Optical Materials 66 (2017) 220e229
Contents lists available at ScienceDirect
Optical Materials journal homepage: www.elsevier.com/locate/optmat
Improvement of photoluminescence properties of Eu3þ doped SrNb2O6 phosphor by charge compensation Junpeng Xue a, Yue Guo a, Byung Kee Moon a, Sung Heum Park a, Jung Hyun Jeong a, *, Jung Hwan Kim b, Lili Wang c a b c
Department of Physics, Pukyong National University, Busan 608-737, Republic of Korea Department of Physics, Dongeui University, Busan 614-714, Republic of Korea Department of Chemistry and Pharmaceutical Science, Qingdao Agricultural University, Qingdao 266109, People's Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 November 2016 Received in revised form 8 January 2017 Accepted 3 February 2017
In this paper, a series of Eu3þ-doped SrNb2O6 phosphors have been successfully prepared by a hightemperature solid-state reaction technique. The as-prepared samples exhibit strong red emission peak at around 612 nm, which is attributed to the 5D0-7F2 transition of the Eu3þ ion. Both the emission intensity and color rendering effect can be obviously improved in SrNb2O6:Eu3þ phosphors by selfcompensation or co-doping with Liþ ions. Meanwhile, the decay time of phosphors can also be extended by charge compensation. The JuddeOfelt theory is used to calculate the optical transition strength parameters and quantum efficiencies of the obtained samples. In addition, Eu3þ and Liþ concentration-dependent excitation and emission spectra are investigated in detail. The critical distance is determined to be about 11.48 Å and the strongest red emission intensity is achieved in the Sr0.7Nb2O6:0.15Eu3þ,0.15Liþ phosphor. The CIE-1931coordinate (0.633, 0.366) of this sample is very close to that of the standard red light (0.67, 0.33). All of the results indicate that charge compensation approach can greatly improve the photoluminescence properties of Eu3þ-doped SrNb2O6 phosphors, which will further promote their applications in solid state lighting. © 2017 Elsevier B.V. All rights reserved.
Keywords: SrNb2O6 Photoluminescence Charge compensation Eu3þ
1. Introduction Rare-earth ions have gained extensive attention due to their abundant energy levels as well as unique and fascinating optical properties. Currently, rare-earth ions doped luminescent materials have been widely used in many aspects, such as displays, solidstate lighting, optical temperature sensors, biomarkers, optical heaters, drug carriers and photovoltaic devices [1e6]. In particular, rare-earth ions based phosphors, which could provide with good spectral characteristics featuring with high efficient luminescent performance, high stability, low energy consumption, long lifetime and environmentally friendly characteristics, are referred as the next generation of illumination source [7e9]. As one of the most frequently used red-emitting activators, Eu3þ ion is widely studied owing to its intense red remission at around 610e615 nm arising from the 5D0 / 7F2 transition [10]. Up to now, the Eu3þ-doped red
* Corresponding author. E-mail address:
[email protected] (J.H. Jeong). http://dx.doi.org/10.1016/j.optmat.2017.02.002 0925-3467/© 2017 Elsevier B.V. All rights reserved.
emitting phosphors have been found in vanadates [11], phosphates [12], borates [13], metasilicate [1,14], tungstates [15] and molybdates [16,17]. However, some available approaches, such as changing composition of host and adjusting the quantity of charger, are still needed to further improve the luminescent properties of Eu3þ ions to enhance their performance in the currently existing applications [11,18]. Among them, charge compensation is the most commonly used method because it can be easily realized. As is well known, the alkali metal ions, such as Liþ, Naþ and Kþ, which have low oxidation states and different particle radius, can significantly enhance the photoluminescent properties of rare-earth ions activated phosphors by co-doping methods [19,20]. In comparison with other alkali metal ions, the Liþ ions are widely employed to enhance the emission intensity of Eu3þ-doped materials, meanwhile, some impressive achievements have been obtained in some systems, such as Sr5(PO4)3F [21], Ca3(PO4)2 [22], Sr2CeO4 [23,24], CaMgSi2O6 [25], CaMoO4 [26] and ZnMoO4 [27]. On the other hand, self-compensation, which can produce vacancy artificially, is another method to achieve charge compensation.
J. Xue et al. / Optical Materials 66 (2017) 220e229
Nowadays, niobates are thought to be promising luminescent hosts due to their unique properties including high chemical stability, good mechanical performance, wide transparency range, commendable electro-optical, photoelastic and nonlinear properties, which endow them with important applications in the fields of photocatalytic technology, microwave resonators pyroelectric, electro-optic, ferroelectric and photorefractive devices [3,28]. Over the last decades, the luminescent behaviors of rare-earth ions based niobates, such as, CaNb2O6 [9], LaNbO4 [29], YNbO4 [30] and GdNbO4 [31], have been intensively studied. In comparison, the interest in the strontium metaniobate (SrNb2O6) is also increasing as a result of its suitability for pyroelectric, electro-optic, ferroelectric and photorefractive devices [32]. However, as far as we know, the research on photoluminescence (PL) and energy transfer behaviors of Eu3þ-doped SrNb2O6 phosphors are rarely reported. In present work, a series of Sr12xNb2O6:xEu3þ,xLiþ (x ¼ 0.00e0.175) phosphors were synthesized by the conventional high-temperature solid-state reaction method. The phase structure, morphology, lifetime and photoluminescent properties of the obtained samples were investigated in detail. Furthermore, the energy transfer mechanism between the Eu3þ was also systematically studied. 