- Email: [email protected]
Optical properties of Eu3+-doped CaAl4O7 synthesized by the Pechini method
Optical Materials 32 (2010) 1117–1122
Contents lists available at ScienceDirect
Optical Materials journal homepage: www.elsevier.com/locate/optmat
Optical properties of Eu3+-doped CaAl4O7 synthesized by the Pechini method M. Puchalska *, Y. Gerasymchuk, E. Zych Faculty of Chemistry, University of Wrocław, 14 F Joliot-Curie Street, 50-383 Wrocław, Poland
a r t i c l e
i n f o
Article history: Received 17 December 2009 Received in revised form 22 March 2010 Accepted 23 March 2010 Available online 22 April 2010 Keywords: Calcium dialuminate Europium(III) Pechini method Luminescence Decay kinetics
a b s t r a c t Ca1 xEuxAl4O7 (x = 0, 0.002, 0.01, 0.02, 0.05, 0.1) and Ca1 2xEuxNaxAl4O7 (x = 0.002, 0.02, 0.05, 0.1) crystalline phosphor powders have been synthesized by modified Pechini method and their spectroscopic properties at room and liquid nitrogen temperature were measured. Excitation spectra of synthesized materials contain broad band attributed to ligand to metal, O2 ?Eu3+, charge transfer absorption and narrow lines related to 7F0?5DJ, 5HJ, 5LJ, 5GJ transitions of europium ions. Upon excitation at 395 nm, the Eu3+-doped calcium dialuminates exhibit red photoluminescence due to parity forbidden f–f intraconfigurational transitions with the most intense component at 612 nm resulting from the 5D0?7F2 hypersensitive transition. The quantum efficiency of the powders emissions does not exceed 12%. The luminescence decay kinetics reveals that some quenching appears for higher concentrations. Site selective spectroscopy disclosed the presence of at least two well defined individual spectroscopic sites of Eu3+ and increasing disturbance of their local symmetries with raising concentration of the activator was observed. The increasing symmetry disorder led to a significant asymmetrical broadening of the luminescence lines at higher concentrations. Concentration dependence of the luminescence kinetics of the two diverse Eu3+ ions is different, as only one of them, with the characteristic luminescence at 579.5 nm, exhibits shortening of the decay time at higher Eu contents. The inexpensive CaAl4O7:Eu powders are potentially good candidates for photonics applications. For Eu, Na co-doped powders the Eu3+ ions producing emission at 579.7 nm dominates strongly over those with luminescence at 577.2 nm. Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction The calcium aluminates continue receiving an interest in a variety of areas, such as materials of construction [1], luminescent materials [2–4], as well as model systems for the study of crystallization phenomena in amorphous precursors [5]. Various crystalline phases of calcium aluminates have been synthesized, including Ca Al2O4, CaAl4O7, Ca3Al2O6 and Ca12Al14O33. All these compositions have been employed as host materials for RE-doped phosphors. Among them, (Eu2+, Dy3+/Nd3+) co-doped CaAl2O4 is a well known, commercially offered persistent luminescence phosphor [6]. CaAl4O7 is an important ceramic material for high temperature refractory applications (melting point: 1765 ± 25 °C) [7]. It is also used as component of high-alumina cement and in steel industry as metallurgical slag. Calcium dialuminate, termed ‘‘grossite” in mineralogical field, is found in natural terrestrial rocks and meteorites [8]. Currently, this compound is of great interest because of its unique crystallographic features, particularly the existence of so-called ‘‘triclusters” consisting of an oxygen atom surrounded by three Al tetrahedra as well as a very low coefficient of thermal expansion [9,10]. CaAl4O7 is also an
* Corresponding author. E-mail address: [email protected] (M. Puchalska). 0925-3467/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.optmat.2010.03.012
attractive host for spectroscopic materials due to a high optical transparency from ultra-violet to near infrared spectral range, excellent mechanical properties and good chemical stability. Eu3+ ion has attracted significant attention as a dopant to design optical amplifiers, display phosphor, electroluminescence devices and optical storage phosphors [11]. Furthermore, Eu3+ ion is also known as a very sensitive site-symmetry probe. Till now, only scant studies have been performed on RE-doped CaAl4O7. The luminescence properties of CaAl4O7 powders activated with Pr3+, Ce3+, Tb3+, (Ce3+, Tb3+) prepared by solid state reaction [3,12,13] and polycrystalline pellet containing 1% of Eu3+ and obtained by sintering have been reported [14]. All these phosphor materials were synthesized at high temperatures (above 1350 °C) using multistep procedures with intermediate grindings. This paper describes details of fabrication of Eu3+doped (0–10 mol%) CaAl4O7 crystalline powders with a modified Pechini method at much lower temperature (950 °C), and presents results of their structural (XRD) and spectroscopic (IR, photoluminescence) analysis. 2. Experimental Calcium aluminates activated with Eu3+ ions, Ca1 xEuxAl4O7 (x = 0, 0.002, 0.01, 0.02, 0.05, 0.1) and Ca1 2xEuxNaxAl4O7
1118
M. Puchalska et al. / Optical Materials 32 (2010) 1117–1122
(x = 0.002, 0.02, 0.05, 0.1) were prepared via modified Pechini citrate process, similarly to the procedure described in [15] for undoped CaAl4O7. The metal nitrates: Al(NO3)39H2O, Ca(NO3)2 4H2O Eu(NO3)36H2O, NaNO3, citric acid (CA) and ethylene glycol (EG) were used as starting materials. The molar ratio of CA, sum of metal cations (Al3+, Eu3+, Ca2+, Na+) and EG was 2:1:8. First, the required amount of CA was dissolved in EG, followed by the addition of stoichiometric amounts of Al(NO3)39H2O, Ca(NO3)24H2O and Eu(NO3)36H2O. The obtained solutions were heated to 80 °C on a hot plate, under continuous stirring. Afterward, the solid resins were gradually heated on hot plate up to 600 °C to obtain raw brownish powders (due to residual carbon). No vigorous combustion was observed. Finally, these materials were calcined in air atmosphere at 950 °C in alumina crucibles for 3 h, resulting in fine, white powders. All samples were checked by powder X-ray diffractometry (DRON-2 Diffractometer, Cu Ka radiation, k = 1.5418 Å). The measurements were performed for 2h ranging from 10o to 100o and with 0.1o step. The infrared absorption spectra were recorded with a Brüker FTIR IFS 113 V spectrophotometer using Nujol to disperse the powder. Photoluminescence measurements were performed at room (298 K) and liquid nitrogen (77 K) temperatures using SpectraPro 750 1-meter monochromator, coupled to Hamamatsu R928 photomultiplier and equipped with a 1200 l/mm grating blazed at 500 nm. A 450 W xenon lamp was used as an excitation source. It was coupled to 275 mm excitation monochromator equipped with a 1800 l/mm grating blazed at 250 nm. Luminescence decay traces were recorded at 298 K and 77 K using a Tektronix TDS 3052B oscilloscope and Nd:YAG Lambda Physics pulsed laser with excitation line of 532 nm. Quantum efficiency of the powders were measured applying the procedure described in [16].
