ARTICLE pubs.acs.org/JPCC
Luminescence Dependence of Pr3+ Activated SiO2 Nanophosphor on Pr3+ Concentration, Temperature, and ZnO Incorporation G. H. Mhlongo,†,‡ O. M. Ntwaeaborwa,*,‡ H. C. Swart,‡ R. E. Kroon,‡ P. Solarz,§ W. Ryba-Romanowski,§ and K. T. Hillie*,†,‡ †
National Centre for Nano-structured Materials, CSIR, PO Box 395, Pretoria, ZA0001, South Africa Department of Physics, University of Free State, Bloemfontein, ZA9300, South Africa § Institute of Low Temperature and Structure Research, Polish Academy of Sciences, Okolna 2, 50-422 Wroclaw, Poland ‡
ABSTRACT: Green-emitting ZnO nanoparticles were successfully embedded in Pr3+doped SiO2 by a sol gel method resulting in a red-emitting ZnO 3 SiO2:Pr3+ nanocomposite phosphor. The particle morphology and luminescent properties of SiO2:Pr3+ phosphor powders, with or without ZnO nanoparticles, were, respectively, investigated by electron microscopy and luminescence spectroscopy. Luminescence of SiO2:Pr3+ was studied as a function of different Pr3+ concentrations and temperature. Both Pr3+ concentration and temperature were found to influence the 1D2 emission from Pr3+ strongly. Defects emission from ZnO nanoparticles was measured at 517 nm compared with the normal peak at 470 nm from micrometer-sized ZnO powders. Only red emission at 614 nm from Pr3+ ions in SiO2:Pr3+ and ZnO 3 SiO2:Pr3+ was observed. The green emission from ZnO in ZnO 3 SiO2:Pr3+ was quenched, and the red emission from Pr3+ was doubly enhanced compared with SiO2:Pr3+. The enhancement of the PL intensity from SiO2:Pr3+ with ZnO incorporation denotes the presence of energy transfer from ZnO to Pr3+ ions.
1. INTRODUCTION Luminescence from trivalent rare earth-doped phosphors has generated much interest in recent years. The intra-4f emission spectra of rare earth ions are characterized by narrow lines with high color purity because the 4f electrons of rare earth ions are shielded by the outer 5s and 5p electrons from external forces.1 In addition, the positions of the 4f configuration energy levels are only slightly dependent on the host matrix, and are roughly the same as the free-ion levels. These unique properties are of interest in development of new materials with the capability to produce visible light with narrow lines for red, green, and blue phosphors. Furthermore, theoretical models have been invented to calculate the energies and intensities for the transitions within the 4fn configuration.2 Among other rare earths, trivalent praseodymium (Pr3+) has been used as an activator in various host matrices to prepare phosphors3 10 that can be used in different types of lightemitting devices. Pr3+ ion has unique features, one of which is the ability to emit efficiently in the visible to the infrared (IR) spectral regions depending on the host matrix and the ion concentration.2 The luminescence from singly doped Pr3+ ion has been widely investigated in different hosts, and it has been discovered that the laser action from the 1G4 state in the IR spectral region can be utilized for fiber optical communication,11,12 whereas the emission from the opposite parity 4f5d state in the UV region can be useful for scintillator applications or tunable UV lasers.6,9 The 3 P0 and/or 1D2 state in the orange-red region is widely used for LEDs13 and field-emission display devices or photoluminescence r 2011 American Chemical Society
(PL) devices.5 Self-quenching of the luminescence from the 1 D2 or 3P0 emitting levels of Pr3+ is a commonly observed phenomenon, and because of that it has been a challenge to optimize the luminescence efficiency resulting from transitions in these levels. A few reasons that could lead to this quenching effect are: (i) multiphonon relaxation, (ii) cross relaxation within Pr3+ ion pairs, and (iii) energy migration to quenching centers.14 The last two processes depend on Pr3+ concentration. The possibility of enhancing luminescence efficiency via energy transfer between activator ions (i.e., the sensitizer and the acceptor) was predicted by Dexter.2 In principle, when a spectral overlap between the emission spectrum of the sensitizer (energy donor) and the absorption spectrum of the activator (energy acceptor) exists, energy transfer takes place.2 In the case of a lack of spectral overlap, energy transfer can take place via resonance condition (equal energy difference between the energy levels of the sensitizer and the donor), phonon-mediated processes, or both. Studies of energy transfer process between Pr3+ and other ions codoped on the same host matrix have been reported. For example, Tripathi et al.15 and Mhlongo et al.17 studied the energy transfer between Pr3+-Eu3+ and Pr3+-Ce3+ ion pairs, respectively, whereas Nie et al.16 reported efficient energy transfer between Pr3+ and Cr3+. Interest has also been focused on semiconductor Received: February 3, 2011 Revised: June 5, 2011 Published: August 10, 2011 17625
