SrAl12O19:Pr3+ nanodisks and nanoplates: New processing technique and photon cascade emission Zhaogang Nie Key Laboratory of Excited State Processes, Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, China; and BK21 Physics Program and Department of Physics, Chungbuk National University, Cheongju 361-763, Korea
Jiahua Zhanga) and Xia Zhang Key Laboratory of Excited State Processes, Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, China
Ki-Soo Limb) BK21 Physics Program and Department of Physics, Chungbuk National University, Cheongju 361-763, Korea (Received 27 October 2008; accepted 13 January 2009)
High-quality SrAl12O19:Pr3+ nanodisks and nanoplates were fabricated via a new processing technique based on a modified polymer steric entrapment method. Serious agglomeration and large particle size distribution of final products, which usually occurred in the conventional method, were eliminated completely. The effects of new synthetic processes on the morphology, crystallization, and yield of products and the relevant mechanisms were discussed. As far as we know, SrAl12O19:Pr3+ nanodisks with mean diameter 60 nm and thickness between 5 and 10 nm were successfully synthesized for the first time by this low-cost technique. The new synthetic method may provide a general route to synthesize other refractory mixed-oxide nanocrystals. Photon cascade emission involving transitions 1S0–1I6 followed by 3P0–3H4 in SrAl12O19:1% Pr3+ nanodisks was investigated. Size-effect-induced blue shift of the 4f5d states of Pr3+ was observed in SrAl12O19:1% Pr3+ nanodisks, in which the quantum efficiency was preserved, as in the bulk counterparts. I. INTRODUCTION
Rare earth doped oxide nanocrystalline phosphors attract considerable attention because of their potential application in lighting and display. With the progress of nanoscience, people are now equipped with various fabrication methods with novel properties on the nanoscale level. These so-called “size-dependent properties” are of great interest and significance. However, the processing methods necessary to produce high-crystallization phosphor, which usually helps to achieve high luminous efficiency, generally involve processing at high temperature, which tends to agglomerate the primary crystallites and thus lose the possible benefits of the nanosize materials. Furthermore, the quantum efficiency (QE) of oxide phosphors fabricated in nanosize regime is usually much lower than their microsize counterparts due to surface effects. They are always assumed as the main hindrances to their practical applications.1–5 Thus, much attention should be paid to developing more new Address all correspondence to these authors. a) e-mail:
[email protected] b) e-mail:
[email protected] DOI: 10.1557/JMR.2009.0193 J. Mater. Res., Vol. 24, No. 5, May 2009
methods to control the morphology of the nanoparticles and preserve their QEs. Some refractory mixed oxides, such as SrAl12O19 (SAO; melting point > 1800 C) or other strontium aluminates, have significant technological importance because of their unique combination of mechanical, thermal, and optical properties, making them candidates for industries of lighting and displays,6 laser,7,8 cement, and even steel.9,10 Despite their importance, the research about their synthesis is still scarce compared with other materials, such as metals and semiconductors, because they are high-temperature phase species that are very difficult to prepare and get in control.11–14 Recently, a polymer steric entrapment (PSE, or other similar names) method using polyvinyl alcohol (PVA) or other longchain polymer as organic carrier was developed to fabricate various multioxide ceramics, but usually not suitable for synthesis of well-dispersed nanocrystals.15–22 Herein, this method was modified, and high-quality SAO:Pr3+ nanodisks and nanoplates were fabricated by this low-cost technique. Large particle sizes, wide particle size distribution, and the phenomenon of agglomeration of final products, which usually occurred in conventional the PSE method,15–22 and some other © 2009 Materials Research Society
1771
Z. Nie et al.: SrAl12O19:Pr3+ nanodisks and nanoplates: New processing technique and photon cascade emission
organic amorphous precursor methods,12,13 were eliminated completely. The effects of new synthesis steps on the morphology, crystallization, and powder yield of the final products and the synthetic mechanisms were discussed. SAO:Pr3+ has been investigated as a promising vacuum-UV (VUV) excited phosphor possessing photon cascade emission (PCE) process.23 Much interest has been attracted to this phosphor for its potential in applications such as plasma displays and mercury-free fluorescent tubes.23–25 In general, when Pr3+ is excited to its 4f5d states, PCE caused by 1S0–1I6 followed by 3P0–3H4 radiative transition could occur in some Pr3+-doped hosts in which the 1S0 level is located below the lowest 4f5d state. In this paper, under VUV synchrotron radiation, the PCE of Pr3+ in SAO:Pr3+ nanodisks were investigated. Unlike familiar spectral properties of other rare earth ions, such as Eu3+ and Tb3+,1–5 in nanomaterials, the impact of “size-dependent properties” on luminescence of SAO:Pr3+ nanodisks is so little that the QE can be retained as in the bulk counterparts. II. EXPERIMENTAL PROCEDURE A. Sample preparation