expected to substitute Ca2+ (r = 1.20 Å) rather than Al3+ (r = 0.53 Å). The replacement of one Ca2+ by one Eu3+ would cause accumulation of a positive net charge, compensation of which could be achieved creating Ca-vacancies or interstitial O2 near the Eu3+ ions. Therefore, an appearance of more than just one Eu3+ site is anticipated and the symmetry of the coordination environment of the activator may further degrade enhancing especially the probability of the electric dipole hypersensitive 5D0?7F2 transition. 3.2. Photoluminescence study Emission spectra of the Eu3+ doped calcium dialuminates as a function of the dopant content have been measured after excitation into 7F0?5L6 transition (395 nm) at 298 K and 77 K and the results are presented in Figs. 2 and 3, respectively. All samples yield red emission with the main component peaking around 612 nm and the spectra consist of characteristic lines due to the 5D0?7FJ transitions. The luminescence arises exceptionally from the 5D0 level; no transitions from the higher excited 5D1 state are observed.
3. Results and discussion 3.1. Structural characterization Fig. 1 shows the X-ray diffraction patterns of all the CaAl4O7: Eu3+ (x = 0, 0.002, 0.01, 0.02, 0.05, 0.1) investigated samples. The spectra prove a single grossite monoclinic phase with a C2/c space group for all compositions (ICSD#14270) [9,17]. In the grossite structure, Ca2+ ion is placed in a sevenfold coordination sphere with C2 (pseudo-C2v) symmetry, whereas Al3+ ion is distributed over two symmetrically independent tetrahedral sites, which are distorted to some degree which leads to virtually C1 site symmetry [9,17]. Because of similar ionic radii the Eu3+ ions (r = 1.15 Å) are
Fig. 1. XRD patterns of the undoped and CaAl4O7:Eu3+ powder phosphor.
Fig. 2. Emission spectra of CaAl4O7:Eu3+ samples under 395 nm excitation at 298 K. Under different excitation wavelengths in the 280–400 nm range no Eu2+ luminescence was observed.
Fig. 3. Emission spectra of CaAl4O7:Eu3+ samples under 395 nm excitation at 77 K. Under different excitation wavelengths in the 280–400 nm range no Eu2+ luminescence was observed.
M. Puchalska et al. / Optical Materials 32 (2010) 1117–1122
The absence of fluorescence from the 5D1 level proves that an efficient multi-phonon relaxation and/or cross-relaxation down to 5D0 level takes place [11,18]. The mid-IR stretching and bending modes (see Fig. 4) of tightly bound tetrahedral AlO4 units are considerably strong and have relatively high frequencies (578–937 cm 1) which facilitates the mentioned non-radiative relaxation to the emitting 5 D0. In contrast to the powder containing 0.2% of Eu3+ ions, it can be observed that those with higher Eu3+ contents (1–10 mol%) show inhomogeneously broadened emission features, which is quite characteristic for activators occupying a variety of sites similar but not identical in terms of their surroundings symmetries. The intensity of a forced electric dipole hypersensitive transition, 5 D0?7F2, producing luminescence around 612 nm is much stronger than that of the 5D0?7F1 magnetic dipole transition (orange emission around 593 nm). This proves a low symmetry around the Eu3+ ions. It is in agreement with the already discussed crystal structure of the host material. In the range of the 5D0?7F0 transition two or even more components, partially overlapping, are seen. It is more clearly seen after lowering the temperature to 77 K (see Fig. 3). At the liquid nitrogen temperature, two distinct well separated lines are observed at 577.2 nm and 579.7 nm for the smallest Eu3+ concentration (0.2 mol%). As the Eu content is increased these lines (asymmetrically) broaden and finally overlap, which may be assigned to the appearance of additional components due to some distortion of symmetry around (a fraction of) the Eu3+ ions. Also the relative intensities of the two emission lines change with concentration. For 0.2 mol% doped powder the band at 579.7 nm is stronger, for 1% and 2% the one at 577.2 nm is more intensive, and finally for the highest concentration (10 mol%) the component at 579.7 nm takes its upper hand again. Altogether the observations confirm the presence of different Eu3+ sites in the crystal lattice of CaAl4O7 as well as the change of their relative populations with an increase of rare earth content. Also energy migration and transfer between the sites cannot be excluded at higher concentrations. Although there is only one Ca2+ site offered by CaAl4O7, Eu3+ ions experience at least two distinctive symmetries of their surroundings for low concentration and even more for higher Eu contents. This is understandable, as balancing the higher charge of Eu3+ compared to Ca2+ requires formation of vacancies at Ca2+ sites or O2 interstitials (or both). Consequently, some structural disorder around the ions of the dopant should appear especially for higher concentrations of the activator. The above observations and conclusions get further support from spectra of Eu, Na co-doped Ca1 2xEuxNaxAl4O7 (x = 0.002, 0.02, 0.05, 0.1) powders. In Fig. 5 their 77 K emissions spectra are presented. It is clearly seen that charge compensation
Fig. 4. FT-IR spectra of CaAl4O7:Eu3+ (1 mol%).
