Photoluminescence investigations on a novel green-emitting phosphor
Ba
3
Sc(BO
3
)
3
:Tb
3+using synchrotron vacuum ultraviolet radiation
De-Yin Wang,
aYi-Chin Chen,
aChien-Hao Huang,
aBing-Ming Cheng
band Teng-Ming Chen*
aReceived 5th December 2011, Accepted 19th March 2012 DOI: 10.1039/c2jm16350k
Vacuum ultraviolet (VUV) spectroscopic properties of undoped and Tb3+-doped Ba
3Sc(BO3)3were
investigated by using synchrotron radiation. The Tb3+-doped Ba
3Sc(BO3)3sample crystallized in
a flower-like shape even was synthesized at 1100C. Upon VUV excitation, Ba3Sc(BO3)3exhibited an
intrinsic broad UV emission centered at 336 nm, which results from radiative annihilation of self-trapped exciton (STE) that may presumably be associated with band-gap excitations or molecular transitions within the BO33group. The maximum host absorption for Ba3Sc(BO3)3was found at about
180 nm. Upon doping of Tb3+ions into Ba
3Sc(BO3)3, an efficient energy transfer from the host
excitations to Tb3+via STE emission was observed, showing host sensitization of Tb3+occurs. The
energy transfer from host to Tb3+via STE emission in Ba
3Sc(BO3)3:Tb3+was studied as a function of
temperature and Tb3+doping concentration. It has been demonstrated that the energy transfer
efficiency was increased with either increasing temperature or Tb3+doping concentration. In the case of
temperature dependent energy transfer, the energy transfer from the STE to Tb3+is thermally activated,
probably due to exciton mobility, while in the case of concentration dependent energy transfer, the energy transfer from the STE to Tb3+is promoted due to a closer distance between the STE and Tb3+at
high Tb3+concentration.
1
Introduction
With the rapid development of mercury-free fluorescent lamps and plasma display panels (PDPs), which use Xe-discharge as the excitation source for phosphor, the demand for phosphors that could be efficiently excited by a vacuum ultraviolet (VUV,l < 200 nm) radiation has been rapidly increasing.1–11Several VUV phosphors have been developed for PDPs applications, such as red-emitting (Y,Gd)BO3:Eu3+, blue-emitting BaMgAl10O17:Eu2+
and green-emitting Zn2SiO4:Mn2+.11–14However, many problems
still remain, including poor color purity for the red phosphor, thermal and aging degradation for the blue phosphor, and long decay time for the green phosphor.11–14In order to improve the performance of PDPs, much attention has been paid to explore and find new color-emitting phosphors. In the case of phosphors excited by VUV radiation, it is widely considered the excitation energy is transferred from host matrix to the activator;15,16that is to say, the activator is not directed excited. As a consequence, for a good VUV phosphor, it is necessary for the excitation energy to be efficiently transferred from the host matrix to the activator. If a host itself emits and its emission overlaps with the direct absorptions of an activator (e.g., f–f or f–d transition), then
nonradiation energy transfer from host to activator would happen according to the Forster–Dexter energy transfer theory,17 giving rise to efficient emission from activator in such host. From this standpoint, with an aim to develop new green-emitting phosphor, we have selected Ba3Sc(BO3)3 as a host lattice with
Tb3+ serving as an activator. The choice of Ba
3Sc(BO3)3 is
motivated by previous results that several of its isostructural compounds such as Ba3Y(BO3)3 and Ba3Lu(BO3)3 exhibit
intrinsic UV emission under X-ray excitation.18In addition, the
f–f and f–d transitions of Tb3+generally located in the UV region.
Therefore, due to spectra overlap, efficient host-to-Tb3+energy
transfer is expected in Tb3+-doped Ba
3Sc(BO3)3. In this paper, we
have investigated the energy transfer from the host excitations to Tb3+in the matrix of Ba
3Sc(BO3)3, and examined the influence of
temperature and Tb3+ doping concentrations on the energy
transfer process.
