Eu3+, Li+-codoped ZnO:Zn phosphor with intense Eu3+ emissions upon indirect excitation of near-UV light has been synthesized under reducing condition. Steady-state and time-resolved photoluminescence and diffuse reflectance spectra are measured to investigate properties of the luminescence. The results suggest that there exists prominent energy transfer from ZnO host to Eu3+ ions. A series of energy levels as temporary storage of excitation energy play a crucial role on this energy transfer process. Two kinds of Eu3+ sites in Eu3+, Li+-codoped ZnO:Zn are distinguished based on the emission and excitation spectra in comparison with pure Eu2O3.
©2008 Optical Society of America
Wurtzite zinc oxide, a direct band semiconductor with wide band-gap (~3.3eV at room temperature) has attracted much attention as a luminescent material for both fundamental research and applications. When fired in a reducing atmosphere, ZnO with oxygen vacancies (usually denoted as ZnO:Zn) can give off efficient green emission that is of interest for the applications on flat panel displays [1-3]. Trivalent rare earth ions (RE3+) doped ZnO phosphors have been studied extensively during the last two decades [4-9], however, the results are quite disappointing, in that the desired characteristic emissions from RE3+ are so weak under the indirect UV excitation of the ZnO host. To achieve the efficient luminescence of ZnO:RE3+ phosphor, enhancing energy transfer (ET) from the semiconductor host to the RE3+ ions is required. The verification of the mechanism behind the ET process is also an important issue.
The relatively large radius of RE3+ and charge mismatching between the RE3+ and divalent Zn2+ ions are blamed to the unsuccessful incorporation of RE3+ into ZnO matrix, hence the inefficient ET is often observed. Lithium and RE3+-codoped phosphors exhibit more intense RE3+ emissions than singly doped RE3+ samples, resulted from the enhancement of solubility for RE3+ inserting the host  or other mechanisms [11-13]. S. Bachir et al. reported that the ET between ZnO host and RE3+ occurred only in the case of Tm3+-doped ZnO and explained the ET by a two-step mechanism involving electron-hole recombination via Tm2+ . W. Jia et al. sintered the sample ZnO:1% EuF3, Li in N2 and found efficient ET from ZnO to Eu3+ evidenced by the enhanced absorption intensity ratio of UV band around 380 nm to f-f transitions of RE3+ in the excitation spectrum .
In comparison with the two methods frequently used to obtain the white light: one is the combination of GaN-based blue light-emitting diode (LED) chip with YAG:Ce3+ yellow phosphor, the other is the mixing of three red-green-blue (RGB) individual LEDs to generate white light, the method involved in InxGa1-xN-based blue and near-UV LED systems incorporating down-converting luminescent phosphors can provide relatively satisfied white light with both high color rendering and luminous efficacy. To achieve the high-performance white LEDs as illuminating sources, many endeavors have been made to improve the color-rendering-index, external quantum efficiency and luminous efficiency. Alternative phosphors to conventional materials for white light generation with strong absorption of n-UV light, high color purity and stability, hence the high luminescence efficiency is expected to be developed.
In this contribution, we report the successful synthesis of Eu3+, Li+-codoped ZnO:Zn phosphor that is suitable candidate for white LED pumped by near-UV (380-410 nm) light. The prominent ET from ZnO host to RE3+ ions is observed. The possible mechanisms are discussed based on steady state and time-resolved spectra and diffuse reflectance data.
Un-doped and Eu3+, Li+-codoped ZnO:Zn samples are fabricated through solid-state reaction technique at 1100°C under reducing atmospheres for 3 hours. Mixed powders of analytical grade (99.9%) raw materials of ZnO, Li2CO3, and Eu2O3 (Beijing Chemicals Co. Ltd.) are carefully ground for 30 minutes, keeping the molar ratios of Eu3+ and Li+ equally in each sample, and then sintered in muffle using an alumina crucible as a container. While in reduced condition, a bigger crucible is filled with carbon sticks surrounding a smaller crucible containing the sample. X-ray diffraction (XRD) patterns are recorded by a Rigaku D/Max 2500V PC diffractometer operated at 18 kW with Cu Kα radiation (λ = 1.5406Å). The Li contents of the samples are determined using the inductively coupled plasma-optical emission spectrometer (ICP-OES) (Thermo Scientific iCAP 6300). Photoluminescence (PL), photoluminescence excitation (PLE), and diffuse reflectance spectra are measured using a Spectra-fluorometer (Hitachi F-4500). For time resolved PL measurements, the third (355 nm) harmonic of a Nd-YAG laser (Spectra-Physics, GCR 130) is used as the excitation source. Signal is detected by a Tektronix digital oscilloscope (TDS 3052).