1119
Fig. 5. Emission spectra of CaAl4O7:Eu3+, Na+ samples under 395 nm excitation at 77 K.
leads to reduction of the number of luminescent lines and their significant narrowing. The effects are especially striking at higher concentrations. In the range of the 5D0?7F0 transition a component peaking at 577.2 nm is hardly seen for the doubly doped powders (for the 2% specimen it is more visible but still much less than for the singly doped counterpart). These results basically prove that Eu3+ substitutes Ca2+. The Eu3+ surrounding symmetry gets significant distortion when only Eu is incorporated into the host lattice. Once Na+ is used to balance the Eu3+ extra charge most ions of the dopant generally experiences one type of symmetry. Thus, the variations of luminescence spectra in both the singly (Eu) and doubly (Eu, Na) doped CaAl4O7 powders are reasonable and convincingly explained. Existence of two Eu3+ sites in CaAl4O7 was also reported in [14] for polycrystalline pellet of CaAl4O7:Eu3+ containing 1% of the activator and prepared at 1350 °C. Also for Ce3+-activated powder the presence of two symmetry sites of the dopant in this aluminate was concluded [3]. Thus, the much lower temperature of making our powders did not alter this property. The effect of Eu3+ content on photoluminescence intensity was also investigated. The intensity of the luminescence increases with Eu concentration but with a clear tendency to saturation, which in fact suggests that at higher Eu contents some quenching of the Eu3+ luminescence occurs. We will return to this effect discussing kinetics of the Eu3+ emission. Fig. 6 presents concentration dependence of the QE of the CaAl4O7:Eu powders. In the range of 1–5% the efficiencies are very similar and do not exceed about 12%. Above 5 mol% of Eu the QE lowers to about 7–8% indicating some concentration-induced quenching of the emission. The below presented results of kinetics measurements will support this conclusion. The rather low QE is slightly disappointing. It is hard to expect that technological improvement could raise this number to a commercially attractive value. The excitation spectra of CaAl4O7:Eu3+ samples were recorded in the 190–550 nm spectral range monitoring the 5D0?7F2 emission around 612 nm. Fig. 7 and 8 present results of measurements for 0.2, 1 and 10 mol% of Eu3+ at room and nitrogen temperature, respectively. Clearly, the excitation spectra may be divided into two regions; one consisting of a broad intense band around 225 nm that is associated with the O2 ?Eu3+ charge transfer transition and another one located at longer wavelengths containing low intensity narrow lines arising from 7F0?5L6, 7F0?5DJ (J = 1–3) and 7F0?5GJ (J = 2–6) intra-configurational transitions of Eu3+ ion. Concentration dependence of the spectra is not significant. However, they show the decrease of the relative intensity of
1120
M. Puchalska et al. / Optical Materials 32 (2010) 1117–1122
Fig. 6. Concentration dependence of the quantum efficiency of the CaAl4O7:Eu3+ powders.
the 7F0?5D2 transition located around 460–465 nm in comparison with the 7F0?5L6 transition as the Eu content increases. Some variations in the position and shape of the CT band, both at 298 K and 77 K, are also observed but can simply come from so called saturation effect, which can easily impart such distortions [19,20]. High concentration of the dopant and high absorption coefficient of the CT transitions make this more than probable. Since the emission spectra revealed the presence of at least two distinctive symmetry sites of the activator ions the excitation spectra of CaAl4O7:Eu3+ (1 mol%) monitoring the 5D0?7F0 emission lines located at 577.2 nm and 579.7 nm were measured at 77 K. These two luminescence lines were already temporary assigned to be connected with different Eu3+ sites. The results are presented in Fig. 9 and Table 1 lists the energies of the excited states for the two Eu3+sites. Indeed, there are significant differences observed between these excitation spectra which confirm that the two emission lines result from Eu3+ ions located in the environments of different symmetries. Clearly, the number of lines, their positions and relative intensities are different in both excitation spectra. These results definitively prove the existence of at least two different Eu3+ sites in the investigated CaAl4O7 powders. A detailed analysis of the results in Fig. 9 reveals that the most intense lines seen in the excitation spectrum of the 579.7 nm emission seem to be also present as low intensity features in the excitation spectrum of the 577.5 nm luminescence. This may indicate appearance of energy transfer between the Eu3+ ions occupying the two positions.
Fig. 7. The excitation spectra of CaAl4O7:Eu3+ samples recorded by monitoring a 5 D0?7F2 emission at 612 nm at 298 K.
Fig. 9. The excitation spectra of CaAl4O7:Eu3+ samples recorded by monitoring a 5 D0?7F0 emission at 577.2 nm and 579.7 nm at 77 K.
Table 1 Energies (cm 1) of the excitation spectra of CaAl4O7:Eu3+ (1%) monitored for kem = 577.2 nm and 579.7 nm. em = 579.7 nm
em = 577.2 nm
27,555 26,759 26,525 26,267 25,374 25,278 25,125 24,900
27,670 26,709 26,539 26,350 25,490 25,406 25,182 25,025 24,912 21,588 21,523 19,029 18,864
21,491 Fig. 8. The excitation spectra of CaAl4O7:Eu3+ samples recorded by monitoring a 5 D0?7F2 emission at 612 nm at 77 K.