2
Experimental
2.1. Materials and synthesis
Powder samples of undoped and Tb3+-doped Ba
3Sc(BO3)3(BSB)
were synthesized by high temperature solid-state reactions. BaCO3 (99.9%, Sigma-Aldrich), H3BO3 (99.99%,
Sigma-Aldrich), Sc2O3 (99.9%, Aldrich) and Tb4O7(99.9%, Aldrich)
were used as reagents. The oxides were mixed according to the desired stoichiometric ratios of each sample, and then
aPhosphors Research Laboratory and Department of Applied Chemistry,
National Chiao Tung University, Hsinchu 30010, Taiwan. E-mail: [email protected]; Tel: + 886-35731695
bNational Synchrotron Radiation Research Center, Hsinchu 30076, Taiwan
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 9957
www.rsc.org/materials
PAPER
thoroughly ground. The obtained mixtures were fired at 1100C in a reducing atmosphere (15% H2/85% Ar) for the Tb3+-doped
samples and in air for the undoped sample. 2.2. Characterization methods
The phase purity of all samples was checked by using powder X-ray diffraction (XRD) analysis with a Bruker AXS D8 advanced automatic diffractometer operated at 40 kV and 40 mA with Cu-Ka radiation (l ¼ 1.5418 A). Particle sizes and shapes were determined by a scan electron microscope (SEM) (JSM-6390LV, JEOL).
The VUV photoluminescence (PL) spectra were recorded at the Beamline 03A at National Synchrotron Radiation Research Center (NSSRC) in Taiwan. The experimental setup for PL spectra was similar to that described elsewhere.2,19In summary,
VUV excitation light from this high-flux beamline attached to the 1.5-GeV storage ring was dispersed with a six-meter cylindrical grating monochromator (CGM). The intensity of the VUV light is monitored with a gold mesh transmitting about 90% and recorded with an electrometer (Keithley 6512). A Jobin-Yvon HR320 equipped with a 1200 lines/mm grating and a Hama-matsu R943-02 photomultiplier (PMT) were used to record the PL spectra. For measurement of PL excitation (PLE) spectra, the dispersive emission was monitored at selected band; in which, the CGM beam line with a 450 lines/mm grating was scanned. All the PLE spectra were normalized with the spectral response curve of the CGM beam line. For measurements of low temperature spectra, the sample holder was attached to a cryo-head of a helium closed-cycle cryostat system (APD HC-4); which was mounted on a rotatable flange so that the sample could be rotated to about 45with respect to both the incident VUV source and the entrance slit of the dispersed mono-chromator. The temperature control unit provided the cold head of the cryostat to better than1 K during the data collection period. Unless otherwise specified, all spectra were measured at room temperature.
3
Results and discussion
3.1 Crystal structure, phase identification and microstructure Ba3Sc(BO3)3 (BSB) crystallizes in hexagonal crystal structure
with space group P63cm. Its crystal structure is built with isolated
BO3 triangles and scandium–oxygen octahedra and barium–
oxygen nine-vertex polyhedra.20 All boron atoms are
coordi-nated to three oxygen atoms, which are distributed layer upon layer along the c-axis. In the crystal structure of Ba3Sc(BO3)3,
there are two non-equivalent Sc sites having C3and C3v point
symmetries and four different Ba sites having C3v, Cs, C3and Cs
point symmetries, respectively.20Shown in Fig. 1 are the XRD
patterns of the undoped and Tb3+doped BSB. The XRD patterns
of the synthesized samples BSB:xTb3+(0# x # 20%) are in good
agreement with reference data (ICSD No.75340) regardless of Tb3+ doping concentration, indicating that these Tb3+ doped
samples retained the same crystal structure as that of the undo-ped BSB. However, when Tb3+doping concentration is increased
to 30% (see Fig. 1), some new peaks marked by the asterisk appear, which probably be ascribed to the formation of Ba3Tb(BO3)3(PDF No. 511848), so in this work only samples
with composition BSB:xTb3+(0# x # 30%) were investigated.
Fig. 2 shows the SEM image of a 20% Tb-doped BSB sample that has flower-liked shape with average diameter around 6mm. Each flower consists of dozens of petals with the averages width of 200 nm and length of 2mm. Taking a close look at a single flower (see the inset in Fig. 2), we can see that all the petals grow out from the central part of a flower and look like icicles.
