- Email: [email protected]
Effect of the PVA (polyvinyl alcohol) concentration on the optical properties of Eu-doped YAG phosphors
Optical Materials 60 (2016) 495e500
Contents lists available at ScienceDirect
Optical Materials journal homepage: www.elsevier.com/locate/optmat
Effect of the PVA (polyvinyl alcohol) concentration on the optical properties of Eu-doped YAG phosphors ^ nica C. Teixeira c, *, Daniela A. Hora a, Adriano B. Andrade a, Nilson S. Ferreira b, Vero Marcos V. dos S. Rezende d ~o Cristo ~o, SE, Brazil va Departamento de Física, Universidade Federal de Sergipe, 49100-000, Sa , 68902-280, Macapa , AP, Brazil Departamento de Física, Universidade Federal do Amapa c rio Nacional de Luz Síncrotron, Centro Nacional de Pesquisa em Energia e Materiais, P.O. Box 6192, 13084-971, Campinas, SP, Brazil Laborato d Grupo de Nanomateriais Funcionais (GNF), Departamento de Física, Universidade Federal de Sergipe, 49500-000, Itabaiana, SE, Brazil a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 July 2016 Received in revised form 22 August 2016 Accepted 6 September 2016
The influence of the polyvinyl alcohol (PVA) concentration on the synthesis and structural, morphological and optical properties of Y3Al5O13: Eu (Eu-doped YAG) was systematically investigated in this work. The final concentration of PVA in the preparation step influenced the crystallite size and also the degree of particle agglomeration in Eu-doped YAG phosphors. X-ray excited optical luminescence (XEOL) emission spectra results indicated typical Eu3þ emission lines and an abnormally intense 5D0 / 7F4. The intensity parameters U2 and U4 were calculated and indicated the PVA concentration affects the ratio U2:U4. X-ray absorption spectroscopy (XAS) results showed Eu valence did not change and the symmetry around the Eu3þ is influenced by the PVA concentration. XEOL-XAS showed the luminescence increases as a function of energy. © 2016 Elsevier B.V. All rights reserved.
Keywords: PVA based sol-gel XEOL Luminescence Phosphors Eu-doped YAG
1. Introduction Yttrium aluminate garnet (Y3Al5O12-YAG) has been largely investigated for application as a host to prepare photonic sources. It has a high thermal stability and its crystalline structure present two trivalent cationic sites able to accommodate doping ions such as rare earths, which are, for example, responsible for luminescence properties [1e5]. Doped YAG has been applied in X-ray tubes, low voltage field emission displays (FEDs), vacuum fluorescent displays [6e10], X-ray digital imaging detectors, X-ray micro-radiography and it is also suitable for X-ray imaging with high spatial resolution [11e13]. Particularly, Eu-doped YAG has been cited as a red emitter phosphor, which has potential applications in photonics, mainly in optical display and lighting, such as light emitting diodes (LEDs), plasma panel displays (PDPs), FEDs [3,14e16], fluorescence thermometry [17], etc. YAG is a versatile host, which can be obtained as crystal or ceramic powders [18e20]. In both cases, it is possible to manipulate its properties by the insertion of impurities on the host structure.
* Corresponding author. E-mail address: [email protected] (V.C. Teixeira). http://dx.doi.org/10.1016/j.optmat.2016.09.011 0925-3467/© 2016 Elsevier B.V. All rights reserved.
However, the ceramics can be produced by simpler methods than the crystals and their characteristics may be strongly affected by the way they are grown. For example, for nanosized grains, when compared to the bulk ones, they can present completely new properties [21]. In this case, the surface/volume ratio is high and the surface effects, which come from the surface dangling bonds, or quantum confinement effects may determine the material properties [21,22]. Several works report methodologies to synthesize luminescent materials in ceramic powder form, for example, solid state reaction [23], co-precipitation [24], solvothermal [25], combustion [26], glycothermal treatment [27], spray pyrolysis [28], conventional solgel [29], sol-gel assisted by organic molecules, polymers [30e35], etc. The conventional solegel route uses alkoxides for the hydrolysis and condensation of the precursors. However, this method has been recently modified by the use of other organic agents containing alcohol and/or carboxylic acid groups, such as ethylene glycol, coconut water [30,31], natural organic matter [32], PVA (polyvinyl alcohol) [33e36], etc. The sol-gel assisted by PVA, [C2H4O]n, for example, has been used in order to produce magnetic, biocompatible, and luminescent materials [33e36] and this method was employed to produce the Eu-doped YAG phosphors
496
D.A. Hora et al. / Optical Materials 60 (2016) 495e500
investigated in this work. The general aim of this work is to study the effect of PVA concentration in the step synthesis on structural, morphological, and optical properties of Eu-doped YAG. For this, the samples' properties were evaluated with X-ray powder diffraction (XRD), scanning electron microscopy (SEM), X-ray excited optical luminescence (XEOL) excitation and emission around the Eu LIII-edge (6.977 keV). XEOL excitation spectra, XEOL yield or XEOL-XAS were collected simultaneously with the X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) signals. Those last signals (XANES and EXFAS) were recorded to investigate the valence and the symmetry of Eu sites on Eu-doped YAG. 2. Methodology Eu-doped YAG powders were synthesized via sol-gel method using PVA as the condensation agent. The starting reactants were polyvinyl alcohol (PVA) (Proquímios), yttrium nitrate tetrahydrate (Sigma-Aldrich, 99.99%), aluminum nitrate nonahydrate (SigmaAldrich, 99.9%), and europium nitrate pentahydrate (Sigma-Aldrich, 99.9%). The first step was the preparation of a PVA solution (0.1 g/ cm3), in which the PVA was dissolved in distilled water and continuously stirred and heated at 100 C until solution was transparent and homogeneous. The second step was the preparation of three starting solutions to get Eu-doped YAG samples. The starting reactants (nitrates), in stoichiometric amounts, were dissolved in distilled water and different volumetric ratios of PVA solution/distilled water (PVA/DW, in ml) 5/25, 10/20 and 15/15 e were added to obtain the final solution. The following step involved heating (100 C) and stirring these solutions in order to get sols; gels were achieved after approximately 6 h. These samples were dried at 100 C/24 h to get xerogels, which were homogenized in agate mortars and calcined in an electric furnace at 1100 C/2 h [37,38] in static air atmosphere. The characterization step started with the powder X-ray diffraction (XRD) and was performed in continuous scan mode, using Co Ka (1.79 Å) radiation in a Rigaku RINT 2000/PC diffractometer in the Bragg-Brentano geometry, operating at 40 kV/ 40 mA. The particle size and shape were evaluated by scanning electron microscopy (SEM) using a Jeol JSM-7500F from the Multiuser Center of Nanotechnology at the Federal University of Sergipe (CMNano/UFS). The samples for SEM measurements were dispersed in isopropyl alcohol (P.A.) and a single drop for each sample was collected and deposited on a conductive substrate. Xray absorption spectroscopy (XAS: XANES and EXAFS) and XEOL experiments were performed at the X-ray absorption fine structure (D08B-XAFS-2) beamline at the Brazilian Synchrotron Light Laboratory (LNLS). The XAS data were recorded around the Eu-LIII edge (6.977 keV) in fluorescence mode and the signal was collected using a Ge15 detector from Specs. The total XEOL yield, or XEOL excitation spectra, and the XANES/EXAFS spectra were obtained simultaneously. This data will be referred as XEOL-XAS in the following text. The XEOL-XAS setup was composed of an optical fiber (1 mm aperture) coupled to a Hamamatsu R928 photomultiplier. The XEOL emission spectra were recorded using a specific excitation energy (6.977 keV). The setup was composed of an optical fiber and an Ocean Optics HR2000 spectrometer.
Fig. 1. XRD patterns of the calcined YAG precursor, as prepared, prepared using different PVA/DW (in ml/ml) compared to the indexed pattern (ICSD 17058) [39].
