Red-Emitting Phosphor Ca2MgSi2O7: Sm3+, Eu3+ for UV White Light-Emitting Diodes

: In this research, a series of Ca 2 MgSi 2 O 7 : 2% Sm 3+ , x % Eu 3+ (x=4, 5, 6, 7, 8) red phosphors were synthesized using the high - temperature solid - phase method. The phase and luminescence properties of the samples were investigated using an X - ray diffractometer and a photoluminescence spectrometer. The results showed that the synthesized samples were of pure phase, and the introduction of small amounts of Sm 3+ and Eu 3+ had no significant impact on the crystal structure of the phosphors. It is observed from the phosphor spectra that the co - doped samples exhibited intense red - light emission corresponding to that of Eu 3+ at 613 nm when excited at the strongest excitation peak of Sm 3+ (401 nm); here, t he emission peak intensity of Eu 3+ increased by a factor of 5.3. It is found that resonant nonradiative energy transfer occurs from Sm 3+ to Eu 3+ in the sample, and the energy transfer efficiency reaches up to 44%. The calculated critical distance for energy transfer is 16.117 Å and the concentration interrupt mechanism is an electric dipole - dipole (d - d) interaction. The color coordinates of the samples are all located in the red region (0.6319, 0.3676) with a color purity of ~ 89.3%. The samples exhibited thermal stability of up to 68.6% of that at room temperature when heated to 150˚C . The LED samples packaged with the phosphors emitted warm white - light with a color temperature (CCT) of 5553 K and a color rendering index ( R a ) of 84.8. The magnesium yellow feldspar silicate phosphor is suitable as the red component in white LED trichromatic phosphors and has potential applications in solid - state lighting.


Introduction
White LEDs are recognized as the fourth generation of energy-efficient "green" light sources and they are widely applied in solid state lighting, display and associated technologies due to features such as high luminous efficiency, long service life and energy efficiency. Critical to the success of these light sources is the development of phosphors with high color rendering index and color purity 1,2 . When compared with other materials, silicate-based phosphors have excellent chemical and thermal stability and offer a broad range of excitation wavelengths. As such, they are widely used as a substrate material in integrated white LED devices.
At present, reports on the development of silicate phosphors compatible with near-UV, UV and blue LED chips are becoming more frequent. Xu et. al. synthesized Ca 15 (PO 4 ) 2 (SiO 4 ) 6 : Sm 3+ , Eu 3+ , Dy 3+ phosphors, and these phosphors exhibited multi-color luminescence 3 . Duan et. al. prepared Ba 2 YAlO 5 : Sm 3+ , Eu 3+ red phosphor using the high-temperature solid-phase method, and they studied the crystal structure, luminescence characteristics and thermal stability of the samples. They demonstrated a red phosphor with thermal stability up to 64.3% , which has significant potential as components in white LEDs 4 . SaHu et. al. prepared Ca 2 MgSi 2 O 7 : Eu 2+ , Dy 2+ phosphor and investigated the dependence of the mechanoluminescence (ML) intensity of the phosphor and the loading proximity relationship; The results showed the potential application of the phosphor as a sensor to detect stress in objects 5 . In recent years, there have however, been relatively few reports on the preparation of synthetic Ca 2 MgSi 2 O 7 matrices doped with rare earth elements for application as trichromatic phosphors in white LEDs.
Basic earth silicates composed of A 2 BSi 2 O 7 (A=Sr, Ba, Ca) (B=Mg, Zn) are luminescent substrates with good physicochemical stability, whose anions and cations are mostly combined by strong covalent ionic bonds and have a variety of cationic lattice sites for different cation substitution. This highly-adjustable and adaptable framework structure with a rich crystal field environment results in substrates with very good luminescence performance 6 . Eu 3+ is a commonly-used activator in luminescent materials in rare-earth ionic systems because of its ability to provide narrow-band red emission in the visible region. Sm 3+ has strong absorption in the UV-Vis region, thus playing a key role in the application of white light LED devices. Both of these ions have small gaps between specific energy levels, which can lead to energy-transfer between them. In general, the energy transfer between Sm 3+ and other rare earth ions is an effective way to increase the luminescence intensity of a phosphor material 7 . In this work, we make use of this energy transfer characteristic between Sm 3+ and Eu 3+ ions to enhance the luminescence intensity in Ca 2 MgSi 2 O 7 : Eu 3+ phosphor.In this study, new Ca 2 MgSi 2 O 7 : 2% Sm 3+ , x% Eu 3+ (x=4,5,6,7,8) red phosphor luminescent materials were prepared using the hightemperature solid-phase method. The XRD diffraction pattern, crystal structure, photoluminescence properties, energy transfer mechanism, colour coordinates, thermal stability, and LED chip test of the samples were analyzed. This comprehensive theoretical and experimental study forms a basis for this material to find application in LED solid-state lighting devices.