1.1. Sample preparation The Sr12xNb2O6:xEu3þ,xLiþ phosphors were prepared through a high-temperature solid-state reaction technique. According to the appropriate stoichiometric ratio, the starting materials, such as SrCO3 (GTSCIEN, 97.00%), Nb2O5 (DAEJUNG, 99.90%), Li2CO3 (ALDRICH, 99.997%) and Eu2O3 (Aladdin, 99.99%), were weighed and ground finely in an agate mortar for 30 min with an appropriate amount of ethanol. Then, the homogeneous mixture was kept in a crucible and sintered in a muffle furnace at 1250 C for 7 h. After cooled to the room temperature, the obtained white samples were ground to a fine powder for further characterization. 1.2. Characterization and optical measurements The X-ray diffraction (XRD) measurement was performed to verify the phase purity by a Philips X'Pert MPD (Philips, Netherlands) X-ray diffractometer at 40 kV and 30 mA. The diffraction patterns were scanned within an angular range of 10e70 (2q). The morphology and particle size of phosphors as well as the energy dispersive X-ray (EDX) spectrum were characterized using a scanning electron microscope (SEM) system (JSM-6490, JEOL Company). X-Ray photoelectron spectroscopy (XPS) measurements were performed on a PHI 5000 VersaProbe spectrometer using a monochromatic Al Ka radiation source. UVevis diffuse reflectance spectra (DRS) were collected by a V-670 (JASCO) UVevis spectrophotometer. The PL and PL excitation (PLE) spectra were recorded by a Photon Technology International (PTI, USA) fluorimeter with a 60 W Xe-arc lamp as the excitation light source. The quantum yields (QY) of the samples were measured with use of an integrating sphere and a FLS 920 fluorescence spectrophotometer. All measurements were performed at room temperature in air atmosphere. 2. Results and discussion 2.1. The phase formation and structure The typical SEM images of pure SrNb2O6, Sr0.85Nb2O6:0.15Eu3þ, Sr0.775Nb2O6:0.15Eu3þ and Sr0.7Nb2O6:0.15Eu3þ,0.15Liþ phosphors are shown in Fig. 1. It can be found that all the samples are made of irregular particles with average particle size around 5 mm, which
221
are caused by the inherent characteristics of the high-temperature solid-state method. It should be noted that the SrNb2O6, Sr0.85Nb2O6:0.15Eu3þ and Sr0.775Nb2O6:0.15Eu3þ phosphors show the similar morphology properties, while the Sr0.7Nb2O6:0.15Eu3þ, 0.15Liþ phosphor shows the agglomeration and coarse surface. The different charge compensation has an obvious impact on the surface and particle size of the resultant phosphors, especially alkali metal ions. In this work, the Li2CO3 does not only act as charge compensating agent but also serves as liquid flux. It can accelerate the grain growth of the particles to generate highly crystalline phosphor. Also, it will help the doped activators to penetrate deep inside the matrix and form homogeneous distribution in the matrix at high temperature [15]. The unit cell structure of SrNb2O6, which possesses an orthorhombic columbite structure (space group P21/c) with cell parameters of a ¼ 7.772 Å, b ¼ 5.592 Å, c ¼ 10.989 Å, V ¼ 474.51 Å3, N ¼ 4, is shown in Fig. 2. Based on the crystal structure, it is clear that there is only one cationic site with Wyckoff position 4e for activators to accommodate, while each cation has different coordination environments, such as the Sr atom, which is the center of hendecahedron, surrounded by eight oxygen atoms with the average distance of 2.605 Å, and the Nb atom, which is the center of octahedron, is surrounded by six oxygen atoms with the average distance of 1.998 Å. Taking into consideration of the ionic radii and charge balance, Eu3þ [r ¼ 1.066 Å for coordination number (CN) ¼ 8] ions are expected to substitute the Sr2þ ions (r ¼ 1.26 Å for CN ¼ 8) instead of the Nb5þ (r ¼ 0.64 Å for CN ¼ 6) ions. Thus, we can conclude that the Eu3þ ions would occupy the sites of Sr2þ ions in the SrNb2O6 host lattices. The structure of [NbO6SrO8NbO6] is the minimum unit cell, in which the SrO8 polyhedra and NbO6 octahedron share the common edge, while two Nb atoms are connected by sharing point, forming a 3D framework. Fig. 3 shows the XRD patterns of the representative samples of pure SrNb2O6, Sr0.85Nb2O6:0.15Eu3þ, Sr0.775Nb2O6:0.15Eu3þ, Sr0.7Nb2O6:0.15Eu3þ,0.15Liþ phosphors together with the standard pattern (JCPDS 28-1243) of SrNb2O6. As illustrated in Fig. 3(a), all the diffraction peaks of compounds are matched well with the standard pattern of SrNb2O6, indicating that the obtained samples are single phase and the Eu3þ ions are completely dissolved in the SrNb2O6 host lattices. From the zoomed XRD patterns (Fig. 3(b)), it is also observed that the XRD peaks move to larger angles with the addition of Eu3þ and Liþ ions. This is because the Sr2þ ions with larger ionic radius are replaced by the smaller Eu3þ and Liþ (0.92 Å) ions. Based on Bragg equation 2d sinq ¼ nl (here d is the interplanar distance, q stands for half diffraction angle, n represents the integer and l is the wavelength of X-ray), the interplanar distance would be shortened when the larger (Sr2þ) ions are substituted by smaller (Eu3þ, Liþ) ions, resulting in the shifting of the diffraction peaks. In order to further determine the structure of the obtained phosphor, the lattice constants of SrNb2O6 phosphors were refined by the GSAS software and the final consequence are showed in Fig. 4. The reliability factors of the refinement are wRp ¼ 3.72%, Rp ¼ 2.72%, and c2 ¼ 4.75, which indicated that the phosphor well possess pure orthorhombic phase without any impurity phases. Fig. 5 shows the EDX spectra of the samples modified with different charge compensation methods. Fig. 5(aec) show the EDX spectra of the Sr0.85Nb2O6:0.15Eu3þ, 3þ Sr0.775Nb2O6:0.15Eu and Sr0.7Nb2O6:0.15Eu3þ,0.15Liþ phosphors, respectively. The elements of Sr, Nb, O, and Eu could be clearly identified, whereas the Li could not be detected due to its light element property (see Fig. 5(aec)). The experimental weight percentages and calculated weight percentages of the elements in the samples are displayed in the inset of Fig. 5. Clearly, the experimental values are close to calculated values, indicating that Eu3þ-doped SrNb2O6 phosphors were successfully synthesized. To
222
J. Xue et al. / Optical Materials 66 (2017) 220e229
Fig. 1. SEM images of prepared phosphors (a) SrNb2O6; (b) Sr0.850Nb2O6:0.15Eu3þ; (c) Sr0.775Nb2O6:0.15Eu3þ; (d) Sr0.700Nb2O6:0.15Eu3þ,0.15Liþ.