The flow chart of this new synthesis method for fabricating SAO:1% Pr3+ nanocrystals is summarized in Fig. 1. The processes are as follows: (1) Precursor preparation: the aqueous solution containing stoichiometric amounts of Sr2+, Al3+, and Pr3+ (Mg2+ was added as the charge compensatory ion of Pr3+)23 nitrates (high-purity reagents, all from Beijing Fine Chemical Company, Beijing, China) was mixed with a amount of 78% hydrolyzed PVA (–[CH2CHOH]n–, n = 1750 50) solution. The amounts of PVA in the solution were adjusted in such a way that there were two times more positively charged valences of cations than negatively charged (–OH) functional ends of the organics. The resulting solutions were heated and stirred until a yellow gel was formed. (2) Bubbling and drying: the yellow gel was dried overnight in an oven at 110 C. It would continue to expand to more than twentyfold volume and be full of bubbles until all the water was evaporated and a yellow soft resin remained. Fine precursor powders were obtained by just hand grinding the resin for a few minutes. (3) Combustion separation: as plotted in Fig. 2, the small crucible containing some precursor powders was put into a big one with a cover. This crucible system was heated in a box furnace with a temperature at about 300 C for 1 h. The precursor will burn with very heavy smoke rolling out. The soot scatters and then sticks to the inside of the big crucible. (4) Presintering and grinding: the residue and the soot were collected, respectively, and then were subjected to the same presintering (500 C/3 h) and grinding (2 h) process for several times 1772
(five times, this work), until the organic materials fully decomposed and the powder was completely white. The powders were milled using an attritor mill with zirconia milling media (rotation rate: 240 rpm, ball diameter: 5 mm). Finally, (5) calcination: the obtained samples were calcined at 1000 to 1500 C for 3 h. SAO:1% Pr3+ bulk materials were synthesized at 1500 C for 10 h by solid-state reactions under reducing atmosphere with corresponding oxides (high-purity reagents, all from Beijing Fine Chemical Company) as the starting materials. The obtained bulk samples were subjected to x-ray diffraction (XRD) analysis, and only single-phase samples were used in further experiments. B. Characterizations
The crystallization behavior and morphology of products were characterized by Rigaku XRD using a Cu ˚ ) and scanning target radiation resource (l = 1.54078 A electron microscopy (SEM; Hitachi S-4800). The roomtemperature VUV excitation and emission spectra were measured at the VUV spectroscopy experimental station on beam line U24 of National Synchrotron Radiation Laboratory, University of Science and Technology of China. The electron energy of the storage ring is 800 MeV, and the beam current is in the range 150 to 250 mA. A Seya-Namioka monochromator (1200 g/mm) was used for the synchrotron radiation excitation spectra, while an ARC-257 monochromator (1200 g/mm) for the emission spectra, and the signal is detected by a Hamamatsu H8259-01 photomultiplier. The pressure in the sample chamber is about 1 104 Pa. The excitation spectra were corrected using the excitation spectra of sodium salicylate as standard. The low-temperature (10 K) luminescence decay curves were measured with a TDS928 Tektronix digital oscillograph (model TDS 3052) using a PSX-100-193 Excimer Laser with light wavelength 193 nm as the excitation source. The samples were cooled by a liquid helium cycling system. III. RESULTS AND DISCUSSIONS A. Synthesis of SAO:Pr3+ nanodisks and nanoplates 1. New synthesis processes compared with the traditional PSE method
The new synthetic flow is described in detail in Sec. II. Usually, the traditional PSE route is composed only of processes (1) and (5),15 in which the water in solution is evaporated completely by stirring and heating in process (1) and then the obtained precursor will be performed the final calcination directly. Compared with the old method, three new processes were added to the conventional flow (see Fig. 1), and the polymeric carrier has two additional requirements:
J. Mater. Res., Vol. 24, No. 5, May 2009
Z. Nie et al.: SrAl12O19:Pr3+ nanodisks and nanoplates: New processing technique and photon cascade emission
particle size and lower the final crystallization temperature. Detailed effects of the additional new processes on the synthesis of SAO:1% Pr3+ nanocrystals are discussed next. 2. Effects of new synthetic processes on the morphology of final products
FIG. 1. Flow chart of fabrication of SAO:Pr3+ via an improved PSE method.
FIG. 2. Schematic illustration of the reactor system for the combustion separation process.