Structure (ICSD) under collection code#17058. This result suggests the PVA/DW ratio did not influence the YAG synthesis and all precursors reacted completely until the oxide single phase formation. According to the literature, at this synthesis process, the three dimensional structure of PVA acts as a confined space, which allows high contact among the solvated ions, favoring the synthesis process and also minimizing the precipitation of the precursors. Consequently, it also minimizes the formation of spurious phases [40]. After the thermal treatment, the YAG particles are formed with high crystallinity. Changes in the particle size are related to the modification of the surface energy. This thermodynamic parameter depends on the surface tension of the curved interfaces and also on the particles surface area [41]. Surface tension is a variable governed by the chemical nature of ions present on the particle surface and the PVA role is associated with the capability of H-bonding with its solvents, inducing modifications on their surface tension and in turn influencing the crystallite and the particle size [42]. To evaluate the YAG particles and crystallite size, both Scanning Electron Microcopy (SEM) and Scherer's calculations were employed. Eu-doped YAG crystallite size was evaluated using Scherrer's equation [43] (Eq. (1)):
d ¼ k$l=b$cos qB
(1)
where d is the crystallite size, k is the shape coefficient for the reciprocal lattice point (in this work it was considered k ¼ 0.89, for spherical crystallites), l is the X-ray wavelength (l ¼ 1.79 Å for CoKalpha), B is the full width at half maximum (FWHM) of the main diffraction peak, which was calculated removing the instrumental broadening obtained using a micrometric LaB6 standard; and qB corresponds to the diffraction angle of the more intense Bragg peak (420), presented in Fig. 1. The summarized results are presented in Table 1, where the biggest crystallite size was observed for the
3. Results Fig. 1 shows the X-ray powder diffraction of the Eu-doped YAG precursor, as prepared calcined at 1100 C/2 h. The diffratograms patterns exhibit only the YAG crystalline phase when compared to the standard pattern [39], indexed at the Inorganic Crystal Database
Table 1 Crystallite sizes calculated using the XRD patterns of Eu-doped YAG samples. (PVA/DW, ml/ml) Crystallite sizes (nm)
5/25 28
10/20 34
15/15 22
D.A. Hora et al. / Optical Materials 60 (2016) 495e500
samples prepared using10/20 (PVA/WD, ml/ml) solution. The smallest one was found for the samples prepared using the 15/15 solution. These results indicated YAG crystallite size can be influenced by the adjustment of PVA concentration in the starting synthesis process. Increasing the PVA concentration should reduce the crystallite size since the particles are more separated from each other. This can be observed when increasing PVA concentration from 5/25 (28 nm) to 15/15 (22 nm). The augment of the crystalline size at the 10/20 concentration (34 nm) is, however, a surprise. It is possible that the tendency of the crystallite size as a PVA concentration function is not linear, but there is no clear explanation, up to now, of the reason. SEM was employed on the investigation of Eu-doped YAG particles size and morphology. In Fig. 2, micrographs of samples produced using solutions of (a) 5/25, (b) 10/20 and (c)15/15 are presented. In all cases, it is possible to observe agglomerates composed by particles with a general elliptical shape. This kind of morphology is similar to the one reported for Eu-doped YAG particles obtained with citrate-based sol-gel [44] and also by combustion method [45]. The PVA concentration has no obvious effect on the particle shape control, but on the other hand, it can influence the particle and crystallite size, as observed in both Fig. 2 and in Table 1. In addition, the samples with the largest crystallite size, i.e., those produced using the 10/20 solution, presented lower particle agglomeration than the samples with smaller crystallite size. According to some authors, the particle agglomeration may occur due to the aggregation of irregular spherical or elliptical shapes [46], and the effect of PVA structure on the distribution of ions at the gel and xerogel steps can lead to a strong agglomeration of the primary particles [47], which grow from a bottom-up model after burning the organic compounds precursors. Thus, the PVA concentration on the starting process may influence the crystallites and particles size and also their degree of agglomeration [35]. A similar agglomeration degree is observed for Eu-doped YAG synthesized via citric acid based sol-gel [44,47]. A report about the Eudoped YAG synthesis via solid-state reaction [48] showed the samples reached an average particle size of about 2e2.5 mm, while those produced with the ethylene glycol based sol-gel produced
497
Fig. 3. XEOL emission spectra of samples produced using different PVA concentration, excited at Eu LIII-edge (6.977 keV).
particles whose sizes reached in the range of 40e200 nm, depending on the calcination temperature [48]. Eu-doped YAG particles between 20 and 30 nm were obtained via co-precipitation and also via sol-gel, using citric acid as starting precursor for the initial solution [47]. In conclusion, it was observed, via the methodology employed in this work that it is possible to produce nanoparticulated ceramic powders and that the PVA concentration can influence strongly the degree of particle agglomeration and, consequently, these features may influence the ceramics optical properties. The XEOL emission spectra are presented in Fig. 3. They were recorded at room temperature and excited with energy corresponding to the Eu LIII edge (6.977 keV). The spectra exhibit the typical Eu3þ emission lines, which correspond to the Laporte
Fig. 2. SEM images of Eu-doped YAG particles produced using different PVA concentration (PVA/DW, ml/ml): (a) 5/25, (b) 10/20 and (c) 15/15.
498
D.A. Hora et al. / Optical Materials 60 (2016) 495e500
forbidden 4f/4f intraconfigurational transitions (5D0 / 7Fj, J ¼ 0,1,2,3,4), observed at the range of 550e750 nm [49]. The 5 D0 / 7Fj, J ¼ 0, 3, which are forbidden by magnetic and forced electric dipole are attributed to very weak transitions around 585 and 655 nm [49,50], respectively. The dominant and strongest emission peak, at 590 nm, is attributed to the magnetic-dipole allowed 5D0 / 7F1 transition. In general, Eu3þ emission spectrum exhibits a very high intense emission, around 615 nm, which arises from the hypersensitive forced electric-dipole allowed 5D0 / 7F2 transition. This transition depends strongly on the structure around Eu3þion and its absence (or very low intensity) in the XEOL emission spectra (Fig. 3) suggests Eu3þ ions are located in sites with inversion center symmetry [49e52]. The 5D0 / 7F4, an electric dipole transition, presents an abnormally strong emission around 710e715 nm. Skaudizus et al. [53] showed systematically that this unusually strong emission may occur due an increase in the electronegativity or optical basicity in a specific octahedral or tetrahedral site in garnet structures such as the YAG one [49,53]. Ferreira et al. [54] suggests that it is a consequence of the polarizability of the chemical environment and also due to the environment around the Eu3þ [49,51,52,54]. Eu3þ in YAG replaces an Y3þ site, which presents D2 symmetry [55]. In the Ca3Sc2Si3O12:Eu [56], where Eu3þ occupies a D2 site symmetry, the high intensity of the 5D0 / 7F4 transition was attributed to a distortion of geometry leading to a higher symmetry around the Eu3þ [49,51,52,54e56]. Some features such as crystallinity, crystallite, particle size, and shape may influence the samples' optical properties and the synthesis condition directly influences these variables, consequently the Eu-doped YAG optical properties can be strongly affected by them. In order to understand this influence, the Eu3þJudd-Ofelt intensity parameters, Ul(l ¼ 2 and 4) [49e54], were calculated from the XEOL emission spectra (Fig. 3) using the 5D0 / 7F2 and 5 D0 / 7F4 transitions. The calculations of Judd-Ofelt intensity parameters were done using the following equation (Eq. (2)):
. X A0/l ¼ 4e2 u3 3Zc3 c Ul h5 D0 jU ðlÞ FJ 2
(2)
l¼2;4
U2(Table 2) is for the material prepared with the 10/20 solution. It may have occurred due to the high crystallinity presented by these samples when compared to the others mentioned in this work. This crystallinity may create an environment close to a perfect centrosymmetric site around the Eu3þ [49,50]. The lower crystallinity of the other samples allowed distortions in Eu3þ site, increasing the U2 values. It is worth noting that the 10/20 material also exhibit the smaller U4 intensity parameter. However, U2 was much smaller than U4 one (Table 2). It explains the higher intensity of the 5 D0 / 7F4 in relation to the 5D0 / 7F2 transition and its origin may come from differences in the site geometry in different garnet structures [56]. Thus, the observed effect on the Eu3þ XEOL emission may be understood as an effect of crystallinity and particle size as well. The symmetry and the neighborhood around the Eu3þ ions located on the particle's surface may also be affected by the particle size due to the surface/volume ratio, which is higher for nanoparticles than for bulk materials. The quenching in the emission intensity is related in this case with the crystallite size, since there are very small changes in the relation of the 5D0-7FJ intensities. It was observed, for example, in Y2O3:Eu3þ [57] that very small particles (until the 50 nm limit) present smaller quantum efficiency. The efficiency only decreases for much larger particles. The chromatic coordinates (x, y) were calculated from the XEOL emission (Fig. 3) spectra using the SpectraLux 2.0 software [58], which is based on the Comission Internationale de l’Elcairage (CIE) chromaticity diagram. The results, which are represented in Fig. 4, indicated a strong red emission with coordinates (x ¼ 0.686, y ¼ 0.314), (x ¼ 0.665, y ¼ 0.333), and (x ¼ 0.662, y ¼ 0.336) for samples produced using: (a) 5/25, (b) 10/20, and (c) 15/15 PV A/DW (in ml/ml) solutions, respectively. This result suggests these materials are red emitting phosphors and can be applied as red scintillator in X-rays detection systems, for example. XANES spectra around the Eu LIII-edge, presented in Fig. 5, are dominated by only one intense, well-resolved absorption peak associated with the Eu3þ 2p3/2 / 5d electronic transition. This is an indication that the Eu ions are in the same oxidizing state in YAG samples and that the PVA concentration does not contribute to
where A0/l ¼ A0/1 ðS0/l =S0/1 Þðsl =s1 Þ, are the spontaneous emission rates in which A0 / 1 refers to the 5D0 / 7F1, a magnetic dipole allowed transition, used as reference; S0/l are the areas under the emission curves for the 5D0 / 7Fl transitions and sl is the energy barycenter of each transition, u is the angular frequency of the incident radiation fieldc ¼ nðn2 þ 2Þ2 =9, is the Lorentz local field correction term which depends on the medium refraction index, n was considered z1.5 for several Eu3þ based solid samples; 7 and 5 D0 U ðlÞ FJ2 is the reduced matrix element, equal to 0.0032 and 0.0023 for the 5D0 / 7F2 and 5D0 / 7F4, respectively, [49,50,52,56,57]. The U2 values are more influenced by the angular part of the chemical environment while the high rank U4 is affected more by the covalence of the bond [49e52]. It is important to note that the U6 parameter was not calculated due to the absence of the 5D0/7F6 transition in the XEOL emission spectra. The lowest value of
Table 2 Judd-Ofelt intensity parameters calculated from the Eu-doped YAG XEOL emission spectra. (PVA/DW, ml/ml)
U2 (1020 cm2)
U4 (1020 cm2)
U2/U4
5/25 10/20 15/15
1.5 0.8 2.2
2.1 1.8 3.9
0.7 0.4 0.6
Fig. 4. Chromaticity diagram showing the emission color samples produced using PVA/DW (in ml/ml): (a) 5/25, (b) 10/20 and (c) 15/15.
