In this work, the persistent luminescence mechanisms of Tb3+ (in CdSiO3) and Eu2+ (in BaAl2O4) based on solid experimental data are compared. The photoluminescence spectroscopy shows the different nature of the inter- and intraconfigurational transitions for Eu2+ and Tb3+, respectively. The electron is the charge carrier in both mechanisms, implying the presence of electron acceptor defects. The preliminary structural analysis shows a free space in CdSiO3 able to accommodate interstitial oxide ions needed by charge compensation during the initial preparation. The subsequent annealing removes this oxide leaving behind an electron trap. Despite the low band gap energy for CdSiO3, determined with synchrotron radiation UV-VUV excitation spectroscopy of Tb3+, the persistent luminescence from Tb3+ is observed only with UV irradiation. The need of high excitation energy is due to the position of 7F6 level deep below the bottom of the conduction band, as determined with the 4f8→4f75d1 and the ligand-to-metal charge-transfer transitions. Finally, the persistent luminescence mechanisms are constructed and, despite the differences, the mechanisms for Tb3+ and Eu2+ proved to be rather similar. This similarity confirms the solidity of the interpretation of experimental data for the Eu2+ doped persistent luminescence materials and encourages the use of similar models for other persistent luminescence materials.
© 2012 OSA
Persistent luminescence materials can emit light for several hours after the removal of the irradiation source. These materials have received special attention lately due to their significant applications in emergency signalization, micro defect sensing, optoelectronics for image storage, detectors of high energy radiation and thermal sensors [1,2]. The persistent luminescence phenomenon is considered a special case of thermally stimulated luminescence, where the excitation energy is stored to traps and then released, induced by thermal energy available at room temperature . The advance in the persistent luminescence materials research arose with the discovery of the phosphor SrAl2O4:Eu2+,Dy3+  in 1996. This breakthrough was followed by the finding of new efficient persistent luminescence materials based on different stable aluminates CaAl2O4:Eu2+,Nd3+ , Sr4Al14O25:Eu2+,Dy3+  and silicate materials like Sr2MgSi2O7:Eu2+,Dy3+ . Presently, the best materials can continue emitting light in excess of 24 h in the dark.
One may observe that the majority of published works are about Eu2+ doped materials. Despite this majority, other materials containing different dopants, such as Tb3+ [5–8], Eu3+ [9,10], Mn2+ [11,12] or Ti [13,14] also show efficient persistent luminescence. Among these materials, those doped with Tb3+ ions perform the best though only very recently the first persistent luminescence mechanism based on solid experimental data for a Tb3+ doped material was developed .
In order to improve the luminescence intensity and duration of the persistent phosphors it is important to know profoundly the mechanism of the phenomenon. The mechanism of Eu2+ persistent luminescence has attracted much attention lately [1,2,15] and is close to achieving general acceptance though some minor details are still contested . Comparing the Eu2+ persistent luminescence materials with those doped with different dopants e.g. Tb3+ may still contribute significantly to the understanding of the persistent luminescence mechanisms.
In this work, the persistent luminescence mechanisms for Eu2+ (in BaAl2O4) [1,16] and Tb3+ (in CdSiO3)  are discussed. Both mechanisms were developed based on the experimental results derived from photoluminescence, synchrotron radiation (SR) UV-VUV spectroscopy and thermoluminescence measurements. The positions of the ground state of the R2+ and R3+ ions (R: La-Nd, Sm-Lu) in the hosts’ band structure were determined using the 4f→5d and the ligand-to-metal charge-transfer (LMCT) transitions following the ideas of the previous empirical model .
2.1 Materials preparation
The BaAl2O4:Eu2+,Dy3+ materials were prepared via combustion synthesis, where the metal nitrates and urea were used as reactants and fuel, respectively. The precursors were dissolved into the smallest possible amount of distilled water. A silica capsule filled with the homogeneous solution was inserted into a furnace pre-heated at 500 °C [16,18–20]. The reaction began after ca. 5 minutes after the introduction of the capsule into the furnace. The mixture was then self-ignited with a white flame and produced a white powder. After the completion of the reaction, the furnace was turned off and was allowed to cool freely. The products were removed from the oven when the temperature had decreased to ca. 25 °C. The nominal concentrations of Eu2+ and Dy3+ were 1 and 2 mole-% (of the Ba2+ amount), respectively.
