Eu^3+/2+ co-doping system induced by adjusting Al/Y ratio in Eu doped CaYAlO₄: preparation, bond energy, site preference and ⁵D₀–⁷F₄ transition intensity

Abstract

CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2) phosphors have been synthesized via a solid-state reaction process. XRD patterns indicate that they are pure phase. The photoluminescence properties of the CaY_1−xAl_1+xO₄:2%Eu phosphors exhibit both the blue emission of Eu²⁺ (4f⁶5d¹–4f⁷) and red-orange emission of Eu³⁺ (⁵D₀–⁷F_1,2,3,4) under UV light excitation, which showed that the Eu^3+/2+ co-doping system was obtained by adjusting the Al/Y ratio. Eu³⁺ ions can be reduced to Eu²⁺ ions when the Al/Y ratio was changed. In this work, the bond energy method was used to determine and explain the mechanism of the site occupation of Eu ions entering the host matrix. Also, the emission spectrum showed an unusual comparable intensity ⁵D₀–⁷F₄ transition peak. The relative intensity of ⁵D₀–⁷F₂ and ⁵D₀–⁷F₄ can be stabilized by changing the relative proportions of Al³⁺ and Y³⁺. Furthermore, this was explained by the J–O theory.

---

1. Introduction

Rare-earth ions, as widely used activators, have been playing an irreplaceable role in lighting-emitting diodes (LEDs),¹ due to their abundant emission colors based on the 4f–4f or 5d–4f transitions.^2,3 The luminescence properties of different rare earth ions deliver an explicit comprehension on the internal correlations between them, which is generally determined by electronic configuration of the dopant and dynamic coupling between the dopant and the host lattices.^4,5 The local structure around the rare earth in the crystal lattice also plays an important role in controlling its luminescence performance especially the 4f–5d transitions.^6,7

Eu ions, including Eu³⁺ and Eu²⁺, are the most commonly used activators in phosphor materials among the rare-earth ions.^8–10 Eu²⁺ ions emit a tunable color ranging from ultraviolet to red due to its 5d–4f transition.^11,12 The 5d orbit of Eu²⁺ is strongly affected by the environment of the crystal, thus the emission of Eu²⁺ is strongly influenced by the crystal field.¹³ Eu³⁺ is one of the most frequently used red-emitting activators, which mainly shows characteristic emissions resulting from the transitions of ⁵D₀–⁷F_J (J = 0, 1, 2, 3, 4).^14,15 However, the partly-forbidden f–f transitions of Eu³⁺ have low oscillator strength, resulting in low absorption efficiency and a low color rendering index (CRI). Therefore, it is a promising method to overcome the limitations of Eu³⁺ activated phosphors via the coexistence of Eu³⁺ and Eu²⁺ in single phase phosphors.¹⁶

The Eu^3+/2+ co-doped phosphors can be prepared through a reduction annealing process in a reducing atmosphere such as H₂, H₂/N₂ mixture or CO.¹⁶ After that, the Eu³⁺/Eu²⁺ co-doped phosphor can be obtained when Eu³⁺ was reduced to Eu²⁺ partially not completely. However, it is very difficult to obtain the Eu³⁺/Eu²⁺ co-doped CaYAlO₄ even through conventional high temperature solid-state reaction under a reducing atmosphere. Another extreme example is Eu ion doped CaAl₂O₄, in which an abnormal reduction of Eu³⁺ → Eu²⁺ was observed in monoclinic phase of CaAl₂O₄: Eu that calcined in air atmosphere at high temperature.^17,18 Comparing the formula of CaYAlO4 and CaAl₂O₄, the only change is their Al/Y ratio. Herein, we investigated the CaY_1−xAl_1+xO₄:2%Eu system by adjusting the Al/Y ratio. In our work, the phenomenon of the reduction of Eu³⁺ to Eu²⁺ in CaY_1−xAl_1+xO₄ systems was observed, and the bond energy method is adopted to theoretically explain the site preferential occupancy of Eu²⁺/Eu³⁺ in the CaY_1−xAl_1+xO₄ systems.

The deviation of its bond energy in different lattice can be compared to determine which site the activators will occupy according to the similar property of chemical bonds and the similar value of their bond energy.¹⁹ The smaller deviation of the bond energy between the host and activators ions, the more easily the lattice site can be replaced by the ion. Therefore, we use this method to solve the site occupancy problem of doping rare earth ions in the matrix. In the previous research, the site preferential occupancy for Eu in Sr₂V₂O₇, Sr₉Gd(VO₄)₇ and Sr₂V₂O₇/Sr₉Gd(VO₄)₇ phosphors²⁰ as well as the site occupancy preference of Bi²⁺ in β-Ca₂P₂O₇ (ref. 21) crystal have been confirmed though the bond energy method.¹⁹ In addition, the bond energy method can connect the relationship between bond energy and preferential occupancy in Eu³⁺ doped in CaAl₂Si₂O₈ crystal.²² Our calculated results are in good agreement with the experimental data and the photoluminescence spectra.

