Color-tunable phosphor of Sr₃YNa(PO₄)₃F:Tb³⁺ via interionic cross-relaxation energy transfer

Abstract

A series of color-tunable Sr₃YNa(PO₄)₃F:Tb³⁺ phosphors with a fluorapatite structure were synthesized by a traditional high-temperature solid state reaction. The emitting color tuning from blue to green can be observed by gradually increasing Tb³⁺ concentrations, which is attributed to the enhanced cross-relaxation (CR) between Tb³⁺ ions, as described by (⁵D₃, ⁷F₆)–(⁵D₄, ⁷F₀). The CR process is analyzed based on the Dexter and Inokuti–Hirayama model, which is assigned to the electric dipole–dipole interaction. The energy transfer critical distance between Tb³⁺ ions is evaluated to be 18.1 Å. In addition, the thermal quenching mechanism of Sr₃YNa(PO₄)₃F:Tb³⁺ is also investigated. At the general working temperature of an LED (423 K), the luminescence intensity still maintains 81% and 92% with the Tb³⁺ concentration of 10 and 30 mol%, respectively, indicating an excellent thermal quenching performance of Tb³⁺. Due to the good optical and thermal properties, the Sr₃YNa(PO₄)₃F:Tb³⁺ phosphor can be used as a promising green emitting phosphor candidate in the field of white light applications.

---

1. Introduction

White light-emitting diodes (WLEDs) considered as the next-generation green lighting sources, have attracted much interest in the solid-state lighting area.^1,2 Compared with the conventional incandescent or fluorescent lamps, WLEDs have excellent physical and chemical characteristics, such as long lifetime, high brightness and efficiency, as well as environmental friendliness.^3,4 The most used commercial WLED is a combination of an InGaN blue LED chip with a Ce-doped yttrium aluminium garnet (YAG:Ce) phosphor.⁵ However, this suffers from a low color rendering index (R_a < 80) and high correlated color temperature (T_c > 4500 K) due to color deficiency in the red and blue-green region.^6,7 To solve the above problem, green and red phosphors with a blue LED, and a UV LED combining tricolor (red, green and blue) phosphors have been regarded as alternatives. Therefore, for an excellent color rendering index, it is necessary to develop efficient green phosphors that possess an excitation wavelength matching the emission wavelength of blue LEDs (440–470 nm) or UV LEDs (350–410 nm).

Recently, great efforts have been made to develop tricolor phosphors with high emission efficiency via doping with suitable ions. Rare earth (RE) ions have great attracted attention due to their abundant emission colors based of 4f–4f or 5d–4f transitions. Moreover, rare earth ions are generally multi-electron structures and have rich energy levels. Especially, when the rare earth ion is doped into the matrix materials, the electronic energy levels of rare earth ions split by the action of the surrounding crystal field. Among the RE ions, Tb³⁺ ion is a famous green-emitting activator due to the emission peaks located at 488, 545, 583 and 620 nm, assigned to the ⁵D₄–⁷F_J (J = 6, 5, 4, 3) multiplet transitions, respectively.⁸ Besides the green emission from ⁵D₄, blue emissions from higher level ⁵D₃ can be also observed. However, owing to the multi-phonon relaxation and cross-relaxation (CR) occurring between Tb³⁺ ions, host lattice with low phonon frequency and low doping concentration of Tb³⁺ are required to detect the blue emissions, which will suppress the ⁵D₃–⁵D₄ nonradiative relaxation.^9–11 Thus, it is essential to choose a suitable host with low phonon frequency doping with appropriate concentration to realize Tb³⁺ activated efficient phosphors.

In recent years, fluorophosphates have been widely investigated on account of their considerable thermal and chemical stability.^12,13 In addition, the largest electronegative of fluorine atoms make them usually exhibit attractive electron ability. Apatite type alkaline-earth halophosphates belongs to the hexagonal symmetrical system (space group of P6₃/m) with a general formula of M₅(PO₄)₃X (M = Na⁺, K⁺, Ca²⁺, Mg²⁺, Ba²⁺, Sr²⁺, Mn²⁺, Gd³⁺, Lu³⁺, Y³⁺, X = F, Cl, OH and O). So far, a variety of phosphates with the formula M₃NaRE(PO₄)₃F have been reported in the literature, such as Sr₃NaGd(PO₄)₃F, Ba₃NaLa(PO₄)₃F, and Sr₃NaLa(PO₄)₃F.^14–16 To the best of our knowledge, there are no reports about the detailed CR property among the Tb³⁺ ions in Sr₃YNa(PO₄)₃F host.