The polycrystalline CdSiO3:R3+ (R: Eu and Tb) materials were prepared with a conventional solid state reaction. Cadmium acetate (Cd(CH3COO)2⋅2H2O), fumed silica (SiO2) and rare earth nitrates (R(NO3)3·6H2O) in stoichiometric amounts were ground intimately. The mixtures were then heated in air at 950 °C for seven hours in aluminosilicate crucibles. The nominal concentration of R3+ (of the Cd2+ amount) was 1 mole-%.
For both materials, the europium, terbium and dysprosium nitrates used for doping were obtained from the respective oxides with a reaction with concentrated nitric acid.
The photoluminescence measurements on the CdSiO3:Tb3+ and BaAl2O4:Eu2+,Dy3+ materials were carried out at room temperature with a SPEX-Fluorolog-2 spectrofluorometer equipped with two 0.22 m SPEX 1680 double grating monochromators. A 450 W Xenon lamp was used as the excitation source. The excitation, emission and persistent luminescence spectra were collected at an angle of 22.5 ° (front face). All spectra were recorded using automatic detector mode correction.
The time-resolved excitation spectra of the BaAl2O4:Eu2+,Dy3+ material were recorded with a SPEX 1934D phosphorimeter accessory coupled to the SPEX-Fluorolog-2 spectrofluorometer. A 50 W pulsed Xenon lamp was used as the excitation source and the spectra were recorded with a 1 ms delay.
The UV-VUV excitation spectra of the CdSiO3:R3+ (R: Eu, Tb) and BaAl2O4:Eu2+,Dy3+ materials were recorded between 80 and 330 nm by using the UV-VUV synchrotron radiation facility at the SUPERLUMI beamline I of HASYLAB (Hamburger Synchrotronstrahlungs-labor) at DESY (Deutsches Elektronen-Synchrotron, Hamburg, Germany) . The polycrystalline materials were mounted on the cold finger of a liquid He flow cryostat and the spectra were recorded at selected temperatures between 10 and 298 K. The setup consisted of a 2-m McPherson type primary (excitation) monochromator with a resolution up to 0.02 nm. The UV-VUV excitation spectra were corrected for the variation in the incident flux of the excitation beam using the excitation spectrum of sodium salicylate as a standard.
3. Results and discussion
3.1 Photoluminescent properties
The emission spectrum of CdSiO3:Tb3+ was obtained under excitation in the host at 247 nm at room temperature (Fig. 1 , left). The spectrum is composed almost exclusively of the line emission due to the intraconfigurational 4f8 transitions of Tb3+, 5D3→7FJ and 5D4→7F6-2. The most intense transition of Tb3+, 5D4→7F5, peaks at ca. 545 nm and yields the green emission color. For practical applications, this is close to the optimal sensitivity range of the human eye though at low illumination conditions blue emission might be even better. Due to the low site symmetry (C1) of the three Cd2+ sites occupied by Tb3+ in CdSiO3 (and to the charge compensation defects), the appearance of the Tb3+ emission is more or less band-like due to many superimposed Tb3+ emission lines. It has been reported  that the CdSiO3:Tb3+ emission, as many of CdSiO3 doped with the R3+ ions with less strong luminescence in visible (e.g. Nd3+ and Gd3+ just to mention a few), contains also a broad band centered at ca. 400 nm. In the present investigation, this band emission is either very weak or entirely absent, indicating that the energy transfer process from the host to Tb3+ is effective. Furthermore, the almost total absence of the 5D3→7FJ transitions from CdSiO3:Tb3+ in the blue (and near UV) range is rather unusual taken into account the low dopant concentration, only 1 mole-%. This may be interpreted in a way that the emission from the 5D3 level(s) is quenched and the excitation energy ends to the 5D4 level(s) instead. The absence of the 5D3 emission indicates that the 5D3 levels are located in the conduction band (CB) of CdSiO3 and the 5D3 emission is quenched to the lowest emitting levels, 5D4 via the host’s CB. The 77 K emission spectrum shows that the 5D3 emission is completely quenched at low temperatures, in agreement with the position of 5D3 close to the CB bottom.