In addition, Eu³⁺-doped inorganic phosphors can be used as efficient orange to red emitting phosphors due to the ⁵D₀–⁷F_J (J = 0, 1, 2, 3 and 4) transitions. Most of the phosphors show the dominant emission either ⁵D₀ → ⁷F₁ (∼593 nm) or ⁵D₀ → ⁷F₂ (∼610 nm) transition of Eu³⁺ ions. The dominated ⁵D₀ → ⁷F₄ (∼703 nm) transition emission is infrequent. Only some numerable phosphors show that they have stronger intensity of ⁵D₀–⁷F₄ than ⁵D₀–⁷F₁, such as Sr_0.99[La_(1−x)Eu_x]_1.01Zn_0.99O_3.495,²³ LaBO₃:Eu³⁺,²⁴ Na₉[EuW₁₀O₃₆]·14H₂O,²⁵ Na₂CaSiO₄:Eu³⁺,²⁶ Ca₂Ga₂SiO₇:Eu³⁺,²⁷ YAsO₄:Eu.²⁸ Similarly, this phenomenon has been found in CaYAlO₄ with Eu doped. But, the stronger emission of ⁵D₀–⁷F₄ disappeared when Eu³⁺ and Tb³⁺ were co-doped into CaYAlO₄.²⁹ However, in our work, the relative intensity of ⁵D₀–⁷F₄ and ⁵D₀–⁷F₂ can be modified by changing the Al³⁺ and Y³⁺ ions ratios in CaY_1−xAl_1+xO₄ and the Judd–Ofelt theory³⁰ was applied to calculate the radiative properties of the prepared materials.

2. Experimental section

2.1 Materials and synthesis

A series of CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2, 0.3, 0.4) phosphors was prepared by a traditional high temperature solid-state reaction route using stoichiometric amounts of CaCO₃ (A.R.), Y₂O₃ (99.99%), Al₂O₃ (99.99%), Eu₂O₃ (99.99%) as raw materials, after the reactants were mixed and well-ground in an agate mortar, they were preheated at 1400 °C for 4 hours. Then the mixtures were reground and calcined at 1500 °C for 4 hours. All of CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2, 0.3, 0.4) phosphors prepared under a reducing atmosphere of H₂ (5%) and N₂ (95%). The final power products were obtained after cooling to room temperature and grinding to powder sample.

2.2 Characterization

The phase purity of the products was examined by X-ray diffraction (XRD) using a BRUKER D8 ADVANCE powder diffractometer with Cu Kα radiation (0.15405 nm) at room temperature, the diffraction data were collected in 2θ range from 5 to 90°. The UV-vis diffuse reflectance spectra were collected through a Cary 5000 UV-vis-NIR spectrophotometer equipped with a double out-of-plane Littrow monochromator using BaSO₄ as a standard reference. All excitation, emission, decay spectra and quantum efficiency were measured on an Edinburgh FLS980 combined fluorescence lifetime, a 450 W xenon lamp was used as the excitation source of emission spectra, while a 60 W μF flash lamp with a pulse width of 1.5–3.0 μs was for the measurements of decay curves.

3. Result and discussion

3.1 Phase characterization

For comparison, the CaYAlO₄:2%Eu was prepared firstly under a reducing atmosphere of H₂ (5%) and N₂ (95%). The XRD patterns of the CaYAlO₄:2%Eu together with the Joint Committee on Powder Diffraction Standards (JCPDS) card no. 24-0221 are shown in Fig. 1(a). The XRD patterns of CaYAlO₄:2%Eu shows that the diffraction peaks of all single samples are matched well with the standard card of CaYAlO₄ (JCPDS: 24-0221), which indicates that the samples are purity phase and doping Eu ions do not cause significant changes to the crystal structure.

Fig. 1 (a) Powder X-ray diffraction (XRD) patterns of CaYAlO₄:2%Eu³⁺ and the standard cards of CaYAlO₄ (JCPDS: 24-0221); (b) experimental (cross), calculated (solid line) and difference (bottom) results of XRD refinements of CaYAlO₄:2%Eu³⁺. (c) The crystal structure of CaYAlO₄. Ca²⁺or Y³⁺ ions are coordinated with nine oxygen atoms to form CaO₉ or YO₉ group and Al³⁺ ions are coordinated with six oxygen atoms and form the AlO₆ octahedron.

In order to get further confirmation and knowledge regarding crystal structure information and sites of CaYAlO₄:2%Eu, the XRD Rietveld refinement of CaYAlO₄:2%Eu were performed by the General Structure Analysis System (GASA) program with the single crystallographic data of CaYAlO₄ as the initial model. Fig. 1(b) presents red lines and black lines stand for experimental and calculated patterns, respectively, which matched well with each other. The obtained converged weighted-profiles of R_p = 8.4% and R_wp = 6.1%, which reveals a good quality of fit. As the crystallographic data of CaYAlO₄:2%Eu shown in Table 1. And Fig. S1† shows the XRD refinements of CaY_0.9Al_1.1O₄:2%Eu and CaY_0.8Al_1.2O₄:2%Eu. This aluminate compound has a tetragonal crystal system with space group I4/mmm, a = b = 3.664 Å, c = 11.889 Å, V = 159.61 ų. However, the lattice constants of CaYAlO₄:2%Eu are a = b = 3.648 Å, c = 11.885 Å, which indicates that its crystal constants increase with increasing the concentration of Eu³⁺.