3. Results and discussion
Figure 1 shows the X-ray diffraction patterns of undoped ZnO:Zn (c) and Eu3+, Li+-codoped samples with various doping concentrations: (a) 5at %, (b) 1at %, respectively. All the diffraction peaks of the undoped ZnO:Zn and codoped sample of 1% are well indexed as the hexagonal wurtzite ZnO (JCPDS card No. 36-1451). The impurity-related diffraction peaks (asterisk-indicated in curve (a)) indicates that new phases occur when the doping concentration increases as high as 5%. The slightly shift of diffraction peaks toward smaller angles with the doping concentration growing comes from the increase of the interplain distance, which is mainly due to the larger radius of Eu3+ (0.095 nm) than that of Zn2+ (0.074 nm). We argue that small amount of rare earth ions are most likely occupying the sites of Zn ions. However, the diffraction peak shift is not the sufficient evidence for this deduction.
The PL and PLE spectra of Eu3+, Li+-codoped ZnO:Zn are shown in Fig. 2. The relative spectra of pure Eu2O3 powders are plotted together for comparison in the lower part of the profile. A broad green emission band centered at 520 nm (curve a), known as self-activated (SA) emission related to defects is observed in the PL spectra of codoped sample when excited under 390, 410 and 270 nm, respectively. The mechanism behind this green emission is not well established yet . Among the mechanisms proposed [3,14-17], oxygen vacancies-caused emission is commonly accepted [3,16,17]. In addition to the broad green emission band, strong emissions from the Eu3+ dominated by the 5D0-7F2 transitions are also observed under the various excitation wavelengths, indicating that the Eu3+ ions are located at the lower symmetric sites. It is worth noting that the peak positions of 5D0-7F2 transitions varies with different excitations: the peak of the transitions is at 615 nm upon indirect near-UV pumping at 390 or 410 nm but changes to 609 nm when directly exciting 5D2 level by 466 nm (see curve a, b and d in Fig. 2). This suggests that Eu3+ ions are located at different sites in which the transitions of 5D0 level to the preferred splitting levels of 7F2 state are different due to the crystal-fields around Eu3+. One can assume that there are two kinds of Eu3+ in the Eu3+, Li+-codoped ZnO:Zn sample: one is inside the lattices of ZnO host and the other is at interstitial sites at the boundary. These different sites for Eu3+ are further confirmed by the PLE spectra when monitored at 609 nm and 615 nm, respectively. In the former case, the PLE spectrum (curve χ in Fig. 2) features the broad absorption band centered at 270 nm, which is attributed to the O2- → Eu3+ charge transfer (CT) transition, and has about 10 nm blue-shift compared to the PLE spectrum of Eu2O3 monitored at the same wavelength (curve δ in Fig. 2). In addition to this CT band, an intense absorption peak of Eu3+ 5D2 level at 466 nm and a broad band centered at around 390 nm are also observed. In the latter, monitored at 615 nm, the CT band is absent in curve β, showing the difference from the PLE spectrum monitored at 609 nm. The curve β also presents a strong excitation peak around 390 nm with full width at half maximum (FWHM) of ~40 nm but a relatively weak absorption peak for 5D2 level of doped Eu3+. The reason behind this difference is that the efficient absorption of direct-band ZnO host dominates over the direct excitation of RE3+ ions inside the host lattices. In view of the band gap absorption of ZnO (~3.3eV) at room temperature and the spectra monitored at 520 and 615 nm (curves α and β in Fig. 2), we are further convinced that there exists a prominent ET process between the ZnO host to the RE3+ activators which are believed to occupy two kinds of sites respectively in the Eu3+, Li+-codoped sample.
Moreover, the absorption peaks marked with downward arrows manifested in curve δ are almost absent in the PLE spectra (curves β and χ in Fig. 2) of Eu3+, Li+-codoped ZnO:Zn samples. Considering the 10 nm shifts between the abovementioned CT bands in curves χ and δ, one can deduce that Eu3+ ions located at boundary sites in co-doped sample are subjected to quite different environment from those in pure Eu2O3 powder. Eu3+ ions possess numerous energy levels in the 370-420 nm range with several discrete peaks as shown in the PLE spectrum of Eu2O3 (curve δ in Fig. 2). However, in the corresponding range for Eu3+, Li+-codoped ZnO:Zn, only a broad band around 390 nm appears when monitored 5D0-7F2 transitions (curves β and χ). Such excitation peaks at ~390 nm must be ascribed to the absorption of ZnO host instead of rare earth ions.