3.2 VUV-excited spectral properties of undoped and Tb3+
-doped BSB
The VUV spectroscopic investigation of undoped BSB was per-formed as a prerequisite for the interpretation of luminescence spectra of BSB:Tb3+. The PL and PLE spectra of the undoped
BSB measured at room temperature are shown in Fig. 3. Upon excitation at 172 nm, BSB emits a broad intrinsic UV-lumines-cence band with maximum at 336 nm. The emission band most likely arises from the recombination of self-trapped excitons (STE) that may be associated with band-gap excitations or molecular transitions within the BO33 group.21,22 The broad
absorption band with maximum at round 180 nm in the PLE spectrum of BSB, obtained by monitoring the intrinsic UV
Fig. 1 XRD patterns of BSB:xTb3+(0# x # 30%).
Fig. 2 SEM image of BSB:20%Tb3+. The inset shows a zoomed image.
emission, is assigned to the host absorption that could be asso-ciated with intramolecular transitions of the borate group. The intrinsic host emission has extended to 450 nm in wavelength, which partially overlaps with the Tb3+4f–4f absorptions (e.g.,7F
6
/5D
3transition). Due to the spectral overlap, when Tb3+ions
are introduced to the host lattice of BSB, efficient energy transfer from host excitation to Tb3+ions is expected. Indeed, as indicated
in Fig. 4, when BSB is doped with 1% Tb3+, the intrinsic UV
emission is greatly reduced and accompanied by an increase in Tb3+emission, suggesting efficient energy transfer from STE to
Tb3+. The PL spectrum of the sample containing 1% Tb3+can be
divided into two spectral regions, an UV emission in 300–400 nm region ascribed to the overlapped emission from the host and Tb3+ 5D
3–7F6 transition, and a green emission in 475–650 nm
region originated from Tb3+ 5D
4–7FJ(J¼ 6, 5, 4 and 3)
transi-tions. Weak5D
3–7F6transition of Tb3+around 380 nm can be
observed in the PL spectra of BSB:1%Tb3+, implying that energy
transfer from host to Tb3+ involves in Tb3+ 5D
3 state.
Furthermore, Fig. 5 is a comparison between the PLE spectrum of BSB:1%Tb3+obtained by monitoring the Tb3+ 5D
4–7F5
emis-sion at 543 nm and that of undoped BSB monitoring the STE emission, respectively. The PLE spectrum of the two samples exhibits similarly features with a broad absorption band below 210 nm, indicating that they both resulted from the same origin, viz. host absorption. The presence of the host absorption band in the PLE spectrum of 1% Tb3+-doped BSB, detecting within the
Tb3+ emission, gives another evidence that the host-to-Tb3+
energy transfer in fact took place. Furthermore, as to the PLE spectrum of 1% Tb3+doped sample, the strong excitation band in
220–270 nm region with maximum around 246 nm is assigned to the spin-allowed 4f8–4f75d1transition of Tb3+, and relative weak
excitation band around 278 nm is assigned to the spin-forbidden 4f8–4f75d1transition of Tb3+.
3.3 Concentration-dependent luminescence of BSB:Tb3+
To give further evidence to support the occurrence of host-to-Tb3+ energy transfer, and determine the optimal Tb3+doping
concentration in BSB:Tb3+, concentration dependent
lumines-cence measurements were performed. Presented in Fig. 6 are the PL spectra of BSB:xTb3+(0# x # 30%) under excitation with
light lying within the host absorption bands. As revealed in Fig. 6, in the presence of Tb3+, the STE emission is greatly
quenched due to energy transfer. At lower Tb3+doping
concen-tration (i.e., 1%), weak STE emission can still be observed, because in this case a portion of excited STE could relax by photon emission. However, at higher Tb3+doping concentration,
the STE emission is significantly quenched. The energy transfer is so efficient that the STE emission almost totally quenched when Tb3+doping concentration is increased to 5%. Assuming that the
decrease of STE is caused by energy transfer from STE to Tb3+
acceptor only, and ignoring the possible energy transfer to other impurity or defect sites, we can roughly estimate the energy transfer efficiency (h) by analogy to the case for Ce3+-Tb3+:23
h ¼ 1 Is/Is0 (1)
Fig. 3 PLE and PL spectra of undoped BSB monitoring the STE
emission at 336 nm and under excitation at 172 nm.
Fig. 4 PL spectra of undoped BSB and 1% Tb3+-doped BSB under
excitation at 172 nm. The inset shows an enlarged portion of the PL spectrum of BSB:1%Tb3+in the 350–470 nm region.