In this paper, we have synthesized the Sr₃Y_1−xNa(PO₄)₃F:xTb³⁺ (SYNPF:xTb³⁺) phosphors (0.002 ≤ x ≤ 0.50) via high temperature solid reaction. The dependence of tunable luminescence properties of SYNPF:Tb³⁺ on Tb³⁺ concentration is investigated. The quantitative theories for the nonradiative energy transfer are studied according to Förster and Dexter. The energy transfer critical distance and the critical concentration between the Tb³⁺ ions are evaluated based on the Dexter model. The thermal quenching mechanism of the phosphor was also studied by observing the luminescence variation of SYNPF:Tb³⁺ at different temperatures. The results indicate that the SYNPF:Tb³⁺ phosphors have potential applications for UV- or NUV-based WLEDs.

2. Experimental section

2.1 Materials and synthesis

The SYNPF:Tb³⁺ phosphors were synthesized by a traditional high temperature solid-state reaction. The starting materials include analytical grade SrCO₃, SrF₂, Na₂CO₃, NH₄H₂PO₄, Y₂O₃ (99.99%), and Tb₄O₇ (99.99%). The SrF₂ (A.R.), Na₂CO₃ (A.R.), and NH₄H₂PO₄ (A.R.) were purchased from Guangfu Co. Ltd, Tianjin (China), SrCO₃ (A.R.), Y₂O₃ (99.99%), and Tb₄O₇ (99.99%) were bought from Sinopharm Chemical Reagent Co. Ltd, Shanghai (China). The raw materials were weighted and thoroughly mixed homogeneously using an agate mortar for 30 min. After mixing and grinding, the mixtures were placed into a crucible and sintered at 1150 °C for 3 h with active carbon. Finally, all the samples were furnace-cooled to room temperature and reground into fine powder for further measurements.

2.2 Characterization

The crystal structure of the sintered samples was identified by powder X-ray diffraction (XRD) analysis (Bruker AXS D8), with Cu Kα radiation (λ = 1.5418 Å) operating at 40 kV and 40 mA. The photoluminescence (PL) and photoluminescence excitation (PLE) spectra were performed using a Hitachi F-4600 spectrometer equipped with a 150 W xenon lamp. The luminescence decay curves were measured by a Horiba Fluorolog FL3-111 spectrometer with a scintillating xenon lamp. The temperature-dependence luminescence properties were characterized on F4600, in which the prepared samples can be accurately heated to the temperature ranging from room temperature to 400 °C.

3. Results and discussion

3.1 Phase identification

Fig. 1 shows that the XRD patterns of SYNPF host and the SYNPF:xTb³⁺ (x = 0, 0.02, and 0.30) together with the standard XRD profile of Sr₃(La,Ce)Na(PO₄)₃(F,OH) (JCPDS 50-1595) as a reference. The XRD patterns of all the samples cannot be indexed to any standard data in JCPDS. Fortunately, we found that all the diffraction peaks of the samples agree well with the standard data of JCPDS 50-1595 except a little shift to larger diffraction angle, which implies the prepared phosphor is isostructural with Sr₃(La,Ce)Na(PO₄)₃(F,OH). The shifts of the diffraction peaks can be assigned to the shrinkage of the unit cell due to the substitution of Y³⁺ (1.015 Å) with La³⁺ (1.061 Å). Furthermore, with increasing of Tb³⁺ concentration, a little shift to the smaller 2θ angle can be observed in comparison with the XRD pattern of SYNPF host, which is due to the larger radius of Tb³⁺ (1.040 Å) than of Y³⁺. No other diffraction peaks were observed, indicating that the obtained phosphor were single phase and the doped ions were successful dissolved in the SYNPF host without inducing significant changes of the crystal structure. To further study the structure of the obtained samples, Rietveld structure refinement of SYNPF host was performed using the general structure analysis system (GSAS) program to obtain the detailed crystal information.¹⁷ Fig. 1(b) illustrates the refinement pattern of SYNPF. The reliability factors R_wp = 7.32%, R_p = 5.93% and χ² = 7.82, indicating that all the observed peaks satisfy the reflection conditions and our prepared phosphor is of single phase. The prepared SYNPF sample possesses a trigonal system and a P3̄ (no. 147) space group with the cell parameters being a = b = 9.623313 Å, c = 7.147825 Å and V = 573.263 ų.