The emission spectrum of BaAl2O4:Eu2+,Dy3+ (Fig. 1, right) presents a broad band assigned to the interconfigurational 4f65d1→4f7 transition of Eu2+. Two maxima, peaking at 435 and 500 nm, were observed due to the transitions from the lowest 2D level of the excited 4f65d1 configuration to the ground 8S7/2 level of the 4f7 configuration of Eu2+. The existence of two bands can be due to the presence of two Ba2+ sites in the BaAl2O4 structure or to the creation of a new Ba2+ site due to the effect of water exposure on BaAl2O4:Eu2+ .
The persistent emission spectra, measured 120 s after ceasing the irradiation, are very similar to the conventional ones. For the Tb3+ material, the 5D3 persistent emission is even less intense than the conventional one (when compared to the 5D4 one) but the differences are too small to justify a different quenching mechanism. The Tb3+ persistent luminescence is also observed under excitation on the 5D3 level (378 nm), indicating that this level is indeed located in host’s CB. In the case of the Eu2+ doped material, the only difference is the absence of the 435 nm band in the persistent luminescence spectrum, indicating that probably one of the Ba2+ sites does not participate on the persistent luminescence process.
One of the most critical points in the proposed mechanisms of persistent luminescence for Eu2+  is the transfer of the charge (electron) from the 4f65d1 levels to the host’s conduction band: both the energetics (position of the 4f65d1 levels vis-à-vis CB) and the time of transfer (transition probability) seem to matter. The lifetime of the Laporte allowed though spin-forbidden 4f65d1→4f7 emission of Eu2+ is ca. 1 µs, which seems to be largely sufficient to allow efficient electron transfer to CB. The case with the Tb3+ doped materials is somewhat similar to the Eu2+ doped ones: the excitation occurs to the 4f75d1 levels which are located, at least partially, in host’s CB. This requires a host with low band gap energy, however, because the 7F6 ground level of Tb3+ is considerably farther away from the host’s CB than the 8S7/2 ground level of Eu2+. To perhaps balance this disadvantage, the lifetime of the Laporte forbidden 5D4→7FJ transitions is much longer, in the range of several ms, thus enabling the capture of electrons to CB in a manner even more efficient than for the Eu2+ ion.
3.2 Trap nature and structure
The persistent luminescence phenomenon occurs due to the presence of charge carriers (electrons or holes) trapped in defects in the host structure. In the mechanisms studied, it is suggested that the electron (e-) is the charge carrier, implying the presence of electron acceptor defects (e.g. oxygen vacancy) . The creation of the oxide vacancies can occur in both cases studied by the evaporation of metal oxides (MO, M: Cd, Ba) at the high preparation temperatures together with a cation vacancy (). Furthermore, the lattice defects are created by the charge compensation necessary when a R3+ ion replaces a M2+ host cation (), either creating a metal vacancy or an interstitial oxide ion . All symbols are according to the Kröger-Vink notation . In the Eu2+ doped materials, R3+ co-doping frequently increases the persistent luminescence efficiency, since R3+ may also act as traps. However, in the Tb3+ doped CdSiO3, there is no need for co-doping because Tb3+ acts both as a luminescence center and a creator of the energy storing defects.
When studying more closely the CdSiO3 structure  as presented in the DIAMOND  view (Fig. 2 ), it is observed that the Cd-Cd distance between the terminal Cd ions belonging to different three-membered ribbons (only two shown) is ca. 5.2 Å whilst within the ribbons composed of three CdO6 octahedra the distances are rather short, ca. 3.4 Å. There is thus a considerable free space that may accommodate an interstitial oxide ion needed to compensate the charge (Tb3+ vs. Cd2+) mismatch. In fact, the distance of 5.2 Å is just sufficient to form two Tb-O bonds of the length 2.5-2.6 Å. The position of the charge compensation defect immediately adjacent to the two Tb3+ ions () gives a rather infrequent clear possibility for clustering of the defects as follows: - - . This should be studied further with e.g. synchrotron radiation EXAFS methods.