18,986
1121
M. Puchalska et al. / Optical Materials 32 (2010) 1117–1122
3.3. Decay kinetics The decay traces of the emissions from the 5D0 state for all compositions were measured at room and liquid nitrogen temperature monitoring the most intense 5D0?7F2 luminescence at 612 nm as well as the two site-distinctive components of the 5D0?7F0 transitions at 577.5 nm and 579.5 nm (298 K) and at 577.2 nm and 579.7 nm (77 K). The traces recorded for the 5D0?7F0 emissions at both temperatures are presented in Fig. 10. Single or, where necessary, biexponential functions have been used to fit the experimental data and obtained radiative lifetimes are presented in Table 2. Luminescence decay traces of the 5D0?7F2 emission at 612 nm could be reasonably reproduced with monoexponential function for all materials at both temperatures. At liquid nitrogen temperature the radiative lifetimes are almost identical (in the range of 1.56–1.69 ms) despite Eu3+ ions of different symmetries have to produce luminescence around 612 nm. At room temperature the 612 nm emission decays faster with increasing Eu content
and the time constant changes from about 1.7 ms for the 0.2% specimen to about 1.3 ms for the highest concentration of the dopant. Decays of the two 5D0?7F0 emissions are significantly different, especially at room temperature. At RT the kinetics of the 579.5 nm luminescence is clearly concentration sensitive becoming noticeably shorter for higher dopant contents. It is striking that the initial lifetime is roughly constant above 0.2%, while at later stages the luminescence decays continuously faster and faster. This may be taken as a support for the previously suggested energy transfer from the Eu3+ producing emission around 579.5 nm to the ions with characteristic luminescence at 577.5 nm. Kinetics of the latter one is perfectly identical within the whole range of Eu concentrations at RT. This is however confusing as the energy would go from a state slightly lower in energy to the one positioned higher. While such a phonon assisted transfer of energy is known and absolutely possible it is not clear at all while the opposite (typical) direction of energy transfer could not occur. On the other hand the shortening of the decay of the 579.5 nm emission may also indicate that
Fig. 10. 5D0 luminescence decay curves of CaAl4O7:Eu3+ samples excited at 532.7 nm and monitored at 577.5 nm and 579.5 nm at 298 K and at 577.2 nm and 579.7 nm at 77 K.
Table 2 Luminescence decay times for CaAl4O7:Eu3+ samples at 298 K and 77 K. Eu3+ conc. mol%
02 1 2 5 10
298 K
77 K
612 nm
577.5 nm
612 nm
577.2 nm
s1
s1 + s2
s
579.5 nm
s1 + s2
s
s1
s1
579.7 nm
s1
1.71 1.56 1.54 1.50 1.32
1.34 1.42 + 0.36 1.47 + 0.43 1.43 + 0.43 1.53 + 0.73
– 1.28 1.33 1.35 1.35
1.61 1.52 + 0.31 1.36 + 0.26 1.23 + 0.27 1.17 + 0.25
– 1.30 1.10 0.98 0.97
1.69 1.58 1.56 1.59 1.58
1.69 1.62 1.58 1.58 1.57
1.43 1.32 1.28 1.28 1.41
1122
M. Puchalska et al. / Optical Materials 32 (2010) 1117–1122
nearby a defect site is located and is able to drain energy from the excited dopant. Than this would imply that the defect locates near the Eu3+ emitting at 579.5 nm only when the dopant content surpasses the concentration of 0.2%. This seems reasonable, yet further work would be needed to understand details of this behavior. After lowering the temperature to 77 K the 579.5 nm luminescence (now positioned at 579.7 nm) is no longer concentration dependent. Interestingly, for the lowest concentrations the decay of the 577.2 nm emission at 77 K shows a significant tail indicating some afterglow. It may well be that a defect serves as a shallow trap capable of intercepting some of the energy delivered to the material and passes it back to the activator with some delay. 4. Conclusions Grossite phase of Eu-doped CaAl4O7 with the activator content up to 10 mol% was synthesized by modified Pechini method at 950 °C. The investigated crystalline powders exhibited a red luminescence with the main component peaking at 612 nm. The calculated quantum efficiencies are rather low, with the highest value reaching about 12% for 1–5% of Eu3+. Emission spectra and decay kinetics proved that Eu3+ ions occupy two distinctive symmetry sites and their surroundings get further distortion when the dopant content increases leading to continuous broadening of the spectral features. Co-doping with Na+ ions leads to basically one emission center. At low temperatures kinetics of emission from either of the sites does not change with Eu concentration. Yet, at room temperature the luminescence at 579.5 nm gets shorter with
increasing Eu content, which could indicate some energy transfer between the two sites or, more probably, a drain of the energy by a defect. References [1] S. Jonas, F. Nadachowski, D. Szwagierczak, Ceram. Int. 25 (1999) 77. [2] T. Aitasalo, J. Holsa, H. Jungner, M. Lastusaari, J. Niittykoski, M. Parkkinen, R. Valtonen, Opt. Mater. 26 (2004) 113. [3] D. Jia, J. Lumin. 117 (2006) 170. [4] Y. Murayama, Luminous Paints, in: S. Shionoya, W.M. Yen (Eds.), Phosphor Handbook, CRC Press, Boca Raton, FL, USA, 1999, p. 651. [5] A. Douy, M. Gervais, J. Am. Ceram. Soc. 83 (2000) 70. [6] J. Holsa, H. Jungner, M. Lastusaari, J. Niittykoski, J. Alloys Compd. 323–324 (2001) 326. [7] B. Hallstedt, J. Am. Ceram. Soc. 73 (1) (1990) 15. [8] D. Weber, A. Bischof, Eur. J. Miner. 6 (4) (1994) 591. [9] E.R. Boyko, L.G. Wisnyi, Acta Crystallogr. 11 (1958) 444. [10] S. Jonas, F. Nadachowski, D. Szwagierczak, Ceram. Int. 24 (1998) 211. [11] G. Blasse, B.C. Grabmaier, Luminescent Materials, Springer, Berlin, 1994. [12] M.L.H. ter Heerdt, E. van der Kolk E, W.M. Yen, A.M. Srivastava, J. Lumin. 100 (2002) 107. [13] D. Jia, J. Zhu, B. Wu, S.E. Paje, J. Lumin. 93 (2001) 107. [14] H. Yuan, W.M. Dennis, L. Lu, W.M. Yen, W. Jia, H. Liu, in: C. Ronda (Ed.), Proceedings of the 6th International Conference on the Physics and Chemistry of Luminescent Materials, Vols. 97–29, Electrochemical Society, 1998, p. 206. [15] X. Yuan, Y. Xu, Y. He, J. Alloys Compd. 441 (2007) 251. [16] E. Zych, A. Meijerink, C. de Mello Donega, J. Phys.-Condens. Mat. 15 (2003) 5145. [17] D.W. Goodwin, A.J. Lindop, Acta Crystallogr. B 26 (1970) 1230. [18] M. Buijs, A. Meijerink, G. Blasse, J. Lumin. 37 (1987) 9. [19] C. de Mello Donega, A. Meijerink, G. Blasse, J. Lumin. 62 (1994) 189. [20] C. de Mello Donega, Vibronic spectroscopy of Pr3+ in solids, Ph.D. Thesis, Utrecht University, The Netherlands, 1994. ISBN: 90-393-0813-6.