Fig. 1 (a) XRD patterns of as-synthesized SYNPF and SYNPF:xTb³⁺, (b) Rietveld structure pattern of the SYNPF host.

3.2 Luminescence properties of Tb³⁺ ions doped phosphors

Fig. 2 depicts the PL (λ_ex = 225 nm) and PLE (λ_em = 545 nm) spectra of the SYNPF:0.02Tb³⁺ sample. The PLE spectrum consists with two broad bands. One is a strong broadband with the central peak located at 225 nm, originating from the parity allowed 4f⁸ → 4f⁷5d¹ transition of Tb³⁺ ions. The other is composed of several weak peaks in the wavelength range of 280–500 nm, which are attributed to the f–d transitions and intra 4f–4f transitions of the Tb³⁺ ions. Upon excitation by 225 nm, the emission spectrum of SYNPF:0.02Tb³⁺ exhibits both blue and green emissions in the regions of 350–475 nm and 475–650 nm, respectively. The blue emission peaks located at 380 nm, 414 nm, 436 nm, 456 nm, 472 nm can be assigned as the characteristic emission from ⁵D₃ to ⁷F_J (J = 6, 5, 4, 3, 2) of Tb³⁺ ions. Besides the blue emission, some green emission peaks located at 490 nm, 545 nm, 586 nm, 621 nm, respectively are also detected, which are attributed to the transition of Tb³⁺ ions from ⁵D₄ to ⁷F_J (J = 6, 5, 4, 3), as shown in Fig. 2.

Fig. 2 PLE (λ_em = 545 nm) and PL (λ_ex = 225 nm) spectra of SYNPF:0.02Tb³⁺.

In order to study the concentration-dependent luminescent properties (0.002 ≤ x ≤ 0.5) of Tb³⁺, a series of SYNPF:xTb³⁺ samples have been synthesized. Fig. 3(a) exhibits the PL spectra of SYNPF:xTb³⁺ phosphors under the excitation of 225 nm. All of the PL spectra of SYNPF:xTb³⁺ are composed of two bands (⁵D₃ → ⁷F_J and ⁵D₄ → ⁷F_J). When Tb³⁺ concentration reaches to 1 mol%, the emission intensity of the ⁵D₃ shows the highest value. With the increasing of Tb³⁺ concentration, the blue emission intensity of ⁵D₃ decreases gradually, meanwhile, the intensities of ⁵D₄ increases, which is due to the reduction of distance of Tb³⁺ ions. With the decreasing of distance of Tb³⁺ ions, the interactions between Tb³⁺ ions becomes stronger, leading to enhancement of the probability of energy transfer from ⁵D₃ to ⁵D₄ level. When the Tb³⁺ concentration is up to 30 mol%, the emission intensity of ⁵D₄ level starts to decrease, and concentration quenching occurred, as shown in Fig. 3(a). Fig. 3(b) shows the dependence of emission peak intensities at 380 nm (⁵D₃ → ⁷F₆) and at 552 nm (⁵D₄ → ⁷F₅) on the Tb³⁺ concentrations upon the 225 nm excitation. It can be clearly observed that the emission of ⁵D₃ and ⁵D₄ shows the inverse tendency with the increasing of Tb³⁺ concentrations, owing to the CR process and concentration quenching of Tb³⁺, respectively. Since the energy of ⁵D₃ level of 5800 cm⁻¹ is much higher than that of ⁵D₄ level, the de-excitation from ⁵D₃ to ⁵D₄ is owing to CR and multi-phonon relaxation processes. As we know that the CR process is generally determined by the doping concentration of Tb³⁺ ions, and the multi-phonon relaxation process is related to the matrix frequency. Previous studies have demonstrated that CR process has taken place only if the host possesses low phonon frequency to achieve the tunable blue-green phosphor. Several researches have demonstrated that fluoride have a low phonon frequency (∼450 cm⁻¹), which can realize the transmission from ⁵D₃ to ⁵D₄ of Tb³⁺ ions as long as the phonon frequency is low enough (∼1000 cm⁻¹).¹⁴

Fig. 3 Dependence of the (a) PL (λ_ex = 225 nm) spectra and (b) emission intensity of Tb³⁺ in SYNPF phosphors with the variation of Tb³⁺ concentrations.