dx.doi.org/10.1021/jp201142d | J. Phys. Chem. C 2011, 115, 17625–17632
The Journal of Physical Chemistry C nanocrystals, “quantum dots”, because they are considered to be another good choice as sensitizing centers for radiative relaxation of activator ions because their excitation cross sections are very high as a result of the efficient band-to-band absorptions. Semiconductor nanocrystals exhibiting quantum size effect have been prepared and studied in many different forms, mainly as dispersed colloidal particles in a liquid or solid matrix and also in thin films.18 Rare earth ion-doped glass matrices containing semiconductor nanocrystals prepared by the sol gel method have been previously reported.19,20 The enhanced luminescence of rare earth ions was achieved by efficient energy transfer from semiconductor quantum dot to rare earth ions. Reisfeld et al.21 reported the significant increase in the emission of Tb3+- and Eu3+-doped ZrO2 in the presence of CdS nanocrystals. Their luminescence decay measurements revealed the lifetime of 0.5 ms for Eu3+activated ZrO2 both with and without CdS nanocrystals, whereas the decay time for pure ZrO2:Tb3+ was twice as long by 0.6 ms as compared with that of ZrO2:Tb3+ containing CdS nanocrystals, which was 0.3 ms. According to Reisfeld et al., 30 the obtained results confirmed that the energy from nonradiative recombination can be transferred to the high-lying excited energy levels of the Tb3+ and Eu3+ ions. This is followed by a nonradiative decay from the excited levels to the long-lived 5D0 level of Eu3+ or 5D4 level of Tb3+, which is expected to increase the population of the emitting levels but not their emission probability to the ground state, as confirmed by similar decay time of Eu, which was observed even after the addition of CdS nanocrystals. Hayakawa et al.22 concluded that the adsorbed semiconductor nanocrystals such as ZnO, CdS, and so on have a great effect on the excitation of 4f electrons in rare earth ions. This conclusion was made after observing the enhanced fluorescence intensity and longer lifetime from Eu3+ in SiO2 with CdS nanoparticles compared with pure SiO2:Eu3+. Again, Chen et al.23 reported the energy transfer between ZnO and Eu3+ and observed the longer decay time of 1.40 ns for ZnO nanowires with Eu3+ ions as compared with pure ZnO nanowires (0.45 ns). In the present work, we report the enhanced emission of Pr3+ induced by energy transfer from ZnO nanoparticles incorporated in situ in sol gel silica. Luminescence decay curves measurements were also perfomed to determine the lifetimes of the 1D2 state of Pr3+ at different temperatures. The effects of temperature on the 1D2 multiplet lifetime as well as the self-quenching of 1D2 emission with increasing Pr3+ concentration were investigated. The possible mechanism of energy transfer between ZnO and Pr3+ in SiO2 host is also discussed.
2. EXPERIMENTAL SECTION ZnO Nanocrystals Preparation by Chemical Precipitation. ZnO nanoparticles were prepared using zinc acetate (Zn(CH3COO)2) and sodium hydroxide (NaOH) as starting materials. In a typical preparation, 0.459 g of Zn(CH3COO)2 3 2H2O dissolved in 30 mL of boiling ethanol, and the solution was cooled in an ice bath. NaOH (0.2 g) dissolved in 10 mL of ethanol in a preheated ultrasonic bath was also cooled in an ice bath. This solution was slowly added to Zn2+ solution under vigorous stirring for 30 min. The resulting solution was then kept at room temperature for 24 h for further nucleation and growth of ZnO nanoparticles. Green emission by an UV excitation was observed from a transparent suspension of ZnO nanoparticles after 24 h of storage at room temperature, indicating the
ARTICLE
Figure 1. XRD patterns of commercial ZnO microparticles and ZnO nanoparticles.