D.A. Hora et al. / Optical Materials 60 (2016) 495e500
Fig. 5. XAS and XEOL-XAS spectra excited in the Eu LIII-edge energy region for samples produced using PVA/DW (in ml/ml): (a) 5/25, (b) 10/20 and (c) 15/15.
Fig. 6. Simulated and experimental FT amplitude of EXAFS signals around the Eu LIIIedge.
their reduction. XANES results are in agreement with the XEOL spectra showed in Fig. 3, which exhibited only typical Eu3þ (4f e 4f intraconfigurational transitions) emission lines. The XEOL-XAS spectra are also shown in Fig. 5. These data were recorded simultaneously with the XANES and normalized by I0. The
499
XEOL-XAS spectra are composed of the area under the optical radioluminescence (RL) calculated for each incident X-ray photon energy, as a function of the incident X-rays photon energies (red curves shown in Fig. 5). The intensity of the XEOL-XAS spectra increases with the excitation photon energy in the range of 6.900e7.600 keV. It also noted that both XAS and XEOL-XAS spectra exhibit positive edges jump. The phenomenological theory for XEOL-XAS spectrum, proposed by Emura et al. [59], suggests that the XEOL edge shape is related to a competitive absorption of specific sites, which are responsible for the luminescence and those ones that are responsible for different physical processes. The positive edge jump at the XEOL-XAS curves is an indication that there is no competition between different excitation channels. That is to say that the recombination of the charge carrier pairs may directly excite optical channels into the material, which are the Eu3þ ions, and they decay from the 4f excited state to the 4f6ground states. The influence of the PVA concentration on the starting synthesis process leads to different observations of the crystallite size, particles agglomeration, and, consequently, on the XEOL emission spectra. Thus, in order to study the influence of PVA concentration on the local Eu environment, EXAFS spectra were collected and simulated in order to obtain structural parameters such as bond distance (Å) and the DebyeeWaller factor (Ds2), that is the crystal lattice disorder factor, for the first EueO bond shell. Simulated and experimental Fourier transforms (FT) of the EXAFS spectra about the Eu LIII-edge are shown in Fig. 6. The radial distribution functions (RDF) were obtained from the Fourier transforms (FT) of K2(k), and the EXAFS fits were performed based on the probability of Eu ions replacing Y ones in the YAG matrix. Both of these ions, Y and Eu, are in the trivalent oxidizing state in the studied material. Furthermore, they present similar ionic radii [60]. The fitted Fourier transform was calculated considering Eu-O1 and Eu-O2 surrounded by the four nearest oxygen neighbors. As shown in Fig. 6, all spectra exhibit a first peak around 2.3 Å, which is attributed to the first EueO interaction. Table 3 presents the structural parameters obtained from the simulated data, which was calculated using the Artemis software [61]. R-factor showed in the Table 3 reveals that the simulated data match well the experimental data. The EXAFS results indicated that the EueO distance and the Debye-Waller factor for the first and second coordination shells change with the PVA concentration. The DebyeeWaller factor and the difference between EueO1 and EueO2 are smaller for the 10/20 and higher for the 15/15 samples. A small Debye-Waller factor value for the 10/20 sample is an indication that the Eu3þ ions are located in a less distorted site, which is in agreement with the previous discussion. 4. Conclusion Eu-doped YAG nanophosphors were synthesized via PVA based sol-gel and the influence of the PVA concentration in the starting
Table 3 Simulated structural parameters of Eu-doped YAG produced with different PVA concentrations. Samples (PVA/DW, ml/ml)
Shell
CN*
Atomic length R (Å)
DebyeeWaller factor (Å2)
5/15
EueO1 EueO2 EueO1 EueO2 EueO1 EueO2
6 6 6 6 6 6
2.3078 2.4383 2.3128 2.4433 2.2974 2.4279
0.00335 0.00018 0.00211 0.00016 0.00368 0.00036
10/20 15/15 *
CN: coordination number
500
D.A. Hora et al. / Optical Materials 60 (2016) 495e500
synthesis process on the structural, morphological, and optical properties were investigated by XRD, SEM, XEOL, and XAS measurements. XRD results showed that the YAG phase was obtained and that the crystallite sizes, in fact, influenced by the PVA concentration. SEM images presented large agglomerates composed of small particles relatively regular in shape, which agrees with some reports. The XEOL-XAS was monitored as an energy function around the Eu LIII-edge. The XEOL emission spectra presented the typical Eu3þ intraconfigurational 4f e 4f lines and an abnormally strong 5D0 / 7F4 transition. The chromatic coordinates indicated that it pure red emitters were produced. XRD, XEOL and EXAFS results indicated that the Eu is located in a site with inversion center symmetry, which is more ordered for the 10/20 PV A/DW, ml/ml samples. Acknowledgments The authors are grateful to the FINEP (Brazilian Innovation Agency), CAPES (Brazilian Federal Agency), CNPq (No. 470.972/ 2013-0) (Brazilian National Council for Scientific and Technological Development), to the LNLS/CNPEM (Brazilian Synchrotron Light Laboratory e Brazilian Center for Research in Energy and Materials) for the facilities and for the financial support (proposal XAFS2#18003/15) and to the CMNano/UFS for the facility. They would also like thank Prof. Dr. Lucas C.V. Rodrigues, from the Universidade de S~ ao Paulo, for the valuable discussions about XEOL. References [1] A. Yousif, H.C. Swart, O.M. Ntwaeaborwa, E. Coetsee, Appl. Surf. Sci. 270 (2013) 331. [2] E.D. Milliken, L.C. Oliveira, G. Denis, E.G. Yukihara, J. Lumin. 132 (2012) 2495. [3] J.-H. In, H.-C. Lee, M.-J. Yoon, K.-K. Lee, J.-W. Lee, C.-H. Lee, J. Supercrit. Fluids 40 (2007) 389. [4] M.V. dos S. Rezende, C.W.A. Paschoal, Opt. Mater. (Amst) 46 (2015) 530. [5] M.V. dos S. Rezende, J. Phys. Chem. Solids 75 (2014) 1113. [6] H.S. Jang, W. Bin Im, D.C. Lee, D.Y. Jeon, S.S. Kim, J. Lumin. 126 (2007) 371. [7] K.Y. Jung, H.W. Lee, J. Lumin. 126 (2007) 469. [8] Y. Zhou, J. Lin, M. Yu, S. Wang, H. Zhang, Mater. Lett. 56 (2002) 628. [9] S. Zhou, Z. Fu, J. Zhang, S. Zhang, J. Lumin. 118 (2006) 179. [10] D. Ravichandran, R. Roy, A.G. Chakhovskoi, C.E. Hunt, W.B. White, S. Erdei, J. Lumin. 71 (1997) 291. [11] J. Tous, M. Horv ath, L. Pína, K. Bla zek, B. Sopko, Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip. 591 (2008) 264. [12] J. Tous, K. Blazek, L. Pína, B. Sopko, Appl. Radiat. Isot.. 68 (n.d.) 651. [13] S.L. David, C.M. Michail, M. Roussou, E. Nirgianaki, A.E. Toutountzis, I.G. Valais, G. Fountos, P.F. Liaparinos, I. Kandarakis, G. Panayiotakis, IEEE Trans, Nucl. Sci. 57 (2010) 951. [14] S. Shikao, W. Jiye, J. Alloys Compd. 327 (2001) 82. [15] Y.H. Zhou, J. Lin, S.B. Wang, H.J. Zhang, Opt. Mater. (Amst) 20 (2002) 13. [16] C.-H. Lu, W.-T. Hsu, B.-M. Cheng, J. Appl. Phys. 100 (2006) 063535. [17] S.W. Allison, G.T. Gillies, A.J. Rondinone, M.R. Cates, Nanotechnology 14 (2003) 859. [18] Y. Li, J. Zhang, Q. Xiao, R. Zeng, Mater. Lett. 62 (2008) 3787. kov vr, K. Kamada, V. Babin, A. Yoshikawa, M. Nikl, 3 (n.d.) 1. [19] E. Miho a, K. Va kova , K. Va vr [20] E. Miho u, K. Kamada, V. Babin, a. Yoshikawa, M. Nikl, Radiat. Meas. 56 (2013) 98. [21] A. Ikesue, Y.L. Aung, T. Taira, T. Kamimura, K. Yoshida, G.L. Messing, Annu. Rev. Mater. Res. 36 (2006) 397. [22] V.C. Teixeira, P.J.R. Montes, M.E.G. Valerio, Opt. Mater. (Amst) 36 (2014) 1580. [23] M. Sekita, H. Haneda, S. Shirasaki, T. Yanagitani, J. Appl. Phys. 69 (1991) 3709.