Table 1 Crystallographic data of CaYAlO₄:0.02Eu³⁺, as determined by the Rietveld refinement of powder XRD data at room temperature

Atom	Site	x	y	z	Occupancy
Ca1	4e	0.0000	0.0000	0.3604	0.500
Y1	4e	0.0000	0.0000	0.3603	0.490
Eu1	4e	0.0000	0.0000	0.5000	0.010
Al1	2a	0.5000	0.5000	0.5000	1.000
O1	4e	0.5000	0.0000	0.5000	1.000
O2	4e	0.0000	0.0000	0.1686	1.000

---

CaYAlO₄ has a tetragonal K₂NiF₄ structure and belongs to a family of compounds with the general formula ABCO₄, where A is an alkaline earth cation, B is Y, Sc, or a trivalent rare earth element and C is Al, Ga or a transition metal ion. In the unit cell, as shown in Fig. 1(c), the Ca and Y cations are distributed almost statistically in the nine coordinated sites and Al³⁺ ions are coordinated with six oxygen atoms and form the AlO₆ octahedron.

Notably, the (Ca/Y)O₉ polyhedron is closely surrounded by AlO₆ octahedrons to form a cage structure (Fig. S2†). Thus, it can be concluded that the local environment of Ca/Y sites is highly compressed due to the rigid structure of CaYAlO₄, which gives rise to the difficulty of reduction of Eu³⁺ activators. The Lin et al. has been reported an effective approach-crystal-site engineering to control Eu³⁺ reduced to Eu²⁺.¹⁶ So, a similar way has been used to change the reduction of Eu³⁺ doped in CaYAlO₄.

Compared with Eu³⁺, the smaller ions will replace the Al³⁺ site (such B³⁺ and Si⁴⁺), therefore, an artificial defect substitution will be used for Al³⁺ doping to CaYAlO₄. Fig. 2 showed the XRD patterns of CaY_1−xAl_1+xO₄:2%Eu (x = 0–0.4). Compared with standard card of CaYAlO₄, When x > 0.2, the XRD pattern diffraction peak appeared a weak miscellaneous peak around 31°, which indicates that a new phase appeared at this time, the diffraction peak at about 31 degrees is described to Ca₃Y₂O₆, the strongest exact diffraction peak of CaAl₂O₄ is at 30 degrees (see Fig. S3†), so the impurity cannot be ascribed to CaAl₂O₄. In addition, the properties of luminescence for Eu ions doped in Ca₃Y₂O₆ crystals aren't detected in research works at present, which indicates that the Eu²⁺ or Eu³⁺ in Ca₃Y₂O₆ has not any emission at room temperature. When x ≤ 0.2, the diffraction peak of the sample is consistent with the standard card diffraction peak, which means that the Al³⁺ actually occupy the Y³⁺ within a certain range. Thus, we restricted the x value to a maximum of 0.20. Table 2 summarizes the lattice parameters and reliability factors of CaY_1−xAl_1+xO₄:2%Eu. In addition, because the Al³⁺ ion radius is smaller than that of Y³⁺, the volume of the unit cell will decrease when Al³⁺ occupies the Y³⁺ site.

Fig. 2 The powder X-ray diffraction patterns of CaY_1−xAl_1+xO₄:2%Eu (x = 0–0.4) and the standard cards of CaYAlO₄ (JCPDS: 24-0221).

Table 2 The lattice parameters and reliability factors of CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2)

Sample	CaYAlO4	CaY0.9Al1.1O4	CaY0.8Al1.2O4
Rwp	0.081	0.092	0.101
Rp	0.061	0.071	0.082
χ^2	2.410	2.030	3.890
a = b (Å)	3.664	3.642	3.610
c (Å)	11.889	11.882	11.859
V (Å^3)	159.61	157.60	154.55