The interstitial oxide may well be accommodated in the regular 2b (½, ½, ½) site in the space group P21/c, yet to be studied by DFT (Density Functional Theory) calculations, giving the Tb-O distances of 2.54 Å. This value is few tenths of Å longer than the optimal bond distance of Tb-O but well within the bond range found for the Tb-O bond. However, this yields one short (2.1 Å) O-O distance between and one oxygen in the SiO4 groups, much shorter than the O-O distances in the SiO4 group (2.56 – 2.80 Å) in e.g. CdSiO3. Because of this increased oxygen-oxygen repulsion the interstitial oxide may have reduced stability, and may be replaced by trapped electrons and thus may have an important role to play in the energy storing in the persistent luminescence. An analogous situation has been reported  to arise in the cuprate superconducting phases. Taken into account that the crystallographic position of the interstitial oxide is only an approximate one, the Tb-O and O-O distances may well be modified with the optimization of the atomic positions. This optimization is being carried out with the DFT calculations but is out of the scope of the present work.
Similar free spaces may probably exist in other hosts as MAl2O4 accepting interstitial oxygens when co-doped with R3+. The analysis of their structure can help the understanding of the effect of the co-dopants in the persistent luminescence of Eu2+ doped materials, too. Furthermore, the reduction mechanism of Eu3+ to Eu2+ without the use of reducing atmosphere proposed for the BaAl2O4 material  involves the release of . Therefore, the reduced stability of may be related to this spontaneous reduction, and further studies with other hosts should be carried out.
3.3 Band gap structure
The Synchrotron Radiation (SR) excitation spectra of the materials (Fig. 3 ) were measured at 10 K in order to avoid the persistent luminescence which may smudge the excitation spectrum, at least at room temperature. A sharp edge is observed at ca. 235 nm (5.28 eV) in the excitation spectra of the Tb3+ doped CdSiO3. The edge is the excitation from the top of the valence (VB) to the bottom of the conduction band, i.e. the band gap energy (Eg). In the excitation spectra of the Eu2+ doped BaAl2O4, the corresponding edge is at a much higher energy, at ca. 190 nm (6.5 eV). As assumed above, the band gap should be smaller for the Tb3+ doped materials to enable persistent luminescence at all or, at least, to let the persistence to be excited with low energy UV radiation (or blue light) and thus improve the efficiency. However, despite the low band gap energy, the persistent luminescence from Tb3+ in CdSiO3 is observed only with UV irradiation.
One may observe in the UV-VUV excitation spectrum of CdSiO3:Tb3+ (Fig. 3) the weakness of the Tb3+ 4f8→4f75d1 transition bands even at 10 K indicating the easy delocalization of electrons to the conduction band of CdSiO3. This was originally shown for the Ce3+ doped Y3Al5O12 (YAG) single crystals . The 4fn→4fn-15d1 transitions can thus lose their superior intensity relative to the 4f-4f transitions when they are within the host’s CB. The 4f75d1 levels of Tb3+ are thus probably entirely inside CB, whilst the 4f65d1 levels of Eu2+ are only partially inside the CB since some Eu2+ 4f7→4f65d1 transition bands have rather high intensity.
In order to determine the R2+/3+ (R: La-Nd, Sm-Lu) 4f ground state positions in the hosts’ band structure, it is important to start with their LMCT transitions . The most accessible CT transition in the whole Rn+ doped CdSiO3 system is for CdSiO3:Eu3+. This information is obtained by the SR UV-VUV excitation spectrum of Eu3+ (Fig. 4 , left), where a broad band is observed between 230 to 270 nm (5.4 to 4.7 eV, respectively). The band was assigned to the LMCT O2-(2p)→Eu3+ transition. The CT energy gives the position of the Eu2+ ground state in the band gap (4.7 eV above the top of the host's VB). The energy difference between the ground levels of the same rare earth but in a different oxidation state, i.e. Eu2+ and Eu3+, depends on the host. This difference is quite high for a free ion (18 eV) but is lowered in solid state, e.g. in fluorides, oxides and sulfides down to >7, 6-7 and <6 eV, respectively . For the Eu2+/Eu3+ pair in CdSiO3, a value of 6.1 eV was used for this energy difference due to the softness of the CdSiO3 host. Once the position of the ground level of Eu3+ is known, following the host independent evolution of the R3+ energy levels , it is possible to determine the position of the ground levels of all other R3+’s. For example, the energy of the Tb3+ ground level (7F6) is at 3.5 eV above that of Eu3+ (7F0). The Tb3+ (7F6) ground level position in CdSiO3 was thus found at ca. 2.1 eV above the VB top. The position of the 7F6 ground level deep in the band gap explains the presence of persistent luminescence only with UV irradiation.