3.3 CR energy transfer

Fig. 4 presents the energy level schematic and the corresponding energy level transitions for SYNPF:xTb³⁺. When NUV light irradiates the fluorescent powder, electrons from the ground state ⁷F₆ level are stimulated to excited state ⁵D₃ level (process 1), then a part of electrons of Tb³⁺ ions shift from ⁵D₃ excited state transitions of to the ground state of ⁷F_J (J = 5, 4, 3, 2, 1) level with blue emission. Furthermore, a part of electrons from ⁵D₃ level decays nonradiatively to ⁵D₄. Simultaneously, a nearby acceptor Tb³⁺ ion is excited from its ground state ⁷F₆ to an intermediate state ⁷F₀, and then returns to ground state via multi-phonon relaxation (process 2). As the similar values of the energy between ⁵D₃ and ⁵D₄ as well as ⁷F₆ and ⁷F₀, an energy transfer is expected to take place through the CR process in the host of SYNPF:Tb³⁺, resulting in the increasing of the green emission and the reduction of blue emission. The process can be described as: ⁵D₃ (Tb³⁺) + ⁷F₆ (Tb³⁺) → ⁵D₄ (Tb³⁺) + ⁷F₀ (Tb³⁺).

Fig. 4 Schematic energy levels of SYNPF:Tb³⁺ for cross-relaxation energy transfer.

In general, the energy transfer mechanism of the luminescent center can be classified into exchange interactions, and electric multipole interaction. As we know that the exchange interaction strongly depends on the distance R between donor and acceptor ions. When R is small enough (3 Å ∼ 4 Å), exchange interaction dominates in the host, otherwise, it is attributed to electrical multipole interaction. Thus, the distance between Tb³⁺ ions (R_Tb) was induced to specify the type of energy transfer mechanism, which can be estimated as follows:^15,16

where V is the volume of one unit cell, N is the number of the cationic sites occupied by activators in one unit cell, and x is the concentration of Tb³⁺ ions. For SYNPF host, V and N equal to 573.263 ų and 6. Therefore, by using eqn (1), the value of R_Tb can be calculated to be 45.2, 26.3, 20.9, 15.4, 12.2, 10.7, 9.7, 8.5 and 7.1 Å for the Tb³⁺ concentration (the molar ration of Tb³⁺ substitute the Y³⁺) from 0.2 mol% to 50 mol%, respectively. According to theory of Van Uitert, it can be deduced that the CR between Tb³⁺ originates from the electric multipole interaction.

In principle, the CR process can shorten the lifetime of ⁵D₃. Thus, the fluorescence lifetimes (τ₁) for the SYNPF:xTb³⁺ with various Tb³⁺ concentrations (x = 0.002–0.50) were measured by monitoring the ⁵D₃–⁷F₆ emission (λ = 380 nm), as shown in Fig. 5. For the low Tb³⁺ concentration of SYNPF, the decay curves can be well fitted into the single exponential function, which can be written as:

Fig. 5 Fluorescence decay curve of ⁵D₃ (380 nm) in SYNPF:xTb³⁺ with various Tb³⁺ concentrations under 350 nm excitation.

With the increasing of Tb³⁺ concentration, the decays speed up and deviate from single exponential function significantly, which is due to the strengthened CR effects. That is to say the doping of the Tb³⁺ ions has gradually change the luminescence dynamics of the Tb³⁺ ions due to the CR energy transfer from the Tb³⁺ to Tb³⁺ ions.