Fig. 2. Crystal structure of SrNb2O6 and the Sr2þ and Nb5þ sites.
further investigate the surface compositions and chemical states of the as-prepared the Eu3þ-doped SrNb2O6 phosphors with different charge compensation methods, the XPS measurements were carried out and the corresponding results are showed in Fig. 5. No obvious peaks for impurities have been observed for all the samples. According to the XPS observations (Fig. 5(def)), Sr, Nb, O and Eu were detected in all the resultant compounds, while the Li was only found in the Sr0.7Nb2O6:0.15Eu3þ,0.15Liþsamples [31,33,34]. The C1s peak at around 284.6 eV can be attributed to the signal from carbon contained in the instrument which was used for calibration [35]. Inset show two strong peak in the high-resolution XPS spectra, are assigned to Li1s, which further confirms that the Li species in the SrNb2O6 is Liþ cation [36].
2.2. Spectrum analysis of Eu3þ-doped SrNb2O6 phosphors with different charge compensation methods Fig. 6 shows the PLE and PL spectra of the SrNb2O6:0.15Eu3þ/ 0.15Liþ phosphors. As presented in the PLE spectrum (monitored at 612 nm), there are two weak broad excitation bands centered at 266 and 328 nm, which are attributed to charge transfer band (CTB) arising from the transitions of Eu3þ/ O2 and O2 / Nb5þ, respectively [3,37]. The sharp lines at 363, 375, 393, 412 and 463 nm are ascribed to the 7F0 / 5D4, 7F0 / 5L7,7F0 / 5L6, 7F0 / 5 D3 and 7F0 / 5D2 transitions of Eu3þ ions, respectively [37,38]. Among these characteristic excitation peaks, the 7F0 / 5L6 (393 nm) transition exhibits the strongest intensity, suggesting that
J. Xue et al. / Optical Materials 66 (2017) 220e229
223
Fig. 3. (a) XRD patterns of SrNb2O6, Sr0.850Nb2O6:0.15Eu3þ, Sr0.775Nb2O6:0.15Eu3þ, Sr0.700Nb2O6:0.15Eu3þ,0.15Liþ phosphors and standard card (JCPDS# 28-1243) (b) Zoomed XRD patterns from 24 to 31.
Fig. 4. Rietveld refinement of XRD pattern of SrNb2O6 host.
the synthesized phosphors can be efficiently excited by NUV light. Under the irradiation of 393 nm light, the emission spectrum consists of the characteristic emissions of Eu3þ ions due to the 5D0 /7FJ (J ¼ 0, 1, 2, etc.) transitions (see Fig. 6). Among those emissions, the red emission at 612 nm corresponding to the forced electric dipole 5D0/7F2 transition of Eu3þ ions are strongest. As is well known, the 5D0 /7F2 transition is very sensitive to the local environment, while the 5D0 / 7F1 transition is not much affected by the crystal field around the Eu3þ ions. Therefore, the intensity ratio (R) between 612 nm emission and 594 nm emission is a method to measure the site of symmetry occupied by the rare-earth ion. In general, lower symmetry crystal field around Eu3þ ions results in higher value of R. In this work, the intensity ratio R of SrNb2O6:0.15Eu3þ,0.15Liþ is 2.56 (see Table 1), indicating that Eu3þ
ions occupy the sites without inversion symmetry [39]. Furthermore, the full width at half-maximum (FWHM) of the red emission at 612 nm (5D0 / 7F2) is quite small (~2.94 nm), which is superior to previous result (~6.49 nm) [39]. As a result, high color purity and excellent chromaticity coordinate properties can be achieved in this Eu3þ-doped SrNb2O6 phosphors. In addition, the position of the emission peaks does not have any obvious changes, expect the emission intensities are closely related to the excitation wavelengths, as shown in Fig. 6. It is known that charge compensation, including selfcompensation and co-doping with alkali metal ions, can improve the emission intensity of rare earth doped luminescent materials. To investigate the effect of the different charge compensation methods on luminescent properties of Eu3þ-doped SrNb2O6 phosphors, the UVeVis diffuse reflectance spectra are measured and the results are represented in Fig. 7. It is clear that all the samples show a remarkable drop in reflection around 300 nm, corresponding to the host absorption band of the SrNb2O6. However, when the Eu3þ ions are introduced into the SrNb2O6 host, some obvious absorption peaks are observed in the range of 300e600 nm, which are assigned to the 4f-4f absorption of the Eu3þ ions, coincide well with result from the PLE spectrum, as shown in Fig. 6. The absorption edge for self-compensation sample is located at the longer wavelength than that of the Eu3þ/Liþ co-doped phosphors. SrNb2O6 has a direct band gap of 3.20 eV with the O 2p states dominating the valence band, whereas the conduction band mainly consists of the Nb 4d orbitals [40]. In accordance with the report of Wood and Tauc [41], the energy band gap (Eg) can be roughly calculated by the following equation:
ahv ¼ A hv Eg
n=2
(1)
where a, hn, Eg, and A are the optical absorption coefficient, phonon energy, band gap and Planck's constant, respectively. Among them, n is associated with different types of transitions, n ¼ 1 and 4 correspond to direct and indirect absorption, respectively. According to Fig. 7, the Eg value for SrNb2O6 is calculated to be
224
J. Xue et al. / Optical Materials 66 (2017) 220e229
Fig. 5. EDX spectra of the (a) Sr0.85Nb2O6:0.15Eu3þ, (b) Sr0.775Nb2O6:0.15Eu3þ, and (c) Sr0.7Nb2O6:0.15Eu3þ,0.15Liþ phosphors. XPS survey spectrum of the (d) Sr0.85Nb2O6:0.15Eu3þ, (e) Sr0.775Nb2O6:0.15Eu3þ, (f) Sr0.7Nb2O6:0.15Eu3þ,0.15Liþ samples. The insets show the experimental weight percentages (Exp Wt%), calculated weight percentages (Cal Wt%) of the elements in the samples and the high resolution XPS spectra position of the Li doped SrNb2O6 phosphors.