Contents lists available at ScienceDirect
Optical Materials journal homepage: www.elsevier.com/locate/optmat
Optical properties of Eu3+-doped CaAl4O7 synthesized by the Pechini method M. Puchalska *, Y. Gerasymchuk, E. Zych Faculty of Chemistry, University of Wrocław, 14 F Joliot-Curie Street, 50-383 Wrocław, Poland
a r t i c l e
i n f o
Article history: Received 17 December 2009 Received in revised form 22 March 2010 Accepted 23 March 2010 Available online 22 April 2010 Keywords: Calcium dialuminate Europium(III) Pechini method Luminescence Decay kinetics
a b s t r a c t Ca1 xEuxAl4O7 (x = 0, 0.002, 0.01, 0.02, 0.05, 0.1) and Ca1 2xEuxNaxAl4O7 (x = 0.002, 0.02, 0.05, 0.1) crystalline phosphor powders have been synthesized by modified Pechini method and their spectroscopic properties at room and liquid nitrogen temperature were measured. Excitation spectra of synthesized materials contain broad band attributed to ligand to metal, O2 ?Eu3+, charge transfer absorption and narrow lines related to 7F0?5DJ, 5HJ, 5LJ, 5GJ transitions of europium ions. Upon excitation at 395 nm, the Eu3+-doped calcium dialuminates exhibit red photoluminescence due to parity forbidden f–f intraconfigurational transitions with the most intense component at 612 nm resulting from the 5D0?7F2 hypersensitive transition. The quantum efficiency of the powders emissions does not exceed 12%. The luminescence decay kinetics reveals that some quenching appears for higher concentrations. Site selective spectroscopy disclosed the presence of at least two well defined individual spectroscopic sites of Eu3+ and increasing disturbance of their local symmetries with raising concentration of the activator was observed. The increasing symmetry disorder led to a significant asymmetrical broadening of the luminescence lines at higher concentrations. Concentration dependence of the luminescence kinetics of the two diverse Eu3+ ions is different, as only one of them, with the characteristic luminescence at 579.5 nm, exhibits shortening of the decay time at higher Eu contents. The inexpensive CaAl4O7:Eu powders are potentially good candidates for photonics applications. For Eu, Na co-doped powders the Eu3+ ions producing emission at 579.7 nm dominates strongly over those with luminescence at 577.2 nm. Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction The calcium aluminates continue receiving an interest in a variety of areas, such as materials of construction [1], luminescent materials [2–4], as well as model systems for the study of crystallization phenomena in amorphous precursors [5]. Various crystalline phases of calcium aluminates have been synthesized, including Ca Al2O4, CaAl4O7, Ca3Al2O6 and Ca12Al14O33. All these compositions have been employed as host materials for RE-doped phosphors. Among them, (Eu2+, Dy3+/Nd3+) co-doped CaAl2O4 is a well known, commercially offered persistent luminescence phosphor [6]. CaAl4O7 is an important ceramic material for high temperature refractory applications (melting point: 1765 ± 25 °C) [7]. It is also used as component of high-alumina cement and in steel industry as metallurgical slag. Calcium dialuminate, termed ‘‘grossite” in mineralogical field, is found in natural terrestrial rocks and meteorites [8]. Currently, this compound is of great interest because of its unique crystallographic features, particularly the existence of so-called ‘‘triclusters” consisting of an oxygen atom surrounded by three Al tetrahedra as well as a very low coefficient of thermal expansion [9,10]. CaAl4O7 is also an
* Corresponding author. E-mail address: [email protected] (M. Puchalska). 0925-3467/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.optmat.2010.03.012
attractive host for spectroscopic materials due to a high optical transparency from ultra-violet to near infrared spectral range, excellent mechanical properties and good chemical stability. Eu3+ ion has attracted significant attention as a dopant to design optical amplifiers, display phosphor, electroluminescence devices and optical storage phosphors [11]. Furthermore, Eu3+ ion is also known as a very sensitive site-symmetry probe. Till now, only scant studies have been performed on RE-doped CaAl4O7. The luminescence properties of CaAl4O7 powders activated with Pr3+, Ce3+, Tb3+, (Ce3+, Tb3+) prepared by solid state reaction [3,12,13] and polycrystalline pellet containing 1% of Eu3+ and obtained by sintering have been reported [14]. All these phosphor materials were synthesized at high temperatures (above 1350 °C) using multistep procedures with intermediate grindings. This paper describes details of fabrication of Eu3+doped (0–10 mol%) CaAl4O7 crystalline powders with a modified Pechini method at much lower temperature (950 °C), and presents results of their structural (XRD) and spectroscopic (IR, photoluminescence) analysis. 2. Experimental Calcium aluminates activated with Eu3+ ions, Ca1 xEuxAl4O7 (x = 0, 0.002, 0.01, 0.02, 0.05, 0.1) and Ca1 2xEuxNaxAl4O7
1118
M. Puchalska et al. / Optical Materials 32 (2010) 1117–1122
(x = 0.002, 0.02, 0.05, 0.1) were prepared via modified Pechini citrate process, similarly to the procedure described in [15] for undoped CaAl4O7. The metal nitrates: Al(NO3)39H2O, Ca(NO3)2 4H2O Eu(NO3)36H2O, NaNO3, citric acid (CA) and ethylene glycol (EG) were used as starting materials. The molar ratio of CA, sum of metal cations (Al3+, Eu3+, Ca2+, Na+) and EG was 2:1:8. First, the required amount of CA was dissolved in EG, followed by the addition of stoichiometric amounts of Al(NO3)39H2O, Ca(NO3)24H2O and Eu(NO3)36H2O. The obtained solutions were heated to 80 °C on a hot plate, under continuous stirring. Afterward, the solid resins were gradually heated on hot plate up to 600 °C to obtain raw brownish powders (due to residual carbon). No vigorous combustion was observed. Finally, these materials were calcined in air atmosphere at 950 °C in alumina crucibles for 3 h, resulting in fine, white powders. All samples were checked by powder X-ray diffractometry (DRON-2 Diffractometer, Cu Ka radiation, k = 1.5418 Å). The measurements were performed for 2h ranging from 10o to 100o and with 0.1o step. The infrared absorption spectra were recorded with a Brüker FTIR IFS 113 V spectrophotometer using Nujol to disperse the powder. Photoluminescence measurements were performed at room (298 K) and liquid nitrogen (77 K) temperatures using SpectraPro 750 1-meter monochromator, coupled to Hamamatsu R928 photomultiplier and equipped with a 1200 l/mm grating blazed at 500 nm. A 450 W xenon lamp was used as an excitation source. It was coupled to 275 mm excitation monochromator equipped with a 1800 l/mm grating blazed at 250 nm. Luminescence decay traces were recorded at 298 K and 77 K using a Tektronix TDS 3052B oscilloscope and Nd:YAG Lambda Physics pulsed laser with excitation line of 532 nm. Quantum efficiency of the powders were measured applying the procedure described in [16].