The energy transfer process between Tb³⁺ ions can also be verified by the fluorescence decay of ⁵D₄ (τ₂). Fig. 6 shows the decay curves of ⁵D₄ → ⁷F₅ transition (λ_em = 545 nm) for SYNPF:xTb³⁺ with different Tb³⁺ concentration upon the excitation of 350 nm. All the samples exhibit the single exponential decay model, even at high Tb³⁺ concentration. When the decay time is limited in a small range of times, such as to 0–8 ms, it can be obviously observed the build-up process in the curves, which presents the trend of first increase and then decrease, especially at low Tb³⁺ concentration, as shown in Fig. 6(b). The above phenomena indicate that there exists a CR energy transfer between Tb³⁺ ions. The energy excited into the 5d levels of Tb³⁺ ion first transfers to the ⁵D₄ level of a neighbor Tb³⁺ ion via CR in the build-up process, and then the ⁵D₄ → ⁷F₅ transition decays slowly. However, the initial rise processes were only observed at low Tb³⁺ concentrations (0.2–30 mol%), and the decay curve presents simple single-exponential decay processes at high Tb³⁺ concentrations (50 mol%), which implies the enhancement of CR with the increasing of Tb³⁺ ions concentration.

Fig. 6 Fluorescence decay curve of ⁵D₄ (545 nm) in SYNPF:xTb³⁺ with various Tb³⁺ concentrations in (a) in a full time scale, and (b) a local time scale under 350 nm excitation.

According to the lifetimes of ⁵D₃ (τ₁) and ⁵D₄ (τ₂), it can be also found that the enhancement of CR decreases the lifetime of the ⁵D₃ level, whereas the lifetime of the ⁵D₄ level is unaffected if the Tb³⁺ concentration is below the quenching point. To analysis the dynamic process of the CR, the average fluorescence lifetimes can be defined by using I–H model. The expression is shown as follows:¹⁴

where I(t) represents the fluorescent intensity at time t. The decay lifetimes of ⁵D₃ for various Tb³⁺ concentrations were calculated to be 1.224, 0.970, 0.813, 0.574, 0.376, 0.281, 0.228, 0.172, and 0.143 ms for the Tb³⁺ concentration increasing from 0.2 to 50 mol%, respectively. For reference, the lifetimes of ⁵D₄ were also calculated, as shown in Fig. 7. It can be clearly observed that the lifetime of ⁵D₄ nearly unchanged below Tb³⁺ concentration of 0.3, which indicates that CR process has little influence on the lifetime of the ⁵D₄. When the Tb³⁺ concentration exceeds 30 mol%, the lifetime of ⁵D₄ decreases from 3.39 to 2.87 ms, due to the concentration quenching, as shown in Fig. 7. In contrast to the lifetime of ⁵D₄, the lifetime of ⁵D₃ level gradually decreases with the increasing of Tb³⁺ ions concentration, which is assigned to the energy transfer via CR between Tb³⁺ ions. To identify the interaction between Tb³⁺ ions, the decay lifetimes of ⁵D₃ level are well fitted with the relationship established by Dexter:¹⁵where τ₀ is the intrinsic decay lifetime of the donor (Tb³⁺), τ(x) is fluorescent lifetime at accepter (Tb³⁺) concentration x, x₀ is the critical concentration with the same dimension as the doping concentration x, and s = 6, 8 or 10, indicating D–D, D–Q and Q–Q interactions, respectively. According to eqn (4), the factor s is defined to be 5.4, which is close to 6, indicating that the energy transfer mechanism is D–D interaction. For the slightly doped of Tb³⁺ (0.2 mol%), the CR energy transfer from ⁵D₃ to ⁵D₄ is negligible, and the multi-phonon relaxation is supposed to be responsible for the ⁵D₄ populating.¹⁴ The lifetime of the ⁵D₃ energy level with the Tb³⁺ concentration of 0.002Tb³⁺ (1.22 ms) is regarded to be the initial life of ⁵D₃ (τ₀). Thus, the critical concentration x₀ can be calculated to be 0.03 mol. By using eqn (1), the average distance between Tb³⁺ is determined with the value of 18.1 Å, which is regarded as the critical distance for the Tb³⁺ in SYNPF.

Fig. 7 Fluorescent lifetime of ⁵D₃ and ⁵D₄ lifetimes of a series of SYNPF:xTb³⁺ as function of Tb³⁺concentration with fitting line.