approximately 3.94 eV. Compared with the band gap gained from DRS, the band structure calculation on a basis of density functional theory (DFT) is much smaller due to the well-known local density approximation (LDA) resulting in an underestimate of the band gaps in semiconductors [42]. The same phenomenon was also found in CdWO4 [43]. Fig. 8 shows the PL spectra of Eu3þ-doped SrNb2O6 phosphors obtained by different charge compensation methods. As displayed in Fig. 8, it can be found that the red emission intensity of the Eu3þ ions can be enhanced by both of these two charge compensation approaches. Meanwhile, compared with that of the samples without charge compensation, we find that the emission intensity of the phosphors co-doped with Liþ ions is increased by 4.23 times and the self-compensated products is increased by 2.97 times, suggesting that charge compensation through co-doping of Liþ ions is an effective approach to improve the emission intensity of Eu3þdoped SrNb2O6 phosphors. Similar phenomena were also observed in Eu3þ-doped ZnB2O4 and Sr3Ti2O7 phosphors [13,18]. The CIE chromaticity coordinates are calculated from the above PL spectra
and the corresponding results are labeled in Fig. 9. Meanwhile, the intensity ratio R of SrNb2O6:0.15Eu3þ phosphors with different charge compensation methods are showed in Table 1. It can be seen that the phosphors co-doped with Liþ ions possess lower symmetry crystal field around Eu3þ ions. Clearly, under 393 nm excitation, the Eu3þ-doped phosphors can exhibit red emission and color rendering effect is close to the standard red light by charge compensation [11,20]. To further study the charge compensation behavior, the luminescence dynamics of the SrNb2O6:0.15Eu3þ samples monitoring at 612 nm emission peak under the excitation of 393 nm are investigated. The luminescence decay curves are depicted in Fig. 10. It shows that these decay curves can be well fitted with the following double exponential decay equation [16]:
IðtÞ ¼ A1 expðt=t1 Þ þ A2 expðt=t2 Þ þ I0
(2)
where I0 and I(t) are the emission intensity at certain time 0 and t, respectively. A1 and A2 are constants. t1 and t2 stand for two
J. Xue et al. / Optical Materials 66 (2017) 220e229
Fig. 6. PL (lex ¼ 328, 393, 463 nm) and PLE (lem ¼ 612) spectra of Sr0.7Nb2O6:0.15Eu3þ,0.15Liþ phosphors.
225
Fig. 8. PL spectra of Eu3þ-doped SrNb2O6 phosphors with different charge compensation approaches excited at 393 nm.
Table 1 CIE chromaticity coordinates, 5D0 / 7F2 relative emission intensity and the intensity ratio R with charge compensation approaches. Point
Samples
CIE coordinate
Emission intensity
R
a1 a2 a3
Sr0.850Nb2O6:0.15Eu3þ Sr0.775Nb2O6:0.15Eu3þ Sr0.700Nb2O6:0.15Eu3þ,0.15Liþ
(0.603,0.391) (0.619,0.378) (0.632,0.366)
1.00 2.97 4.23
2.46 1.72 2.56
The intensity of Sr0.85Nb2O6:0.15Eu3þ is regarded as 1.0.
t* ¼ A1 t21 þ A2 t22
Fig. 7. UVevis DRS of SrNb2O6;Sr0.85Nb2O6:0.15Eu3þ; Sr0.775Nb2O6:0.15Eu3þ; Sr0.7Nb2O6:0.15Eu3þ,0.15Liþ phosphors. Inset shows the band gap of pure SrNb2O6.
components of the luminescence lifetime, which correspond to the fast and slow lifetimes for exponential components, respectively, indicating that the decay time can be related to two different probabilities of radiative decay [44]. The fast component is mainly attributed to the Eu3þ ions on the surface of the particles, whereas the slow component corresponds to the Eu3þ ions in the inner side of particles [44,45]. The average decay times (t*) can be expressed by the formula (3) [46,47]:
. ðA1 t1 þ A2 t2 Þ
(3)
From these decay curves, the fluorescent lifetimes are calculated and listed in Table 2. The values of decay times for Sr0.85Nb2O6:0.15Eu3þ, Sr0.775Nb2O6:0.15Eu3þ, Sr0.7Nb2O6:0.15Eu3þ, 0.15Liþ phosphors are determined to be 180.09, 434.50 and 883.72 ms, respectively, indicating the lifetime of the samples can be prolonged by charge compensation. As demonstrated in Table 2, the lifetime for the Eu3þ-doped materials can be significantly influenced by the host materials. For SrNb2O6 (this work), CaNb2O6 [9], CaMoO4 [16], LaNbO4 [29], YNbO4 [30], GdNbO4 [31] and Ca9Gd(VO4)7 [48], the lifetime is in microseconds, whereas the decay times for K2Y(WO4) (PO4) [49] and CaYAlO4 [50] are in milliseconds. Therefore, it is reasonable to conclude that the decay time of Eu3þ ions based materials is greatly dependent on the host materials owing to their various refractive indices and local fields [51,52]. Therefore, we can deduce that the charge compensation is an effective approach to strengthen the red emission intensity of Eu3þ ions as well as its fluorescent lifetime. In order to comprehensively study the effect of the charge compensation behavior on the photoluminescence properties, the Judd-Ofelt (J-O) theory is applied to calculate optical transition strength parameters, such as U2, U4 and U6 [53,54]. Here, we will use some formulas to calculate experimental parameters by analyzing the PL emission spectrum of Eu3þ-doped SrNb2O6 phosphors. For Eu3þ ions, 5D0 / 7FJ (J ¼ 2,4,6) are electric-dipole transitions, while 5D0 / 7F1 is magnetic-dipole transition. Based on the J-O theory [53,54], the magnetic-dipole spontaneous emission probability Amd is expressed as
226
J. Xue et al. / Optical Materials 66 (2017) 220e229 Table 2 The decay time of Eu3þ-doped SrNb2O6 and some other reported phosphors. Sample
Decay time (ms)
Reference
Sr0.850Nb2O6:0.15Eu3þ Sr0.775Nb2O6:0.15Eu3þ Sr0.700Nb2O6:0.15Eu3þ,0.15Liþ CaNb2O6:0.12Eu3þ CaMoO4:0.05Eu3þ LaNbO4:0.04Eu3þ YNbO4:0.05Eu3þ GdNbO4:0.05Eu3þ Ca9Gd(VO4)7:0.03Eu3þ K2Y(WO4) (PO4):0.07Eu3þ CaYAlO4:0.01Eu3þ
180.09 434.50 883.72 636.00 455.00 690.00 740.00 799.00 76.00 2382.00 1570.00
This work This work This work [9] [16] [29] [30] [31] [48] [49] [50]
a constant and independent of hosts. The electric-dipole spontaneous emission probability AJ is expressed as
D E X 64p4 e2 KJ3 n n2 þ 2 2 l 0 0 2 AJ ¼ U j J j J ∪ l 3hð2J 0 þ 1Þ 9
(5)
l¼2;4; 6
where e is elementary charge, KJ is average energy of 5D0 / 7FJ (J ¼ 2,4,6) transition, Ul is intensity parameters of transition, hjJ U J j0 J 0 i2 is the square of reduced matrix elements. On the basis of the square of reduced matrix elements, as presented in Table 3, as well as Eq. (5), the following equation is given: Fig. 9. CIE chromaticity diagram for SrNb2O6:Eu3þ samples with different charge compensation approaches.