formation of the ZnO nanoparticles. The solution was then centrifuged and repeatedly washed with heptane to remove unwanted Na+ and CH3COO ions. The resulting ZnO precipitate was then redispersed in ethanol or dried at 90 °C for 2 h for characterization. SiO2:Pr3+ and ZnO 3 SiO2:Pr3+ Preparation by Sol Gel. We mixed 0.05 mol of tetra-ethylorthosilicate (TEOS), 0.1 mol of water (H2O), 0.1 mol of ethanol, and 0.145 mol of dilute nitric acid (HNO3) under vigorous stirring at room temperature for 1 h. A desired amount of Pr(NO)3 3 6H2O dissolved in 0.1 mol of ethanol was added to the TEOS solution and stirred for another 1 h until a less viscous SiO2:Pr3+ gel formed. This was divided into two parts: the first part was transferred to the Petri dish for drying, and the second part was combined with 1 mol % of ZnO nanoparticles suspended in ethanol. This solution was stirred vigorously until a thick viscous gel formed. This was also dried at room temperature for 8 days. The dried SiO2:Pr3+ and ZnO-SiO2:Pr3+ gels were ground into fine powders and then heat-treated at 600 °C for 2 h in ambient air. The particle morphology of ZnO nanoparticles, pure SiO2, and ZnO 3 SiO2:Pr3+ was analyzed using a JEOL JSM-7500F field-emission scanning electron microscope (FE-SEM) and JEOL 2100 high-resolution transmission electron microscope (HRTEM). An energy-dispersive spectrometer (EDS) was used to analyze the chemical composition of the samples. PL spectra from the micro- and nanosized ZnO particles suspended in ethanol were recorded using an LS 55 fluorescence spectrometer. The excitation and luminescence spectra of the SiO2:Pr3+ and ZnO 3 SiO2:Pr3+ phosphor powders were recorded at the Deutsche Elektronen Synchrotron (DESY) in Hamburg, Germany using the setup at SUPERLUMI experimental station of HASYLAB. Lifetime measurements were obtained using an optical parametric oscillator (OPO Continuum Surelite) pumped by third harmonics of Nd:YAG laser as an excitation source. The decay signals were detected and stored with a Tektronix TDS 3052 digital oscilloscope. The measurements were carried out at liquid-helium temperatures using a cryostat (CF-1204 Oxford Instruments) on a continuous flow mode. All measurements were taken at room temperature. 17626
dx.doi.org/10.1021/jp201142d |J. Phys. Chem. C 2011, 115, 17625–17632
The Journal of Physical Chemistry C
ARTICLE
Figure 2. TEM image of the clustered ZnO nanoparticles.
Figure 3. TEM images of (a) SiO2 nanoparticles and (b) ZnO 3 SiO2:Pr3+ nanophosphor and the inset of HRTEM image of ZnO 3 SiO2:Pr3+.
3. RESULTS AND DISCUSSION 3.1. Particle Morphology. Figure 1 compares the XRD patterns of commercial micrometer-size ZnO and ZnO nanoparticles prepared in this study. Except for broadening of the diffraction peaks due to smaller particle sizes, the XRD patterns of ZnO nanoparticles resemble those of the commercial micrometersize ZnO. These patterns are consistent with the well-known wurtzite hexagonal phases of ZnO referenced in JCPDS card number 36-1451. The highly crystalline and broadened diffraction peaks of the well-known wurtzite hexagonal structure of ZnO nanoparticles. The average particle size of the ZnO nanoparticles estimated using the Scherrer’s equation was 4 nm. Figure 2 shows the transmission electron microscopy (TEM) image of the ZnO nanoparticles. Shown in the image is an
agglomeration of hexagonal particles. The inset is the HRTEM image showing bands with different directions indicating that individual particles were randomly distributed. The estimated size of the particles was in the range of 4 to 6 nm in diameter, consistent with the XRD data. The presence of clusters of doped ions or particles inside the amorphous silica matrix was confirmed by the EDS and HRTEM data. Figure 3a is the TEM image of amorphous SiO2. The highresolution scanning electron microscopy image (not shown) showed that SiO2 was constituted by an agglomeration of spherical particles. Figure 3b shows the TEM image of ZnO 3 SiO2:Pr3+. It shows a distribution of nanometer-scale particles enveloped in amorphous SiO2. The inset of Figure 3b shows that these particles were spherical with sizes ranging from 2 to 10 nm in diameter. The EDS spectrum (Figure 4) recorded from the 17627
dx.doi.org/10.1021/jp201142d |J. Phys. Chem. C 2011, 115, 17625–17632
The Journal of Physical Chemistry C
ARTICLE
Figure 4. Energy-dispersive X-ray spectrum of 1 mol % ZnO 3 SiO2:0.2 mol % Pr3+ phosphor.
Figure 5. VUV excitation spectrum and the emission spectrum of 0.2 mol % Pr3+ single-doped SiO2.