Base on the above discussion, the interaction mechanism of Tb³⁺, critical concentration and distance were determined. However, the above results were based on an intuitive method, and the values were approximated. For further verify the above investigations, I–H model was used to explain the CR energy transfer in SYNPF:xTb³⁺. As shown in Fig. 6(a), the decay of SYNPF:0.002Tb³⁺ is single exponential, whereas, with the increasing of Tb³⁺ concentration, the decay curves become non-exponential. Since the I–H model has succeeded in describing the non-exponential fluorescent decay, the intensity can be expressed as follows:¹⁸

where I_D(t) is the fluorescent intensity at time t for the system in the presence of energy transfer, and s = 6, 8, and corresponding to 10, are for D–D, D–Q, and Q–Q interactions, respectively. Γ(1 − 3/s) is a gamma function, n_A is the number of acceptor ions per unit volume, and R₀ is the critical distance. According to eqn (2) and (5), we can obtain that

The ratio I_D(t)/I_D0(t) characterizes the decay of excited donors to the acceptors via CR. From eqn (6), one has

and

According to eqn (8), log₁₀{ln[I_D0(t)/I_D(t)]} acts as a linear function of log₁₀ t with a slope of 3/s. Fig. 8 depicts the log–log plot of ln[I_D(t)/I_D0(t)] as a function of t for sample of SYNPF:0.05Tb³⁺. By using the value of the slope (0.49), the parameter s was calculated to be 6, which implies that the interaction of CR energy transfer is the dipole–dipole interaction, which agrees well with the results obtained by the Dexter model.

Fig. 8 The relationship of ln(I_D0(t)/I_D(t)) vs. t plotted in a double logarithmic coordinate for sample SYNPF:0.05Tb³⁺.

In order to further confirm the process of energy transfer, the CR rate (W_CR) and CR efficiency were calculated, as shown in Fig. 9. The W_CR of different Tb³⁺ concentration can be determined by using eqn (9):^19,20

where τ₁₀ is the fluorescence lifetime of ⁵D₃ at the lowest doping concentration of Tb³⁺ (x = 0.002) where the CR is negligible.¹⁴ τ₁₀ is then written as:¹⁹where W₀ is the multi-phonon relaxation rate and γ₁ is the radiative transition rate of ⁵D₃, both of which are independent of Tb³⁺ ions concentration. Fig. 9(a) illustrates the dependence of W_CR on Tb³⁺ concentration with a nearly linear relationship. Thus, the CR rate can be written as:

W_CR = Ax (11)

where A is a proportional constant and x is the concentration of Tb³⁺ ions. As we know that the multi-phonon relaxation rate is independent of concentration for the luminescent centers. Therefore, via using eqn (9)–(11), τ₁ can be fitted by the function as τ₁ = 1/(Ax + B), where B is the sum of W₀ and τ₁. In addition, the CR efficiency of the ⁵D₃ to ⁵D₄ of the Tb³⁺ ions of SYNPF:xTb³⁺ is also calculated by using eqn (12):²⁰

Fig. 9 Dependence of (a) CR rate (W_CR), and (b) CR efficiency η_CR in SYNPF:xTb³⁺ on Tb³⁺ concentration.

And shown in Fig. 9(b), it can be seen that the CR efficiency increases gradually and reaches to 88% with the increase of Tb³⁺ ions concentration.

Due to the CR between Tb³⁺, the intensity ratios of green and blue emissions show significant dependence on W_CR, as shown in Fig. 10. The relationship between R_G/B and W_CR can be written as:¹⁹

where R₀ is the initial value of the ratio and τ₂₀ is the lifetime of ⁵D₄ at the lowest concentration of Tb³⁺ ions. τ₂ and τ₂₀ can be eliminated due to τ₂ remains nearly unchanged with different Tb³⁺ concentrations, as shown in Fig. (7). Therefore, eqn (13) can be rewritten as:where R_G/B and R₀ at different Tb³⁺ concentrations can be calculated from the emission spectra. It can be found that the dependence of R_G/B/R₀ on W_CR is close to a linear relationship, indicating a strong dependence of R_G/B on Tb³⁺ concentration.

Fig. 10 Dependence of R_G/B/R₀ as a function of W_CR for SYNPF:xTb³⁺ with various Tb³⁺ concentration.

