Particle Size Effect on the Packaging Performance of
YAG:Ce Phosphors in White LEDs
Shih Chieh Huang, Jui Kung Wu, and Wei-Jen Hsu
Hsiung Din Technologies Corporation, 248-43, Sinsheng Road, L.E.P.Z., Kaohsiung 806, Taiwan
Hsin Hsiung Chang, Hsien Yen Hung, and Chi Lian Lin
Nanching Optotech Corporation, 888, Ching Kuo Road, Taoyuan 330, Taiwan
Hung-Yuan Su
Lite-On Technology Corporation, 90, Chien I Road, Taipei 235, Taiwan
Nitin Bagkar, Wei-Chih Ke, Hui Tung Kuo, and Ru-Shi Liu*
Department of Chemistry, National Taiwan University, Taipei 106, Taiwan
The synthesis of cerium-doped yttrium aluminum garnet (YAG:Ce) phosphor of different sizes with uniform size dis-
tribution was carried out using solid-state reaction followed by grinding and sieving method. The effect of particle size dis-
tribution of YAG:Ce phosphors on the photoluminescence (PL) properties was investigated. The results demonstrate that the
uniform size distribution and particle size affects the packaging performance in white light emitting diode (LED) applications.
The YAG:Ce phosphors with different particle sizes were packaged in white LEDs using different amounts of each phosphors
in order to get similar efficiency as that of commercially available YAG:Ce phosphors. It was observed that minimum amount
of phosphor material is required for smaller particle size for getting the similar efficiency as that exhibited by commercially
available YAG:Ce phosphors. The results are particularly interesting in view of reducing the cost of current LEDs by lowering
the amount of phosphors without compromising the efficiencies of final LED package. A systematic study of YAG:Ce phos-
phors on the packing performance in white LEDs is reported.
Introduction
White light emitting devices (LEDs) have been
considered as a replacement for conventional lighting
systems because of their excellent properties such as high
energy efficiency, long lifetime, low-power consumption
Int. J. Appl. Ceram. Technol., 6 [4] 465–469 (2009)
DOI:10.1111/j.1744-7402.2009.02367.x
Ceramic Product Development and Commercialization
The authors thank the National Science Council (contract no. NSC 96-2120-M-002-019)
and the Ministry for Economic Affairs (contract no. 96-EC-17-A-07-S1-043) for financial
support.
*rsliu@ntu.edu.tw
r 2009 The American Ceramic Society
and environmental protection.
1–3
Generally, the white
light was produced by a combination of blue LED with
yellow luminescence from Y
3
Al
5
O
12
:Ce
31
yttrium alu-
minum garnet (YAG:Ce) phosphor materials.
4
The yel-
low emission is sufficient enough to complement the
residual blue light that escapes through the phosphor to
produce white light. Thus, YAG:Ce has been the most
suitable phosphor that can be utilized in the white LED
market. The perform ance of phosphors in lighting ap-
plications is usually related to emission efficiency,
quenching of luminescent centers and internal scatter-
ing. There have been numerous studies to impro ve
YAG/Ce efficiency with focus on improving the YAG
host quality, reducing defect quenching centers and re-
ducing concentration quenching.
5–7
The size reduction
of phosphors was found to be the most effective way to
improve the LED efficiency. The smaller particles pos-
sess a higher surface to volume ratio, which enhances the
efficiency of absorption and emission.
It is desirable to have a fine particle size for high-
resolution, optimum chromaticity, and brightness in
white LEDs. The effect of particle size of YAG:Ce
phosphors on the luminescent properties has been ex-
tensively studied.
8,9
However, the effect of particle size
of phosphors on the packaging performance of white
LEDs and the cost reduction have not been studied in
the literature. The cost reduction by improvement in
the manufacturing process as well as in the luminescent
properties is extremely valuable for appli cations in white
LEDs. In this paper, we have synthesized four different
sizes of YAG:Ce phosphors with narrow size distribu-
tion and studied the effect of particle size on the final
packaging performance in white LEDs for the first time.
One of the synthesized YAG:Ce samples (PS-1) resem-
bles the particle size of the comme rcially available
YAG:Ce phosphors. The characterization of YAG:Ce
phosphors were carried out by X-ray diffraction, scan-
ning electron microscope (SEM), and PL measure-
ments. A systematic study of YAG phosphors on the
packaging performance in white LEDs is reported.