flammable and smoky, the prerequisite of process (3), and sufficient dosage to ensure that process (3) is effective. More organic carrier is needed here to help process (2) produce precursors with a soft and porous structure, making them more flammable, and therefore expedite process (3). In this work, the precursor with a cationic valence to PVA end group (–OH) ratio (abbreviated to cation/PVA ratio) of 2:1 was adopted instead of the ratio of 4:1,15–17 which is usually used in the old method and cannot supply sufficient organic fuel for process (3). In addition, process (4) will remove the remaining organic materials before the final calcination process (5), and powerful milling or prolonging the grinding time properly will be helpful to reduce the final
The effects of new synthetic processes (3) and (4) on the morphology of final SAO:1% Pr3+ products are displayed in Fig. 3. Except for Figs. 3(f), 3(g), and 3(h), all the SEM photographs are obtained from the samples calcined at 1200 C/3 h. As shown in Fig. 3(a), the scanning electron micrograph of SAO:1% Pr3+ powders synthesized by conventional route reveals a porous agglomerated structure with a large size distribution of 1 to 100 mm. As can be seen in the inset, the larger particle is in fact made up of primary smaller platelike particles with thickness much more than 100 nm. The results are similar when we change the cations/PVA ratio from 2:1 to 4:1 and are also similar to other mixed oxides obtained by the old PSE method.15–22 If we add only process (4) into the old flow, as displayed in Fig. 3(b), the phenomenon of agglomeration, which usually occurred in PSE, is eliminated completely. Most of the particles have hexagonal platelike morphologies with regular corners and edges, which is in accord with the hexagonal magnetoplumbite structure of SAO.26,27 The thickness of the nanoplates is between 30 and 150 nm, and the distance between the opposite sides has a size distribution of 0.1 of 1 mm. Similar morphology was also observed by other researchers, but the dispersibility of products was not described.28 If the two new processes (3) and (4) are both added, the SEM images of SAO:Pr3+ from the residue are given in Fig. 3(c) and those from the soot in Figs. 3(d) and 3(e), respectively. For the residue, most of the particles also adopt regular hexagonal shape, only they are rather massive. The distance between the opposite sides is about 1 mm, and the thickness is between 200 and 300 nm. For the soot, however, well-dispersed nanodisks, round in shape, with mean diameter 60 nm and thickness between 5 and 10 nm are obtained. As far as we know, this is the first observation of SAO nanodisks. With increasing calcination temperature, the crystalline grains from the soot will grow in size without a doubt. However, the nanodisks will not retain their original shape. Raising the calcining temperature to 1400 C, the hexagonal nanoplates with mean thickness 75 nm appear, as shown in Figs. 3(f) and 3(g). Interestingly, we find some particles, displayed in Fig. 3(f), that present a transition shape to hexagonal structure. The corners of the hexagonal particles have not grown up completely while the particle size increases, which indicates that the nanodisks in Figs. 3(d) and 3(e) are also in a transition shape to the hexagonal
J. Mater. Res., Vol. 24, No. 5, May 2009
1773
Z. Nie et al.: SrAl12O19:Pr3+ nanodisks and nanoplates: New processing technique and photon cascade emission
FIG. 3. Scanning electron micrographs of SAO:1% Pr3+ synthesized by different synthetic processes, (a): (1) and (5); (b): (1), (2), (4), and (5); (c): (1)–(5) (from the residue); (d)–(h): (1)–(5) (from the soot). The products are from the samples calcined at different conditions, (a)–(e): 1200 C/4 h; (f) and (g): 1400 C/4 h; (h): 1500 C/4 h. The relevant synthetic processes are presented in Sec. II.
structure. With a further increase of the calcination temperature to 1500 C, only the particle size increases and the hexagonal morphologies will not change again, as shown in Fig. 3(h). As can be seen from Figs. 3(e)–3(h), with increasing calcination temperature the increment speed of the radial size of nanoparticles is much faster than that of the thickness. Thus through the five-step synthesis route, SAO:1% Pr3+ nanodisks and nanoplates 1774
can be fabricated in sequence by controlling the calcination temperature. 3. Crystallization behavior and powder yield
Figure 4 presents the XRD patterns of the SAO:1% Pr3+ powders from the residue and the soot, respectively. Single hexagonal SAO phases (JCPDS No. 84-1613) are
J. Mater. Res., Vol. 24, No. 5, May 2009
Z. Nie et al.: SrAl12O19:Pr3+ nanodisks and nanoplates: New processing technique and photon cascade emission
obtained at 1200 C for both samples and also for other products calcined under the same condition, and no additional phase is observed. As a result of size effect, the XRD peaks of the soot are wider than those of the residue and became stronger and sharper with the increase of calcination temperature. For the residue, the width of the sharp peaks cannot change with calcination temperature. In fact, the larger sizes of the residue, as shown in Fig. 3(c), are found to be outside of the reliable nanoregime and can be regarded as bulk materials. A double dose of polymer carrier is used relative to the traditional PSE method, and process (4) does not affect the powder yield (powder yield = weight of calcined powder/weight of organics used)16 greatly. Thus, the powder yield should be about half of that obtained by the old method. A total powder yield of 85% is obtained by this improved process. After the organic materials are removed completely from the soot and residue, respectively, the remaining weight of the two
FIG. 4. XRD spectra of SAO:1% Pr3+ powders from the residue and the soot, respectively, calcined at 1000–1400 C/4 h.