It is noteworthy that the broadening of the excitation peaks centered at ~390 nm makes the application possible for the phosphors in near-UV excited white LED. The left pattern of Fig. 3 presents the diffuse reflectance spectra of ZnO, ZnO:Zn, Eu3+,Li+-codoped ZnO and Eu3+,Li+-codoped ZnO:Zn samples, respectively. The broadening effects is also observed when Eu3+ and Li+ ions are introduced into the semiconductors, and the codoped sample sintered in the reducing condition shows stronger broadening effects compared to that in ambient. It has been reported that enhancement of luminescence can be achieved by Li+ -doping based on some reasonable mechanisms [10-13], and the lithium prefers to occupy the interstitial sites acting as shallow donors in ZnO phosphors [18-20]. By means of the ICP-OES method, Li contents in the samples codoped with 5at % and 1at % Eu3+, Li+ can be determined to be 0.90% and 0.12%, respectively, showing the same decreasing tendency after sintered at higher temperature . Therefore, the enhanced red luminescence from RE3+ intraconfigurational transitions and the excitation broadening from the ZnO host can be explained by the Eu3+, Li+ incorporation and the formation of a series of shallow donor levels due to the reducing condition. The reduced atmosphere increases the quantity of oxygen vacancies and the incorporation of activators, hence followed by the increase of the lattice perturbation. Also, Li+ ions are most likely to be reduced to Li atoms, leading to the preferable occupation of interstitial sites. All these factors result in a series of energy levels near below the conduction band as temporary excitation energy storage. The right pattern in Fig. 3 shows another evidence of our arguments aforementioned, in which the diffuse reflectance spectra of Eu3+,Li+-codoped ZnO:Zn and Li+-free ZnO:Zn,Eu3+ samples are presented. The red shift of the reflectance edge in the lower curve indicates that the introduction of Li+ can enhance the absorption of near-UV light, supporting our interpretation based on the formation of shallow donor levels in the semiconductor host.
In our current work, ET is observed from the semiconductor host to the RE3+ no mater where they are located (inside the lattices or at the interstitial sites of grain boundary). Thus, the shallow donor levels originated from reduction condition play a crucial role as temporary energy storage which transfers the excitation energy to the luminescence centers yielding the green broad emission and characteristic sharp lines for Eu3+. Figure 4 depicts intensity decays of undoped ZnO:Zn and Eu3+,Li+-codoped ZnO:Zn samples. The lifetimes of broad green emission (λem=500 nm) and Eu3+ red emission (λem=615 nm) are obtained by fitting the decay curves with two exponential components, and their values are listed in Table 1. It is important to mention that after codoping RE3+ and Li+ as activators and compensators, the lifetimes of green emission become longer (from τ1= 0.2 μs, τ2= 1.8 μs to τ1= 0.6 μs, τ2= 4.3 μs), indicating that the electrons can also be trapped by the shallow donor levels then subsequently released with radiation through recombination with the holes trapped by defects like oxygen vacancies. In addition, these arguments can well explain the red shift of green emissions from 500 nm in ZnO:Zn (data not shown here but well-documented elsewhere) to 517 nm in Eu3+,Li+-codoped sample (see curve a in Fig. 2). The green emission of ZnO:Zn arises from the recombination of intrinsic shallower donors or electrons in conduction band, whereas in the case of Eu3+,Li+-codoped ZnO:Zn sample the green emission is from the recombination of the shallow donors created by extrinsic impurities like interstitial Li with the holes trapped in deep centers like oxygen vacancies.
In summary, we report the synthesis of Eu3+, Li+-codoped ZnO:Zn phosphor as a promising candidate for potential application in near-UV white LEDs. Both broad green band and intense sharp emissions from 4f-4f transitions are observed under the indirect excitation of near-UV light. Energy transfer occurs between the ZnO host to RE3+ located at different sites of semiconductor matrix. The ET process is enhanced due to the formation of shallow donor levels that act as temporary energy storage for excited electrons.
This work is supported by the National Natural Science Foundation of China under Grant No. 10504031, 10574128 and the MOST of China (2006CB601104, 2006AA03A138).