expected to substitute Ca2+ (r = 1.20 Å) rather than Al3+ (r = 0.53 Å). The replacement of one Ca2+ by one Eu3+ would cause accumulation of a positive net charge, compensation of which could be achieved creating Ca-vacancies or interstitial O2 near the Eu3+ ions. Therefore, an appearance of more than just one Eu3+ site is anticipated and the symmetry of the coordination environment of the activator may further degrade enhancing especially the probability of the electric dipole hypersensitive 5D0?7F2 transition. 3.2. Photoluminescence study Emission spectra of the Eu3+ doped calcium dialuminates as a function of the dopant content have been measured after excitation into 7F0?5L6 transition (395 nm) at 298 K and 77 K and the results are presented in Figs. 2 and 3, respectively. All samples yield red emission with the main component peaking around 612 nm and the spectra consist of characteristic lines due to the 5D0?7FJ transitions. The luminescence arises exceptionally from the 5D0 level; no transitions from the higher excited 5D1 state are observed.
3. Results and discussion 3.1. Structural characterization Fig. 1 shows the X-ray diffraction patterns of all the CaAl4O7: Eu3+ (x = 0, 0.002, 0.01, 0.02, 0.05, 0.1) investigated samples. The spectra prove a single grossite monoclinic phase with a C2/c space group for all compositions (ICSD#14270) [9,17]. In the grossite structure, Ca2+ ion is placed in a sevenfold coordination sphere with C2 (pseudo-C2v) symmetry, whereas Al3+ ion is distributed over two symmetrically independent tetrahedral sites, which are distorted to some degree which leads to virtually C1 site symmetry [9,17]. Because of similar ionic radii the Eu3+ ions (r = 1.15 Å) are
Fig. 1. XRD patterns of the undoped and CaAl4O7:Eu3+ powder phosphor.
Fig. 2. Emission spectra of CaAl4O7:Eu3+ samples under 395 nm excitation at 298 K. Under different excitation wavelengths in the 280–400 nm range no Eu2+ luminescence was observed.
Fig. 3. Emission spectra of CaAl4O7:Eu3+ samples under 395 nm excitation at 77 K. Under different excitation wavelengths in the 280–400 nm range no Eu2+ luminescence was observed.
M. Puchalska et al. / Optical Materials 32 (2010) 1117–1122
The absence of fluorescence from the 5D1 level proves that an efficient multi-phonon relaxation and/or cross-relaxation down to 5D0 level takes place [11,18]. The mid-IR stretching and bending modes (see Fig. 4) of tightly bound tetrahedral AlO4 units are considerably strong and have relatively high frequencies (578–937 cm 1) which facilitates the mentioned non-radiative relaxation to the emitting 5 D0. In contrast to the powder containing 0.2% of Eu3+ ions, it can be observed that those with higher Eu3+ contents (1–10 mol%) show inhomogeneously broadened emission features, which is quite characteristic for activators occupying a variety of sites similar but not identical in terms of their surroundings symmetries. The intensity of a forced electric dipole hypersensitive transition, 5 D0?7F2, producing luminescence around 612 nm is much stronger than that of the 5D0?7F1 magnetic dipole transition (orange emission around 593 nm). This proves a low symmetry around the Eu3+ ions. It is in agreement with the already discussed crystal structure of the host material. In the range of the 5D0?7F0 transition two or even more components, partially overlapping, are seen. It is more clearly seen after lowering the temperature to 77 K (see Fig. 3). At the liquid nitrogen temperature, two distinct well separated lines are observed at 577.2 nm and 579.7 nm for the smallest Eu3+ concentration (0.2 mol%). As the Eu content is increased these lines (asymmetrically) broaden and finally overlap, which may be assigned to the appearance of additional components due to some distortion of symmetry around (a fraction of) the Eu3+ ions. Also the relative intensities of the two emission lines change with concentration. For 0.2 mol% doped powder the band at 579.7 nm is stronger, for 1% and 2% the one at 577.2 nm is more intensive, and finally for the highest concentration (10 mol%) the component at 579.7 nm takes its upper hand again. Altogether the observations confirm the presence of different Eu3+ sites in the crystal lattice of CaAl4O7 as well as the change of their relative populations with an increase of rare earth content. Also energy migration and transfer between the sites cannot be excluded at higher concentrations. Although there is only one Ca2+ site offered by CaAl4O7, Eu3+ ions experience at least two distinctive symmetries of their surroundings for low concentration and even more for higher Eu contents. This is understandable, as balancing the higher charge of Eu3+ compared to Ca2+ requires formation of vacancies at Ca2+ sites or O2 interstitials (or both). Consequently, some structural disorder around the ions of the dopant should appear especially for higher concentrations of the activator. The above observations and conclusions get further support from spectra of Eu, Na co-doped Ca1 2xEuxNaxAl4O7 (x = 0.002, 0.02, 0.05, 0.1) powders. In Fig. 5 their 77 K emissions spectra are presented. It is clearly seen that charge compensation
Fig. 4. FT-IR spectra of CaAl4O7:Eu3+ (1 mol%).