[24] X. Li, H. Liu, J.Y. Wang, H.M. Cui, F. Han, X.D. Zhang, R.I. Boughton, Mater. Lett. 58 (2004) 2377. [25] X. Li, H. Liu, J. Wang, H. Cui, S. Yang, I.R. Boughton, J. Phys. Chem. Solids 66 (2005) 201. [26] S. Ramanathan, M.B. Kakade, S.K. Roy, K.K. Kutty, Ceram. Int. 29 (2003) 477. [27] M. Inoue, H. Otsu, H. Kominami, T. Inui, J. Alloys Compd. 226 (1995) 146. [28] M. Nyman, J. Caruso, M.J. Hampden-Smith, T.T. Kodas, J. Am. Ceram. Soc. 80 (1997) 1231. [29] C.H. Lu, W.T. Hsu, C.H. Hsu, H.C. Lu, B.M. Cheng, J. Alloys Compd. 456 (2008) 57. [30] M.V. dos S. Rezende, M.E.G. Valerio, R.A. Jackson, Mater. Res. Bull. 61 (2015) 348. [31] M.V. dos S. Rezende, P.J.R. Montes, F.M. dos S. Soares, C. dos Santos, M.E.G. Valerio, J. Synchrotron Radiat. 21 (2014) 143. ~o, Z.S. Macedo, Powder Technol. 254 (2014) 344. [32] G. da C. Cunha, L.P.C. Roma [33] T. Liu, Y. Xu, Y. Li, Z. Wang, J. Zhao, Ceram. Int. 37 (2011) 3361. [34] T. Liu, Y. Xu, C. Zeng, Mater. Sci. Eng. B Solid-State Mater. Adv. Technol. 176 (2011) 535. [35] M. Sivakumar, S. Kanagesan, R. Suresh Babu, S. Jesurani, R. Velmurugan, C. Thirupathi, T. Kalaivani, J. Mater. Sci. Mater. Electron 23 (2012) 1045. [36] O. Kaygili, S. Keser, R.H.A. Orainy, T. Ates, F. Yakuphanoglu, Mater. Sci. Eng. C 35 (2014) 239. [37] X. Xu, X. Sun, H. Liu, J.-G. Li, X. Li, D. Huo, S. Liu, J. Am. Ceram. Soc. 95 (2012) 3821. [38] J. Li, F. Chen, W. Liu, W. Zhang, L. Wang, X. Ba, Y. Zhu, Y. Pan, J. Guo, J. Eur. Ceram. Soc. 32 (2012) 2971. [39] L. Dobrzycki, E. Bulska, D.A. Pawlak, Z. Frukacz, K. Wo zniak, Inorg. Chem. 43 (2004) 7656. [40] J. Feng, T. Liu, Y. Xu, J. Zhao, Y. He, Ceram. Int. 37 (2011) 1203. [41] A. Rusanov, Surf. Sci. Rep. 58 (2005) 111. [42] A. Bhattacharya, P. Ray, J. Appl. Polym. Sci. 93 (2004) 122. [43] B.D. Cullity, Elements of X-Ray Diffraction, Addison-Wesley Publishing Company Inc., Massachusetts, 1956. [44] F. Yu, D. Yuan, X. Cheng, X. Duan, X. Wang, L. Kong, L. Wang, Z. Li, Mater. Lett. 61 (2007) 2322. [45] K. Guo, X.-M. Zhang, H.-H. Chen, X.-X. Yang, X. Guo, J.-T. Zhao, Mater. Res. Bull. 45 (2010) 1157. [46] Z. Sun, D. Yuan, X. Duan, X. Wei, H. Sun, C. Luan, Z. Wang, X. Shi, D. Xu, M. Lv, J. Cryst. Growth 260 (2004) 171. [47] P. Ramanujam, B. Vaidhyanathan, J. Binner, A. Anshuman, C. Spacie, Ceram. Int. 40 (2014) 4179. [48] C.-H. Hsu, C.-H. Huang, B.-M. Cheng, C.-H. Lu, Mater. Chem. Phys. 124 (2010) 632. [49] K. Binnemans, Coord. Chem. Rev. 295 (2015) 1. [50] H.F. Brito, O.M.L. Malta, M.C.F.C. Felinto, E.E.S. Teotonio, Luminescence phenomena involving metal enolates, in: J. Zabicky (Ed.), The Chemistry of Metal Enolates Part 1, John Wiley & Sons Ltd, West Sussex, 2009. [51] A.P. Souza, L.C.V. Rodrigues, H.F. Brito, S. Alves, O.L. Malta, J. Lumin. 130 (2010) 181. [52] C.V. Rodrigues, L.L. Luz, J.D.L. Dutra, S.A. Junior, O.L. Malta, C.C. Gatto, H.C. Streit, R.O. Freire, C. Wickleder, M.O. Rodrigues, Phys. Chem. Chem. Phys. 16 (2014) 14858. [53] R. Skaudzius, A. Katelnikovas, D. Enseling, A. Kareiva, T. Jüstel, J. Lumin. 147 (2014) 290. [54] R.A.S. Ferreira, S.S. Nobre, C.M. Granadeiro, H.I.S. Nogueira, L.D. Carlos, O.L. Malta, J. Lumin. 121 (2006) 561. [55] I.E. Kolesnikov, D.V. Tolstikova, A.V. Kurochkin, A.A. Manshina, M.D. Mikhailov, Opt. Mater. 37 (2014) 306. [56] M. Bettinelli, A. Speghini, F. Piccinelli, A.N.C. Neto, O.L. Malta, J. Lumin. 131 (2011) 1026. € lsa €, [57] I.G.N. Silva, L.C.V. Rodrigues, E.R. Souza, J. Kai, M.C.F.C. Felinto, J. Ho H.F. Brito, O.L. Malta, Opt. Mater. 40 (2015) 41. ^ntico [58] P.A. Santa-Cruz, F.S. Teles, Spectra Lux Software V. 2.0 Beta, Ponto Qua Nanodispositivos, RENAMI, 2003. [59] S. Emura, H. Maeda, M. Nomura, Phys. B Condens. Matter. 208e209 (1995) 108. [60] R.D. Shannon, Acta Crystallogr. Sect. A 32 (1976) 751. [61] B. Ravel, M. Newville, J. Synchrotron Radiat. 12 (2005) 537.