References and links
1. S. Shionoya and W. M. Yen, eds., Phosphor Handbook (CRC Press, Boca Raton, 1999)
2. R. N. Bhargava, V. Chhabra, T. Som, A. Ekimov, and N. Taskar, “Quantum confined atoms of doped ZnO nanocrystals,” Phys. Status Solidi B 229, 897–901 (2002). [CrossRef]
3. A. van Dijken, E. A. Meulenkamp, D. Vanmaekelbergh, and A. Meijerink, “The kinetics of the radiative and nonradiative processes in nanocrystalline ZnO particles upon photoexcitation,” J. Phys. Chem. B. 104, 1715–1723 (2002). [CrossRef]
4. S. Bachir, K. Azuma, J. Kossanyi, P. Valat, and J. C. Ronfard-Haret, “Photoluminescence of polycrystalline zinc oxide co-activated with trivalent rare earth ions and lithium. Insertion of rare-earth ions into zinc oxide,” J. Lumin. 75, 35–49 (1997). [CrossRef]
5. J. Kossanyi, D. Kouyate, J. Pouliquen, J. C. Ronfard-Haret, P. Valat, D. Oelkrug, U. Mammel, G. P. Kelly, and F. Wilkinson, “Photoluminescence of semiconducting zinc oxide containing rare earth ions as impurities,” J. Lumin. 46, 17–24 (1990). [CrossRef]
6. Y. K. Park, J. I. Han, M. G. Kwak, H. Yang, S. H. Ju, and W. S. Cho, “Effect of coupling structure of Eu on the photoluminescent characteristics for ZnO:EuCl3 phosphors,” Appl. Phys. Letts. 72, 668–670 (1998). [CrossRef]
7. W. Jia, K. Monge, and F. Fernandez, “Energy transfer from the host to Eu3+ in ZnO,” Opt. Mater. 23, 27–32 (2003). [CrossRef]
8. S. A. M. Lima, F. A. Sigoli, M. R. Davolos, and M. Jafelicci Jr., “Europium(III)-containing zinc oxide from Pechini method,” J. Alloy. Compd. 344, 280–284 (2002). [CrossRef]
9. D. Kouyate, J.-C. Ronfard, and J. Kossanyi, “Photo- and electro-luminescence of rare earth-doped semiconducting zinc oxide electrodes: Emission from both the dopant and the support,” J. Lumin. 50, 205–210 (1991). [CrossRef]
10. S. M. Yeh and C. S. Su, “Mixing LiF in Gd2O3:Eu to enhance ultraviolet radiation induced thermoluminescent sensitivity after sintering process,” Mater. Sci. Eng. B. 38, 245–249 (1996). [CrossRef]
11. H. Zhang, X. Fu, S. Niu, G. Sun, and Q. Xin, “Luminescence properties of Li+ doped nanosized SnO2:Eu,” J. Lumin. 115, 7–12 (2005). [CrossRef]
12. O. A. Lopez, J. McKittrick, and L. E. Shea, “Fluorescence properties of polycrystalline Tm3+-activated Y3Al5O12 and Tm3+-Li+ co-activated Y3Al5O12 in the visible and near IR ranges,” J. Lumin. 71, 1–11 (1997). [CrossRef]
13. F. Gu, S. F. Wang, M. K. Lü, G. J. Zhou, D. Xu, and D. R. Yuan, “Structure Evaluation and Highly Enhanced Luminescence of Dy3+-Doped ZnO Nanocrystals by Li+ Doping via Combustion Method,” Langmuir. 20, 3528–3531 (2004). [CrossRef]
14. M. Liu, A. H. Kitai, and P. Mascher, “Point defects and luminescence centres in zinc oxide and zinc oxide doped with manganese,” J. Lumin. 54, 35–42 (1992). [CrossRef]
15. R. Dingle, “Luminescent transitions associated with divalent copper impurities and the green emission from semiconducting zinc oxide,” Phys. Rev. Lett. 23, 579–581 (1969). [CrossRef]
16. F. A. Kroger and H. J. Vink, “The origin of the fluorescence in self-activated ZnS, CdS, and ZnO,” J. Chem. Phys. 22, 250–252 (1954). [CrossRef]
17. K. Vanheusden, W. L. Warren, C. H. Seager, D. R. Tallant, J. A. Voigt, and B. E. Gnade, “Mechanisms behind green photoluminescence in ZnO phosphor powders,” J. Appl. Phys. 79, 7983–7990 (1996). [CrossRef]
18. C. H. Park, S. B. Zhang, and S. H. Wei, “Origin of p-type doping difficulty in ZnO: The impurity perspective,” Phys. Rev. B. 66, 073202-1-3 (2002). [CrossRef]
19. S. B. Orlinskii, J. Schmidt, P. G. Baranov, D. M. Hofmann, C. de M. Donega, and A. Merjerink, “Probing the wave function of shallow Li and Na donors in ZnO nanoparticles,” Phys. Rev. Lett. 92, 047603-1-4 (2004). [CrossRef] [PubMed]
20. M. G. Wardle, J. P. Goss, and P. R. Briddon, “Theory of Li in ZnO,A limitation for Li-based p-type doping,” Phys. Rev. B. 71, 155205-1-10 (2005). [CrossRef]
21. S. H. Byeon, M. G. Ko, J. C. Park, and D. K. Kim, “Low-temperature crystallization and highly enhanced photoluminescence of Gd2-xYxO3:Eu3+ by Li doping,” Chem. Mater. 14, 603–608 (2002). [CrossRef]