---

3.2 Photoluminescence properties

For Eu doped samples, one can clearly distinguish the different emission bands from the photoluminescence (PL) properties of the two valence states of Eu, i.e., the broad ones are attributed to the parity-allowed 4f–5d transitions of Eu²⁺ and the sharp ones are ascribed to the parity–forbidden transitions of Eu³⁺ (⁵D₀ → ⁷F_J, J = 0–4). Fig. 3(a), (c) and (f) illustrate the emission spectra of CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2) samples under the NUV excitation of 365 nm. When x = 0, as shown in Fig. 3(b), the photoluminescence excitation (PLE) spectrum monitored at 622 nm reveals a broad band with the peak at 277 nm in the range of 200–350 nm due to the charge transfer band (CTB) from O to the Eu ions, along with some sharp peaks in the range of 350–500 nm related to the 4f–4f transitions of Eu³⁺ ions. Under the excitation of 277 or 365 nm, only the sharp f–f transitions can be found. The sharp emission lines of CaYAlO₄:Eu can be assigned to f–f transition of Eu³⁺, which indicates that Eu³⁺ could not be directly reduced to Eu²⁺ in CaYAlO₄:Eu system under a reducing atmosphere. All the emissions spectra are intense. The most intense emission at around 622 nm is attributed to the hypersensitive transition ⁵D₀ → ⁷F₂, indicating that the Eu³⁺ ions mainly occupy the sites without inversion symmetry. The more intense emission at 702 nm is rarely reported. This peak comes from the ⁵D₀ → ⁷F₄ transition of Eu³⁺. As shown in Fig. S2,† the excitation peaks are similar under the different monitoring wavelength (λ_em = 592, 621, 702 nm), which indicates that the emission peaks of Eu³⁺ at 592, 621, 702 nm comes from the same site of Eu³⁺.

Fig. 3 (a), (c) and (f) are the emission spectra of CaYAlO₄:Eu, CaY_0.9Al_1.1O₄:Eu and CaY_0.8Al_1.2O₄:Eu under the excitation wavelength at 365 nm, respectively. (b), (d) and (g) are the excitation spectra under the monitoring wavelength at 622 nm and the emission spectra under the excitation wavelength at 277 nm under excitation wavelength of CaYAlO₄:Eu, CaY_0.9Al_1.1O₄:Eu and CaY_0.8Al_1.2O₄:Eu, respectively. (e) and (h) are the excitation spectra under the monitoring wavelength at 445 nm and the emission spectra under the excitation wavelength at 327 nm of CaY_0.9Al_1.1O₄:Eu and CaY_0.8Al_1.2O₄:Eu, respectively.

Fig. 3(b and f) show the emission spectra of CaY_0.9Al_1.1O₄:Eu and CaY_0.8Al_1.2O₄:Eu under the excitation of 365 nm, respectively. It is surprising to find an appearance of a broad band with the peak at 445 nm except for some sharp peaks in the range of 350–500 nm related to the 4f–4f transitions of Eu³⁺ ions. Fig. 3(e and h) show the PL spectra of CaY_0.9Al_1.1O₄:Eu and CaY_0.8Al_1.2O₄:Eu under the monitoring wavelength at 445 and the PLE spectra under the excitation wavelength at 327. These correspond to the 4f⁶5d¹–4f⁷ transition of Eu²⁺. This indicates that Eu³⁺ had been directly reduced to Eu²⁺ in CaY_0.9Al_1.1O₄:Eu and CaY_0.8Al_1.2O₄:Eu system under a reducing atmosphere. The luminescence intensity of Eu²⁺ increases with increasing x in CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2) system (Fig. S4†), which can be attributed to the increase of x value. These results suggest that Eu³⁺ is partially transformed to Eu²⁺ in CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2).

The corresponding CIE chromaticity diagram for CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2) has been shown in Fig. 4. The points A, B and C stand for the CIE coordinate position when x = 0, 0.1 and 0.2, respectively. With increasing of the Al/Y ratio in CaY_1−xAl_1+xO₄:2%Eu, the emission color changed from A (0.583, 0.322) red to C (0.213, 0.083) blue.

Fig. 4 The corresponding CIE chromaticity diagram for CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2) under the 365 nm excitation. (A) x = 0; (B) x = 0.1; (C) x = 0.2.

The diffuse reflectance in the UV-vis region was used to calculate the experimental band-gap value of CaY_1−xAl_1+xO₄:Eu³⁺(x = 0, 0.1, 0.2). Fig. 5 illustrates the diffuse reflectance of CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2). The results show that the samples begin to exhibit low reflectance below 330 nm due to the high radiation absorption. This behavior is assigned to the edge absorption, corresponding to the electronic transition from the valence band to the conduction band of CaYAlO₄. The matrix of yttrium aluminate and calcium shows optical transparency in the visible region between 330 and 750 nm, making it a candidate for applications in photonic devices.

Fig. 5 (a) Diffuse reflectance of CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2). (b), (c) and (d): The (hνF(R_∞))² was plotted against the (hν) using the Kubelka–Munk function.

In Fig. 5b–d, the (hνF(R_∞))² of CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2) was plotted against the (hν) using the Kubelka–Munk function.³¹ The following relational expression proposed by Tauc, Davis, and Mott is used:^32,33

[αhv]ⁿ = A(hv − E_g) (1)

where h is Planck's constant, ν is frequency of vibration, α is absorption coefficient, E_g is band gap, and A is proportional constant. hv represents the energy per photon, E_g is the value of the band gap, n = 1/2 means a indirect allowed transition, 2 represents a direct allowed transition, 3/2 stands for a direct forbidden transition, or 3 indicates as indirect forbidden transition. Since the direct allowed transition is used in the experiment, n = 2 is used for these samples.