TEM image of Figure 3b confirms the presence of Si, O, Zn, and Pr elements in our samples. Similar results were reported by Yu et al.24 when they observed a uniform dispersion of particles of Eu3+ and ZnO codoped in SiO2. 3.2. Photoluminescence Properties of Pr3+ Single-Doped SiO2. The PL excitation and emission spectra of SiO2:0.2 mol % Pr3+ are presented in Figure 5. The excitation spectrum was measured monitoring Pr3+ emission at 614 nm using synchrotron radiation. This spectrum shows the weak bands between 140 and 240 nm. These bands can be ascribed to transitions from the 4f2 ground state to the lowest 4f5d sublevels of Pr3+ considering the fact that the Pr3+ f O2 charge transfer transitions that might occur appear at substantially short wavelengths. The assignment of these bands compares well to the results of RybaRomanowski et al.,25 van der Kolk et al.,26 Fu et al.,27 Ivanovskikh et al.,28 and Bo et al.29 when Pr3+ was doped in different host matrices. The excitation efficiency increases again below 100 nm for excitation at 90 nm. This agrees well with the observation of Ivanovskikh et al.,28 where they found the excitation efficiency Ca3Sc2Si3O 12:0.2%Pr3+ to increase significantly for excitation at
75 nm, whereas it was very weak in the spectral region between 85 and 160 nm. In principle, there are two types of emission transitions from Pr3+ that can occur with VUV excitation, namely, interconfigurational 4f5d f 4f2 or intraconfigurational 4f2f 4f2. The energetic location of the 4f5d states relative to the 1S0 state strongly depends on the host matrix. If the lowest 4f5d state lies below the 1 S0 state, then the high energetic excitation will stimulate the broad emission from the lowest 4f5d state, which is parityallowed interconfigurational 4f5df4f2 transition. For example, Vink et al.30 reported the strong emission bands in the UV region from 200 to 300 nm due to parity-allowed 4f15d1 f 4f2 transitions of Pr3+ for CaSO4:Pr3+ phosphor after VUV excitation at 190 nm. According to Vink et al.,30 such results suggest that the 4f15d1 levels lie below the 1S0 level. However, when Pr3+ was doped in SrSO4 and BaSO4 systems, the narrow emission peak around 400 nm from 1S0 level was observed, and this suggests that the 4f15d1 levels lie above the 1S0 level. In the current study, the first broad emission peak centered at 309 nm in the UV spectral region presented in the emission spectrum in Figure 4 corresponding to the parity-allowed transitions from the lowest excited state of the 4f5d configuration to the 3H4,5,6 and 3F2 of the 4f2 configuration was observed.25 27,30 These results are in good agreement with the results of You et al.31 for Pr3+ doped in the YBO3 system. This blue emission peak is followed by a prominent red emission peak centered at 614 nm that can be assigned to the 1D2 f 3H4 transition of Pr3+. The small shoulder around 660 nm corresponds to the 3P0 f 3F2 transition. The assignment of these emission peaks compares well with the results reported by Tan et al.32 and Rai et al.3 for Pr3+ doped in CaTiO3 and tellorite glass, respectively. The emission peaks centered at 890 and 1005 nm correspond to the 1D2 f 3F2 and 1D2 f 3F3 transitions. These were also reported by Biswas et al.4 for Pr3+ doped in SiO2 around 884 and 1060 nm. Figure 6 shows the luminescence decay curve of Pr3+1D2 f 3 H4 emission. The lifetime computed from the decay curve data obtained using the first-order exponential fitting was 179 μs for the 1D2 level of Pr3+. The self-quenching of the 1D2 f 3H4 luminescence of Pr3+ has been observed in many Pr3+-doped systems including crystals 17628
dx.doi.org/10.1021/jp201142d |J. Phys. Chem. C 2011, 115, 17625–17632
The Journal of Physical Chemistry C
ARTICLE
Figure 6. Luminescence decay curves of the 3P0 and 1D2 levels of Pr3+ (0.2 mol %) in SiO2 obtained at 8 K. The solid black line corresponds to the experimental data and the solid red line is the fit to a single exponential.
Figure 8. Lifetimes of the 1D2 f 3H4 transition of Pr3+ (0.2 mol %) doped in SiO2 plotted against the temperature in the 8 300 K regions.
Figure 7. Decay curves of luminescence originating from the 1D2 level of Pr3+ ions in SiO2.