For the BaAl2O4 system, the trivalent Eu3+ doped material is not easily obtained, since the spontaneous reduction from Eu3+ to Eu2+ occurs even in non-reducing atmospheres [16,31]. Furthermore, the 4f-4f transitions of Eu3+ are absent in the steady-state spectra due to non-radiative processes (quenching) by Eu2+, whose 4f65d1→4f7 transitions are allowed, having a much shorter lifetime (ca. 1 μs) than the 4f→4f transitions of Eu3+ (several ms) and thus larger transition probability. Consequently, in order to determine the Eu3+ LMCT in BaAl2O4, the time-resolved excitation spectra of BaAl2O4:Eu2+,Dy3+ were measured at 77 K with a delay of 1 ms to avoid both the persistent luminescence and the Eu2+ emission (Fig. 4, right). A broad band is observed between 220 to 300 nm (5.6 to 4.1 eV) readily assigned to the Eu3+ O2-(2p)→Eu3+ LMCT transition, though overlapped by several Eu3+ 7F0→2S+1LJ bands. The higher energy position of the Eu2+ ground state in the BaAl2O4 band gap, as in the majority of the hosts  allows the excitation of persistent luminescence with blue light even in wide band gap hosts.
3.4 The mechanisms of persistent luminescence
Once the collection of vital information about the energy level systems in both the Tb3+ doped CdSiO3 and the Eu2+ doped BaAl2O4 was completed, the mechanisms of the persistent luminescence could be constructed. As a first observation, despite some differences in the photoluminescence processes, band gap energy and ground state positions of the luminescence centers, the mechanisms for Tb3+ and Eu2+ are quite similar. The duration of the persistent luminescence is comparable in both cases, longer than 1 h.
In general, the charging mechanisms for both the Eu2+ and Tb3+ doped materials consider that (Fig. 5 ): i) under irradiation of the material, in addition to conventional luminescence, some electrons escape to the host’s conduction band, along with a simultaneous formation of the pairs Eu2+−h+ or Tb3+−h+; ii) some of these electrons are trapped from CB to the defects created by charge compensation and the release of the unstable interstitial oxide, thus storing part of the excitation energy. The reverse decharging process of freeing the electrons from the traps to the excited states via CB precedes the radiative relaxation of the system back to the ground states of Eu2+ or Tb3+ via the excited states 4f65d1 (Eu2+) or 5D3,4 (Tb3+) − thus creating the persistent luminescence.
In closing the present work, plausible mechanisms coherent with experimental evidence could be constructed for the persistent luminescence from both Eu2+ and Tb3+. The mechanisms were found rather similar, despite many seemingly different basic properties of the systems, including the nature of the inter- and intraconfigurational transitions. In both systems, the electrons were found as the charge carriers and the nature of the traps is thus similar (electron traps). Nevertheless, the R3+ co-dopants may act in different ways in the two systems. For CdSiO3, the structural analysis showed that the interstitial oxide ions created by charge compensation can be accommodated in the free space. The low stability of these interstitial oxide ions can cause their substitution by trapped electrons thus acting as electron traps creators. More profound structural analyses should be carried out to find if the possible free space for exists in other hosts as MAl2O4. Only sparse data is available about the general properties for the Tb3+ doped persistent luminescence materials, thus more detailed studies are needed on these systems, too.
Financial support is acknowledged from the Coimbra Group, Turku University Foundation, Academy of Finland (Finland, contract #137333/2010), FAPESP, CAPES, CNPq, inct-INAMI and Nanobiotec-Brasil RH-INAMI (all Brazil). The synchrotron radiation studies were supported by the European Community – Research Infrastructure Action under the FP6 Structuring the European Research Area Programme, RII3-CT-2004-506008 (IA-SFS). Dr. Aleksei Kotlov (HASYLAB) is gratefully acknowledged for his assistance during the synchrotron measurements.