Experimental results and discussion 2.1 Physical phase characterization
The phase purity of the prepared samples was measured by XRD. Fig. 1(a) shows the X-ray diffraction patterns of the Ca 2 MgSi 2 O 7 : 2% Sm 3+ , x% Eu 3+ (x=4, 5, 6, 7, 8) series of samples prepared using the high-temperature solid-phase method. As can be seen from the figure, the diffrac-tion peaks of the prepared sample are consistent with all the diffraction peaks of the standard card PDF#83-1815 in the International Center for Diffraction Data (ICCD) file, and no excess spurious phases are detected. For this reason, it can be concluded that the synthesized series of samples have high crystallinity and are of pure phase. The addition of small amounts of Sm 3+ and Eu 3+ had no significant effect on the structure of the main lattice Ca 2 MgSi 2 O 7 . Fig. 1(b) shows an enlarged view of the strongest diffraction peak at 31.15°. It is notable that the intensity of the strongest diffraction peak is slightly shifted to a higher angle. According to Bragg's formula (1) 8 : The decline in radius r yields a decreased crystal plane spacing d, which shifts the strongest diffraction peak in the high-angle direction.   9 . The Ca atoms are interspersed with these groups and are closely packed together to form a highly dense skeleton. The eight-coordinated Ca 2+ (r=1.12 Å), Eu 3+ (r=1.06 Å), and Sm 3+ (r=1.29 Å) all have similar effective ionic radii. And the effective ionic radii of Mg 2+ and Si 4+ in each coordination environment are relatively small compared with those of Eu 3+ and Sm 3+ . Therefore, considering the effective ionic radii of cations with different coordination numbers, it can be concluded that the two doped rare earth ions should occupy Ca 2+ instead of Mg 2+ (r=0.57 Å) and Si 4+ (r=0.26 Å), which have small ionic radii.  Fig. 3(a) shows the excitation and emission spectra of the Ca 2 MgSi 2 O 7 : 2% Sm 3+ phosphor. It is observed that several characteristic excitation peaks of Sm 3+ at 344 nm, 360 nm, 375 nm, 401 nm, 415 nm, 424 nm, 460 nm and 469 nm appear in the range of 300-500 nm for the sample monitored at 600 nm, which originate from the ground state 6 H 5/2 → 4 L 17/2 , 4 D 3/2 , 6 P 7/2 , 4 K 11/2 , 6 P 5/2 , 4 G 9/2 , 4 I 13/2 , 4 I 11/2 electron leaps. Several characteristic emission peaks of Sm 3+ at 562 nm , 600 nm , 646 nm , and 706 nm appear in the sample in the range of 500-750 nm under excitation at 401 nm, and they are attributed to the jump from the excited state 4 G 5/2 to the ground state 6 H J (J=5/2, 7/2, 9/2, 11/2), respectively 10 . Fig. 3(b) shows the excitation and emission spectra of Ca 2 MgSi 2 O 7 :7% Eu 3+ . It can be seen that several characteristic excitation peaks of Eu 3+ at 319 nm, 360 nm, 381 nm, 393 nm, 413 nm, 465 nm and 525 nm appear in the range of 300-550 nm for the samples monitored at 613 nm. They correspond to the 7 F 0 → 5 H 6 ， 5 D 4 ， 5 G 2 ， 6 P 5/2 ， 5 L 6 ， 5 D 3 ， 5 D 2 and 5 D 1 transition of Eu 3+ , accordingly. This means that the characteristic excitation peaks of this sample can be obtained in the UV and near-UV wavelengths. The emission spectrum shown under excitation at 393 nm contains five characteristic emission peaks at 578 nm, 592 nm, 613 nm, 652 nm and 702 nm, respectively, which are attributed to the 5 D 0 → 7 F J (J=0, 1, 2, 3, 4)transition of Eu 3+ . Similar results can be seen in the work of Wang 11 et. al.. The emission intensity of the characteristic emission peaks corresponding to different J magnitudes generally depends on the symmetry of the local crystal field environment of Eu 3+ ions. The emission intensity at 613 nm corresponding to the transition from 5 D 0 to 7 F 2 is stronger than that at 592 nm corresponding to the transition from 5 D 0 to 7 F 1 . This is because the 5 D 0 → 7 F 1 transition is a magnetic dipole transition, which is much weaker than the electric dipole transition and is largely unaffected by the symmetry of the coordination environment. On the opposite side, the 5 D 0 → 7 F 2 leap is an electric dipole transition, which is very sensitive to the symmetry of the crystal field environment and only occurs when the Eu 3+ ion occupies a lattice position that is not the center of inverse symmetry 12,13 . After comparing spectra in Fig. 3(a) and Fig. 3(b), we found that the excitation spectrum of Sm 3+ and the emission spectrum of Eu 3+ have a large range of overlap, so an energy transfer phenomenon from Sm 3+ to Eu 3+ may exist.