Contents lists available at ScienceDirect
Optical Materials journal homepage: www.elsevier.com/locate/optmat
Effect of the PVA (polyvinyl alcohol) concentration on the optical properties of Eu-doped YAG phosphors ^ nica C. Teixeira c, *, Daniela A. Hora a, Adriano B. Andrade a, Nilson S. Ferreira b, Vero Marcos V. dos S. Rezende d ~o Cristo ~o, SE, Brazil va Departamento de Física, Universidade Federal de Sergipe, 49100-000, Sa , 68902-280, Macapa , AP, Brazil Departamento de Física, Universidade Federal do Amapa c rio Nacional de Luz Síncrotron, Centro Nacional de Pesquisa em Energia e Materiais, P.O. Box 6192, 13084-971, Campinas, SP, Brazil Laborato d Grupo de Nanomateriais Funcionais (GNF), Departamento de Física, Universidade Federal de Sergipe, 49500-000, Itabaiana, SE, Brazil a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 July 2016 Received in revised form 22 August 2016 Accepted 6 September 2016
The influence of the polyvinyl alcohol (PVA) concentration on the synthesis and structural, morphological and optical properties of Y3Al5O13: Eu (Eu-doped YAG) was systematically investigated in this work. The final concentration of PVA in the preparation step influenced the crystallite size and also the degree of particle agglomeration in Eu-doped YAG phosphors. X-ray excited optical luminescence (XEOL) emission spectra results indicated typical Eu3þ emission lines and an abnormally intense 5D0 / 7F4. The intensity parameters U2 and U4 were calculated and indicated the PVA concentration affects the ratio U2:U4. X-ray absorption spectroscopy (XAS) results showed Eu valence did not change and the symmetry around the Eu3þ is influenced by the PVA concentration. XEOL-XAS showed the luminescence increases as a function of energy. © 2016 Elsevier B.V. All rights reserved.
Keywords: PVA based sol-gel XEOL Luminescence Phosphors Eu-doped YAG
1. Introduction Yttrium aluminate garnet (Y3Al5O12-YAG) has been largely investigated for application as a host to prepare photonic sources. It has a high thermal stability and its crystalline structure present two trivalent cationic sites able to accommodate doping ions such as rare earths, which are, for example, responsible for luminescence properties [1e5]. Doped YAG has been applied in X-ray tubes, low voltage field emission displays (FEDs), vacuum fluorescent displays [6e10], X-ray digital imaging detectors, X-ray micro-radiography and it is also suitable for X-ray imaging with high spatial resolution [11e13]. Particularly, Eu-doped YAG has been cited as a red emitter phosphor, which has potential applications in photonics, mainly in optical display and lighting, such as light emitting diodes (LEDs), plasma panel displays (PDPs), FEDs [3,14e16], fluorescence thermometry [17], etc. YAG is a versatile host, which can be obtained as crystal or ceramic powders [18e20]. In both cases, it is possible to manipulate its properties by the insertion of impurities on the host structure.
* Corresponding author. E-mail address: [email protected] (V.C. Teixeira). http://dx.doi.org/10.1016/j.optmat.2016.09.011 0925-3467/© 2016 Elsevier B.V. All rights reserved.
However, the ceramics can be produced by simpler methods than the crystals and their characteristics may be strongly affected by the way they are grown. For example, for nanosized grains, when compared to the bulk ones, they can present completely new properties [21]. In this case, the surface/volume ratio is high and the surface effects, which come from the surface dangling bonds, or quantum confinement effects may determine the material properties [21,22]. Several works report methodologies to synthesize luminescent materials in ceramic powder form, for example, solid state reaction [23], co-precipitation [24], solvothermal [25], combustion [26], glycothermal treatment [27], spray pyrolysis [28], conventional solgel [29], sol-gel assisted by organic molecules, polymers [30e35], etc. The conventional solegel route uses alkoxides for the hydrolysis and condensation of the precursors. However, this method has been recently modified by the use of other organic agents containing alcohol and/or carboxylic acid groups, such as ethylene glycol, coconut water [30,31], natural organic matter [32], PVA (polyvinyl alcohol) [33e36], etc. The sol-gel assisted by PVA, [C2H4O]n, for example, has been used in order to produce magnetic, biocompatible, and luminescent materials [33e36] and this method was employed to produce the Eu-doped YAG phosphors
496
D.A. Hora et al. / Optical Materials 60 (2016) 495e500
investigated in this work. The general aim of this work is to study the effect of PVA concentration in the step synthesis on structural, morphological, and optical properties of Eu-doped YAG. For this, the samples' properties were evaluated with X-ray powder diffraction (XRD), scanning electron microscopy (SEM), X-ray excited optical luminescence (XEOL) excitation and emission around the Eu LIII-edge (6.977 keV). XEOL excitation spectra, XEOL yield or XEOL-XAS were collected simultaneously with the X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) signals. Those last signals (XANES and EXFAS) were recorded to investigate the valence and the symmetry of Eu sites on Eu-doped YAG. 2. Methodology Eu-doped YAG powders were synthesized via sol-gel method using PVA as the condensation agent. The starting reactants were polyvinyl alcohol (PVA) (Proquímios), yttrium nitrate tetrahydrate (Sigma-Aldrich, 99.99%), aluminum nitrate nonahydrate (SigmaAldrich, 99.9%), and europium nitrate pentahydrate (Sigma-Aldrich, 99.9%). The first step was the preparation of a PVA solution (0.1 g/ cm3), in which the PVA was dissolved in distilled water and continuously stirred and heated at 100 C until solution was transparent and homogeneous. The second step was the preparation of three starting solutions to get Eu-doped YAG samples. The starting reactants (nitrates), in stoichiometric amounts, were dissolved in distilled water and different volumetric ratios of PVA solution/distilled water (PVA/DW, in ml) 5/25, 10/20 and 15/15 e were added to obtain the final solution. The following step involved heating (100 C) and stirring these solutions in order to get sols; gels were achieved after approximately 6 h. These samples were dried at 100 C/24 h to get xerogels, which were homogenized in agate mortars and calcined in an electric furnace at 1100 C/2 h [37,38] in static air atmosphere. The characterization step started with the powder X-ray diffraction (XRD) and was performed in continuous scan mode, using Co Ka (1.79 Å) radiation in a Rigaku RINT 2000/PC diffractometer in the Bragg-Brentano geometry, operating at 40 kV/ 40 mA. The particle size and shape were evaluated by scanning electron microscopy (SEM) using a Jeol JSM-7500F from the Multiuser Center of Nanotechnology at the Federal University of Sergipe (CMNano/UFS). The samples for SEM measurements were dispersed in isopropyl alcohol (P.A.) and a single drop for each sample was collected and deposited on a conductive substrate. Xray absorption spectroscopy (XAS: XANES and EXAFS) and XEOL experiments were performed at the X-ray absorption fine structure (D08B-XAFS-2) beamline at the Brazilian Synchrotron Light Laboratory (LNLS). The XAS data were recorded around the Eu-LIII edge (6.977 keV) in fluorescence mode and the signal was collected using a Ge15 detector from Specs. The total XEOL yield, or XEOL excitation spectra, and the XANES/EXAFS spectra were obtained simultaneously. This data will be referred as XEOL-XAS in the following text. The XEOL-XAS setup was composed of an optical fiber (1 mm aperture) coupled to a Hamamatsu R928 photomultiplier. The XEOL emission spectra were recorded using a specific excitation energy (6.977 keV). The setup was composed of an optical fiber and an Ocean Optics HR2000 spectrometer.
Fig. 1. XRD patterns of the calcined YAG precursor, as prepared, prepared using different PVA/DW (in ml/ml) compared to the indexed pattern (ICSD 17058) [39].