The acquired diffuse reflectance spectra of CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2) in Fig. 5 are converted to Kubella–Munk equation:^32,34

where R_∞ is the diffuse reflectance of the layer relative to the standard. Thus, the vertical axis is converted to quantity F(R_∞), which is proportional to the absorption coefficient α. The α in the Tauc equation is substituted with F(R_∞). Thus, in the actual experiment, the relational expression (1) becomes:^23,34

(hvF(R_∞))² = C(hv − E_g) (3)

where C is a proportional constant. Using the Kubelka–Munk function, the was plotted against the hv. The curves that plots the value of (hv − (hvF(R_∞)²) on the horizontal axis hv and vertical axis (hvF(R_∞))² are drawn in Fig. 5(b–d). Here, the unit for hv is eV (electron volts), and its relationship to the wavelength λ (nm) becomes hv = 1239.7/λ. A red lines are drawn tangent to the point of inflection on the black curves. The values associated with the point of intersection of the lines tangent to the plotted curve inflection point with the horizontal axis (hv axis) becomes the band gap E_g value. Their band gaps of CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2) are 4.10, 4.10 and 4.11 eV. The results show that a significant changing cannot be found in the band-gap edge with increasing value of x. In fact, the R. V. Perrella's³⁴ reported that the difference of energy band-gap between samples doped with 1 and 10 mol% of Eu³⁺ is in the order of magnitude of the phonon energy of the lattice, the Eu³⁺ substitute Y³⁺ cannot change the band gap. Here, when the x increases and cause the band gap also unchanged, so we infer that Al³⁺ also substitute the Y³⁺ site rather than Ca²⁺ site.

The decay curves of Eu³⁺ shown in Fig. 6a–c are accurately fitted using the following single-exponential equation:

where I_t and I₀ are the intensity at time t and time zero, respectively, the τ is the lifetime. When under the excitation wavelength of 394 nm, the lifetimes of Eu³⁺ in CaYAlO₄:Eu, CaY_0.9Al_1.1O₄:Eu and CaY_0.8Al_1.2O₄:Eu are 1.198 ms, 1.190 ms and 1.201 ms, respectively.

Fig. 6 (a), (b) and (c) are the decay curves of Eu³⁺ ions in CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2) (λ_ex = 394 nm, λ_em = 624 nm). The (d) and (e) are the decay curves of Eu²⁺ ions in the CaY_1−xAl_1+xO₄:2%Eu (x = 0.1, 0.2) sample (λ_ex = 327 nm, λ_em = 445 nm).

The decay curves of Eu²⁺ in Fig. 6d and e are accurately fitted well to a second-order exponential decay model based on the following formula:

where I(t) is the luminescence intensity at times t. A₁ and A₂ are fitting constants, τ₁ and τ₂ are exponential component of the decay time and t is the time, respectively. The average decay time can be described by the following formula:

According to the equation, the lifetimes of Eu²⁺ in CaY_0.9Al_1.1O₄:Eu and CaY_0.8Al_1.2O₄:Eu are calculated to be 1.07 ns and 1.14 ns, respectively.

3.3 Bond energy method and preferential occupancy of Eu ions

The dopant occupancy in the matrix can be determined by comparing the deviation of its bond energy in different lattice location. The similar value of the deviation for its bond energy, the more stable the dopants in matrix are. Thus, to determine which site Eu³⁺ and Eu²⁺ will occupy, the bond-energy method has been used to study to the local structure Eu³⁺/Eu²⁺ in CaYAlO₄ crystals. The bond energy of different dopants in CaYAlO₄ crystallographic frame can be estimated through the following equation:^19,21,22V_Eu is valence state of Eu (including +2 or +3), both J and d₀ are constant usually they can be found in reference. E_Ca/Y/Al–O is mean the bond energy of Ca–O, Y–O or Al–O. In the case of pure CaYAlO₄ crystals without any dopant, V_Ca/Y/Al/V_Eu is equal to 1. The eqn (7) can be expressed as:^19,21,22and if is not equal to 1, then it means the valence state affects the crystal bond energy effectively.

It is assumed that the Eu³⁺–O²⁻ bonds have the similar bond lengths as Ca/Y/Al–O bonds when the dopant substitutes Ca/Y/Al cations. In this regard, the value of various un-doped and doped CaYAlO₄ crystals can be evaluated. The displacement of dopants leads to a large variation of the crystal composition and bond energy, which may affect not only the crystal properties but also the crystal stability. In such a case, it is convenient to measure the variation of bond energy by the following expression when the Eu locates at Ca/Y/Al sites. Based on chemical viewpoint, the dopants preferentially occupy the sites with smaller alterations of bond energy, the sites with smaller absolute values can be expressed as below,^19,21,22

ΔE^Ca/Y/Al_Eu = |E_Ca/Y/Al–O − E_Eu–O| (9)

Here ΔE^Ca/Y/Al_Eu is the variation of bond energy when the Eu locates at Ca/Y/Al sites. Based on chemical viewpoint, the dopants preferentially occupy the sites with smaller deviation of bond energy (E_Ca/Y/Al–O), the sites with smaller absolute values of ΔE^Ca/Y/Al_Eu.