Photoluminescence properties
To further investigate the energy transfer phenomenon between Sm 3+ to Eu 3+ , we chose the strongest excitation peak of Sm 3+ at 401 nm to excite the co-doped sample. Fig. 3(c) shows the emission spectra of Ca 2 MgSi 2 O 7 : 2% Sm 3+ , x% Eu 3+ (x=4, 5,6,7,8) under 401 nm excitation. The insets show enlarged views of the emission peaks at 562 nm and 614 nm, respectively. The luminescence intensity of the characteristic emission peaks of Sm 3+ at 562 nm, 600 nm and 646 nm gradually decreased with the increase of Eu 3+ doping concentration under 401 nm excitation. In contrast, the intensity of the Eu 3+ emission increased, with the strongest emission peak occurring at 613 nm (emission here increased significantly by 5.3 times). When the Eu 3+ doping concentration was 7% , its luminous intensity was the strongest; after that, as the Eu 3+ concentration was further increased, the emission intensity decreased due to concentration quenching. This demonstrates that Sm 3+ can transfer some of the absorbed energy to Eu 3+ , leading to a decrease in its own luminescence intensity and enhancement of the red emission of Eu 3+ . It is therefore possible to surmise that there is non-radiative energy transfer 14 (NR) from Sm 3+ to Eu 3+ . However, from the excitation spectra in Fig. 3(a) and 3(b), it can be seen that there is considerable overlap between the excitation spectra of the two in the range of 390 nm-425 nm, so the enhanced luminescence intensity of Eu 3+ in the emission spectra of the co-doped samples could be the reason for the co-excitation of Sm 3+ and Eu 3+ . To exclude the possibility of co-excitation, we choose the characteristic excitation peak 344 nm belonging to Sm 3+ only, to excite the co-doped sample. Fig. 3(d) shows the emission spectra of Ca 2 MgSi 2 O 7 : 2% Sm 3+ , x% Eu 3+ (x=4, 5, 6, 7, 8) under 344 nm excitation. A strong characteristic emission peak belonging to Eu 3+ was found at 613 nm, ruling out the possibility of co-excitation, and tentatively proving that a strong energytransfer phenomenon does exist between Sm 3+ to Eu 3+ . We have taken the question of whether there would be energy reversal from Sm 3+ to Eu 3+ to investigate the energy transfer from Eu 3+ to Sm 3+ . We excited the co-doped samples with the characteristic excitation peak of Eu 3+ at 393 nm as the excitation wavelength. Fig. 3(e) shows the emission spectra of Ca 2 MgSi 2 O 7 : 2% Sm 3+ , x% Eu 3+ (x=4, 5, 6, 7, 8) under 393 nm excitation. This result is almost identical to the single-doped emission spectrum of Eu 3+ in Fig. 3(b). The co-doped sample showed only the characteristic emission peak of Eu 3+ and no characteristic emission peak of Sm 3+ . This indicates that there is no energy transfer from Eu 3+ to Sm 3+ in this sample. This may be due to the difference between the specific energy levels of Sm 3+ and Eu 3+ . The 5 D 0 energy level of Eu 3+ is lower than the 4 G 5/2 energy level of Sm 3+ , so it is difficult to conduct energy-transfer to the 4 G 5/2 energy level of Sm 3+ . Hence we conclude that there is no energy transversion from Sm 3+ to Eu 3+ in the present study.