increasing Pr3+ concentration (0.1, 0.5, and 1 mol %) in fluorophosphates glass. The characteristic decay process of the 1D2-emitting level of 3+ Pr ions is governed by the sum of probabilities for several competing processes, namely, radiative decay, nonradiative decay by multiphonon relaxation, and nonradiative decay by energy transfer to the Pr3+ ions.33 Because of the large energy gap between the 1D2 and 1G4 energy levels, the nonradiative 1D2 f 1 G4 multiphonon relaxation is expected to be small.8,14,33 Furthermore, the concentration dependence may indicate that in the depopulation of the 1D2 state the energy transfer processes are dominant over multiphonon processes involving a single ion. In highly concentrated systems, ion ion interaction brings about cross-relaxation among Pr3+ ions and migration of the excitation energy from one Pr3+ ion to another through energy transfer processes among Pr3+ ions and finally to a quenching center thus leading to self-quenching of luminescence.14,25,34,35 This process, in particular, becomes efficient when the energy level structure of luminescent ions allows for cross-relaxation process. However, if the rate of migration of excitation energy is small with respect to cross-relaxation rate, then the luminescence decay curves are no longer exponential, as documented by Ryba-Romanowski et al.25 In a case where excitation migration energy is fast, the donor acceptor system becomes more uniform while the luminescence decay curves follow the single exponential time dependence. In the current results, the single exponential luminescence decays were recorded with 0.05, 0.1, 0.2, and 0.5 mol % Pr3+ concentrations in SiO2 samples, suggesting that the rate of excitation energy migration is higher than the rate of cross-relaxation processes involved. Migration of the excitation energy through resonance energy transfer among Pr3+ ions and finally to the quencher center could therefore be the main reason for concentration quenching of fluorescence leading to fast lifetime decrease in Pr3+ concentrations above 0.2 mol %. Figure 8 shows the luminescence decay profiles of the Pr3+1D2 f 3H4 emission captured when the temperature was varied from 8 to 300 K under direct excitation of the 1D2 using a pulsed laser. As shown in the Figure, the luminescence decay curves followed the single exponential with the 1D2 lifetime gradually decreasing from 179 μs at 8 K to 148 μs at 300 K.
and glasses.8,14,33 The luminescence decay curves of the Pr3+1D2 f 3H4 transition were also recorded as a function of different Pr3+ concentrations in SiO2 at room temperature, as shown in Figure 7. The 1D2 level was directly excited to avoid the overlap in emission from the 1D2 and 3P0 states, which occurs when exciting into the 3PJ states. It was noticed that the fluorescence intensity from the Pr3+1D2 f 3H4 emission decayed slowly for 0.05, 0.1, and 0.2 Pr3+ concentrations, and the lifetimes computed from the decay curves using a single exponential fitting were 138, 141, and 144 for 0.05, 0.1, and 0.2, respectively. However, when the Pr3+ concentration was further increased to 0.5 mol %, the Pr3+1D2 f 3H4 emission decayed faster to a great extent with the lifetime of 123 μs. Similar observations have been reported.8,14 Del Longo et al.8 reported the lifetime decrease in the 1D2 level with increasing Pr3+ concentration (0.5, 1, and 2%) in zinc borate glass. Balda et al.14 also reported the shortening of lifetimes of the 1D2 level with
17629
dx.doi.org/10.1021/jp201142d |J. Phys. Chem. C 2011, 115, 17625–17632
The Journal of Physical Chemistry C
Figure 9. PL emission spectra of ZnO nanoparticles and ZnO microparticles.
Temperature dependence of the 1D2 lifetime has previously been reported.14,33,34,36 The temperature-dependent process is predicted when the condition of resonance is satisfied for transitions originating in thermally populated crystal field components of the ground states of acceptor ions. If not, the energy mismatch has to be assisted by lattice phonons while the temperature influences the rate of the cross relaxation.36 The decrease in the 1D2 lifetime with increasing temperature between 8 and 300 K cannot be assigned to nonradiative decay through multiphonon relaxation but rather to contribution of nonresonant cross relaxation process because the rate of multiphonon decay of 1D2 level is negligible in most of the hosts matrices compared with that of radiative decay due to the large energy gap between 1D2 and 1 G4.14,36 The cross relaxation mechanism involved during the self-quenching process is believed to proceed through intermediate 1G4 and 3F4 levels. 3.3. Energy Transfer from ZnO Nanoparticles to Pr3+ Ions in SiO2. The PL emission spectra of ZnO nanoparticles and ZnO microparticles recorded when the samples were excited at 325 nm with a monochromatized xenon of the LS 55 spectrophotometer are shown in Figure 9. A similar spectrum of the ZnO microparticle was recorded using a SUPERLUMI experimental station exciting at 90 nm. Note that the emission spectrum of the ZnO nanoparticles could not be recorded using the SUPERLUMI experimental station because of excessive charging. It is, however, well known that ZnO nanoparticles display dual emission in the UV and visible regions irrespective of the type of excitation (UV photons or high-energy electrons). It is therefore reasonable to assume that our result would not be different if our samples were not charging. The green emission from the ZnO nanoparticles associated with recombination of delocalized electrons at singly occupied oxygen vacancies with deep trapped holes19 was observed at 517 nm. This emission is red-shifted from the defects emission of the ZnO microparticles at 470 nm, which is in good agreement with the results of Ntwaeaborwa et al.19 It was also noticed that the green emission from the ZnO nanoparticles at 517 nm was more intense compared with that of micrometer-sized ZnO particles, and this is a result of the increase in the surface-area volume ratio due to smaller particle
ARTICLE
Figure 10. PL emission spectra of SiO2:Pr3+ and ZnO 3 SiO2:Pr3+ after VUV excitation at 90 nm using synchrotron radiation.