References and links
1. T. Aitasalo, J. Hölsä, H. Jungner, M. Lastusaari, and J. Niittykoski, “Thermoluminescence study of persistent luminescence materials: Eu2+- and R3+-doped calcium aluminates, CaAl2O4:Eu2+,R3+.,” J. Phys. Chem. B 110(10), 4589–4598 (2006). [CrossRef]
2. T. Matsuzawa, Y. Aoki, N. Takeuchi, and Y. Murayama, “A new long phosphorescent phosphor with high brightness, SrAl2O4:Eu2+,Dy3+,” J. Electrochem. Soc. 143(8), 2670–2673 (1996). [CrossRef]
3. Y. Lin, Z. Tang, Z. Zhang, and C. W. Nan, “Anomalous luminescence in Sr4Al14O25:Eu,Dy phosphors,” Appl. Phys. Lett. 81(6), 996–998 (2002). [CrossRef]
4. Y. Lin, Z. Tang, Z. Zhang, X. Wang, and J. Zhang, “Preparation of a new long afterglow blue-emitting Sr2MgSi2O7-based photoluminescent phosphor,” J. Mater. Sci. Lett. 20(16), 1505–1506 (2001). [CrossRef]
5. L. C. V. Rodrigues, H. F. Brito, J. Hölsä, R. Stefani, M. C. F. C. Felinto, M. Lastusaari, M. Malkamäki, and L. A. O. Nunes, “Discovery of the persistent luminescence mechanism of CdSiO3:Tb3+,” J. Phys. Chem. C (to be published).
6. J. Trojan-Piegza, E. Zych, J. Hölsä, and J. Niittykoski, “Spectroscopic properties of persistent luminescence phosphors: Lu2O3:Tb3+,M2+ (M = Ca, Sr, Ba),” J. Phys. Chem. C 113(47), 20493–20498 (2009). [CrossRef]
7. T. Kinoshita, M. Yamazaki, H. Kawazoe, and H. Hosono, “Long lasting phosphorescence and photostimulated luminescence in Tb-ion-activated reduced calcium aluminate glasses,” J. Appl. Phys. 86(7), 3729–3733 (1999). [CrossRef]
8. C. Liu, G. Che, Z. Xu, and Q. Wang, “Luminescence properties of a Tb3+ activated long-afterglow phosphor,” J. Alloys Compd. 474(1-2), 250–253 (2009). [CrossRef]
9. J. Hölsä, T. Laamanen, M. Lastusaari, M. Malkamäki, J. Niittykoski, and E. Zych, “Effect of Mg2+ and TiIV doping on the luminescence of Y2O2S:Eu3+,” Opt. Mater. 31(12), 1791–1793 (2009). [CrossRef]
10. J. Fu, “Orange and red emitting long-lasting phosphors MO:Eu3+ (M = Ca, Sr, Ba),” Electrochem. Solid-State Lett. 3(7), 350–351 (1999). [CrossRef]
11. T. Aitasalo, A. Hietikko, D. Hreniak, J. Hölsä, M. Lastusaari, J. Niittykoski, and W. Stręk, “Luminescence properties of BaMg2Si2O7:Eu2+,Mn2+,” J. Alloys Compd. 451(1-2), 229–231 (2008). [CrossRef]
12. X. Qu, L. Cao, W. Liu, G. Su, and P. Wang, “Luminescence properties of CdSiO3: Mn2+,RE3+ (RE = Sm, Dy, Eu) phosphors,” J. Alloys Compd. 487(1-2), 387–390 (2009). [CrossRef]
13. Y. Cong, B. Li, B. Lei, and W. Li, “Long lasting phosphorescent properties of Ti doped ZrO2,” J. Lumin. 126(2), 822–826 (2007). [CrossRef]
14. J. M. Carvalho, L. C. V. Rodrigues, J. Hölsä, M. Lastusaari, L. A. O. Nunes, M. C. F. C. Felinto, O. L. Malta, and H. F. Brito, “Influence of titanium and lutetium on the persistent luminescence of ZrO2,” Opt. Mater. Express 2(3), 331–340 (2012). [CrossRef]
15. K. Van den Eeckhout, P. F. Smet, and D. Poelman, “Persistent luminescence in Eu2+-doped compounds: a review,” Materials 3(4), 2536–2566 (2010). [CrossRef]