parts is about equal. Then the nanodisk powder yield of 42% from the soot is obtained. The PSE method is derived from the improvement of the traditional Pechini process.15–17 After improvement, the old PSE method is able to support a higher amount of cations and result in a higher powder yield. In our work, after further modifying the old PSE method, the powder yield of this new method is still much higher than those reported, such as the cordierite (27%),22 when using the genetic Pechini method. 4. Synthetic mechanism
With the optimum amount and chain length of PVA, the mechanism of the old PSE method represents a perfect cationic entrapment in uniform polymeric network structures, as depicted in Figs. 5(a) and 5(b), in which a pure, highly reactive, and homogeneous cationic distribution at a molecular level is expected.15–17 This is after all an ideal representation, however. In theory, the reactiveness of primary particles is related to their particle size.20,21 More fine particle size will result in more specific surface and, hence, higher reactivity. It can be observed in Fig. 3(a); different particles display different agglomerate degrees, which indicate that even under the best conditions the size distribution of primary particles is not actually uniform. Thus, the reasonable situation is that the polymer chains will inevitably entangle each other, as displayed in Figs. 5(a)–5(c), so an absolutely uniform dispersion of polymeric network structures is impossible. Based on these considerations, process (3) is added. With this step, the small and light particles in the smaller network spaces are taken out by the heavy smoke and the convective airflow from a specific circumstance (see Fig. 2). Then the light/small and the heavy/big parts can be separated respectively. As shown in Figs. 3(c) and 3(d), a more uniform particle size distribution was obtained relative to that in Fig. 3(a). In addition, the particle size of the residue is greater, while the soot is smaller, than that of the products obtained by adding only process (4) in Fig. 3(b), which provides further evidence of the nonuniform dispersion of the network structures in precursor.
FIG. 5. Schematic illustrations of the old (ideal) cationic entrapment mechanism (a ! b) and the new (practical) one (a ! c) in polymer network structures with (a) not enough and (b) optimal and (c) amount polymer. J. Mater. Res., Vol. 24, No. 5, May 2009
1775
Z. Nie et al.: SrAl12O19:Pr3+ nanodisks and nanoplates: New processing technique and photon cascade emission
Moreover, the polymer process in the old method can produce carbonaceous material that gives heat through its combustion, so that fine and single-phase powders can be formed at a relatively low external temperature.15–17 However, it is reported that the formation of carbonates may also cause the disproportionation of cations in precursor and lead to unwanted phases.20 Further, in our opinion, the decomposition and phase-state transition of organic materials at high temperature, similar to fluxes used in solid-state reactions, may strengthen the ion exchange among the primary particles in different network structures and increase their agglomerate possibility. As shown in Fig. 3(a), the porous structure actually results from the conglutination of primary particles. Hereupon, process (4) was added to remove organic materials before final calcination. As can be seen in Fig. 3(b), using process (4), the phenomenon of agglomeration is eliminated completely and the mean particle size decreases greatly relative to the old method. Furthermore, it is interesting that although the carbonaceous material is removed in advance, the crystallization temperature still moves down relative to that of the bulk calcined by solid-state reaction. In fact, it should be attributed to the unstable amorphous precursor powder, which has an extremely high reactivity at low calcination temperatures. First, as discussed before, more fine particle size, greater reactiveness. The mechanical attrition milling in process (4) is very effective in crushing the already agglomerated particles in the precursor and increasing their specific surface areas.20,21 As a result, the primary particles in the network structures are able to keep very fine (or uncoagulated, mostly in nanosize)20 before crystallization. Second, the precursor powder has a high surface energy and contains significant internal stresses, caused by the powerful impact during the milling process. It has been reported that stresses in the ground powders from powerful milling will also result in the decrease of crystallization temperature.21 Thus powerful milling or prolonging the grinding time properly will be helpful to reduce the final particle size and lower the final crystallization temperature. A similar process was also used to prepare SAO nanoplates by other researchers28; however, a marked decrease of the crystallization temperature was not observed because the milling process was not valued enough. In general, with the additional new processes, the main drawbacks of the old PSE method, including large particle size distribution and agglomerated morphology of final products, were ultimately overcome. B. PCE properties of SAO:Pr3+ nanocrystals 3+
The corrected emission spectra of SAO:1% Pr nanodisk and bulk samples on the lowest 4f5d state excitation at 198 nm are presented in Fig. 6. The inset is the 1776
energy-level diagram of SAO:1% Pr3+. Compared with the bulk, the characteristic emissions that display the PCE process, 1S0–1I6 (402 nm) followed by 3P0–3H4 (485 nm) radiative transitions, are also observed,23 and the branching ratios of 1S0 emissions and the intensity ratio of 3 P0–3H4 to 1S0–1I6 are statistically identical in SAO:1% Pr3+ nanodisks. Similar results are also obtained for other SAO:1% Pr3+ nanoplates calcined at a temperature higher than 1200 C. Figure 7 displays the VUV excitation spectra of SAO:1% Pr3+ nanodisk (dashed) and bulk (solid line) samples monitoring the 1S0–1I6 emission at 402 nm. The similar bands above 47,600 cm1 (210 nm) and the sharp lines at 46,729 cm1 (214 nm) for both samples can be assigned to the 4f 2–4f5d and 3H4–1S0 transitions of Pr3+, respectively. No obvious position changes can be detected for the sharp lines because the relevant transitions are within the 4f shell, which is shielded from
FIG. 6. Emission spectra upon the lowest 4f5d state excitation at 198 nm for SAO:1% Pr3+ nanodisk (solid) and bulk (dashed) materials. The spectra have been offset vertically for clarity.