Fig. 5 PLE spectra of 1% Tb3+-doped BSB monitoring Tb3+emission at
543 nm and the STE emission of undoped BSB at 336 nm. The PLE spectra are normalized on the host absorption intensity.
where Isand Is0are the STE emission intensity in the presence
and absence of Tb3+acceptor, respectively. According to eqn (1),
the energy transfer efficiency for 1% Tb3+-doped sample is
roughly estimated to be about 75%. With further increasing Tb3+
doping concentration, the corresponding STE emission intensity deceases further, and accordingly, the energy transfer efficiency at high Tb3+doping concentration will be much greater than 75%
and close to 100%. As the Tb3+doping concentration increases,
there are more Tb3+ ions that will be distributed around the
created STE and the distance between the center of STE and Tb3+
is shortened and, consequently, thus the energy transfer proba-bility between them is increased, then the STE emission is quenched. Partially owing to the host-to-Tb3+energy transfer, it
is found that the emission from Tb3+ 5D
4 level increases upon
increasing Tb3+ dopant concentrations. In principle, an
increasing in the Tb3+concentration will increase its absorption
efficiency, which will simultaneously increase the emitted light intensity of Tb3+.24 The optimal Tb3+doping concentration in
BSB:Tb3+was found to be 20%. Upon 172 nm excitation, the
relative quantum efficiency of the 20% Tb3+doped sample was
estimated to be about 39% by comparing its PL spectrum with that of sodium salicylate whose quantum efficiency is about 58% over the excitation wavelength range from 140 to 320nm.21With further increasing Tb3+ concentration, the Tb3+ 5D
4 emission
starts to decease due to concentration quenching effect. In addition, it is found that the Tb3+ 5D
3–7F6emission (380 nm) is
deceased with increasing Tb3+dopant concentration (see the inset
in Fig. 6), which is caused by the cross relaxation process between
5D
3,7F0 / 5D4,7F6 levels.11,25 An increasing in Tb3+ doping
concentration has also shorten the distance between two adjacent Tb3+ions, resulting in an increase in the cross relaxation
prob-ability; thus the Tb3+ 5D
3–7F6emission is greatly reduced at high
Tb3+concentration.
3.4 Temperature-dependent luminescence of BSB:Tb3+
To obtain a better insight in the energy transfer process involves in STE emission in BSB:Tb3+, and examine the influence of
temperature on the energy transfer, we have the temperature dependent PL spectra for a 1% Tb3+-doped BSB sample and
presented the results in Fig. 7. In the whole temperature range (15–295 K) under study, the PL spectra of BSB:1%Tb3+ are
composed of a broad emission band from STE and line emissions from Tb3+. As the temperature drops, the original almost
quenched STE emission at room temperature increases in intensity and is again clearly observed. In the meantime, the emission intensity of Tb3+is gradually decreased. These
obser-vations suggest that the efficiency of the host-to-Tb3+ energy
transfer decreased with temperature cooling and implies that the energy transfer is inhibited at low temperatures. On one hand, it is generally recognized that due to efficient energy transfer, the excitation energy will migrate about large number of centers before being emitted.24However, at low temperature, in the case
of 1% Tb3+doping, it is statistically unlikely that the STE center
would be created close to a Tb3+ ion. In addition, as the
temperature drops, it is likely that the STE is getting less mobile and even become immobilized at low temperature; as a conse-quence, the STE migration rate is slow down and the STE migration range is getting smaller either,26,27causing the energy
transfer rate to Tb3+is sufficiently reduced at low temperature.
On the other hand, even for the purest crystals, there is always a certain concentration of defects that can act as acceptors, so that the STE emission could transfer energy to them either. These defects can relax to their ground state by multiphonon emission, and in this situation, the STE emission will be quenched either. Similar to the discussion on temperature dependent host-to-Tb3+
energy transfer, energy transfer efficiency from STE emission to these defects will be deceased with decreasing temperature, and therefore the intensity of the STE emission increased with decreasing temperature.