E2 D 64p4 e2 KJ3 n n2 þ 2 2 UJ jJ U J j0 J 0 AJ ¼ 0 3hð2J þ 1Þ 9
(6)
According to the J-O theory, the transition rate of energy level is proportional with integral strength of emission spectrum, so we can get the formula, as shown below:
Z AJ ¼Z Amd
IJ ðyÞdy
¼
Imd ðyÞdy
2 3 E D e2 KJ n2 þ 2 J 0 0 2 U j J j J ∪ J Smd K 3 9n2 md (7)
Fig. 10. Decay curves of SrNb2O6:Eu3þ samples with different charge compensation approaches.
Amd ¼
3 64p4 Kmd n3 Smd 3hð2J 0 þ 1Þ
(4)
where n, h, Kmd, J0 and Smd are refractive index of phosphors, Planck's constant, average energy of 5D0 / 7F1 transition, the total angular momentum of initial transition state and the magneticdipole line strength, respectively. Among them, the Smd of Eu3þ is
Combining with the formula (7), we can calculate the UJ by utilizing the presented values, as given in Tables 3 and 4. However, U6 cannot be calculated due to 5D0 / 7F6 transition is too weak to observe in the emission spectrum of Eu3þand it lies in infrared region. Generally speaking, the intensity parameter U2 has been regard as a probe to detect the asymmetry behaviors of the rareearth ion sites and the degree of structural order, U4 and U6 are associated with the bulk properties and the rigidity of the host [55,56]. Table 5 shows the U2 and U4 parameters with different charge compensation approaches. It is evident that with different charge compensation methods, U2 value is different and the asymmetrical environment around the Eu3þ ions is connected with charge compensation technique. The maximum U2 exists in phosphors co-doped with Liþ ions, which indicates that 0.15Eu3þ/ 0.15Liþ co-doped SrNb2O6 phosphor has the weakest asymmetrical environment around the Eu3þ ions. However, the U2 of the selfcompensated phosphors is the smallest and the generation of
Table 3 The square of reduced matrix elements of Eu3þ ions. Matrix [U [U [U
(2) 2
] ] ]
(4) 2 (6) 2
5
D0 / 7F2
5
D0 / 7F4
5
D0 / 7F6
0.0032 0.0000 0.0000
0.0000 0.0023 0.0000
0.0000 0.0000 0.0002
J. Xue et al. / Optical Materials 66 (2017) 220e229 Table 4 Average energy of 5D0 /7FJ transitions of Eu3þ ions at room temperature. Transitions of Eu3þ ions
Central wavelength
Average energy
5
594 nm 612 nm 706 nm
16835.02 16339.87 14164.31
D0 / 7F1 5 D0 / 7F2 5 D0 / 7F4
Sr2þ vacancy maybe responsible for this phenomenon. The variation of U2 are well coincided with the intensity ratio R with different charge compensation methods. The value of U4 is always smaller than that of U2, which suggests that the sensitivity of 5D0 / 7 F4 transition for testing the asymmetry around the Eu3þ ions is much lower than the 5D0 / 7F2 transition. The PL emission spectrum and fluorescence lifetime of Eu3þdoped SrNb2O6 phosphors can also be used to calculate quantum efficiency through the following expressions [53,54]:
1
t
¼ AR þ AN
(8)
h ¼ AR =ðAR þ AN Þ X
AR ¼
(9)
AJ
(10)
J¼0;1;2;3;4
where t is the decay time of 5D0 energy level, AR is radiative transition rate, AN is non-radiative transition rate and h is quantum efficiency. Among them, AR is the sum of 5D0 / 7FJ transition rate and the decay time of 5D0 energy level can be obtained under 393 nm excitation. Combining the formulas (8), (9) and (10), the quantum efficiency is expressed as:
h¼t
X
AJ
(11)
J¼0;1;2;3;4
The quantum efficiency with different charge compensation methods are depicted in Table 5. It can be seen that the quantum efficiency of phosphors co-doped with 0.15Liþ ions reaches 52.05%, the self-compensated phosphor is 17.81% and the phosphors without charge compensation is as low as 10.33%, indicating that charge compensation can improve the quantum efficiency, especially 0.15Eu3þ/0.15Liþ co-doped SrNb2O6 phosphors. The emergence of this phenomenon is not accidental and this case can be well explained with the formation of Eu3þ aggregation or clusters. As for Eu3þ-doped Sr0.85Nb2O6 phosphors, one Sr2þ ion will be superseded by one Eu3þ ion in crystal lattice inducing an extra positive charge. Along with the augment of the substitution amount of Eu3þ ions for Sr2þ ions, these extra positive charges will form repelling force and hinder the further substitution, namely, some Eu3þ ions cannot enter into the sites of Sr2þ ions. Therefore, these Eu3þ ions, which stay out of the host lattices, may form clusters or aggregate together on account of electrostatic effect, leading to the energy quenching as well as weak luminescent performance. For Sr0.775Nb2O6:0.15Eu3þ (self-compensated compound), the decrement of the mole number of Sr2þ ions can relieve the charge unbalance and makes more Eu3þ ions enter into host
227