FIG. 7. VUV excitation spectra of 1S0–1I6 emission at 402 nm for SAO:1% Pr3+ nanodisk (dashed) and bulk (solid) materials. The excitation spectrum of SAO:2% Ce3+ (gray line, the line has been offset vertically for clarity) has been shifted to the high-energy side by 12,240 + 690 cm1.
J. Mater. Res., Vol. 24, No. 5, May 2009
Z. Nie et al.: SrAl12O19:Pr3+ nanodisks and nanoplates: New processing technique and photon cascade emission
environmental effects by the outer shell electrons.29 However, unlike the sharp lines, the onset of the lowest 4f5d state of Pr3+ in nanodisks has a blue shift, compared with that in the bulk (see inset). For the rare earth ions in a crystal, the positions of the 4f n15d levels are in theory influenced much more by the crystal field interaction than are those of the 4f n levels.30,31 The effect of crystal field on the 5d orbit will depress the level energies in a specific host lattices compared with the free ions. Thus, the blue shift of the lowest 4f5d state indicates that the crystal-field strength in nanodisks decreases relative to that in the bulk because of size effects. As for the SAO:1% Pr3+ nanoplates, the results are similar to some other reports that no obvious shift of the lowest 4f5d state can be observed because of the increase in particle size.28 In addition, it is well established that the 4f5d configuration of Pr3+ is always found at 12,240 750 cm1 higher energy than the 5d configuration of Ce3+, and their main configuration splittings are always same.30,31 This situation is independent on the type of host lattice. Then from the energetic positions of the 5d level of Ce3+, the position of the main 4f5d levels of Pr3+ in the same host can be estimated. In Fig. 7 the excitation spectrum of SAO:2% Ce3+ (gray line) has been shifted to the highenergy side by 12,240 + 690 cm1. The similarities between the two spectra reveals the main 4f5d configuration of Pr3+ in nanodisks is approximately between 50,000 cm1 (200 nm) and 59,000 cm1 (170 nm). The arrows indicate the possible locations of five main 5d bands of Ce3+, which are very consistent with the observed splittings of Pr3+ in nano and bulk materials. The luminescent dynamics of 1S0 in SAO:1% Pr3+ nanodisks were also measured and compared with that in the bulk in Fig. 8. The relaxations of 1S0 can be fit using single-exponential decays with relaxation times of 563 2 ns and 614 1 ns for the nanodisk and the bulk samples, respectively. The differential decay time is only about 9% between the two samples.
FIG. 8. Fluorescence decay curves of Pr3+ 1S0 emissions in SAO:1% Pr3+ nanodisk (gray) and bulk (black) materials at 10 K after pulsed excitation at 193 nm. The detection wave length is 402 nm.
The visible QE of PCE in SAO:1% Pr3+ nanodisks, ZPCE, upon the lowest 4f5d state excitation can be written as ZPCE ¼ að1 þ Zp bÞ
;
ð1Þ
where a (0.39) and b (0.42) is the branch ratio of 1 S0–1I6 and 3P0–3H4 to all 1S0 and 3P0 transitions, respectively. Zp is the efficiency of 3P0 as the intermediate state in the PCE process to yield the second emission.32 Neglecting the infrared transitions of 3PJ and taking the 1S0–1I6 fluorescence intensity as a measure of the initial 3P0 population, Zp is equal to the intensity ratio of 3P0–3H4 to 1S0–1I6 (0.23). According to the luminescence spectra in Fig. 6, if the visible transition of 1S0–1I6 is only included in Eq. (1), ZPCE is evaluated to be 42%, which is close to that of the SAO:1% Pr3+ bulk samples (43%) synthesized by solid-state reaction in our lab.33 The results above confirm the spectra properties of bulk materials, fluorescence branching ratio, luminescent dynamics, and QE, can be preserved in a 5 to 10 nm particle. This fact is much different from other rare earth ions, such as Eu3+ and Tb3+,1–5 doped nanoparticles, in which usually strongly reduced QE and sharp decreased lifetime and/or nonexponential relaxation curves are observed. We attribute the particularities of SAO:1% Pr3+ nanodisks to several factors, including (i) the excellent synthetic method, (ii) the low maximumphonon-energy (MPE) of the host, and (iii) the unique optical properties of 1S0 in a special crystal field environment. Firstly, the almost unchanged QE of PCE should be credited to the excellent synthetic process. On one hand, process (4) results in the absence of –OH or other organic groups, which usually present at the surface and lead to strong quenching of the luminescent ions through coupling with these high-frequency surface groups in low-temperature chemical methods.34 Thus, this step helps to reduce the effects of surface quenching on the luminescence of Pr3+. As discussed previously, 3P0 is the intermediate state in the PCE process to yield the second emission. The statistically identical intensity ratio of 3 P0–3H4 to 1S0–1I6 demonstrates that in this system there is no additional quenching effect caused by surfaceinduced loss mechanisms, commonly reported in other nanocrystalline systems.1–5 On the other hand, the final calcination process (5) yields excellent crystallization nanoparticles, so that the size effects do not have an obvious effect on the local symmetry of the rare earth ions in nanocrystals. According to the Judd–Ofelt theory, the transition intensities between different J levels depend on the local symmetry of rare earth ions.35,36 Thus the identical branching ratio of 1S0 emissions indicates that the local symmetry of Pr3+ is truly maintained in this nanomaterial.