3.5 Luminescence performance comparison between BSB:Tb3+
and Zn2SiO4:Mn2+and LaPO4:Ce3+,Tb3+
To further evaluate the luminescence performance of BSB:Tb3+,
we have compared the PL and PLE spectra, and the CIE coor-dinates of BSB:20%Tb3+with those of Zn
2SiO4:Mn2+(P1-G1S,
Kasei Optonix Ltd.) and LaPO4:Ce3+,Tb3+(NP-220, Nichia Ltd.)
(see Fig. 8 and Fig. 9). The Zn2SiO4:Mn2+phosphor is the most
commonly used green-emitting phosphor in PDPs, and the LaPO4:Ce3+,Tb3+phosphor is a typical green-emitting phosphor Fig. 6 Comparison of PL spectra of BSB:xTb3+(0# x # 30%) under
excitation at 172 nm. The inset shows an enlarged portion for the PL spectra of BSB:xTb3+(1# x # 30%) in the 300–400 nm region.
Fig. 7 Temperature dependent PL spectra of BSB:1%Tb3+ (l ex ¼
172 nm).
widely used in CCFL. The results indicate that the PLE intensity of BSB:20%Tb3+ at 172 nm is about 1.25 times of that of
Zn2SiO4:Mn2+, and 58% of that of LaPO4:Ce3+, Tb3+,
respec-tively. Upon excitation at 172 nm, the integrated emission intensity of BSB:20%Tb3+in the 470–630 nm range is about 45%
of that of Zn2SiO4:Mn2+, and 65% of that of LaPO4:Ce3+, Tb3+,
respectively. Comparing with the green CIE coordinates defined by NSTC (see the inset table in Fig. 9), it is found that BSB:20% Tb3+ has an inferior green color purity than that of
Zn2SiO4:Mn2+(P1-G1S), but a superior green color purity than
that of LaPO4:Ce3+,Tb3+(NP-220), which is consistent with what
is reflected in the inset figures in Fig. 9. In addition, heating the phosphors up to 500C to seal the PDP panels is inevitable, and is also essential for good adhesion between phosphor and substrate, and for complete elimination of organic additives.28To
examine the thermal stability of BSB:Tb3+, we have
post-annealed the 20%-Tb3+doped sample at 500C in the air for 3 h,
then cool it down naturally to room temperature. Under 172 nm excitation, the emission intensity of the sample with post-annealing is around 95% of that without post-post-annealing, sug-gesting Tb3+is quite stable in BSB:Tb3+.
4
Conclusions
In summary, we have investigated the VUV-excited luminescence of the pristine- and Tb3+-doped Ba
3Sc(BO3)3, and demonstrated
the sensitization of Tb3+emission by energy transfer from the
self-trapped exciton. We also examined the influence of temper-ature and Tb3+doping concentration on the host-to-Tb3+energy
transfer in Ba3Sc(BO3)3:Tb3+. Upon host excitation,
Ba3Sc(BO3)3exhibited a broad UV emission that is attributed to
the recombination of self-trapped excitons. From the excitation spectrum of the STE emission, the band gap of Ba3Sc(BO3)3can
be roughly estimated to be about 6.9 eV. When Tb3+ions were
incorporated into Ba3Sc(BO3)3, an efficient energy transfer from
the host excitations to Tb3+via STE emission was observed. The
strong quenching of STE emission and similar excitation spectra of the Tb3+-doped and undoped samples gave conclusive
evidence for such energy transfer. In addition, it has demon-strated that host-to-Tb3+ energy transfer efficiency will be
increased by increasing Tb3+ concentration and reduced by
cooling temperature. As the Tb3+ doping concentration
increases, the distance between the STE and Tb3+is shortened
consequently, thus the energy transfer probability is increased; while as the temperature drops, the STE is getting less mobile, as a consequence, the STE migration rate slows down and the STE migration range is getting shorter, leading to an observed reduction of energy transfer rate to Tb3+at low temperature.
Acknowledgements
We gratefully thank the National Science Council of Taiwan for financial support under Contract Nos. NSC100-2811-M-009-068 (D.-Y. W.) and NSC98-2113-M-009-005-MY3 (T.-M. C.), and the group members in Dr B.-M. Cheng’s Laboratory at NSRRC for their assistance with the VUV spectra measurements.