Figure 11. PL excitation spectra of SiO2:Pr3+:0.2 mol % with and without ZnO nanophosphors monitoring the emission peak at 614 nm.
size, which could increase the density of surface defect states in ZnO.37 Direct bandgap emission, which can be due to recombination of excitonic centers in ZnO, was also observed in both ZnO microparticles and ZnO nanoparticles at 380 and 365 nm, respectively, except that one for ZnO nanoparticles was blueshifted to lower wavelengths (higher energies) than that of micrometer-sized ZnO particles. This behavior could be due to widening of the ZnO nanoparticles bandgap due to quantum confinement of charge carriers in the restricted volume of smaller particles.19,20 Figure 10 presents the PL emission spectra of SiO2:0.2 mol % Pr3+ and ZnO 3 SiO2:0.2 mol % Pr3+ phosphors under excitation at 90 nm using synchrotron radiation. It was observed that upon VUV excitation on the 4f5d state of Pr3+ only characteristic emissions from Pr3+ ion with the main red emission centered at 614 nm could be detected from both SiO2:Pr3+ and ZnO 3 SiO2: Pr3+ phosphors. The green emission from ZnO nanoparticles was suppressed in ZnO 3 SiO2:Pr3+, and the intensity of the Pr3+ emission was approximately twice the intensity of the SiO2:Pr3+ despite lower Pr3+ concentration. The quenching of green emission 17630
dx.doi.org/10.1021/jp201142d |J. Phys. Chem. C 2011, 115, 17625–17632
The Journal of Physical Chemistry C
ARTICLE
Figure 12. Luminescence decay curves of (a) ZnO 3 SiO2:Pr3+ nanophosphor and (b) ZnO nanoparticles.
Figure 13. Schematic energy level diagrams showing the transition localized within Pr3+ ions and ZnO nanoparticles and energy transfer from ZnO to Pr3+.
from ZnO as a result of enhanced Pr3+ emission demonstrates sensitization of Pr3+ emission centers by ZnO nanoparticles. Figure 11 compares the PL excitation (λemission = 614 nm) spectra of SiO2:Pr3+ and ZnO 3 SiO2:Pr3+. The transitions were assigned according to the literature. The excitation spectra of ZnO-SiO2:Pr3+ have additional absorption peaks, and they are more resolved than those from SiO2:Pr3+, which further confirms the incorporation and the subsequent effects of ZnO nanoparticles in SiO2:Pr3+. Luminescence decay curves of SiO2:0.2 mol % Pr3+ and ZnO 3 SiO2:0.2 mol % Pr3+ and ZnO emissions recorded at room temperature are presented in Figure 12a,b. It can be seen from Figure 11a that the Pr3+1D2 f 3H4 emission from SiO2:Pr3+ decayed faster with the lifetime of 144 μs, which is almost two times shorter than that of ZnO 3 SiO2:Pr3+, which had a lifetime of 236 μs. For pure ZnO nanoparticles, the decay curve also
followed the first-order exponential profile, and the lifetime was determined to be 2 μs, as shown in Figure 11b, which is even much shorter than that of SiO2:Pr3+ with and without ZnO. These results compare well with the results observed by Hayakawa et al.22 and Chen et al.23 In principle, the radiative recombination rate should be similar before and after ZnO incorporation in SiO2:Pr3+. Moreover, when the nonradiative decay time remains the same, the decay time should be shorter, indicating that the energy transfer exists as a result of additional nonradiative pathways.22,38 In the present results, the longer lifetime observed from ZnO 3 SiO2:Pr3+ may be explained by the formation of defects at the surface or boundaries of ZnO nanocrystals, which can serve as energy trap centers thus supporting the energy transfer in the system. It cannot be ignored that for pure ZnO nanocrystals electron-or hole- trapped surface levels exists due to widening of the bandgap (365 nm) with smaller particle sizes. Also, in ZnO some intrinsic defects such as oxygen vacancies exist. Such defects create some sites that can then trap free electrons and serve as nonradiative recombination centers. However, incorporation of ZnO into SiO2:Pr3+ decreases the nonradiative decay rate significantly as photogenerated trapped electron or hole at the surface strongly interacts with the Pr3+ ion located close to the boundary between ZnO and SiO2. Hence the longer lifetime for SiO2:Pr3+ with ZnO was observed irrespective of additional nonradiative decay pathways. Simplified energy level diagrams of Pr3+ and ZnO are presented in Figure 13 to illustrate the possible mechanism of energy transfer from ZnO to Pr3+ resulting in enhanced emission of red photons from Pr3+. According to this illustration, the excitation energy is absorbed in the band gap of ZnO, followed by nonradiative relaxation to the defects states and a subsequent transfer of energy to Pr3+. The energy transfer from ZnO to Pr3+ is speculated to be through phonon-mediated processes because the spectral overlap between the emission of the donor (ZnO) and the excitation of the acceptor (Pr3+) that could result in energy transfer from ZnO to Pr3+ was not observed. Bang et al.20 also proposed similar mechanism for Eu3+-doped SiO2 with embedded ZnO nanoparticles.