Structure (ICSD) under collection code#17058. This result suggests the PVA/DW ratio did not influence the YAG synthesis and all precursors reacted completely until the oxide single phase formation. According to the literature, at this synthesis process, the three dimensional structure of PVA acts as a confined space, which allows high contact among the solvated ions, favoring the synthesis process and also minimizing the precipitation of the precursors. Consequently, it also minimizes the formation of spurious phases [40]. After the thermal treatment, the YAG particles are formed with high crystallinity. Changes in the particle size are related to the modification of the surface energy. This thermodynamic parameter depends on the surface tension of the curved interfaces and also on the particles surface area [41]. Surface tension is a variable governed by the chemical nature of ions present on the particle surface and the PVA role is associated with the capability of H-bonding with its solvents, inducing modifications on their surface tension and in turn influencing the crystallite and the particle size [42]. To evaluate the YAG particles and crystallite size, both Scanning Electron Microcopy (SEM) and Scherer's calculations were employed. Eu-doped YAG crystallite size was evaluated using Scherrer's equation [43] (Eq. (1)):
d ¼ k$l=b$cos qB
(1)
where d is the crystallite size, k is the shape coefficient for the reciprocal lattice point (in this work it was considered k ¼ 0.89, for spherical crystallites), l is the X-ray wavelength (l ¼ 1.79 Å for CoKalpha), B is the full width at half maximum (FWHM) of the main diffraction peak, which was calculated removing the instrumental broadening obtained using a micrometric LaB6 standard; and qB corresponds to the diffraction angle of the more intense Bragg peak (420), presented in Fig. 1. The summarized results are presented in Table 1, where the biggest crystallite size was observed for the
3. Results Fig. 1 shows the X-ray powder diffraction of the Eu-doped YAG precursor, as prepared calcined at 1100 C/2 h. The diffratograms patterns exhibit only the YAG crystalline phase when compared to the standard pattern [39], indexed at the Inorganic Crystal Database
Table 1 Crystallite sizes calculated using the XRD patterns of Eu-doped YAG samples. (PVA/DW, ml/ml) Crystallite sizes (nm)
5/25 28
10/20 34
15/15 22
D.A. Hora et al. / Optical Materials 60 (2016) 495e500
samples prepared using10/20 (PVA/WD, ml/ml) solution. The smallest one was found for the samples prepared using the 15/15 solution. These results indicated YAG crystallite size can be influenced by the adjustment of PVA concentration in the starting synthesis process. Increasing the PVA concentration should reduce the crystallite size since the particles are more separated from each other. This can be observed when increasing PVA concentration from 5/25 (28 nm) to 15/15 (22 nm). The augment of the crystalline size at the 10/20 concentration (34 nm) is, however, a surprise. It is possible that the tendency of the crystallite size as a PVA concentration function is not linear, but there is no clear explanation, up to now, of the reason. SEM was employed on the investigation of Eu-doped YAG particles size and morphology. In Fig. 2, micrographs of samples produced using solutions of (a) 5/25, (b) 10/20 and (c)15/15 are presented. In all cases, it is possible to observe agglomerates composed by particles with a general elliptical shape. This kind of morphology is similar to the one reported for Eu-doped YAG particles obtained with citrate-based sol-gel [44] and also by combustion method [45]. The PVA concentration has no obvious effect on the particle shape control, but on the other hand, it can influence the particle and crystallite size, as observed in both Fig. 2 and in Table 1. In addition, the samples with the largest crystallite size, i.e., those produced using the 10/20 solution, presented lower particle agglomeration than the samples with smaller crystallite size. According to some authors, the particle agglomeration may occur due to the aggregation of irregular spherical or elliptical shapes [46], and the effect of PVA structure on the distribution of ions at the gel and xerogel steps can lead to a strong agglomeration of the primary particles [47], which grow from a bottom-up model after burning the organic compounds precursors. Thus, the PVA concentration on the starting process may influence the crystallites and particles size and also their degree of agglomeration [35]. A similar agglomeration degree is observed for Eu-doped YAG synthesized via citric acid based sol-gel [44,47]. A report about the Eudoped YAG synthesis via solid-state reaction [48] showed the samples reached an average particle size of about 2e2.5 mm, while those produced with the ethylene glycol based sol-gel produced
497
Fig. 3. XEOL emission spectra of samples produced using different PVA concentration, excited at Eu LIII-edge (6.977 keV).
particles whose sizes reached in the range of 40e200 nm, depending on the calcination temperature [48]. Eu-doped YAG particles between 20 and 30 nm were obtained via co-precipitation and also via sol-gel, using citric acid as starting precursor for the initial solution [47]. In conclusion, it was observed, via the methodology employed in this work that it is possible to produce nanoparticulated ceramic powders and that the PVA concentration can influence strongly the degree of particle agglomeration and, consequently, these features may influence the ceramics optical properties. The XEOL emission spectra are presented in Fig. 3. They were recorded at room temperature and excited with energy corresponding to the Eu LIII edge (6.977 keV). The spectra exhibit the typical Eu3þ emission lines, which correspond to the Laporte
Fig. 2. SEM images of Eu-doped YAG particles produced using different PVA concentration (PVA/DW, ml/ml): (a) 5/25, (b) 10/20 and (c) 15/15.
498
D.A. Hora et al. / Optical Materials 60 (2016) 495e500
forbidden 4f/4f intraconfigurational transitions (5D0 / 7Fj, J ¼ 0,1,2,3,4), observed at the range of 550e750 nm [49]. The 5 D0 / 7Fj, J ¼ 0, 3, which are forbidden by magnetic and forced electric dipole are attributed to very weak transitions around 585 and 655 nm [49,50], respectively. The dominant and strongest emission peak, at 590 nm, is attributed to the magnetic-dipole allowed 5D0 / 7F1 transition. In general, Eu3þ emission spectrum exhibits a very high intense emission, around 615 nm, which arises from the hypersensitive forced electric-dipole allowed 5D0 / 7F2 transition. This transition depends strongly on the structure around Eu3þion and its absence (or very low intensity) in the XEOL emission spectra (Fig. 3) suggests Eu3þ ions are located in sites with inversion center symmetry [49e52]. The 5D0 / 7F4, an electric dipole transition, presents an abnormally strong emission around 710e715 nm. Skaudizus et al. [53] showed systematically that this unusually strong emission may occur due an increase in the electronegativity or optical basicity in a specific octahedral or tetrahedral site in garnet structures such as the YAG one [49,53]. Ferreira et al. [54] suggests that it is a consequence of the polarizability of the chemical environment and also due to the environment around the Eu3þ [49,51,52,54]. Eu3þ in YAG replaces an Y3þ site, which presents D2 symmetry [55]. In the Ca3Sc2Si3O12:Eu [56], where Eu3þ occupies a D2 site symmetry, the high intensity of the 5D0 / 7F4 transition was attributed to a distortion of geometry leading to a higher symmetry around the Eu3þ [49,51,52,54e56]. Some features such as crystallinity, crystallite, particle size, and shape may influence the samples' optical properties and the synthesis condition directly influences these variables, consequently the Eu-doped YAG optical properties can be strongly affected by them. In order to understand this influence, the Eu3þJudd-Ofelt intensity parameters, Ul(l ¼ 2 and 4) [49e54], were calculated from the XEOL emission spectra (Fig. 3) using the 5D0 / 7F2 and 5 D0 / 7F4 transitions. The calculations of Judd-Ofelt intensity parameters were done using the following equation (Eq. (2)):
. X A0/l ¼ 4e2 u3 3Zc3 c Ul h5 D0 jU ðlÞ FJ 2
(2)
l¼2;4
U2(Table 2) is for the material prepared with the 10/20 solution. It may have occurred due to the high crystallinity presented by these samples when compared to the others mentioned in this work. This crystallinity may create an environment close to a perfect centrosymmetric site around the Eu3þ [49,50]. The lower crystallinity of the other samples allowed distortions in Eu3þ site, increasing the U2 values. It is worth noting that the 10/20 material also exhibit the smaller U4 intensity parameter. However, U2 was much smaller than U4 one (Table 2). It explains the higher intensity of the 5 D0 / 7F4 in relation to the 5D0 / 7F2 transition and its origin may come from differences in the site geometry in different garnet structures [56]. Thus, the observed effect on the Eu3þ XEOL emission may be understood as an effect of crystallinity and particle size as well. The symmetry and the neighborhood around the Eu3þ ions located on the particle's surface may also be affected by the particle size due to the surface/volume ratio, which is higher for nanoparticles than for bulk materials. The quenching in the emission intensity is related in this case with the crystallite size, since there are very small changes in the relation of the 5D0-7FJ intensities. It was observed, for example, in Y2O3:Eu3þ [57] that very small particles (until the 50 nm limit) present smaller quantum efficiency. The efficiency only decreases for much larger particles. The chromatic coordinates (x, y) were calculated from the XEOL emission (Fig. 3) spectra using the SpectraLux 2.0 software [58], which is based on the Comission Internationale de l’Elcairage (CIE) chromaticity diagram. The results, which are represented in Fig. 4, indicated a strong red emission with coordinates (x ¼ 0.686, y ¼ 0.314), (x ¼ 0.665, y ¼ 0.333), and (x ¼ 0.662, y ¼ 0.336) for samples produced using: (a) 5/25, (b) 10/20, and (c) 15/15 PV A/DW (in ml/ml) solutions, respectively. This result suggests these materials are red emitting phosphors and can be applied as red scintillator in X-rays detection systems, for example. XANES spectra around the Eu LIII-edge, presented in Fig. 5, are dominated by only one intense, well-resolved absorption peak associated with the Eu3þ 2p3/2 / 5d electronic transition. This is an indication that the Eu ions are in the same oxidizing state in YAG samples and that the PVA concentration does not contribute to
where A0/l ¼ A0/1 ðS0/l =S0/1 Þðsl =s1 Þ, are the spontaneous emission rates in which A0 / 1 refers to the 5D0 / 7F1, a magnetic dipole allowed transition, used as reference; S0/l are the areas under the emission curves for the 5D0 / 7Fl transitions and sl is the energy barycenter of each transition, u is the angular frequency of the incident radiation fieldc ¼ nðn2 þ 2Þ2 =9, is the Lorentz local field correction term which depends on the medium refraction index, n was considered z1.5 for several Eu3þ based solid samples; 7 and 5 D0 U ðlÞ FJ2 is the reduced matrix element, equal to 0.0032 and 0.0023 for the 5D0 / 7F2 and 5D0 / 7F4, respectively, [49,50,52,56,57]. The U2 values are more influenced by the angular part of the chemical environment while the high rank U4 is affected more by the covalence of the bond [49e52]. It is important to note that the U6 parameter was not calculated due to the absence of the 5D0/7F6 transition in the XEOL emission spectra. The lowest value of
Table 2 Judd-Ofelt intensity parameters calculated from the Eu-doped YAG XEOL emission spectra. (PVA/DW, ml/ml)
U2 (1020 cm2)
U4 (1020 cm2)
U2/U4
5/25 10/20 15/15
1.5 0.8 2.2
2.1 1.8 3.9
0.7 0.4 0.6
Fig. 4. Chromaticity diagram showing the emission color samples produced using PVA/DW (in ml/ml): (a) 5/25, (b) 10/20 and (c) 15/15.