Experimental Procedures
Synthesis of YAG:Ce phosphors were carried out by
conventional solid-state reac tion in which the reactants
such as Y
2
O
3
,Al
2
O
3
, and CeO
2
were mixed in a sto-
ichiometric amounts in order to get nominal composi-
tion of (Y
2.95
Ce
0.05
)Al
5
O
12
at around 15001C for 2 h in
reducing atmoshpere of 5% H
2
in 95% N
2
. The thor-
oughly grinded samples were then passed through
different sieves in order to separate different particle
sizes of YAG:Ce phosphors and obtain uniform size
distribution. In this way, four dirrerent samples were
collected with different particle sizes by repeatative siev-
ing. The four YAG:Ce samples with decreasing particle
sizes were desiganted as PS-1, PS-2, PS-3, and PS-4,
respectively. The phase purity was analyzed by X-ray
powder diffraction (XRD) measurements using an
X’Pert PRO advanced automatic diffractometer
(X’Pert PRO, Almelo, The Netherlands) with CuKa
radiation operating at 45 kV an d 40 mA. The SEM
measurements were carried out by using Hitachi S2400
electron microscope (Hitachi, Tokyo, Japan). The par-
ticle size distribution was analyzed using Fritsch particle
sizer ‘‘analysette 22’’ (Fritsch, Idar-Oberstein, Ger-
many). The UV photoluminescence (PL) and photolu-
minescent excitation (PLE) spectra were collected at
room temperature (RT) using a FluoroMax-3 and Flu-
oroMax-P (Jobin Yvon, Edison, NJ) in the range of
300–500 and 470–650 nm, respectively.
Results and Discussions
The powder XRD measurements were carried out
on all the YAG samples and as a representative, the
XRD pattern of PS-1 along with the standard pattern of
YAG is shown in Fig. 1. The observed peaks are in well
agreement with the standard pattern with a cubic unit
cell (space group: Ia3d) from ICSD file 882048 and
lattice parameters of B12 A
˚
, which indicate the forma-
tion of single-phase compositions. SEM measurements
Fig. 1. X-ray diffraction patterns of PS-1 along with the standard
pattern of yttrium aluminum garnet phosphor.
466 International Journal of Applied Ceramic Technology—Huang, et al. Vol. 6, No. 4, 2009
were carried out in order to observe the morphology of
YAG samples. The particle size distribution of the YAG
samples is shown in Fig. 2 an d inset shows the SEM
of PS1 and PS4 samples, which suggest different
morphologies for YAG samples. The particle sizes
with D10%, D50%, and D90% size distribution are
shown in Table I. It was observed that the particle mor-
phology is almost identical for all the samples with vari-
ation in particle sizes. The particle size of PS-1 was
found to be 22.0 mm for D90% size distribution, which
is similar to the commercially available YAG sample.
The other samples show the particle sizes of 10.5, 9.3,
and 4.4 mm for PS-2, PS-3, and PS-4, respectively.
Fig. 2. Particle size distribution of (a) PS-1, (b) PS-2, (c) PS-3, and (d) PS-4 yttrium aluminum garnet (YAG:Ce) phosphors. Inset shows
scanning electron microscopic images of (a) PS-1 and (d) PS-4 YAG:Ce phosphors.
Table I. The Particle Sizes Distribution of D10,
D50, and D90%. Wt % of Yttrium Aluminum Garnet
(YAG:Ce) Mean % of YAG Phosphor in Silicone
No. Sample
P/N
(D10%)
(lm)
P/N
(D50%)
(lm)
P/N
(D90%)
(lm)
Wt % of
YAG:Ce
1 PS-1 8.3 15.6 22.0 5.7
2 PS-2 — 7.2 10.5 4.5
3 PS-3 — 4.1 9.3 4.4
4 PS-4 — 2.2 4.4 3.7
www.ceramics.org/ACTParticle Size Effect on the Packaging Performance of YAG:Ce Phosphors in White LEDs 467
Figure 3 shows the photoluminescence excitation
(PLE) of YAG :Ce phosphors with different particle
sizes. The excitation spectra between 300 and 500 nm
show two excitation peaks, which are simila r to previous
reports on YAG:Ce system. The excitation peaks were
observed at 340 and 457 nm, which can be attributed to
the Ce
31
transitions from 4f ground state to the 5d field
splitting excited states. The broad excitation peaks sug-
gest that YAG phosphors are similar with commercially
available YAG phosphors for excitation efficiency and
can be excited by a blue chip in the range of 450–
470 nm. Figure 4 shows the emission spectra of YAG:Ce
phosphors at 460 nm excitation. It was observed that the
532 nm emission peak was slightly blue-shifted as com-
pared with bulk counterpart. This blue-shift was asso-
ciated with the size of the YAG phosphors. For
phosphors with smaller sizes, the lattice parameters
were usually smaller than those of the bulk materials
because of the huge surface stress. For a smaller lattice
parameter consisting of a smaller atomic spacing, the
ligands field was usually stronger and hence the blue
shift was often observable for field-sensitive emission
centers.
In order to understand the effect of particle sizes on
the final packaging performance of YAG:Ce phosphors
500 550 600 650 700
Intensity (a.u.)