lattices. Thus, it displays a better luminescence performance than uncompensated one. However, there still exists some shortcomings in self-compensated phosphors, for example, the generation of the Sr2þ vacancy because of the inequality substitution. That is, three Sr2þ ions are replaced by two Eu3þ ions and it maybe enhance the asymmetrical environment around the Eu3þ sites. In comparison, the introduction of Liþ ions can not only maintain charge balance, but also solve the problem of the number of moles. The charge compensation of Liþ ions can minimize the quantity of these aggregated Eu3þ ions and reduce the asymmetrical environment around the Eu3þ sites, which is beneficial to increase the emission intensity of the phosphors. Hence, the Liþ ions compensated samples exhibit the best photoluminescence performance [11,13,19]. More Eu3þ enter into the host and then increase the energy transfer rate between the luminescence center, which implies that charge compensation behavior can improve the quantum efficiency. Also, the emission intensity of phosphors is related to the radiant velocity and fluorescent lifetime. For the same luminescent center, the radiant velocity can be treated as a constant, so the fluorescent lifetime is proportional with the emission intensity of phosphors [13,16]. According to the aforementioned theory, we found that the charge compensation plays an important role in enhancing emission intensity of Eu3þ ions, extending the decay time, changing the asymmetrical environment around the Eu3þ sites as well as improving quantum efficiency. 2.3. Eu3þ/Liþ concentration dependent photoluminescence emissions of SrNb2O6 phosphors Upon 393 nm excitation, the PL emission spectra of SrNb2O6:xEu3þ, xLiþ phosphors with different concentrations are shown in Fig. 11. It can be found that the emission spectra exhibit no distinct difference except that emission intensity increases gradually with the increase of the Eu3þ ion concentration, reaching a maximum value at x ¼ 0.15, after that it begins to decrease with further increasing the Eu3þ ions concentration owing to the
Fig. 11. PL (lex ¼ 393 nm) spectra of Sr12xNb2O6:xLiþ,xEu3þ (x ¼ 0.05, 0.075, 0.1, 0.125, 0.15 and 0.175) phosphors with different Eu3þ ion concentrations. The inset shows the relationship between the emission intensity and Eu3þ ion concentration.
Table 5 J-O parameters, transition rates, the reciprocal of fluorescence lifetime and quantum efficiency of samples with different charge compensation. Samples 3þ
Sr0.850Nb2O6:0.15Eu Sr0.775Nb2O6:0.15Eu3þ Sr0.700Nb2O6:0.15Eu3þ,0.15Liþ
U2/1020 cm2
U4/1020 cm2
AR/s1
AN þ AR/s1
h/%
3.19 2.03 3.24
2.22 1.43 2.42
573.51 409.73 588.93
5555.75 2301.50 1131.58
10.33 17.81 52.05
228
J. Xue et al. / Optical Materials 66 (2017) 220e229
concentration quenching effect. Generally, the concentration quenching is due to the energy migration among the activators at the high concentrations and the excitation energy will be lost at quenching site during the energy migration process, resulting in the decrease of PL emission intensity. In order to approximately evaluate the energy transfer mechanism between Eu3þ ions, it is necessary to know the critical distance (Rc) between Eu3þ ions in the host lattices. The Rc can be figured out by the following expression (4) [57]:
Rc ¼ 2½3V=4pXc N1=3
(12)
where V is the volume per unit cell, Xc is the critical concentration of Eu3þ ions, N represents the number of host cations in the unit cell. For the SrNb2O6 host, N ¼ 4, V ¼ 474.51 Å3 and Xc is 0.15 for Eu3þ ions. Therefore, the critical distance (Rc) was calculated to be about 11.48 Å. As is known to all that if the energy transfer via exchange interaction, the range of critical distance is strictly 5e8 Å [58]. Since 11.48 Å is much larger than 8 Å, energy transfer between Eu3þ ions in the SrNb2O6 host is not exchange interaction. As a result, it can be inferred that the electric multipole interaction contributes to the non-radiation concentration quenching between Eu3þ luminescent centers. Table 6 shows that CIE chromaticity coordinates of Sr12xNb2O6:xEu3þ,xLiþ with different concentrations and the best CIE chromaticity coordinates is x ¼ 0.633, y ¼ 0.366 when the x value is 0.15. Furthermore, to identify the luminescent efficiency of the synthesized samples, the quantum efficiency (QE) of the Sr12xNb2O6:xEu3þ,xLiþ phosphors with optimal doping concentration was recorded. The QE of the Sr0.7Nb2O6:0.15Eu3þ,0.15Liþ sample was measured by the integrated sphere method at room temperature under the excitation of 393 nm. Clearly, the QE of Sr0.7Nb2O6:0.15Eu3þ,0.15Liþ phosphors was determined to be 27.2% under the excitation of 393 nm. Compared with other red-emitting phosphors, as displayed in Table 7, it is found that the QE of Sr0.7Nb2O6:0.15Eu3þ,0.15Liþ is lower than that of Y2O3:Eu3þ, Sr2Si5N8:Eu2þ and Ba2Tb(BO3)2Cl:Eu2þ, Eu3þ, while it is comparable with that of commercial Y2O2S:Eu3þ red-emitting phosphors and much better than that of Ba5(BO3)2(B2O5):Sm3þ,Liþ. Note that, the experimental value (27.2%) is lower than the calculated value (52.05%) and the
Table 6 CIE chromaticity coordinates of Sr1-2xNb2O6:xLiþ,xEu3þ phosphors with different concentrations. Points Samples
b1 b2 b3 b4 a3 b5
Excitation wavelength CIE coordinates
Sr0.90Nb2O6:0.050Eu3þ,0.050Liþ Sr0.85Nb2O6:0.075Eu3þ,0.075Liþ Sr0.80Nb2O6:0.100Eu3þ,0.100Liþ Sr0.75Nb2O6:0.125Eu3þ,0.125Liþ Sr0.70Nb2O6:0.150Eu3þ,0.150Liþ Sr0.65Nb2O6:0.175Eu3þ,0.175Liþ
393 393 393 393 393 393
nm nm nm nm nm nm
x
y
0.620 0.623 0.625 0.627 0.633 0.628
0.377 0.375 0.373 0.372 0.366 0.370
Table 7 The QE values of Sr0.7Nb2O6:0.15Eu3þ,0.15Liþ and some other reported red-emitting phosphors. Sample
QE
CIE coordinates
Reference
Sr0.7Nb2O6:0.15Eu3þ,0.15Liþ Y2O3:Eu3þ Y2O2S:Eu3þ Sr2Si5N8:Eu2þ Ba2Tb(BO3)2Cl:Eu2þ,Eu3þ Ba5(BO3)2(B2O5):Sm3þ, Liþ
27.2% 72.5% 35.0% 79.0% 56.0% 16.0%
Red Red Red Red Orange orange-red
This work [59] [60] [61] [62] [63]
Fig. 12. Energy level scheme for all observed bands of the Eu3þ-doped SrNb2O6 phosphors.