3.4 Color chromaticity

Due to the energy transfer between Tb³⁺ and Tb³⁺ ions, the emission color tunes of SYNPF:xTb³⁺ phosphors can be realized from blue to green in the single Tb³⁺ doped upon UV excitation. As shown in Fig. 11, the values of x and y of SYNPF:xTb³⁺ color coordinates and correlated color temperatures are calculated. It can be seen that the CIE chromaticity coordinates of SYNPF:xTb³⁺ are changed from (0.2481, 0.209) to (0.3004, 0.5554) corresponding to light purple blue to green emission. The CIE chromaticity coordinate of SYNPF:0.05Tb³⁺ (0.2511, 0.3721) is very close to an ideal white chromaticity coordinate (0.33, 0.33). Additionally, the color coordinate of SYNPF:0.3Tb³⁺ (0.2885, 0.5423) is close to commercial fluorescent powder MgAl₁₁O₁₉: 0.67Ce³⁺, 0.33Tb³⁺ (0.3300, 0.5950), suggesting that SYNPF:Tb³⁺ is a potential candidate in the solid white light field. The above results indicate that SYNPF:Tb³⁺ has a potential application in the field of lighting with low correlation color temperature, which can be emitted from white light.

Fig. 11 CIE chromaticity diagram for SYNPF:xTb³⁺ with various Tb³⁺ concentrations.

3.5 Thermal quenching

As we know that the stable working temperature of a phosphor converted LED is generally at 150 °C.^21,22 Thus, the thermal characteristics of the SYNPF:Tb³⁺ phosphor were also investigated. Fig. 12(a) and (b) exhibit the temperature-dependent PL spectra of SYNPF:Tb³⁺ with Tb³⁺ concentration of 10 mol% and 30 mol% under the excitation of 225 nm. It can be observed that with the increasing of the temperature from 298 K to 673 K, the emission intensity of all the samples decrease very slowly. Fig. 12(c) and (d) show the emission intensity of the ⁵D₃ and ⁵D₄ of Tb³⁺ ions under different temperature. When the temperature ups to 423 K, the intensity of the ⁵D₃ and ⁵D₄ emission for Tb³⁺ concentration of 0.1 mol% and 0.3 mol% decreased to 80% and 90%, as well as 82% and 94% of the initial value, respectively. For the best sample (SYNPF:0.3Tb³⁺), the intensity of green emission still exceeds 85% as the temperature ups to 675 K. The inset of Fig. 12(c) and (d) show the detailed tendency of the entire emission intensity of SYNPF:Tb³⁺ under different temperature. According to the above results, SYNPF:Tb³⁺ phosphor exhibits excellent thermal quenching performance with the increase of concentration.

Fig. 12 PL spectra of (a) SYNPF:0.1Tb³⁺, and (b) SYNPF:0.3Tb³⁺ with different temperatures in the range from 298 K to 673 K. Integrated intensity of both the emission intensity peak of ⁵D₃ and ⁵D₄ for (c) SYNPF:0.1Tb³⁺ and (d) SYNPF:0.3Tb³⁺ under different temperatures. The inset is the integrated intensity of the entire spectra for SYNPF:0.1Tb³⁺ and 0.3 Tb³⁺.

4. Conclusion

The Tb³⁺ doped SYNPF phosphors have been prepared by a traditional high-temperature solid-state reaction. The PL spectra of SYNPF: Tb³⁺ exhibit two groups of emissions originating from the ⁵D₃ and ⁵D₄ energy level. With increasing of Tb³⁺ concentrations, SYNPF:Tb³⁺ exhibits tunable emissions from the blue to green region, and the intensity ratio of blue to green emission is reduced via the CR described by (⁵D₃, ⁷F₆)–(⁵D₄, ⁷F₀). According to the I–H model and critical distance, the electronic dipole–dipole interaction between Tb³⁺ governs the dynamic CR process. The CR parameters are determined with the critical CR distance R₀ = 18.1 Å and the critical concentration x₀ = 0.03 mol. Moreover, the CR efficiency and the CR rate increase rapidly with the increasing of Tb³⁺ doping concentration. When the doping concentration is up to 50 mol%, the CR efficiency is as high as 86%, and the CR rate increases linearly. The thermal quenching of SYNPF:Tb³⁺ has also been studied. No obvious thermal quenching of SYNPF:Tb³⁺ can be observed. At the LED operating temperature of 423 K, the luminescence intensity of ⁵D₃ and ⁵D₄ for SYNPF:0.1Tb³⁺ and 0.3Tb³⁺ decreases to the 80% and 90%, and 82% and 94% compared to the values at room temperature, respectively. The above results indicate that due to the good luminescence characteristics and thermal properties of SYNPF:Tb³⁺, it is one of the potential candidates for fluorescence in the field of white light.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work is financially supported by the National Natural Science Foundation of China (Grant No. 61604029, and 11774042), the high-level personnel in Dalian innovation support program (Grant No. 2017RQ070), and Fundamental Research Funds for the Central Universities (Grant No. DUT18LK48, and 3132018239).