According to the bond energy method, the values of bond energy of CaY_1−xAl_1+xO₄:2%Eu (x = 0.1, 0.2) have been shown in Table 3. The order of variation of bond energy under the assumption that Ca²⁺/Y³⁺/Al³⁺ ions are substituted by Eu³⁺ is (3.236 kcal mol⁻¹) < (6.533 kcal mol⁻¹) ≪ (118.789 kcal mol⁻¹), which means that Eu³⁺ will preferentially replaces Y³⁺. The calculation result is consistent with the experiment. In addition, the order of variation of bond energy Eu²⁺ substituted Y³⁺, Ca²⁺ and Al³⁺ is (12.386 kcal mol⁻¹) ≈ (12.633 kcal mol⁻¹) ≪ (67.149 kcal mol⁻¹), which means that the site of Ca²⁺ and Y³⁺ could be replaced by Eu²⁺ equally. Considering the valence of Eu²⁺, we can determine that the Eu²⁺ is preferentially occupied the site of Ca²⁺.

Table 3 The bond parameters of the central atom and values of bond energy when Eu³⁺ locates at Ca, Y, and Al sites in CaYAlO₄. All of the bond energy unit are kcal mol⁻¹

Central atom	Coordination atom	Count	d (Å)	EM–O	EEu^3+	EEu^2+
Ca1	O2	1×	2.3013	51.291	39.457	28.407	6.533	12.633
	O1	4×	2.5152	28.772	22.133	15.935
	O2	4×	2.6125	22.119	17.015	12.250
Al1	O1	4×	1.8375	73.757	207.306	149.253	118.784	67.149
	O2	2×	1.9866	49.294	138.549	99.750
Y1	O2	1×	2.3013	65.045	59.186	42.611	3.236	12.385
	O1	4×	2.5152	36.488	33.200	23.903
	O2	4×	2.6125	28.050	25.523	18.376

---

3.4 The ⁵D₀–⁷F₄ emission for CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2)

The emission spectrum showed an unusual comparable intensity ⁵D₀–⁷F₄ transition peak. The relative intensity of ⁵D₀–⁷F₂ and ⁵D₀–⁷F₄ can be stable by changing the relative proportions of Al³⁺ and Y³⁺. To gain some insight into the nature of the luminescence behavior of Al³⁺ ions in crystal CaYAlO₄:2%Eu powders, the Judd–Ofelt model was applied to the determination of spontaneous emission coefficients. The intensity parameters Ω₂ and Ω₄ were determined from luminescence spectra using the method proposed by Kodaira et al.^28,35

The emission intensity, I = ℏωAN, is expressed in terms of the surface under the emission curve, where ℏω is the transition energy, N is the population of the emitting level (⁵D₀) and the Einstein's coefficient of spontaneous emission can be given by²⁵

where χ = n₀(n₀² + 2)²/9 is a Lorentz local field correction. The square reduced matrix elements are 〈⁵D₀‖U⁽²⁾‖⁷F₂〉² = 0.003289; 〈⁵D₀‖U⁽²⁾‖⁷F₄〉² = 0.002365 in eqn (10) and an average index of refraction equal to 1.5 was used.³⁶ In this case the A_0–λ values are obtained by using the relation:^25,37where S_0–λ is the area under the curve related to the ⁵D₀–⁷F_λ transition obtained from the spectral data, σ_λ is the energy barycenter of the 0–λ transition and A_0–1 is the Einstein's coefficient for the 0–1 magnetic dipole transition. The A_0–1 value is estimated to be around of 50 s⁻¹.

The lifetime (τ), non-radiative (A_nrad) and radiative (A_rad) rates are related through the following equation:³⁵

where the radiative (A_rad) rates was obtained by summing over the radiative rates A_0–J for each ⁵D₀–⁷F_J transitions. It can be given by:³⁵

The emission quantum efficiency of the emitting ⁵D₀ level is given by:³⁵

The values of the Ω₂ and Ω₄ parameters as well as other quantities derived from analysis of the luminescence spectra of Al³⁺ replace the Y³⁺ in CaYAlO₄ are given in Table 4.

