

To elucidate the underlying mechanisms responsible for the observed thermoluminescence behavior of Y_2-xSm_xMgTiO₆ series phosphors, three-dimensional thermoluminescence (3D-TL) spectra were acquired for four representative samples irradiated with 100 Gy of X-rays, as shown in Fig. 5. Figure 5a shows that the thermoluminescence range of Y₂MgTiO₆ matrix is between 480 K and 640 K, with peak temperatures at approximately 530 K and 600 K, consistent with the TL glow curve shown in Fig. 4. The characteristic emission band for Y₂MgTiO₆ matrix is centered at 698 nm. Figure 5b presents the three-dimensional thermoluminescence spectrum of the sample doped with 0.01 Sm³⁺. The introduction of Sm³⁺ enhances the thermoluminescence intensity of Y₂MgTiO₆ matrix without altering the peak temperature or emission band position of the matrix thermoluminescence. A new thermoluminescence peak appears around 610 K, with emission bands around 570 nm, 600 nm, and 650 nm, consistent with the characteristic, the emission band is around 570 nm, 600 nm, 650 nm, which is consistent with the characteristic Sm³⁺ emissions obtained in PL. This confirms that this TL peak originates from Sm³⁺. The Sm³⁺ -related TL intensity is significantly higher than that of the matrix, meaning that the peak intensity around 610 K is larger than that of 530 K, consistent with the curve in Fig. 4. The three-dimensional thermoluminescence spectrum for the Sm³⁺ doping concentration of x=0.02 is shown in Fig. 5c. As the Sm³⁺ doping concentration increases, the thermoluminescence intensity intensity of both Sm³⁺ and the matrix is enhanced. At this concentration x=0.02, the thermoluminescence peak intensity of Sm³⁺ (610 K) is comparable to that of the matrix (530 K), achieving a balanced state. Further increases inSm³⁺ concentration result in a significant decrease in the Sm³⁺ -related thermoluminescence intensity, while the matrix thermoluminescence intensity also decreases, but to a lesser extent. This indicates that the Sm³⁺ thermoluminescence emission is strongly affected by concentration quenching, while the matrix TL is less affected. This explains why the 530 K thermoluminescence peak becomes more intense than the 610 K peak at higher Sm³⁺ concentrations Th

Fig. 5Full-screen View

(Color online) The thermoluminescence spectra of Y₂MgTiO₆ (a),Y_1.99Sm_0.01MgTiO₆ (b),Y_1.98Sm_0.02MgTiO₆ (c) and Y_1.9Sm_0.1MgTiO₆ (d) samples by irradiate with X-Ray source for 100 Gy

To determine the energy dependence of the thermoluminescence under X-ray and β -particles irradiation, the thermoluminescence spectra of four samples under X-ray irradiation were measured using the Risø instrument and compared with the thermoluminescence spectra obtained under β -particles irradiation. The results are presented in Fig. 6. The thermoluminescence spectra under both X-ray and β -particle irradiation are broadly similar, with only minor differences observed in the low-temperature region.

Fig. 6Full-screen View

(Color online) Thermoluminescence spectra of Y_2-xSm_xMgTiO₆ phosphors irradiated by X-ray and -particles at 100 Gy

T_m-T_stop method

To confirm the number of overlapping peaks and their respective peak temperatures in the thermoluminescence (TL) spectrum of Y_1.98Sm_0.02MgTiO₆, fitting was performed using the CGCD method, and the thermal luminescence mechanism and kinetic parameters of the sample were further explored. Peak-separation experiments were performed using T_m-T_stop. The test steps for the T_m-T_stop method are as follows.

1. The samples were preheated to 773 K and held for 10 s.

2. The samples were then cooled to room temperature.

3. The samples were irradiated using a ⁹⁰Srβ radiation source at a dose of 100 Gy.

4. The samples were then heated to a sufficiently high temperature (T_stop), to clear the TL signal prior to the T_stop temperature value.

5. The samples were rapidly cooled to room temperature.

6. The samples were then reheated at the same heating rate (5 K/s), recording the remaining TL curve and noting the position of the first maximum, T_m, on the TL curve was recorded. This process was repeated with successively lower T_stop values (decreasing by approximately 3 K each time). Figure 7 shows the relationship between T_m and T_stop, revealing seven plateaus, each corresponding to a distinct TL peak. The T_m values for these overlapping peaks were determined to be 384, 419, 449, 532, 581, 610, and 640 K. Within the temperature range up to 449 K, the peak temperatures of the three lowest-temperature peaks varied with T_stop, indicating a significant trap recapture process characteristic of second-order kinetics. Above 532 K, the peak temperatures of the remaining four peaks remained relatively constant with changing T_stop, suggesting that trap recapture is negligible and consistent with first-order kinetics.