Fluorescence decay curve and energy transfer mechanism
To further investigate the energy-transfer efficiency between Sm 3+ to Eu 3+ , we measured the lifetime-decay curve of the phosphor, which conforms to the double-exponential decay equation (2) 15 . The double-exponential fluorescence decay curve indicates that the emission peak of this phosphor comes from two luminescent centers. Fig. 4(a) shows the lifetime decay curves of Sm 3+ in this series of samples. Here, the decay time τ decreases monotonically with increasing x. The lifetime of Sm 3+ at 600 nm (the lifetime of the representative sample of Ca 2 MgSi 2 O 7 : 2% Sm 3+ ) is 2.9 ms. The average fluorescence lifetimes of Sm 3+ in the co-doped samples were derived from the average lifetime value equation (3) 16,17 as 2.6 ms, 2.5 ms, 2.3 ms, and 1.65 ms, respectively (Fig.  (4b)). Fig. 4(c) shows the lifetime decay curves of Eu 3+ in this series of samples. It can be seen that the lifetime value shows an increasing trend with the increase in Eu 3+ concentration (the lifetime of the representative sample Ca 2 MgSi 2 O 7 :2%Eu 3+ is 0.87 ms). The averaged fluorescence lifetimes of Eu 3+ were calculated to be 0.92 ms, 1.08 ms, 1.25 ms, and 1.31 ms, respectively (Fig.  (4d)). The results indicated that the doping of Eu 3+ promoted the decay of Sm 3+ and improved the energy transfer efficiency from Sm 3+ to Eu 3+ . We conclude that Sm 3+ and Eu 3+ have strong nonradiative energy transfer. Where:  is the emission intensity; t is the actual lifetime when measuring light intensity; τ 1 is the short lifetime, τ 2 is the long lifetime and A 1 , A 2 are fitting parameters. τ * is the average lifetime obtained after the calculation. The fitting coefficients of A 1 , τ 1 , A 2 , and τ 2 were obtained by fitting to the fluorescence curves, and were all in the range of 1.0-1.3. The energy transfer efficiency between Sm 3+ to Eu 3+ was further calculated using the energy-transfer efficiency equation (4) 18 .
Where η is the energy transfer efficiency, τ 0 and τ correspond to the lifetimes of Ca 2 MgSi 2 O 7 : 2% Sm 3+ and Ca 2 MgSi 2 O 7 : 2% Sm 3+ , x% Eu 3+ , respectively. By calculation, the energy transfer efficiency from Sm 3+ to Eu 3+ is 11%, 14%, 21%, and 44%, respectively. It is demonstrated that the high energy transfer efficiency, wide excitation band, and other characteristics of Ca 2 MgSi 2 O 7 : Sm 3+ , Eu 3+ red phosphor material are expected to make it a red luminescent material with significant prospects for application as a component in white LED trichromatic phosphors. Fig. 4  Non-radiative energy transfer (NR) can take place through mechanisms including resonant energy transfer, cross-chip energy transfer, and phonon-assisted energy transfer 19 . Since the energy difference from the 4 G 5/2 to 6 F 9/2 energy levels is approximately equal to the energy difference from the 6 H 5/2 to 6 F 9/2 energy levels, resonant energy transfer is likely to occur. This form of energy transfer can be achieved in three possible ways: exchange interactions, radiation reabsorption, and multipole-multipole interactions. The key transport distance (R c ) between luminescent center ions is generally considered as a criterion to determine which mechanism it takes place. The multipole-multipole interactions and the critical transport distance (R c ) are closely related. Commonly, multipole-multipole interactions dominate when R c is greater than 5 Å. Otherwise, exchange interactions dominate. According to Blasse's theory, the critical distance for energy transfer between ions can be calculated from Eq. (5) 20,21 .