Table 4 Decay rates of radiative (A_rad), nonradiative (A_nrad), and total (A_tot) processes of ⁵D₀–⁷F_J transitions, luminescence lifetime (τ), intensity parameters (Ω₂, Ω₄), quantum efficiencies (η) and actual quantum efficiency (η_a) of CaY_1−xAl_1+xO₄:2%Eu

Samples	Arad	Anrad	Atot	τ	Ω2	Ω4	η	ηa
CaYAlO4	409	427	836	1.198	4.12	2.97	0.489	47.69%
CaY0.9Al1.1O4	406	434	840	1.190	3.99	3.09	0.483	50.42%
CaY0.8Al1.2O4	403	429	832	1.201	3.91	3.10	0.484	55.43%

---

Table 4 shows intensity parameters Ω₂ and Ω₄. Ω_λ is the emission intensity parameters; Ω₂ is the emission intensity parameter of ⁵D₀–⁷F₂. Ω₄ means the emission intensity parameter of ⁵D₀–⁷F₄. Here, the Ω₂ intensity parament values are higher than Ω₄ for CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2). The Ω₂ parameter depends rather on the lower rank components of crystal field and dynamic coupling interactions, while the Ω₄ parament depend rather on the corresponding higher components. It suggests that site symmetry occupied by Eu³⁺ ions does not have a character of centrosymmetric chemistry environment considering that the ⁵D₀–⁷F₂ transitions is formally forbidden due to the electric dipole selection rule. Meanwhile, it has been commented in literature that the luminescence spectra of compounds with D_4d (see Fig. 1(c)) symmetry are often dominated by the ⁵D₀–⁷F₄ transition of Eu³⁺ because of the absence of central symmetry. An undistorted square anti-prism has D_4d symmetry, so a site with symmetry lowed than D_4d but the coordination polyhedron close to a square anti-prism, is expected to have an intense ⁵D₀–⁷F₄ transition, In CaY_1−xAl_1+xO₄:2%Eu, Eu³⁺ is nine-fold coordinated, this coordination polyhedron can be regarded as close to mono-capped square-anti-prism. As a result, in this compound the ⁵D₀–⁷F₄ transition is less intense than the ⁵D₀–⁷F₂ transition but much more intense than ⁵D₀–⁷F₁ magnetic dipole transition. The same remarks have also been for LaBO₃:Eu³⁺, [Eu(DOTA)(H₂O)]⁻ and Sr_0.99La_1.01Zn_0.99O_3.495:Eu³⁺. In addition, the values of Ω₄/Ω₂ is 0.72, 0.77, 0.79 for CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2) respectively, which indicated the ⁵D₀–⁷F₄ emission can be enhanced by properly adjusting the Al/Y ratio, at the same time it does not change the quantum efficiency (η). The quantum efficiency (η_a) measured by the integrating sphere is nearly the result by calculation.

4. Conclusions

A series of CaY_1−xAl_1+xO₄:2%Eu³⁺ (x = 0, 0.1, 0.2, 0.3) phosphors have been identified. The powder XRD patterns and Rietveld refinement reveal that CaY_1−xAl_1+xO₄:2%Eu³⁺ have a tetragonal crystal structure with the space group I4/mmm (No. 139). The photoluminescence property of the CaY_1−xAl_1+xO₄:2%Eu (x = 0, 0.1, 0.2) phosphors exhibit both blue emission of Eu²⁺ (4f⁶5d¹–4f⁷) and red-orange emission of Eu³⁺ (⁵D₀–⁷F_1,2,3,4) under UV light excitation, which showed that Eu^3+/2+ co-doping system was obtained by adjusting Al/Y ratio. Eu³⁺ ions can be reduced to Eu²⁺ ions when Al/Y ratio was changed. Calculated by the bond-energy theory found that Eu³⁺ takes priority over the Y³⁺ site, in addition, the site of Ca²⁺ and Y³⁺ all can be occupied by Eu²⁺. Considering the valence of Eu²⁺, it can determine that the Eu²⁺ is preferentially occupied the site of Ca²⁺. The Judd–Ofelt model was applied to the determination of spontaneous emission coefficients. The intensity parameters Ω₂ and Ω₄ were determined from luminescence spectra.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work is financially supported by the National Natural Science Foundations of China (Grant no. 21301053 and 21571165) and Hubei Natural Science Foundations from Science and Technology Department of Hubei Province (2018CFB517). This research was supported by the Basic Science Research Program though the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning (No. 2018R1A2B6005179). “Eu^3+/2+ co-doping system induced by adjusting Al/Y ratio in CaY_1−xAl_1+xO₄:2%Eu: preparation, bond energy, site preferential and ⁵D₀–⁷F₄ transition intensity” is supplied by the Functional Phosphor Bank at Pukyong National University.