Fig. 7Full-screen View

T_m-T_stop diagram of Y_1.98Sm_0.02MgTiO₆ sample

Computerized glow curve deconvolution (CGCD)

Computerized glow curve deconvolution (CGCD) is a widely used technique for analyzing complex TL mechanisms. It allows for determining the position, shape, and kinetic parameters of overlapping peaks within a TL glow curve[44]. In this study, CGCD, in conjunction with the results obtained from three-dimensional thermoluminescence spectroscopy and the T_m-T_stop method, was used to fit the TL glow curve of the Y_1.98Sm_0.02MgTiO₆ sample irradiated with 100 Gy from a ⁹⁰Srβ radiation source. Each overlapping peak can be described by Eq. 1 [45]: $I(t)=s{n}_0\text{exp}\left(-\frac{E}{kT}\right)\cdot {\left[1+\frac{s(b-1)}{\beta }\cdot {\displaystyle {\int }_{T_0}^T}\text{exp}\left(-\frac{E}{k{T}^{\prime }}\right)d{T}^{\prime }\right]}^{-\frac{b}{b-1}}.$ (1) In Eq. 1: n₀ represents the initial number of captured electrons in the trap level; E is the activation energy of the captured electrons, in eV; s is the frequency factor, in Hz; k is the Boltzmann constant, which is 0.862×10^–4 eV/K; β is the heating rate of the sample, in K/s, which is 5 K/s in this experiment; T is the absolute temperature (in units of K); b is the kinetic order. The CGCD fitting results are shown in Fig. 8. The fitting curve shows good agreement with the experimental data points, and the fitted parameters are presented in Table 1. These kinetic parameters were used to calculate the energy-level lifetime for each overlapping peak, which is a crucial parameter for estimating the saturation dose of each peak and is of significant importance for subsequent dose-response studies. The energy-level lifetimes for different overlapping peaks can be determined using Eq. 2 [46]: $\tau =s^{-1}\cdot e^{\left(\frac{E}{kT}\right)}.$ (2) In Eq. 2: τ represents the energy level lifetime of the overlapping peak; s is the frequency factor for the overlapping peak; E denotes the trap depth of the overlapping peak; k is the Boltzmann constant; T stands for the ambient temperature (typically 300 K). After performing the calculations, the results are summarized in Table 1. The low-temperature peak at 378 K exhibits a relatively short lifetime of only 980 s, while deeper traps have longer lifetimes, as electrons trapped at shallower levels are more easily released. Figure 8 shows the results of CGCD fitting. The peak temperatures of the seven overlapping peaks are 378 K, 415 K, 477 K, 534 K, 578 K, 615 K, and 654 K, corresponding to trap depths ranging from 0.69 eV to 1.49 eV. The kinetic order (b) is 2.0 for the 378 K, 415 K, and 477 K peaks, 1.6 for the 534 K peaks, and approximately 1.0–1.2 for the 578 K, 615 K and 654 K peaks, consistent with the results obtained using the T_m-T_stop method. Furthermore, the fitted peak temperatures for the overlapping peaks at 477, 534, 578, 615, and 654 K closely match those in Fig. 5c. Specifically, the three overlapping peaks at 477 K, 534 K, and 578 K are attributed to the Y₂MgTiO₆ matrix, while the overlapping peaks at 615 K and 654 K are associated with Sm³⁺. The absence of the 378 K and 415 K peaks in Fig. 5c is likely due to experimental differences arising from the different excitation sources. These results demonstrate the scientific basis and reliability of the peak positions and kinetic parameters obtained through CGCD fitting based on investigating the system's internal mechanisms.

Fig. 8Full-screen View

(Color online) The TL curve fitting of Y_1.98Sm_0.02MgTiO₆ phosphor by irradiate with ⁹⁰Srβ radiation source for 100 Gy was carried out by CGCD method

Dose response

The TL dose response test of Y_1.98Sm_0.02MgTiO₆ phosphor was investigated using the following procedure: (1) preheating the sample to 773 K for 10 s; (2) cooling to room temperature; (3) irradiating with ⁹⁰Sr radiation source at a dose of 2 Gy; and (4) measuring the TL emission (heating rate: 5 K/s). These steps were repeated, varying the irradiation dose to 5 Gy, 10 Gy, 20 Gy, 50 Gy, 60 Gy, 70 Gy, 80 Gy, 90 Gy, 100 Gy, 110 Gy, 120 Gy, 150 Gy, 200 Gy, 300 Gy, 400 Gy, 500 Gy, 700 Gy, 800 Gy, 900 Gy, 1000 Gy, 2000 Gy, 5000 Gy, 10,000 Gy, 20,000 Gy and 25,000 Gy. The resulting TK grow curves were recorded and are shown in Fig. 9. It can be seen that as the irradiation dose increased, the shape of the TL curve of the sample exhibited apparent changes. When the irradiation dose is less than 90 Gy, the peak value around 530 K is stronger, and the irradiation dose is more than 90 Gy, the peak value around 610 K is stronger, which indicates that the characteristic thermoluminescence of Sm³⁺in Y_1.98Sm_0.02MgTiO₆ exhibits significantly higher sensitivity to radiation dose than the Y₂MgTiO₆ matrix.