J. Mater. Res., Vol. 24, No. 5, May 2009
1777
Z. Nie et al.: SrAl12O19:Pr3+ nanodisks and nanoplates: New processing technique and photon cascade emission
Secondly, the relatively low MPE of SAO (600 cm1) contributes to retaining the radiative efficiencies of the two-step transitions of Pr3+ in nanomaterials. The energy-level diagram of SAO:1% Pr3+ is shown in Fig. 6. An estimation of the transition probability can be made with the “van Dijk and Schuurmans” revised energy gap law37: WNR ¼ be exp½aðDE 2h omax Þ
;
ð2Þ
where WNR is the nonradiative transition probability, be = 107 s1, a = 4.5 103 cm1. DE is the energy difference between the relative levels [2.4 104 cm1 from 1S0 to 1I6 (or 3PJ) and 3700 cm1 from 3P0 to 1D2] and homax is the MPE of host. Inserting the MPE of SAO into Eq. (1) results in a nonradiative transition probability of about 2.8 1038 s1 from 1S0 to 1I6 (or 3 PJ) and 130 s1 from 3P0 to 1D2, respectively, which are much smaller than the radiative probabilities from 1S0 (1.6 106 s1, t = 614 ns) and 3P0 (2.9 104 s1, t = 35 ms)32 levels, respectively, especially for that from 1S0, which practically can be ignored. Anyway, the low MPE of the host decreases the influence of nonradiative process because of surface/size-effects on the radiative efficiency of Pr3+ in this nanosize system. Thirdly, Pr3+ ions replace Sr2+ sites with a local symmetry of D3h in SAO.23,26 The unique optical properties of 1S0 in this crystal field environment also play a role in preserving the radiative efficiency of 1S0. The 1S0 state of Pr3+ in SAO lies close to the 4f5d band (see Fig. 7). Theoretically, the amount of the opposite parity 5d components mixed into 4f states is affected critically by their energetic separation. Thus, one could expect the optical properties of the 1S0 state would be affected greatly if the separation between the nearby 4f5d states and 1 S0 state is changed. Size-effect-induced blue shift of the lowest 4f5d states of Pr3+ was observed; however, no obvious changes in optical properties of the 1S0 state can be detected. This is the result of the special local symmetry of Pr3+ in SAO. Huang et al. investigated the mixing of the Pr3+ 4f2 1S0 state with the 4f5d states in the site with a D3h symmetry.38 The results showed that in this symmetric site the mixed 4f5d states are essentially not near the 1S0 state. Only two 4f5d states, 21/2(|fdG1 1F3 3i |fdG1 1F3 3i) and 21/2 (|fdG1 1H5 3i |fdG1 1H5 3i), can be mixed into the 1S0, |4f2 1S0i, state. The nonzero density of the two states appears at 3000 cm1 above the bottom of the 4f5d configuration, and the maxima of the state densities are 10,000 cm1 or higher above the bottom. Finally, only about a 0.003 admixture of 4f5d wavefunctions was contained into 1S0 state. Thus, the optical properties of the 1S0 state in SAO, fluorescence branching ratios, and dynamics process will not be affected so much that an obvious change cannot be observed compared with those in the bulk if only a slight shift, whether it is blue or red, of the 4f5d configuration relative to 1S0 state is achieved. 1778
In summary, a combination of various factors above results in the excellent optical properties of SAO:Pr3+ nanodisks, which are basically the same as those of the bulk counterparts. Additional investigations into the synthesis of various other nanosize mixed oxides is underway using this low-cost technique to find more oxide nanocrystalline phosphors, the QEs of which can be compared with their bulk counterparts. IV. CONCLUSIONS
We have developed a new facile synthetic route to prepare well-dispersed SAO:1% Pr3+ nanocrystals. High-quality SAO:1% Pr3+ nanodisks and nanoplates can be fabricated in sequence by adjusting the calcination temperature through this new synthesis method. With the additional processes, large particle size distribution and the phenomenon of agglomeration, which usually occurred in conventional PSE method, were eliminated completely, and the crystallization temperature moved down relative to that of the bulk counterparts calcined by solid-state reaction. Size-effect-induced blue shift of the lowest 4f5d states of Pr3+ was observed in SAO:1% Pr3+ nanodisks, and their visible QE of PCE can be compared with their bulk counterparts under the lowest 4f5d state excitation. This technique may provide a basis to produce other high-temperature-phase oxide nanocrystals for functional and structural use. ACKNOWLEDGMENTS