Notes and references

1. S. Liu, G. Zhao, H. Ying, J. Wang and G. Han, Opt. Mater., 2008, 31, 47–50.
2. T. Selvalakshmi, S. Sellaiyan, A. Uedono and A. Chandra Bose, Mater. Chem. Phys., 2015, 166, 73–81.
3. K. Biswas, A. D. Sontakke, R. Sen and K. Annapurna, J. Fluoresc., 2012, 22, 745–752.
4. L. Yang, Y. Wan, Y. Huang, X. Wang, H. Cheng and H. J. Seo, Mater. Lett., 2016, 172, 23–26.
5. B. Wiendlocha, S. P. Kim, Y. Lee, B. He, G. Lehr, M. G. Kanatzidis, D. T. Morelli and J. P. Heremans, Phys. Chem. Chem. Phys., 2017, 19, 9606–9616.
6. P. L. Roeder, D. MacArthur, X. P. Ma, G. R. Palmer and A. N. Mariano, Am. Mineral., 1987, 72, 801–811.
7. Y. Su, L. Li and G. Li, Chem. Mater., 2008, 20, 6060–6067.
8. H. You and C. Shi, Chin. Sci. Bull., 1996, 41, 123–126.
9. J.-G. Kang, J.-S. Jung, J.-P. Hong, S.-J. Won, Y. Sohn and C. K. Rhee, J. Phys.: Condens. Matter, 2001, 13, 2835–2843.
10. Z. Pei, Q. Zeng and Q. Su, J. Phys. Chem. Solids, 1999, 61, 9–12.
11. J. S. Zhong, H. B. Gao, Y. J. Yuan, L. F. Chen, D. Q. Chen and Z. G. Ji, J. Alloys Compd., 2018, 735, 2303–2310.
12. Z. Jiang, X. Yu, J. Gou, L. Duan, X. Su, G. Fan and Y. Duan, J. Mater. Sci.: Mater. Electron., 2017, 28, 3630–3636.
13. H. Chen, X. Huang and W. Huang, Chin. J. Phys., 2016, 54, 931–939.
14. K. Li, H. Lian, M. Shang and J. Lin, Dalton Trans., 2015, 44, 20542–20550.
15. Y. Wang, J. Ding, Y. Li, L. Yang, X. Ding and Y. Wang, RSC Adv., 2016, 6, 42618–42626.
16. Y. Zhang, X. Li, K. Li, H. Lian, M. Shang and J. Lin, ACS Appl. Mater. Interfaces, 2015, 7, 2715–2725.
17. K. Kumar, A. K. Singh and S. B. Rai, Spectrochim. Acta, Part A, 2013, 102, 212–218.
18. Y. Zhang, J. Chen, C. Xu, Y. Li and H. J. Seo, Phys. B, 2015, 472, 6–10.
19. Y. He and D. Xue, J. Phys. Chem. C, 2007, 111, 13238–13243.
20. L. Li, W. Wang, Y. Pan, Y. Zhu, X. Liu, H. M. Noh, B. K. Moon, B. C. Choi and J. H. Jeong, RSC Adv., 2018, 8, 1191–1202.
21. L. Li, J. Cao, B. Viana, S. Xu and M. Peng, Inorg. Chem., 2017, 56, 6499–6506.
22. L. Li, Y. Pan, W. Wang, Y. Zhu, W. Zhang, H. Xu, L. Zhou and X. Liu, J. Alloys Compd., 2018, 731, 496–503.
23. R. Shi, B. Li, C. Liu and H. Liang, J. Phys. Chem. C, 2016, 120, 19365–19374.
24. Z. Babakhanova, M. Aripova and E. Bernardo, Glass Ceram., 2016, 73, 124–127.
25. R. A. Sa Ferreira, S. S. Nobre, C. M. Granadeiro, H. I. S. Nogueira, L. D. Carlos and O. L. Malta, J. Lumin., 2006, 121, 561–567.
26. M. Xie, Y. Li and R. Li, J. Lumin., 2013, 136, 303–306.
27. G. K. Behrh, R. Gautier, C. Latouche, S. Jobic and H. Serier-Brault, Inorg. Chem., 2016, 55, 9144–9146.
28. A. Strzep, A. Watras, K. Zawisza, P. Boutinaud and R. J. Wiglusz, Inorg. Chem., 2017, 56, 10914–10925.
29. D. Geng, G. Li, M. Shang, C. Peng, Y. Zhang, Z. Cheng and J. Lin, Dalton Trans., 2012, 41, 3078–3086.
30. S. Lv, Y. Wang, Z. Zhu, Z. You, J. Li, S. Gao, H. Wang and C. Tu, Appl. Phys. B: Lasers Opt., 2014, 116, 83–89.
31. J. Chen, W. Zhao, J. Wang and N. Wang, J. Mater. Sci.: Mater. Electron., 2016, 27, 237–244.
32. Y. Su, L. Peng, C. Du and X. Wang, J. Phys. Chem. C, 2012, 116, 15–21.
33. Y. Su, L. Peng, J. Guo, S. Huang, L. Lv and X. Wang, J. Phys. Chem. C, 2014, 118, 10728–10739.
34. R. V. Perrella, C. S. Nascimento Jr, M. S. Goes, E. Pecoraro, M. A. Schiavon, C. O. Paiva-Santos, H. Lima, M. A. Couto dos Santos, S. J. L. Ribeiro and J. L. Ferrari, Opt. Mater., 2016, 57, 45–55.
35. C. A. Kodaira, H. F. Brito, O. L. Malta and O. A. Serra, J. Lumin., 2003, 101, 11–21.
36. R. D. L. Gaspar, E. M. Rodrigues, I. O. Mazali and F. A. Sigoli, RSC Adv., 2013, 3, 2794.
37. Q. Zhang, X. Wang, X. Ding and Y. Wang, Inorg. Chem., 2017, 56, 6990–6998.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c8ra04432e

---