Where R c is the critical distance for energy transfer, X e is the critical concentration, V is the cell volume, and N is the number of lattice sites that can be substituted in a cell. For this series of samples, V=306.9 Å 3 ，X e =0.07，and N=2. Here, a value of R c =16.117 Å is obtained, which is greater than 5 Å . From this, we determine that the effective energy transfer between Sm 3+ and Eu 3+ is dominated by electrical multilevel interactions, leading to concentration quenching. The relationship between the luminescence intensity I of the sample and the activator concentration x after the concentration of rare earth ions in the luminescent material is doped to a certain value is given by Eq. (6) 22 .
Where c is a constant, and the different values of θ indicate different concentration quenching mechanisms. The values of θ can be taken as 6, 8, and 10 which belong to the dipole-dipole, dipole-quadrupole, and quadrupole-quadrupole interactions, respectively. The relationship between lg (I/x) and lg (x) at 613 nm is shown in Fig. 5(a). It was found that the fitting slope isθ/3≈-2.13. Hence, θ≈6, which corresponds to the dipole -dipole interaction. Fig. 5(b) shows the energy-level transition between Sm 3+ and Eu 3+ . Under 401 nm excitation, the electrons of Sm 3+ in the 6 H 5/2 ground state are elevated to the 4 K 11/2 excited state by absorbing photon energy. A non-radiative leap to the 4 G 5/2 sub-stable state then takes place. However Sm 3+ transfers part of the energy of the 4 G 5/2 energy level to the 5 D 0 energy level of Eu 3+ through the resonance energy between the two energy levels. With the narrow energy-gap between the 4 G 5/2 energy level of Sm 3+ and the 5 D 0 energy level of Eu 3+ , the energy transfer (ET) from Sm 3+ to Eu 3+ is almost irreversible. This matches the above emission spectra results. The possibility of emitting phonons in Sm 3+ →Eu 3+ is higher than the possibility of capturing phonons. Therefore ET from Sm 3+ to Eu 3+ is effective 23 . Energy level transition diagram for Sm 3+ and Eu 3+ .

CIE color coordinates, color temperature and color purity
The CIE color coordinates are one of the basic parameters for assessing the luminescence performance of a sample. Fig. 6 shows the CIE color coordinates corresponding to this series of samples, and the inset is an enlarged view of the color coordinates. It can be seen that the color of the sample gradually moved from orange-red to dark red with the increase in Eu 3+ concentration. The average color coordinate of the series samples was (0.6319, 0.3676), which is very close to the standard red coordinate (0.67, 0.33). As the coordination environment of Ca 2+ provides beneficial conditions for the formation of stable luminescent centers of Sm 3+ and Eu 3+ , the higher lattice matching leads to an enhancement of the emission intensity, which makes the color coordinates of the samples closer to the standard red color. The label inset shows the body color transformations of the series of samples under near-UV (365 nm) lamp-irradiation. It is also evident that the body color changes are more obvious. At the same time, the color purity is also an important parameter for the study of phosphor luminescence performance. To calculate the color purity (C p ) of this phosphor, we used the following equation (7) 24 : Where ( Based on the CIE color coordinates, the CCT value can be calculated according to equation (8) Where n = (x -0.3320)/(y -0.1858). As can be seen from

Thermal stability analysis
It is well known that thermal stability is one of the key indicators for measuring the luminescence performance of phosphors. Fig. 7(a) shows the variable temperature emission spectra of representative samples of Ca 2 MgSi 2 O 7 : 2% Sm 3+ , 7% Eu 3+ at different temperatures (25˚C~275˚C) under the monitoring wavelength of 401 nm. It can be seen that the emission intensity of the phosphor gradually decreases as the temperature increases. Fig. 7(b) shows the normalized integrated intensity, peak intensity and half-height width (FWHM) of the Ca 2 MgSi 2 O 7 : 2% Sm 3+ , 7% Eu 3+ spectra as a function of temperature. At 150˚C, the integrated intensity of the phosphor remains at 68.6% of that at room temperature and the peak intensity remains at 60.2% of that at room temperature, which indicates that the phosphor has good thermal stability. The variation in the integrated and peak intensity of the emission peaks are attributed to temperatureinduced broadening of the emission band. The FWHM of the spectrum is broadened from 20.43 nm to 34.36 nm, the effect of broadening is due to the fact that more phonons are generated when the temperature increases, which leads to stronger electro-phonon coupling, this broadening the emission spectrum.