D.A. Hora et al. / Optical Materials 60 (2016) 495e500
Fig. 5. XAS and XEOL-XAS spectra excited in the Eu LIII-edge energy region for samples produced using PVA/DW (in ml/ml): (a) 5/25, (b) 10/20 and (c) 15/15.
Fig. 6. Simulated and experimental FT amplitude of EXAFS signals around the Eu LIIIedge.
their reduction. XANES results are in agreement with the XEOL spectra showed in Fig. 3, which exhibited only typical Eu3þ (4f e 4f intraconfigurational transitions) emission lines. The XEOL-XAS spectra are also shown in Fig. 5. These data were recorded simultaneously with the XANES and normalized by I0. The
499
XEOL-XAS spectra are composed of the area under the optical radioluminescence (RL) calculated for each incident X-ray photon energy, as a function of the incident X-rays photon energies (red curves shown in Fig. 5). The intensity of the XEOL-XAS spectra increases with the excitation photon energy in the range of 6.900e7.600 keV. It also noted that both XAS and XEOL-XAS spectra exhibit positive edges jump. The phenomenological theory for XEOL-XAS spectrum, proposed by Emura et al. [59], suggests that the XEOL edge shape is related to a competitive absorption of specific sites, which are responsible for the luminescence and those ones that are responsible for different physical processes. The positive edge jump at the XEOL-XAS curves is an indication that there is no competition between different excitation channels. That is to say that the recombination of the charge carrier pairs may directly excite optical channels into the material, which are the Eu3þ ions, and they decay from the 4f excited state to the 4f6ground states. The influence of the PVA concentration on the starting synthesis process leads to different observations of the crystallite size, particles agglomeration, and, consequently, on the XEOL emission spectra. Thus, in order to study the influence of PVA concentration on the local Eu environment, EXAFS spectra were collected and simulated in order to obtain structural parameters such as bond distance (Å) and the DebyeeWaller factor (Ds2), that is the crystal lattice disorder factor, for the first EueO bond shell. Simulated and experimental Fourier transforms (FT) of the EXAFS spectra about the Eu LIII-edge are shown in Fig. 6. The radial distribution functions (RDF) were obtained from the Fourier transforms (FT) of K2(k), and the EXAFS fits were performed based on the probability of Eu ions replacing Y ones in the YAG matrix. Both of these ions, Y and Eu, are in the trivalent oxidizing state in the studied material. Furthermore, they present similar ionic radii [60]. The fitted Fourier transform was calculated considering Eu-O1 and Eu-O2 surrounded by the four nearest oxygen neighbors. As shown in Fig. 6, all spectra exhibit a first peak around 2.3 Å, which is attributed to the first EueO interaction. Table 3 presents the structural parameters obtained from the simulated data, which was calculated using the Artemis software [61]. R-factor showed in the Table 3 reveals that the simulated data match well the experimental data. The EXAFS results indicated that the EueO distance and the Debye-Waller factor for the first and second coordination shells change with the PVA concentration. The DebyeeWaller factor and the difference between EueO1 and EueO2 are smaller for the 10/20 and higher for the 15/15 samples. A small Debye-Waller factor value for the 10/20 sample is an indication that the Eu3þ ions are located in a less distorted site, which is in agreement with the previous discussion. 4. Conclusion Eu-doped YAG nanophosphors were synthesized via PVA based sol-gel and the influence of the PVA concentration in the starting
Table 3 Simulated structural parameters of Eu-doped YAG produced with different PVA concentrations. Samples (PVA/DW, ml/ml)
Shell
CN*
Atomic length R (Å)
DebyeeWaller factor (Å2)
5/15
EueO1 EueO2 EueO1 EueO2 EueO1 EueO2
6 6 6 6 6 6
2.3078 2.4383 2.3128 2.4433 2.2974 2.4279
0.00335 0.00018 0.00211 0.00016 0.00368 0.00036
10/20 15/15 *
CN: coordination number
500
D.A. Hora et al. / Optical Materials 60 (2016) 495e500
synthesis process on the structural, morphological, and optical properties were investigated by XRD, SEM, XEOL, and XAS measurements. XRD results showed that the YAG phase was obtained and that the crystallite sizes, in fact, influenced by the PVA concentration. SEM images presented large agglomerates composed of small particles relatively regular in shape, which agrees with some reports. The XEOL-XAS was monitored as an energy function around the Eu LIII-edge. The XEOL emission spectra presented the typical Eu3þ intraconfigurational 4f e 4f lines and an abnormally strong 5D0 / 7F4 transition. The chromatic coordinates indicated that it pure red emitters were produced. XRD, XEOL and EXAFS results indicated that the Eu is located in a site with inversion center symmetry, which is more ordered for the 10/20 PV A/DW, ml/ml samples. Acknowledgments The authors are grateful to the FINEP (Brazilian Innovation Agency), CAPES (Brazilian Federal Agency), CNPq (No. 470.972/ 2013-0) (Brazilian National Council for Scientific and Technological Development), to the LNLS/CNPEM (Brazilian Synchrotron Light Laboratory e Brazilian Center for Research in Energy and Materials) for the facilities and for the financial support (proposal XAFS2#18003/15) and to the CMNano/UFS for the facility. They would also like thank Prof. Dr. Lucas C.V. Rodrigues, from the Universidade de S~ ao Paulo, for the valuable discussions about XEOL. References [1] A. Yousif, H.C. Swart, O.M. Ntwaeaborwa, E. Coetsee, Appl. Surf. Sci. 270 (2013) 331. [2] E.D. Milliken, L.C. Oliveira, G. Denis, E.G. Yukihara, J. Lumin. 132 (2012) 2495. [3] J.-H. In, H.-C. Lee, M.-J. Yoon, K.-K. Lee, J.-W. Lee, C.-H. Lee, J. Supercrit. Fluids 40 (2007) 389. [4] M.V. dos S. Rezende, C.W.A. Paschoal, Opt. Mater. (Amst) 46 (2015) 530. [5] M.V. dos S. Rezende, J. Phys. Chem. Solids 75 (2014) 1113. [6] H.S. Jang, W. Bin Im, D.C. Lee, D.Y. Jeon, S.S. Kim, J. Lumin. 126 (2007) 371. [7] K.Y. Jung, H.W. Lee, J. Lumin. 126 (2007) 469. [8] Y. Zhou, J. Lin, M. Yu, S. Wang, H. Zhang, Mater. Lett. 56 (2002) 628. [9] S. Zhou, Z. Fu, J. Zhang, S. Zhang, J. Lumin. 118 (2006) 179. [10] D. Ravichandran, R. Roy, A.G. Chakhovskoi, C.E. Hunt, W.B. White, S. Erdei, J. Lumin. 71 (1997) 291. [11] J. Tous, M. Horv ath, L. Pína, K. Bla zek, B. Sopko, Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip. 591 (2008) 264. [12] J. Tous, K. Blazek, L. Pína, B. Sopko, Appl. Radiat. Isot.. 68 (n.d.) 651. [13] S.L. David, C.M. Michail, M. Roussou, E. Nirgianaki, A.E. Toutountzis, I.G. Valais, G. Fountos, P.F. Liaparinos, I. Kandarakis, G. Panayiotakis, IEEE Trans, Nucl. Sci. 57 (2010) 951. [14] S. Shikao, W. Jiye, J. Alloys Compd. 327 (2001) 82. [15] Y.H. Zhou, J. Lin, S.B. Wang, H.J. Zhang, Opt. Mater. (Amst) 20 (2002) 13. [16] C.-H. Lu, W.-T. Hsu, B.-M. Cheng, J. Appl. Phys. 100 (2006) 063535. [17] S.W. Allison, G.T. Gillies, A.J. Rondinone, M.R. Cates, Nanotechnology 14 (2003) 859. [18] Y. Li, J. Zhang, Q. Xiao, R. Zeng, Mater. Lett. 62 (2008) 3787. kov vr, K. Kamada, V. Babin, A. Yoshikawa, M. Nikl, 3 (n.d.) 1. [19] E. Miho a, K. Va kova , K. Va vr [20] E. Miho u, K. Kamada, V. Babin, a. Yoshikawa, M. Nikl, Radiat. Meas. 56 (2013) 98. [21] A. Ikesue, Y.L. Aung, T. Taira, T. Kamimura, K. Yoshida, G.L. Messing, Annu. Rev. Mater. Res. 36 (2006) 397. [22] V.C. Teixeira, P.J.R. Montes, M.E.G. Valerio, Opt. Mater. (Amst) 36 (2014) 1580. [23] M. Sekita, H. Haneda, S. Shirasaki, T. Yanagitani, J. Appl. Phys. 69 (1991) 3709.