4. CONCLUSIONS Red emission originating from both 3P0- and 1D2-emitting levels of Pr3+ was observed from SiO2:Pr3+ both with and without 17631
dx.doi.org/10.1021/jp201142d |J. Phys. Chem. C 2011, 115, 17625–17632
The Journal of Physical Chemistry C embedded ZnO nanoparticles. 1D2 level showed the longer lifetime compared with the 3P0 level as observed from luminescence decay data. Investigation of the luminescence of Pr3+ in SiO2 as a function of Pr3+ concentration revealed the shortening of 1D2 lifetime with increasing Pr3+ concentration, that is, above 0.2 mol %. The concentration quenching effect due to migration of excitation energy among Pr3+ ions was speculated to be the main effect of self-quenching of the 1D2 level. The 1D2 level was also found to be influenced by temperature as its lifetime slowly decreased with increasing temperature from 8 to 300 K. Enhanced red emission from Pr3+-doped SiO2 was achieved by incorporating ZnO nanoparticles, indicating that ZnO acted to harvest excitation energy and transfer it nonradiatively to Pr3+ emission. The possible mechanism of energy transfer from ZnO to Pr3+ was discussed.
’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected] (O.M.N.);
[email protected] (K.T.H.). Tel: +2751 401 2193. Fax: +2751 401 3507.
’ ACKNOWLEDGMENT This project is financially supported by both the Department of Science and Technology of South Africa and the Council for Scientific and Industrial Research of South Africa. We would like to thank the SUPERLUMI experimental station HASYLAB, DESY in Hamburg, Germany for giving us the beam time to conduct some PL measurements using synchrotron radiation. ’ REFERENCES (1) van Pieterson, L.; Reid, M. F.; Wegh, R. T.; Soverna, S.; Meijerink, A. Phys. Rev. B 2002, 65, 045113. (2) Blasse, G.; Grabmaier, B. C. Luminescent Materials; SpringerVerlag: Berlin, 1994. (3) Rai, V. K.; Kumar, K.; Rai, S. B. Opt. Mater. 2007, 29, 873–878. (4) Biswas, A.; Chakrabarti, S.; Acharya, H. N. Mater. Sci. Eng. 1997, B 49, 191. (5) Okamoto, S.; Kobayashi, H. J. Appl. Phys. 1999, 86, 10. (6) Nikl, M.; Ren, G.; Ding, D.; Mihokova, E.; Jary, V.; Feng, H. Chem. Phys. Lett. 2010, 493, 72–75. (7) Nemec, P.; Frumarova, B.; Frumar, M. J. Non-Cryst. Solids 2000, 270, 137–146. (8) Del Longo, L.; Ferrari, M.; Zanghellini, E.; Bettinelli, M.; Capobianco, J. A.; Montagna, M.; Rossi, F. J. Non-Cryst. Solids 1998, 231, 178–188. (9) Pidol, L.; Viena, B.; Kahn-Harari, A.; Bessiere, A.; Dorenbos, P. Nucl. Instrum. Methods Phys. Res. 2005, 537, 125–129. (10) Strek, W.; Legendziewicz, J.; Lukowiak, E.; Maruszewski, K.; Sokolnicki, J.; Boiko, A. A.; Borzechowska, M. Spectrochim. Acta, Part A 1998, 54, 2215–2221. (11) Kravesta, V.; Machewirth, D.; Sigel, G. H., Jr. J. Non-Cryst. Solids 1997, 213&214, 304–310. (12) Simons, D. R.; Faber, A. J.; de Waal, H. J. Non-Cryst. Solids 1995, 185, 283–288. (13) Birkhahn, R.; Garter, M.; Steckl, A. J. Appl. Phys. Lett. 1999, 74, 15. (14) Balda, R.; Fernandez, J.; Saez de Ocariz, I.; Adam, J. L.; Mendioroz, A.; Montoya, E. Opt. Mater. 1999, 13, 159–165. (15) Tripathi, H. B; Agarwal, A. K; Kandpal, H. C; Belwal, R. Solid State Commun. 1978, 28, 807–814. (16) Nie, Z.; Lim, K.-S.; Zhang, J.; Wang, X. J. Lumin. 2009, 129, 844–849.