References

1. Y. L. Zhu, Y. J. Liang, M. F. Zhang, M. H. Tong, G. G. Li and S. Wang, RSC Adv., 2015, 5, 98350–98360.
2. K. X. Song, J. X. Zhang, Y. F. Liu, C. H. Zhang, J. Jiang, H. C. Jiang and H. B. Qin, J. Phys. Chem. C, 2015, 119, 24558–24563.
3. M. Y. Chen, Z. G. Xia, M. S. Molokeev and Q. L. Liu, J. Mater. Chem. C, 2015, 3, 12477–12483.
4. M. M. Jiao, Q. F. Xu, C. L. Yang and H. P. You, RSC Adv., 2017, 7, 28647–28654.
5. C. Liang, H. P. You, Y. B. Fu, X. M. Teng, K. Liu and J. H. He, Dalton Trans., 2015, 44, 8100–8106.
6. C. Zeng, H. W. Huang, Y. M. Hu, S. M. Jun and J. Zhou, Mater. Res. Bull., 2016, 76, 62–66.
7. J. S. Zhang, Z. D. Hao, X. Zhang, Y. S. Luo, X. G. Ren, X. J. Wang and J. H. Zhang, J. Appl. Phys., 2009, 106, 034915.
8. S. Li, N. Guo, Q. M. Liang, Y. Ding, H. T. Zhou, R. Z. Ouyang and W. Lv, Spectrochim. Acta, Part A, 2018, 190, 246–252.
9. Y. Zhang, X. J. Zhang, H. R. Zhang, L. L. Zheng, Y. Zeng, Y. Lin, Y. L. Liu and B. F. Lei, RSC Adv., 2018, 8, 3530–3535.
10. D. W. Wen, J. J. Feng, J. H. Li, J. X. Shi, M. W. Wu and Q. Su, J. Mater. Chem. C, 2015, 3, 2107–2114.
11. Y. J. Hua, D. W. Zhang, H. P. Ma, D. G. Deng and S. Q. Xu, RSC Adv., 2016, 6, 113249–113259.
12. M. M. Jiao, N. Guo, W. Lv, Y. C. Jia, W. Z. Lv, Q. Zhao, B. Q. Shao and H. P. You, Inorg. Chem., 2013, 52, 10340–10346.
13. Z. F. Yang, D. H. Xu, J. N. Du, X. D. Gao and J. Y. Sun, RSC Adv., 2016, 6, 87493–87501.
14. J. S. Zhang, B. J. Chen, Z. Q. Liang, X. P. Li, J. S. Sun, R. X. Zhong, L. H. Cheng and H. Y. Zhong, J. Fluorine Chem., 2012, 144, 1–6.
15. D. L. Dexter and J. H. Schulman, J. Chem. Phys., 1954, 22, 1063–1070.
16. W. Lenth, G. Huber and D. Fay, Phys. Rev. B: Condens. Matter Mater. Phys., 1981, 23, 3877–3885.
17. A. C. Larson and R. B. Von Dreele, Los Alamos Natl. Lab., 1994, pp. 1–221.
18. M. Inokuti and F. Hirayama, J. Chem. Phys., 1965, 43, 1978–1989.
19. Z. D. Hao, J. H. Zhang, X. Zhang, S. Z. Lu and X. J. Wang, J. Electrochem. Soc., 2009, 156, 193–196.
20. Y. F. Liu, J. X. Zhang, C. H. Zhang, J. Jiang and H. C. Jiang, J. Phys. Chem. C, 2016, 120, 2362–2370.
21. J. Y. Chen, N. M. Zhang, C. F. Guo, F. J. Pan, X. J. Zhu, H. Suo, X. Q. Zhao and E. M. Goldys, ACS Appl. Mater. Interfaces, 2016, 8, 20856–20864.
22. X. G. Zhang, J. L. Zhang and M. L. Gong, Opt. Mater., 2014, 36, 850–853.

---