Fig. 9Full-screen View

(Color online) The thermoluminescence curves of Y_1.98Sm_0.02MgTiO₆ samples by irradiate with ⁹⁰Srβ radiation source for different doses

Figure 10 shows the change in integrated TL intensity with increasing dose. Each data point in the figure represents the area under the TL glow curve. The dose response appears to reach saturation and can be fitted using Eq. 3 [47]: $I(D)=A\cdot \left[1-\text{exp}\left(-\frac{D}{D_0}\right)\right].$ (3) In Eq. 3: I represents the thermoluminescence integral intensity; A denotes the number of thermoluminescent sensitive units in the measured sample; D stands for the radiation dose; D₀ corresponds to the saturation dose. After fitting, it was found that the saturation dose of the sample was 9956 Gy. Compared with commonly used thermoluminescence dosimeters (e.g., LiF:(Mg,Cu,P),BeO,Al₂O₃:C), this material has a higher saturation dose, and the dose-response can still maintain good linearity at 1 kGy. Due to the easier release of trapped electrons at low temperatures, the decay times for the trapped electrons at 378 K and 415 K are relatively short (0.01 days and 0.3 days, respectively). However, the decay times for trapped electrons at 477 K and above are significantly longer (more than 23 days), indicating better signal stability. Therefore, this material's high saturation dose and long decay time suit high-dose radiation dosimetry applications, such as fruit irradiation preservation.

Fig. 10Full-screen View

Dose response curve of Y_1.98Sm_0.02MgTiO₆ sample

While it is possible to fit the dose response of a TL curve with multiple overlapping peaks using the total integrated TL intensity at various irradiation doses, the significant differences in energy-level lifetimes among these peaks can introduce theoretical deviations. However, these deviations may be negligible within specific dose ranges. To investigate the dose-response characteristics of each individual overlapping peak in the Y_1.98Sm_0.02MgTiO₆ phosphor system, the CGCD method was employed to fit TL curves at various irradiation doses. The results are shown in Fig. 11. In Fig. 11, it can be observed that the peak temperatures T_m of Peaks 1, 2, and 3 shift toward lower temperatures as the irradiation dose increases, consistent with second-order kinetic behavior. Peaks 4, 5, 6, and 7, on the other hand, exhibit no significant change in T_m with increasing irradiation dose, aligning with first-order kinetic behavior and the results are consistent with those of T_m-T_stop method.

Fig. 11Full-screen View

(Color online) Overlapping peak curves at different irradiation doses

With increasing irradiation dose, the thermoluminescence intensity of peak 1 plateaus beyond 200 Gy. The thermoluminescence integral intensity of each overlapping peak under different irradiation doses was determined to explore the dose-response curves of the different overlapping peaks, as shown in Fig. 11. The results are presented in Fig. 12. The other overlapping peaks, except for Peak 1, exhibit good linear dose-response relationships. Peaks 6 and 7 show similar slopes, as do Peaks 3, 4, and 5, suggesting that Peaks 6 and 7 originate from Sm3+, while Peaks 3, 4, and 5 are associated with the matrix. The slope of peak 6 is significantly larger than that of peak 4, indicating that the characteristic thermoluminescence of Sm3+ is much more sensitive to the irradiation dose than the matrix. The dose response of Peak 1 appears to reach saturation. After fitting this response using Eq. 3, the saturation dose for Peak 1 was determined to be 102.5 ± 3.2 Gy. Considering that the dose rate of the ⁹⁰Srβ radioactive source is 0.1 Gy/s, and the energy level lifetime of Peak 1 is 980 seconds, it is estimated that Peak 1 reaches saturation at an irradiation dose of approximately 98 Gy, which agrees well with the experimental dose-response measurements. To increase the saturation dose of Peak 1, a higher dose-rate radiation source could be used. For precise measurements below 1 kGy, it would be suitable to eliminate Peak 1 and integrate the TL signal above 378 K. For measurements above 1 kGy, eliminating Peaks 1 and 2 (with a saturation dose of approximately 2.6 kGy) and integrating the TL signal above 415 K would be more appropriate.

Fig. 12Full-screen View

(Color online) Dose response curves of different overlapping peak

Repeatability test

To verify the repeatability of the thermoluminescence (TL) signal, the TL curves of the Y_1.98Sm_0.02MgTiO₆ phosphor was subjected to repeated measurements under 100 Gy irradiation. Figure 13 shows the results of 12 measurements, where the total integral of the TL curve from each measurement was selected as the outcome and normalized. The experimental results indicated a relative standard error of approximately 0.06% for the measured TL signal, indicating minimal influence of radiation and temperature on the TL measurements.

Fig. 13Full-screen View

Repetitive experiment of thermoluminescence