1119
Fig. 5. Emission spectra of CaAl4O7:Eu3+, Na+ samples under 395 nm excitation at 77 K.
leads to reduction of the number of luminescent lines and their significant narrowing. The effects are especially striking at higher concentrations. In the range of the 5D0?7F0 transition a component peaking at 577.2 nm is hardly seen for the doubly doped powders (for the 2% specimen it is more visible but still much less than for the singly doped counterpart). These results basically prove that Eu3+ substitutes Ca2+. The Eu3+ surrounding symmetry gets significant distortion when only Eu is incorporated into the host lattice. Once Na+ is used to balance the Eu3+ extra charge most ions of the dopant generally experiences one type of symmetry. Thus, the variations of luminescence spectra in both the singly (Eu) and doubly (Eu, Na) doped CaAl4O7 powders are reasonable and convincingly explained. Existence of two Eu3+ sites in CaAl4O7 was also reported in [14] for polycrystalline pellet of CaAl4O7:Eu3+ containing 1% of the activator and prepared at 1350 °C. Also for Ce3+-activated powder the presence of two symmetry sites of the dopant in this aluminate was concluded [3]. Thus, the much lower temperature of making our powders did not alter this property. The effect of Eu3+ content on photoluminescence intensity was also investigated. The intensity of the luminescence increases with Eu concentration but with a clear tendency to saturation, which in fact suggests that at higher Eu contents some quenching of the Eu3+ luminescence occurs. We will return to this effect discussing kinetics of the Eu3+ emission. Fig. 6 presents concentration dependence of the QE of the CaAl4O7:Eu powders. In the range of 1–5% the efficiencies are very similar and do not exceed about 12%. Above 5 mol% of Eu the QE lowers to about 7–8% indicating some concentration-induced quenching of the emission. The below presented results of kinetics measurements will support this conclusion. The rather low QE is slightly disappointing. It is hard to expect that technological improvement could raise this number to a commercially attractive value. The excitation spectra of CaAl4O7:Eu3+ samples were recorded in the 190–550 nm spectral range monitoring the 5D0?7F2 emission around 612 nm. Fig. 7 and 8 present results of measurements for 0.2, 1 and 10 mol% of Eu3+ at room and nitrogen temperature, respectively. Clearly, the excitation spectra may be divided into two regions; one consisting of a broad intense band around 225 nm that is associated with the O2 ?Eu3+ charge transfer transition and another one located at longer wavelengths containing low intensity narrow lines arising from 7F0?5L6, 7F0?5DJ (J = 1–3) and 7F0?5GJ (J = 2–6) intra-configurational transitions of Eu3+ ion. Concentration dependence of the spectra is not significant. However, they show the decrease of the relative intensity of
1120
M. Puchalska et al. / Optical Materials 32 (2010) 1117–1122
Fig. 6. Concentration dependence of the quantum efficiency of the CaAl4O7:Eu3+ powders.
the 7F0?5D2 transition located around 460–465 nm in comparison with the 7F0?5L6 transition as the Eu content increases. Some variations in the position and shape of the CT band, both at 298 K and 77 K, are also observed but can simply come from so called saturation effect, which can easily impart such distortions [19,20]. High concentration of the dopant and high absorption coefficient of the CT transitions make this more than probable. Since the emission spectra revealed the presence of at least two distinctive symmetry sites of the activator ions the excitation spectra of CaAl4O7:Eu3+ (1 mol%) monitoring the 5D0?7F0 emission lines located at 577.2 nm and 579.7 nm were measured at 77 K. These two luminescence lines were already temporary assigned to be connected with different Eu3+ sites. The results are presented in Fig. 9 and Table 1 lists the energies of the excited states for the two Eu3+sites. Indeed, there are significant differences observed between these excitation spectra which confirm that the two emission lines result from Eu3+ ions located in the environments of different symmetries. Clearly, the number of lines, their positions and relative intensities are different in both excitation spectra. These results definitively prove the existence of at least two different Eu3+ sites in the investigated CaAl4O7 powders. A detailed analysis of the results in Fig. 9 reveals that the most intense lines seen in the excitation spectrum of the 579.7 nm emission seem to be also present as low intensity features in the excitation spectrum of the 577.5 nm luminescence. This may indicate appearance of energy transfer between the Eu3+ ions occupying the two positions.
Fig. 7. The excitation spectra of CaAl4O7:Eu3+ samples recorded by monitoring a 5 D0?7F2 emission at 612 nm at 298 K.
Fig. 9. The excitation spectra of CaAl4O7:Eu3+ samples recorded by monitoring a 5 D0?7F0 emission at 577.2 nm and 579.7 nm at 77 K.
Table 1 Energies (cm 1) of the excitation spectra of CaAl4O7:Eu3+ (1%) monitored for kem = 577.2 nm and 579.7 nm. em = 579.7 nm
em = 577.2 nm
27,555 26,759 26,525 26,267 25,374 25,278 25,125 24,900
27,670 26,709 26,539 26,350 25,490 25,406 25,182 25,025 24,912 21,588 21,523 19,029 18,864
21,491 Fig. 8. The excitation spectra of CaAl4O7:Eu3+ samples recorded by monitoring a 5 D0?7F2 emission at 612 nm at 77 K.
18,986
1121
M. Puchalska et al. / Optical Materials 32 (2010) 1117–1122
3.3. Decay kinetics The decay traces of the emissions from the 5D0 state for all compositions were measured at room and liquid nitrogen temperature monitoring the most intense 5D0?7F2 luminescence at 612 nm as well as the two site-distinctive components of the 5D0?7F0 transitions at 577.5 nm and 579.5 nm (298 K) and at 577.2 nm and 579.7 nm (77 K). The traces recorded for the 5D0?7F0 emissions at both temperatures are presented in Fig. 10. Single or, where necessary, biexponential functions have been used to fit the experimental data and obtained radiative lifetimes are presented in Table 2. Luminescence decay traces of the 5D0?7F2 emission at 612 nm could be reasonably reproduced with monoexponential function for all materials at both temperatures. At liquid nitrogen temperature the radiative lifetimes are almost identical (in the range of 1.56–1.69 ms) despite Eu3+ ions of different symmetries have to produce luminescence around 612 nm. At room temperature the 612 nm emission decays faster with increasing Eu content
and the time constant changes from about 1.7 ms for the 0.2% specimen to about 1.3 ms for the highest concentration of the dopant. Decays of the two 5D0?7F0 emissions are significantly different, especially at room temperature. At RT the kinetics of the 579.5 nm luminescence is clearly concentration sensitive becoming noticeably shorter for higher dopant contents. It is striking that the initial lifetime is roughly constant above 0.2%, while at later stages the luminescence decays continuously faster and faster. This may be taken as a support for the previously suggested energy transfer from the Eu3+ producing emission around 579.5 nm to the ions with characteristic luminescence at 577.5 nm. Kinetics of the latter one is perfectly identical within the whole range of Eu concentrations at RT. This is however confusing as the energy would go from a state slightly lower in energy to the one positioned higher. While such a phonon assisted transfer of energy is known and absolutely possible it is not clear at all while the opposite (typical) direction of energy transfer could not occur. On the other hand the shortening of the decay of the 579.5 nm emission may also indicate that
Fig. 10. 5D0 luminescence decay curves of CaAl4O7:Eu3+ samples excited at 532.7 nm and monitored at 577.5 nm and 579.5 nm at 298 K and at 577.2 nm and 579.7 nm at 77 K.