Notes and references
1 R. T. Wegh, H. Donker, K. D. Oskam and A. Meijerink, Science, 1999, 283, 663.
2 C. H. Huang, T. M. Chen and B. M. Cheng, Inorg. Chem., 2011, 50, 6552.
3 K. V. Ivanovskikh, A. Meijerink, F. Piccinelli, A. Speghini, C. Ronda and M. Bettinelli, J. Lumin., 2010, 130, 893.
4 W. B. Im, Y. Kim, H. S. Yoo and D. Y. Jeon, Inorg. Chem., 2009, 48, 557.
5 D. Y. Wang, N. Kodama and L. Zhao, J. Electrochem. Soc., 2010, 157, J233.
6 L. Beauzamy, B. Moine, R. S. Meltzer, Y. Zhou, K. J. Kim and P. Gredin, Phys. Rev. B, 2008, 78, 184302.
7 M. B. Xie, Y. Tao, Y. Huang, H. B. Liang and Q. Su, Inorg. Chem., 2010, 49, 11317.
8 S. Kubota and M. Shimada, Appl. Phys. Lett., 2002, 81, 2749. 9 R. L. Rabinovitz, K. J. Johnston and A. L. Diaz, J. Phys. Chem. C,
2010, 114, 13884.
10 I. Y. Jung, Y. Cho, S. G. Lee, S. H. Sohn, D. K. Kim, D. K. Lee and Y. M. Kweon, Appl. Phys. Lett., 2005, 87, 191908.
11 Z. F. Tian, H. B. Liang, W. P. Chen, Q. Su, G. B. Zhang and G. T. Yang, Opt. Express, 2009, 17, 956.
12 T. J€ustel, J. C. Krupa and D. U. Wiechert, J. Lumin., 2001, 93, 179. Fig. 8 Comparison of PLE spectra of BSB:20%Tb3+(l
em¼ 543 nm),
commodities Zn2SiO4:Mn2+(lem¼ 525 nm) and LaPO4:Ce3+,Tb3+(lem¼
543 nm).
Fig. 9 PL spectra of BSB:20%Tb3+and commodities Zn
2SiO4:Mn2+and
LaPO4:Ce3+,Tb3+excited at 172 nm. The inset figures are the photographs
the three phosphors under 254 nm UV lamp radiation, and the inset table is the CIE coordinates of the three phosphors.
13 C. H. Kim, I. E. Kwon, C. H. Park, Y. J. Hwang, H. S. Bae and B. Y. Yu, J. Alloys Compd., 2000, 311, 33.
14 S. X. Zhang, IEEE Trans. Plasma Sci., 2006, 34, 294. 15 B. Moine and G. Bizarri, Opt. Mater., 2006, 28, 58.
16 N. Yokosawa, K. Suzuki and E. Nakazawa, Jpn. J. Appl. Phys., 2003, 42, 5656.
17 D. L. Dexter, J. Chem. Phys., 1953, 21, 836.
18 C. J. Duan, J. L. Yuan and J. T. Zhao, J. Solid State Chem., 2005, 178, 3698.
19 H. C. Lu and B. M. Cheng, Anal. Chem., 2011, 83, 6539.
20 J. R. Cox, D. A. Keszler and J. F. Huang, Chem. Mater., 1994, 6, 2008.
21 S. P. Feofilov, Y. Zhou, J. Y. Jeong, D. A. Keszler and R. S. Meltzer, J. Lumin., 2007, 125, 80.
22 I. N. Ogorodnikov, V. A. Pustovarov, A. V. Kruzhalov, L. I. Isaenko, M. Kirm and G. Zimmerer, Phys. Solid State, 2000, 42, 464.
23 C. H. Hsu and C. H. Lu, J. Mater. Chem., 2011, 21, 2932.
24 B. Henderson and G. F. Imbusch, Optical Spectroscopy of Inorganic Solids, 1989.
25 G. C. Kim, H. L. Parka and T. W. Kim, Mater. Res. Bull., 2001, 36, 1603.
26 K. Tanimura and N. Itoh, Nucl. Instrum. Methods Phys. Res., Sect. B, 1988, 32, 211.
27 J. Krupa and N. A. Kulagin, Physics of Laser Crystals, V.126, Kluwer Academic Publishers, The Netherlands, 2002.
28 K. S. Sohn, S. S. Kim and H. D. Park, Appl. Phys. Lett., 2002, 81, 1759.