[24] X. Li, H. Liu, J.Y. Wang, H.M. Cui, F. Han, X.D. Zhang, R.I. Boughton, Mater. Lett. 58 (2004) 2377. [25] X. Li, H. Liu, J. Wang, H. Cui, S. Yang, I.R. Boughton, J. Phys. Chem. Solids 66 (2005) 201. [26] S. Ramanathan, M.B. Kakade, S.K. Roy, K.K. Kutty, Ceram. Int. 29 (2003) 477. [27] M. Inoue, H. Otsu, H. Kominami, T. Inui, J. Alloys Compd. 226 (1995) 146. [28] M. Nyman, J. Caruso, M.J. Hampden-Smith, T.T. Kodas, J. Am. Ceram. Soc. 80 (1997) 1231. [29] C.H. Lu, W.T. Hsu, C.H. Hsu, H.C. Lu, B.M. Cheng, J. Alloys Compd. 456 (2008) 57. [30] M.V. dos S. Rezende, M.E.G. Valerio, R.A. Jackson, Mater. Res. Bull. 61 (2015) 348. [31] M.V. dos S. Rezende, P.J.R. Montes, F.M. dos S. Soares, C. dos Santos, M.E.G. Valerio, J. Synchrotron Radiat. 21 (2014) 143. ~o, Z.S. Macedo, Powder Technol. 254 (2014) 344. [32] G. da C. Cunha, L.P.C. Roma [33] T. Liu, Y. Xu, Y. Li, Z. Wang, J. Zhao, Ceram. Int. 37 (2011) 3361. [34] T. Liu, Y. Xu, C. Zeng, Mater. Sci. Eng. B Solid-State Mater. Adv. Technol. 176 (2011) 535. [35] M. Sivakumar, S. Kanagesan, R. Suresh Babu, S. Jesurani, R. Velmurugan, C. Thirupathi, T. Kalaivani, J. Mater. Sci. Mater. Electron 23 (2012) 1045. [36] O. Kaygili, S. Keser, R.H.A. Orainy, T. Ates, F. Yakuphanoglu, Mater. Sci. Eng. C 35 (2014) 239. [37] X. Xu, X. Sun, H. Liu, J.-G. Li, X. Li, D. Huo, S. Liu, J. Am. Ceram. Soc. 95 (2012) 3821. [38] J. Li, F. Chen, W. Liu, W. Zhang, L. Wang, X. Ba, Y. Zhu, Y. Pan, J. Guo, J. Eur. Ceram. Soc. 32 (2012) 2971. [39] L. Dobrzycki, E. Bulska, D.A. Pawlak, Z. Frukacz, K. Wo zniak, Inorg. Chem. 43 (2004) 7656. [40] J. Feng, T. Liu, Y. Xu, J. Zhao, Y. He, Ceram. Int. 37 (2011) 1203. [41] A. Rusanov, Surf. Sci. Rep. 58 (2005) 111. [42] A. Bhattacharya, P. Ray, J. Appl. Polym. Sci. 93 (2004) 122. [43] B.D. Cullity, Elements of X-Ray Diffraction, Addison-Wesley Publishing Company Inc., Massachusetts, 1956. [44] F. Yu, D. Yuan, X. Cheng, X. Duan, X. Wang, L. Kong, L. Wang, Z. Li, Mater. Lett. 61 (2007) 2322. [45] K. Guo, X.-M. Zhang, H.-H. Chen, X.-X. Yang, X. Guo, J.-T. Zhao, Mater. Res. Bull. 45 (2010) 1157. [46] Z. Sun, D. Yuan, X. Duan, X. Wei, H. Sun, C. Luan, Z. Wang, X. Shi, D. Xu, M. Lv, J. Cryst. Growth 260 (2004) 171. [47] P. Ramanujam, B. Vaidhyanathan, J. Binner, A. Anshuman, C. Spacie, Ceram. Int. 40 (2014) 4179. [48] C.-H. Hsu, C.-H. Huang, B.-M. Cheng, C.-H. Lu, Mater. Chem. Phys. 124 (2010) 632. [49] K. Binnemans, Coord. Chem. Rev. 295 (2015) 1. [50] H.F. Brito, O.M.L. Malta, M.C.F.C. Felinto, E.E.S. Teotonio, Luminescence phenomena involving metal enolates, in: J. Zabicky (Ed.), The Chemistry of Metal Enolates Part 1, John Wiley & Sons Ltd, West Sussex, 2009. [51] A.P. Souza, L.C.V. Rodrigues, H.F. Brito, S. Alves, O.L. Malta, J. Lumin. 130 (2010) 181. [52] C.V. Rodrigues, L.L. Luz, J.D.L. Dutra, S.A. Junior, O.L. Malta, C.C. Gatto, H.C. Streit, R.O. Freire, C. Wickleder, M.O. Rodrigues, Phys. Chem. Chem. Phys. 16 (2014) 14858. [53] R. Skaudzius, A. Katelnikovas, D. Enseling, A. Kareiva, T. Jüstel, J. Lumin. 147 (2014) 290. [54] R.A.S. Ferreira, S.S. Nobre, C.M. Granadeiro, H.I.S. Nogueira, L.D. Carlos, O.L. Malta, J. Lumin. 121 (2006) 561. [55] I.E. Kolesnikov, D.V. Tolstikova, A.V. Kurochkin, A.A. Manshina, M.D. Mikhailov, Opt. Mater. 37 (2014) 306. [56] M. Bettinelli, A. Speghini, F. Piccinelli, A.N.C. Neto, O.L. Malta, J. Lumin. 131 (2011) 1026. € lsa €, [57] I.G.N. Silva, L.C.V. Rodrigues, E.R. Souza, J. Kai, M.C.F.C. Felinto, J. Ho H.F. Brito, O.L. Malta, Opt. Mater. 40 (2015) 41. ^ntico [58] P.A. Santa-Cruz, F.S. Teles, Spectra Lux Software V. 2.0 Beta, Ponto Qua Nanodispositivos, RENAMI, 2003. [59] S. Emura, H. Maeda, M. Nomura, Phys. B Condens. Matter. 208e209 (1995) 108. [60] R.D. Shannon, Acta Crystallogr. Sect. A 32 (1976) 751. [61] B. Ravel, M. Newville, J. Synchrotron Radiat. 12 (2005) 537.