Table 2 Luminescence decay times for CaAl4O7:Eu3+ samples at 298 K and 77 K. Eu3+ conc. mol%
02 1 2 5 10
298 K
77 K
612 nm
577.5 nm
612 nm
577.2 nm
s1
s1 + s2
s
579.5 nm
s1 + s2
s
s1
s1
579.7 nm
s1
1.71 1.56 1.54 1.50 1.32
1.34 1.42 + 0.36 1.47 + 0.43 1.43 + 0.43 1.53 + 0.73
– 1.28 1.33 1.35 1.35
1.61 1.52 + 0.31 1.36 + 0.26 1.23 + 0.27 1.17 + 0.25
– 1.30 1.10 0.98 0.97
1.69 1.58 1.56 1.59 1.58
1.69 1.62 1.58 1.58 1.57
1.43 1.32 1.28 1.28 1.41
1122
M. Puchalska et al. / Optical Materials 32 (2010) 1117–1122
nearby a defect site is located and is able to drain energy from the excited dopant. Than this would imply that the defect locates near the Eu3+ emitting at 579.5 nm only when the dopant content surpasses the concentration of 0.2%. This seems reasonable, yet further work would be needed to understand details of this behavior. After lowering the temperature to 77 K the 579.5 nm luminescence (now positioned at 579.7 nm) is no longer concentration dependent. Interestingly, for the lowest concentrations the decay of the 577.2 nm emission at 77 K shows a significant tail indicating some afterglow. It may well be that a defect serves as a shallow trap capable of intercepting some of the energy delivered to the material and passes it back to the activator with some delay. 4. Conclusions Grossite phase of Eu-doped CaAl4O7 with the activator content up to 10 mol% was synthesized by modified Pechini method at 950 °C. The investigated crystalline powders exhibited a red luminescence with the main component peaking at 612 nm. The calculated quantum efficiencies are rather low, with the highest value reaching about 12% for 1–5% of Eu3+. Emission spectra and decay kinetics proved that Eu3+ ions occupy two distinctive symmetry sites and their surroundings get further distortion when the dopant content increases leading to continuous broadening of the spectral features. Co-doping with Na+ ions leads to basically one emission center. At low temperatures kinetics of emission from either of the sites does not change with Eu concentration. Yet, at room temperature the luminescence at 579.5 nm gets shorter with
increasing Eu content, which could indicate some energy transfer between the two sites or, more probably, a drain of the energy by a defect. References [1] S. Jonas, F. Nadachowski, D. Szwagierczak, Ceram. Int. 25 (1999) 77. [2] T. Aitasalo, J. Holsa, H. Jungner, M. Lastusaari, J. Niittykoski, M. Parkkinen, R. Valtonen, Opt. Mater. 26 (2004) 113. [3] D. Jia, J. Lumin. 117 (2006) 170. [4] Y. Murayama, Luminous Paints, in: S. Shionoya, W.M. Yen (Eds.), Phosphor Handbook, CRC Press, Boca Raton, FL, USA, 1999, p. 651. [5] A. Douy, M. Gervais, J. Am. Ceram. Soc. 83 (2000) 70. [6] J. Holsa, H. Jungner, M. Lastusaari, J. Niittykoski, J. Alloys Compd. 323–324 (2001) 326. [7] B. Hallstedt, J. Am. Ceram. Soc. 73 (1) (1990) 15. [8] D. Weber, A. Bischof, Eur. J. Miner. 6 (4) (1994) 591. [9] E.R. Boyko, L.G. Wisnyi, Acta Crystallogr. 11 (1958) 444. [10] S. Jonas, F. Nadachowski, D. Szwagierczak, Ceram. Int. 24 (1998) 211. [11] G. Blasse, B.C. Grabmaier, Luminescent Materials, Springer, Berlin, 1994. [12] M.L.H. ter Heerdt, E. van der Kolk E, W.M. Yen, A.M. Srivastava, J. Lumin. 100 (2002) 107. [13] D. Jia, J. Zhu, B. Wu, S.E. Paje, J. Lumin. 93 (2001) 107. [14] H. Yuan, W.M. Dennis, L. Lu, W.M. Yen, W. Jia, H. Liu, in: C. Ronda (Ed.), Proceedings of the 6th International Conference on the Physics and Chemistry of Luminescent Materials, Vols. 97–29, Electrochemical Society, 1998, p. 206. [15] X. Yuan, Y. Xu, Y. He, J. Alloys Compd. 441 (2007) 251. [16] E. Zych, A. Meijerink, C. de Mello Donega, J. Phys.-Condens. Mat. 15 (2003) 5145. [17] D.W. Goodwin, A.J. Lindop, Acta Crystallogr. B 26 (1970) 1230. [18] M. Buijs, A. Meijerink, G. Blasse, J. Lumin. 37 (1987) 9. [19] C. de Mello Donega, A. Meijerink, G. Blasse, J. Lumin. 62 (1994) 189. [20] C. de Mello Donega, Vibronic spectroscopy of Pr3+ in solids, Ph.D. Thesis, Utrecht University, The Netherlands, 1994. ISBN: 90-393-0813-6.