To further evaluate the potential of these phosphors for near-UV LED application, sample Ca 2 MgSi 2 O 7 :2% Sm 3+ , 7% Eu 3+ along with a commercial blue phosphor BaMgAl 10 O 17 : Eu 2+ and green phosphor (Sr, Ba) 2 SiO 4 : Eu 2+ were packaged with a 395 nm UV chip. Fig. 8 shows the electro-luminescence spectra and luminescence photos of the packaged LED device operating at a current of 20 mA. It presents a low CCT (~5553 K ) and high R a (~84.8), which is superior to those of the commercial white LED composed of blue chip and YAG: Ce 3+ phosphor (CCT~7756 K ; R a~7 5) .The CIE chromaticity coordinates are calculated as (0.3313, 0.3358). The properties of the white LEDs packaged in this work indicate that the red phosphor Ca 2 MgSi 2 O 7 : 2% Sm 3+ , 7% Eu 3+ can be used very effectively as a phosphor in UV-white LED lighting devices 26

Conclusion
In this study, a series of new red phosphors with Sm 3+ , Eu 3+ co-doped with Ca 2 MgSi 2 O 7 were prepared using the high-temperature solid-phase method. The XRD, crystal structure, photoluminescence spectra, fluorescence lifetime, energy transfer mechanism and thermal stability of the samples were systematically investigated. The findings indicate that Sm 3+ , Eu 3+ incorporation did not change the crystal structure of the magnesia yellow feldspar phosphor. It was observed from the spectra that the co-doped samples showed not only the characteristic emission peak of Sm 3+ but also the characteristic emission peak of Eu 3+ , under the excitation of the strongest characteristic excitation peak of Sm 3+ at 401 nm. The emission intensity of Sm 3+ gradually weakened with the increase of Eu 3+ concentration; and the fluorescence lifetime of Sm 3+ decreased gradually with the increase of Eu 3+ concentration. These phenomena demonstrated the presence of resonant non-radiative energy transfer from Sm 3+ to Eu 3+ within the sample, with this transfer occurring with an efficiency of up to 44%. The fabricated samples exhibited color purity of 89.3%, thermal stability of 68.6% (of that observed at room temperature), and a color rendering index (R a ) of 84.8. It was confirmed that this magnesium yellow feldspar silicate phosphor works effectively as the red component in trichromatic phosphors and is highly suitable for application in solid-state illumination sources.

Experiment 4.1 Sample preparation
In this work, Ca 2 MgSi 2 O 7 :2% Sm 3+ , x% Eu 3+ (x=4,5,6,7,8) powder samples were prepared by the commonly used high-temperature solid-phase method. The raw materials were purchased from Shanghai Aladdin Biochemical Technology Co. The chemical formulas are CaCO 3 (99.99%), MgO(99.99%), SiO 2 (99.99%), Sm 2 O 3 (99.9%), Eu 2 O 3 (99.9%). The quantities of raw materials were calculated according to the stoichiometric molar ratio. The raw ingredients were weighed according to the calculated molar ratio mass with an AL104 type electronic scale and put in an agate mortar for full grinding until the raw materials were well mixed. The mixture was then calcined in a MF-1750C Beyecker high-temperature chamber furnace at a heating rate of 5°C/min, rising to 1300°C and holding at this temperature for 230 minutes. The sample was then removed when it cooled to room temperature and was then fully polished in order for characterization.

Instruments and parameters
XRD measurement of the sample was performed using a Shimadzu XRD-6100 X-ray powder diffractometer with Cu Kα rays as the radiation source. A scanning range 2θ of 10° ~ 70° was used, with a step size of 0.02°, and the voltage and current were 40 KV and 30 mA, respectively. Diamond software was used for the crystal structure analysis of the substrates. The emission spectra, excitation spectra and lifetime decay curves of the samples were analyzed using an Edinburgh FLS-920 steady-state/transient fluorescence spectrometer installed with a 450 W xenon lamp. Measurement of sample luminescence characteristics as a function of temperature was performed using the same fluorescence spectrometer with the addition of a PXF4 high temperature fluorescence controller. The color coordinates of a series of samples were evaluated using CIE 1931 software. The packaged phosphor was then combined with commercial blue and green phosphors and integrated into LED packages with 395 nm UV chips.