ARTICLE
(17) Mhlongo, G. H.; Ntwaeaborwa, O. M.; Dhlamini, M. S.; Swart, H. C.; Hillie, K. T. J. Alloys Compd. 2011, 509, 2986–2992. (18) Kamat, P. V.; Patrick, B. J. Phys. Chem. 1992, 96, 6829–6834. (19) Ntwaeaborwa, O. M.; Swart, H. C.; Kroon, R. E.; Ngaruiya, J. M.; Botha, J. R.; Holloway, P. H. Enhanced Photoluminescence of Rare-Earth Activators in Sol-Gel Derived SiO2 by Energy Transfer from ZnO Nanoparticles and Co-Activators. In Photoluminescence Research Progress; Wright, H. K., Edwards, G. V., Eds.; Nova Science Publishers: New York, 2008; Chapter 11, pp 287 306. (20) Bang, J.; Yang, H.; Holloway, P. H. J. Chem. Phys. 2005, 123, 084709. (21) Reisfeld, R.; Gaft, M.; Saridarov, T.; Panczer, G.; Zelner, M. Mater. Lett. 2000, 45, 154–156. (22) Hayakawa, T.; Selvan, S. T.; Nogami, M. J. Sol-Gel Sci. Technol. 2000, 19, 779–783. (23) Chen, R.; Shen, Y. Q.; Xiao, F.; Liu, B.; Gurzadyan, G. G.; Dong, Z .L.; Sun, X. W.; Sun, H. D. J. Phys. Chem. C 2010, 114, 18081–18084. (24) Yu, Y.; Wang, Y.; Chen, D.; Huang, P.; Ma, E.; Bao, F. Nanotechnology 2008, 19, 055711. (25) Ryba-Romanowski, W.; Dominiak-Dzik, G.; Solarz, P.; Klimesz, B.; Zelechower, M. J. Non-Cryst. Solids 2004, 348&346, 391–395. (26) van der Kolk, E.; Dorenbos, P.; Vink, A. P.; Perego, R. C.; van Eijk, C. W. E. Phys. Rev. B 2001, 64, 195129. (27) Fu, Y.; Zhang, G.; Qi, Z.; Wu, W.; Shi, C. J. Lumin. 2007, 124, 370–374. (28) Ivanovskikh, K. V.; Meijerink, A.; Piccinelli, F.; Speghini, A.; Zinin, E. I.; Ronda, C.; Bettinelli, M. J. Lumin. 2010, 130, 893–901. (29) Bo, L.; Chao-Shu, S.; Ze-Ming, Q.; Ye, T. Chin. Phys. Lett. 2005, 22, 2677–2679. (30) Vink, A. P.; van der Kolk, E.; Dorenbos, P.; van Eijk, C. W. E. J. Alloys Compd. 2002, 341, 338–341. (31) You, F.; Huang, S.; Meng, C.; Wang, D.; Xu, J.; Huang, Y.; Zhang, G. J. Lumin. 2007, 122, 58–61. (32) Tan, S.; Yang, P.; Li, C.; Wang, W.; Wang, J.; Zhang, M. Solid State Sci. 2010, 12, 624–629. (33) Mazurak, Z.; Pisarski, W. A.; Gabrys-Pisarska, J.; Zelechower, M. Phys. Status Solidi B 2003, 237, 581–591. (34) Huang, Y.; Tsuboi, T.; Seo, H. J. J. Phys. Chem. A 2008, 112, 5839–5845. (35) Solarz, P.; Gajek, Z. J. Phys. Chem. C 2010, 114, 10937–10946. (36) Guwoski, M. A.; Dominiak-Dzik, G.; Solarz, P.; Lisiecki, R.; Ryba-Romanowski, W. J. Alloys Compd. 2007, 438, 72–76. (37) Dhlamini, M. S.; Ntwaeaborwa, O. M.; Swart, H. C.; Ngaruiya, J. M.; Hillie, K. T. Physica B 2009, 404, 4409–4410. (38) Vergeer, P.; Babin, V.; Meijerink, A. J. Lumin. 2005, 114, 267–274.
17632
dx.doi.org/10.1021/jp201142d |J. Phys. Chem. C 2011, 115, 17625–17632