This work was financially supported by the National Nature Science Foundation of China (10834006 and 10774141) and the Ministry of Knowledge and Economy of Korea through the Ultrashort Quantum Beam Facility Program. REFERENCES 1. K. Riwotzki and M. Haase: Wet-chemical synthesis of doped colloidal nanoparticles: YVO4:Ln (Ln = Eu, Sm, Dy). J. Phys. Chem. B 102, 10129 (1998). 2. K. Riwotzki, H. Meyssamy, A. Kornowski, and M. Haase: Liquid-phase synthesis of doped nanoparticles: Colloids of luminescing LaPO4:Eu and CePO4:Tb particles with a narrow particle size distribution. J. Phys. Chem. B 104, 2824 (2000). 3. R. Schmechel, M. Kennedy, H. von Seggern, H. Winkler, M. Kolbe, R.A. Fischer, L. Xaomao, A. Benker, M. Winterer, and H. Hahn: Luminescence properties of nanocrystalline Y2O3: Eu3+ in different host materials. J. Appl. Phys. 89, 1679 (2001). 4. K. Ko¨mper, H. Borchert, J. Storz, A. Lobo, S. Adam, T. Mo¨ller, and M. Haase: Green-emitting CePO4:Tb/LaPO4 core-shell nanoparticles with 70% photoluminescence quantum yield. Angew. Chem. Int. Ed. 42, 5513 (2003). 5. Z.G. Wei, L.D. Sun, X.C. Jiang, C.S. Liao, and C.H. Yan: Correlation between size-dependent luminescent properties and local structure around Eu3+ ions in YBO3:Eu nanocrystals: An XAFS study. Chem. Mater. 15, 3011 (2003).
J. Mater. Res., Vol. 24, No. 5, May 2009
Z. Nie et al.: SrAl12O19:Pr3+ nanodisks and nanoplates: New processing technique and photon cascade emission
6. S. Shionoya and W.M. Yen: Phosphor Handbook, edited by R. Stern and A.W. Sterkweather (Chemical Rubber Company Press, London, 1995), p. 623. 7. L.D. Merkle, B. Zandi, R. Moncorge, Y. Guyot, H.R. Verdun, and B. McIntosh: Spectroscopy and laser operation of Pr, Mg: SrAl12O19. J. Appl. Phys. 79, 1849 (1996). 8. B. Zandi, L.D. Merkle, J.B. Gruber, D.E. Wortman, and C.A. Morrison: Optical spectra and analysis for Pr3+ in SrAl12O19. J. Appl. Phys. 81, 1047 (1997). 9. S. Maschio, E. Lucchuni, and V. Sergo: Piezospectroscopic analysis of the residual stresses in the strontium hexaluminate/zirconia (SrAl12O19/ZrO2) system. J. Am. Ceram. Soc. 82, 3145 (1999). 10. L. Di Filippo, E. Lucchuni, V. Sergo, and S. Maschio: Synthesis of pure monolithic calcium, strontium, and barium hexaluminates for catalytic applications. J. Am. Ceram. Soc. 83, 1254 (2000). 11. A. Douy and M. Capron: Crystallisation of spray-dried amorphous precursors in the SrO–Al2O3 system: A DSC study. J. Eur. Ceram. Soc. 23, 2075 (2003). 12. L. Chen, X. Sun, Y. Liu, K. Zhou, and Y. Li: Porous ZnAl2O4 synthesized by a modified citrate technique. J. Alloys Compd. 376, 257 (2004). 13. Y. Xu, W. Peng, S. Wang, X. Xiang, and P. Lu: Synthesis of SrAl12O19 via citric acid precursor. Mater. Sci. Eng., B 123, 139 (2005). 14. Y. Xu, W. Peng, S. Wang, X. Xiang, and P. Lu: Synthesis of SrAl2O4 and SrAl12O19 via ethylenediaminetetraacetic acid precursor. Mater. Chem. Phys. 98, 51 (2006). 15. S.J. Lee, E.A. Benson, and W.M. Kriven: Preparation of Portland cement components by poly(vinyl alcohol) solution polymerization. J. Am. Ceram. Soc. 82, 2049 (1999). 16. M.A. Gu¨lgun, M.H. Nguyen, and W.M. Kriven: Polymerized organic–inorganic synthesis of mixed oxides. J. Am. Ceram. Soc. 82, 556 (1999). 17. M.H. Nguyen, S.J. Lee, and W.M. Kriven: Synthesis of oxide powders by way of a polymeric steric entrapment precursor route. J. Mater. Res. 14, 3427 (1999). 18. M.L. Balmer, F.F. Lange, V. Jayaram, and C.G. Levi: Development of nano-composite microstructures in ZrO2–Al2O3 via the solution precursor method. J. Am. Ceram. Soc. 78, 1489 (1995). 19. X. Li, H. Zhang, F. Chi, S. Li, B. Xu, and M. Zhao: Synthesis of nanocrystalline composite oxide La1xSrxFe1yCoyO3 with the perovskite structure using polyethylene glycol-gol method. Mater. Sci. Eng., B 18, 209 (1993). 20. S.J. Lee and W.M. Kriven: Crystallization and densification of nano-size amorphous cordierite powder prepared by a PVA solution-polymerization route. J. Am. Ceram. Soc. 81, 2605 (1998). 21. S.J. Lee, C.H. Lee, and W.M. Kriven: Synthesis of low-firing anorthite powder by the steric-entrapment route. Ceram. Eng. Sci. Proc. 23, 33 (2002). 22. S.J. Lee and W.M. Kriven: Preparation of ceramic powders by a solution-polymerization route employing PVA solution. Ceram. Eng. Sci. Proc. 19, 469 (1998).