16. L. C. V. Rodrigues, R. Stefani, H. F. Brito, M. C. F. C. Felinto, J. Hölsä, M. Lastusaari, T. Laamanen, and M. Malkamäki, “Thermoluminescence and synchrotron radiation studies on the persistent luminescence of BaAl2O4:Eu2+,Dy3+,” J. Solid State Chem. 183(10), 2365–2371 (2010). [CrossRef]
17. P. Dorenbos, “Locating lanthanide impurity levels in the forbidden band of host crystals,” J. Lumin. 108(1-4), 301–305 (2004). [CrossRef]
18. Z. Qiu, Y. Zhou, M. Lu, A. Zhang, and Q. Ma, “Combustion synthesis of long-persistent luminescent MAl2O4:Eu2+, R3+ (M = Sr, Ba, Ca, R = Dy, Nd and La) nanoparticles and luminescence mechanism research,” Acta Mater. 55(8), 2615–2620 (2007). [CrossRef]
19. S. Ekambaram, K. C. Patil, and M. Maaza, “Synthesis of lamp phosphors: facile combustion approach,” J. Alloys Compd. 393(1-2), 81–92 (2005). [CrossRef]
20. B. M. Mothudi, O. M. Ntwaeaborwa, J. R. Botha, and H. C. Swart, “Photoluminescence and phosphorescence properties of MAl2O4:Eu2+,Dy3+ (M=Ca, Ba, Sr) phosphors prepared at an initiating combustion temperature of 500 °C,” Physica B 404(22), 4440–4444 (2009). [CrossRef]
21. HASYLAB, Beamline I: SUPERLUMI, http://hasylab.desy.de/facilities/doris_iii/beamlines/i_superlumi (accessed on Dec. 20, 2011).
22. Y. Liu, J. Kuang, B. Lei, and C. Shi, “Color-control of long-lasting phosphorescence (LLP) through rare earth ion-doped cadmium metasilicate phosphors,” J. Mater. Chem. 15(37), 4025–4031 (2005). [CrossRef]
23. A. V. S. Lourenço, L. C. V. Rodrigues, C. A. Kodaira, R. Stefani, H. F. Brito, M. C. F. C. Felinto, and J. Hölsä, “Persistent luminescence BaAl2O4:Eu2+,Dy3+ phosphor encapsulated in silica: water resistance,” in Proc. Int. Conf. Adv. Mater. (Braz. Mater. Res. Soc.), Rio de Janeiro-RJ, Brazil, September 20–25, 2009, BB556. http://www.sbpmat.org.br/icam2009/pdf/BB556.pdf.
24. F. A. Kröger and H. J. Vink, in Proc. Intern. Colloq. “Semiconductors and phosphors” (Interscience Publishers, Inc., New York, 1958), p. 17.
25. M. Weil, “Parawollastonite-type Cd3[Si3O9],” Acta Crystallogr. Sect. E Struct. Rep. Online 61(6), i102–i104 (2005). [CrossRef]
26. K. Brandenburg, Diamond v.3.2g, Crystal Impact: Bonn, Germany, 2011.
27. P. G. Radaelli, J. D. Jorgensen, A. J. Schultz, J. L. Peng, and R. L. Greene, “Evidence of apical oxygen in Nd2CuOy determined by single-crystal neutron diffraction,” Phys. Rev. B Condens. Matter 49(21), 15322–15326 (1994). [CrossRef]
28. T. Tomiki, H. Akamine, M. Gushiken, Y. Kinjoh, M. Miyazato, T. Miyazato, N. Toyokawa, M. Hiraoka, N. Hirata, Y. Ganaha, and T. Futemma, “Ce3+ centres in Y3Al5O12 (YAG) single crystals,” J. Phys. Soc. Jpn. 60(7), 2437–2445 (1991). [CrossRef]
29. P. Dorenbos, “Lanthanide charge transfer energies and related luminescence, charge carrier trapping, and redox phenomena,” J. Alloys Compd. 488(2), 568–573 (2009). [CrossRef]
30. P. Dorenbos, “Systematic behaviour in trivalent lanthanide charge transfer energies,” J. Phys. Condens. Matter 15(49), 8417–8434 (2003). [CrossRef]
31. M. Peng and G. Hong, “Reduction from Eu3+ to Eu2+ in BaAl2O4:Eu phosphor prepared in an oxidizing atmosphere and luminescent properties of BaAl2O4:Eu,” J. Lumin. 127(2), 735–740 (2007). [CrossRef]