reasons are needed to further investigated in the future work. Although the values are not very high, the QE can be further improved through modifying the particle surfaces, compositions or codoping with other elements [64e66]. The energy level diagram of Eu3þ ions in SrNb2O6 host lattices is presented in Fig. 12. Under the excitation of 393 nm NUV light, electrons are promoted from the ground state (7F0) to the excited state (5L6), and then the nonradiation (NR) process takes place, as a result, the 5D0 level is populated. Finally, the red (612 nm), orange (594) emissions are observed corresponding respectively to the radiative 5D0 / 7F2 and 5 D0 / 7F1 transitions of Eu3þ. 3. Conclusions In this work, the Eu3þ-doped SrNb2O6 phosphors with different charge compensation methods have been fabricated by conventional high-temperature solid-state reaction method. Their PL properties, charge compensation mechanism and luminescence mechanism were investigated in detail. Different charge compensation methods can obviously enhance the emission intensity, change the optical transition strength parameters (U2 and U4) and improve the quantum efficiency. Compared with Eu3þ ions single doped SrNb2O6 sample, the co-doping with Liþ ions can increase the emission intensity about 4.23 times while self-compensation route enhances ahout 2.5 times. The J-O theory results suggest that the 0.15Eu3þ/0.15Liþ co-doped SrNb2O6 phosphors possess the weakest asymmetrical environment around the Eu3þ ions and highest quantum efficiency. Moreover, charge compensation can also improve the color rendering effect of Eu3þ ions as well as prolong the decay time. From the Eu3þ and Liþ concentrationdependent emission spectra, the critical distance (Rc) was determined to be about 11.48 Å and the strongest red emission is found in Sr0.7Nb2O6:0.15Eu3þ,0.15Liþ phosphor. The CIE-1931 chromaticity coordinate (0.633, 0.366) of this sample was very close to that of the standard red light (0.67, 0.33). These results suggest that charge compensation method is an efficient method to improve the luminescence properties of Eu3þ-doped SrNb2O6 phosphors, especially the introducing of Liþ ions, and they may have potential applications for white LEDs as red-emitting phosphors. Acknowledgment This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning (No.
J. Xue et al. / Optical Materials 66 (2017) 220e229
2015060315). The Eu3þ doped SrNb2O6 phosphors were supplied by the Display and Lighting Phosphor Bank at Pukyong National University. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32]
H. Dong, L.D. Sun, C.H. Yan, Chem. Soc. Rev. 44 (2015) 1608e1634. X. Wang, Q. Liu, Y. Bu, C. Liu, T. Liu, X. Yan, RSC Adv. 5 (2015) 86219e86236. P. Du, L. Luo, H.-K. Park, J.S. Yu, Chem. Eng. J. 306 (2016) 840e848. D.K. Chatterjee, M.K. Gnanasammandhan, Y. Zhang, Small 6 (2010) 2781e2795. S.V. Eliseeva, J.-C.G. Buenzli, Chem. Soc. Rev. 39 (2010) 189e227. B.M. van der Ende, L. Aarts, A. Meijerink, Phys. Chem. Chem. Phys. 11 (2009) 11081e11095. J. Zhou, Z. Xia, J. Mater. Chem. C 3 (2015) 7552e7560. H. Zhou, Q. Wang, Y. Jin, J. Mater. Chem. C 3 (2015) 11151e11162. K. Li, X. Liu, Y. Zhang, X. Li, H. Lian, J. Lin, Inorg. Chem. 54 (2015) 323e333. S. Yao, L. Chen, Y. Huang, W. Li, Mater. Sci. Semicond. Process 41 (2016) 265e269. X. He, M. Guan, Z. Li, T. Shang, N. Lian, Q. Zhou, J. Am. Ceram. Soc. 94 (2011) 2483e2488. Z. Fu, X. Wang, Y. Yang, Z. Wu, D. Duan, X. Fu, Dalton Trans. 43 (2014) 2819e2827. Z. Mu, Y. Hu, L. Chen, X. Wang, Opt. Mater. (Amst) 34 (2011) 89e94. X. Zhang, L. Zhou, Q. Pang, J. Shi, M. Gong, J. Phys. Chem. C 118 (2014) 7591e7598. F. Hong, L. Zhou, L. Li, Q. Xia, X. Luo, Opt. Commun. 316 (2014) 206e210. Z. Hou, R. Chai, M. Zhang, C. Zhang, P. Chong, Z. Xu, G. Li, J. Lin, Langmuir 25 (2009) 12340e12348. L. Wu, X. Tian, K. Deng, G. Liu, M. Yin, Opt. Mater. (Amst) 45 (2015) 28e31. L. Zhang, B. Sun, Q. Liu, N. Ding, H. Yang, L. Wang, Q. Zhang, J. Alloys Compd. 657 (2016) 27e31. F. Mo, L. Zhou, Q. Pang, Y. Lan, Z. Liang, Curr. Appl. Phys. 13 (2013) 331e335. F. Zhang, T. Zhang, G. Li, W. Zhang, J. Alloys Compd. 618 (2015) 484e487. K.N. Shinde, S.J. Dhoble, Adv. Mater. Lett. 1 (2010) 254e258. I.M. Nagpure, S. Saha, S.J. Dhoble, J. Lumin. 129 (2009) 898e905. Z. Mu, Y. Hu, L. Chen, X. Wang, J. Electrochem. Soc. 158 (2011) 287e290. R. Ghildiyal, P. Page, K.V.R. Murthy, J. Lumin. 124 (2007) 217e220. R. Wang, Appl. Mech. Mater. 57 (2011) 413e417. X. Wang, C. Liu, Z. Liang, G. Peng, X. Han, Adv. Mater. Res. 654 (2013) 563e566. W. Ran, L. Wang, L. Tan, D. Qu, J. Shi, Sci. Rep. 6 (2016) 27657. M. Pastor, S. Goenka, S. Maiti, K. Biswas, I. Manna, Ceram. Int. 36 (2010) 1041e1045. K. Li, Y. Zhang, X. Li, M. Shang, H. Lian, J. Lin, Phys. Chem. Chem. Phys. 17 (2015) 4283e4292. X. Liu, Y. Lu, C. Chen, S. Luo, Y. Zeng, X. Zhang, J. Phys. Chem. C 118 (2014) 27516e27524. M. Yang, X. Zhao, Y. Ji, F. Liu, W. Liu, J. Sun, X. Liu, New J. Chem. 38 (2014) 4249e4257. I.-S. Cho, S. Lee, J.H. Noh, D.W. Kim, H.S. Jung, D.-W. Kim, K.S. Hong, Cryst. Growth Des. 10 (2010) 2447e2450.