23. A.M. Srivastava and W.W. Beers: Luminescence of Pr3+ in SrAl12O19: Observation of two photon luminescence in oxide lattice. J. Lumin. 71, 285 (1997). 24. A.M. Srivastava: Encyclopedia of Physical Science and Technology, edited by R.A. Meyers, 3rd ed. (Academic Press, San Diego, CA, 2002), p. 855. 25. A.M. Srivastava: Handbook of Luminescence, Display Materials, and Devices, edited by H.S. Nalwa and L.S. Rowher (American Scientific Publishers, Stevenson Ranch, CA, 2003), p. 79. 26. K. Kimura, M. Ohgaki, K. Tanaka, H. Morikawa, and F. Marumo: Study of the bipyramidal site in magnetoplumbitelike compounds, SrM12O19 (M = Al, Fe, Ga). J. Solid State Chem. 87, 186 (1990). 27. S.R. Jansen, H.T. Hintzen, R. Metselaar, J.W. de Haan, L.J.M. van de Ven, A.P.M. Kentgens, and G.H. Nachtegaal: Multiple quantum 27Al magic-angle-spinning nuclear magnetic resonance spectroscopic study of SrAl12O19: Identification of a 27 Al resonance from a well-defined AlO5 Site. J. Phys. Chem. B 102, 5969 (1998). 28. S.M. Loureiro, A. Setlur, W. Heward, S.T. Taylor, H. Comanzo, M. Manoharan, A. Srivastava, P. Schmidt, and U. Happek: First observation of quantum splitting behavior in nanocrystalline SrAl12O19:Pr, Mg phosphor. Chem. Mater. 17, 3108 (2005). 29. G.H. Dieke and H.M. Crosswhite: The spectra of the doubly and triply ionized rare earths. Appl. Opt. 2, 675 (1963). 30. P. Dorenbos: The 4f n $ 4f n15d transitions of the trivalent lanthanides in halogenides and chalcogenides. J. Lumin. 91, 91 (2000). 31. P. Dorenbos: The 5d level positions of the trivalent lanthanides in inorganic compounds. J. Lumin. 91, 155 (2000). 32. S. Huang, X. Wang, R.S. Meltzer, A.M. Srivastavad, A.A. Setlur, and W.M. Yen: Photon cascade emission and quantum efficiency of the 3P0 level in Pr3+-doped SrAl12O19 system. J. Lumin. 94–95, 119 (2001). 33. Z.G. Nie, J.H. Zhang, X. Zhang, X.G. Ren, G.B. Zhang, and X.J. Wang: Evidence for visible quantum cutting via energy transfer in SrAl12O19:Pr,Cr. Opt. Lett. 32, 991 (2007). 34. J.K. Krebs, S.P. Feofilov, A.A. Kaplyanskii, R.I. Zakharchenia, and U. Happek: Non-radiative relaxation of Yb3+ in highly porous g-Al2O3. J. Lumin. 83, 209 (1999). 35. B.R. Judd: Optical absorption intensities of rare-earth ions. Phys. Rev. 127, 750 (1962). 36. G.S. Ofelt: Intensities of crystal spectra of rare-earth ions. J. Chem. Phys. 37, 511 (1962). 37. J.M.F. van Dijk and M.F.H. Schuurmans: On the nonradiative and radiative decay rates and a modified exponential energy gap law for 4f–4f transitions in rare-earth ions. J. Chem. Phys. 78, 5317 (1983). 38. S. Huang, X. Wang, R.S. Meltzer, A.M. Srivastavad, A.A. Setlur, and W.M. Yen: The mixing of the 4f2 1S0 state with the 4f5d states in Pr3+ doped SrAl12O19. J. Lumin. 94–95, 119 (2001).
J. Mater. Res., Vol. 24, No. 5, May 2009
1779