229
[33] S. Liang, X. Wang, Y. Chen, J. Zhu, Y. Zhang, X. Wang, Z. Li, L. Wu, Nanoscale 2 (2010) 2262e2268. [34] J. Chen, Y. Liu, H. Liu, H. Ding, M. Fang, Z. Huang, Opt. Mater. (Amst) 42 (2015) 80e86. [35] Z. He, Y. Shi, C. Gao, L. Wen, J. Chen, S. Song, J. Phys. Chem. C 118 (2013) 389e398. [36] L. Suo, Y. Hu, H. Li, M. Armand, L. Chen, Nat. Commun. 4 (2013) 1e9. [37] J. Zhou, F. Huang, J. Xu, H. Chen, Y. Wang, J. Mater. Chem. C 3 (2015) 3023e3028. [38] Y. Kumar, M. Pal, M. Herrera, X. Mathew, Opt. Mater. (Amst) 60 (2016) 159e168. [39] Y. Guo, B. Kee Moon, S. Heum Park, J. Hyun Jeong, J. Hwan Kim, K. Jang, R. Yu, J. Lumin. 167 (2015) 381e385. [40] I.S. Cho, S.T. Bae, D.H. Kim, K.S. Hong, J. Hydrogen Energy 35 (2010) 12954e12960. [41] D.L. Wood, J. Tauc, Phys. Rev. B 5 (1972) 3144e3151. [42] J.P. Perdew, M. Levy, Phys. Rev. Lett. 51 (1983) 1884e1887. [43] L. Wang, B.K. Moon, S.H. Park, J.H. Kim, J. Shi, K.H. Kim, J.H. Jeong, New J. Chem. 40 (2016) 3552e3560. € rger, L.M. Goldenberg, [44] T. Ninjbadgar, G. Garnweitner, Alexander Bo O.V. Sakhno, J. Stumpe, Adv. Funct. Mater. 19 (2009) 1819e1825. [45] Y. Tian, H. Yang, K. Li, X. Jin, J. Mater. Chem. 22 (2012) 22510e22516. [46] F. Kang, Y. Zhang, M. Peng, Inorg. Chem. 54 (2015) 1462e1473. [47] Z. Xia, D. Chen, J. Am. Ceram. Soc. 1401 (2010) 1397e1401. [48] R.V. Perrella, C.S. Nascimento Júnior, M.S. Goes, E. Pecoraro, M.A. Schiavon, C.O. Paiva-santos, H. Lima, M.A. Couto, S.J.L. Ribeiro, J.L. Ferrari, Opt. Mater. (Amst) 57 (2016) 45e55. [49] R.S. Meltzer, S.P. Feofilov, B. Tissue, H.B. Yuan, Phys. Rev. B 60 (1999) 12e15. [50] C. Duan, M.F. Reid, Curr. Appl. Phys. 6 (2006) 348e350. [51] L. Li, H.M. Noh, B.K. Moon, J.H. Jeong, B.C. Choi, X. Liu, Opt. Express. 4 (2014) 3883e3888. [52] X. Zhang, M. Chen, J. Zhang, X. Qin, M. Gong, Mater. Res. Bull. 73 (2016) 219e225. [53] B.R. Judd, Phys. Rev. 127 (1962) 750e761. [54] G.S. Ofelt, J. Chem. Phys. 37 (1962) 511e520. [55] A. Balakrishna, D. Rajesh, Y.C. Ratnakaram, J. Lumin. 132 (2012) 2984e2991. [56] X. Wang, C. Liu, T. Yu, X. Yan, Phys. Chem. Chem. Phys. 16 (2014) 13440e13446. [57] G. Blasse, Phys. Lett. A 28 (1968) 444e445. [58] G. Blasse, B. Grabmaier, Luminescent Materials, Springer-Verlag, Berlin, Heidelberg, 1994. [59] D. Hou, H. Liang, M. Xie, X. Ding, J. Zhong, Q. Su, Y. Tao, Y. Huang, Z. Gao, Opt. Express 19 (2011) 11071e11083. [60] C. Zhang, H. Liang, S. Zhang, C. Liu, D. Hou, L. Zhou, G. Zhang, J. Shi, J. Phys. Chem. C 116 (2011) 15932e15937. [61] Z. Tao, Y. Huang, H.J. Seo, Dalton Trans. 42 (2013) 2121e2129. [62] P.S. Dutta, A. Khanna, ECS J. Solid. State. Sci. Technol. 2 (2013) R3153eR3167. [63] J. Zheng, Q. Cheng, W. Chen, Z. Guo, C. Chen, ECS J. Solid. State. Sci. Technol. 4 (2015) 72e77. [64] D.L. Dexter, J.H. Schulman, J. Chem. Phys. 22 (1954) 1063e1070. [65] V. Bachmann, C. Ronda, O. Oeckler, W. Schnick, A. Meijerink, Chem. Mater. 21 (2009) 316e325. [66] Y. Wu, Y. Nien, Y. Wang, I. Chen, J. Am. Ceram. Soc. 95 (2012) 1360e1366.