Wavelen
g
th (nm)
PS-1
PS-2
PS-3
PS-4
Fig. 4. Photoluminescence spectra of (a) PS-1, (b) PS-2, (c) PS-3,
and (d) PS-4 yttrium aluminum garnet phosphors at 460 nm
excitation.
0
20
40
60
80
100
PS-4
PS-3PS-2
Iv conversion (%)
PS-1
Sample
Fig. 5. I
v
comparison (%) of white package LEDs of (a) PS-1, (b)
PS-2, (c) PS-3, and (d) PS-4 yttrium aluminum garnet phosphors.
Fig. 6. Plots of phosphor emission, excitation efficiency after
packaging and amount of phosphors for packaging (wt %) against
the particle sizes of (a) PS-1, (b) PS-2, (c) PS-3, and (d) PS-4
YAG:Ce phosphors.
300 350 400 450 500
PS-1
PS-2
PS-3
PS-4
Intensity (a.u.)
Wavelength (nm)
Fig. 3. Photoluminescent excitation spectra of (a) PS-1, (b) PS-2,
(c) PS-3, and (d) PS-4 yttrium aluminum garnet phosphors.
468 International Journal of Applied Ceramic Technology—Huang, et al. Vol. 6, No. 4, 2009
in white LEDs, all the phosphor samples were mixed
with silicone and uniformly applied over InGN chip.
The different amounts of PS-1, PS-2, PS-3, and PS-4
were used in order to get the similar efficiency as that of
commercially available YAG:Ce phosphor in white
LEDs. The comparison of I
v
(%) of white LEDs is
shown in Fig. 5. It was observed that all the samples
showed similar efficiencies in comparison with com-
mercially available YAG:Ce phosphors. The different
amounts of PS-1, PS-2, PS-3, and PS-4 phosphors are
plotted against the particle sizes of the phosphors as
shown in Fig. 6. Interestingly, the amount of phosphors
required to get similar I
v
efficiency depends on the par-
ticle sizes of the samples. The PS-4 sample required the
lowest amount, which is 3.7 wt% as against 5.7 wt% for
PS-1 sample. These results suggest that the reduction in
particle size will require minimum amount of phosphor
samples in comparison with bulk YAG:Ce phosphor
without loss of efficiency of white LEDs. Although the
emission efficiencies exhibited by smaller size particles
are lower (as in Fig. 6), their amount required for pack-
aging in white LEDs was found to be less. For white
LED applications, it was observed that YAG:Ce phos-
phors with smaller partic le size was better than fine-
powder phosphors with a larger particle size because
smaller particle size could reduce internal scattering
when they were coated onto a bare LED surface. In
brief, the studies on the effect of particle size on the
packaging performance to investigate the amounts of
phosphors required will be helpful in lowering the cost
of white LEDs. The YAG:Ce phosphors with smaller
particle size will be cost effective in terms of the reduc-
tion in the amount of phosphors utilized in white LEDs.
Conclusions
In summary, YAG:Ce phosphors exhibiting differ-
ent sizes having uniform distribution were prepared by
solid-state reaction followed by grinding and sieving
method. The effect of particle size distribution of YAG
phosphors on the PL properties was investigated, which
showed that the uniform size distribution and particle
size affects the packaging performance in white LED
applications. Interestingly, the amount of phosphors re-
quired to get similar I
v
efficiency showed dependence on
the particle sizes of the samples. The PS-4 sample hav-
ing smaller particles required the lowest amount, which
is 3.7 wt% without compromising the efficiency of
white LEDs. The YAG:Ce phosphors with smaller par-
ticle size will be cost effective in terms of the reduction
in the amount of phosphors utilized in white LEDs.
References
1. J.S Kim, P. E. Jeon, J. C. Choi, H. L. Park, S. I. Mho, and G. C. Kim,
‘‘Warm-white-Light Emitting Diode Utilizing a Single-phase Full-Color
Ba
3
MgSi
2
O
8
:Eu
21
,Mn
21
Phosphor,’’ Appl. Phys. Lett., 84 [15] 2931–
2933 (2004).
2. M. G. Craford, N. Holonyak, and F. A. Kish, ‘‘In Pursuit of the Ultimate
Lamp,’’ Sci. Am., 284 [2] 62–68 (2001).
3. T. Nishida, T. Ban, and N. Kobayashi, ‘‘High-Color-Rendering Light
Sources Consisting of a 350-nm Ultraviolet Light-Emitting Diode and
Three-Basal-Color Phosphors,’’ Appl. Phys. Lett., 82 [22] 3817–3819 (2003).
4. S. Nakamura and G. Fasol, The Blue Laser Diode: GaN Based Light Emitters
and Lasers, Springer-Verlag, Heidelberg, Germany, 1997.
5. Y. Shimizu, K. Sakano, Y. Noguchi, and T. Moriguchi, U.S. Patent no. 5,
998, 925.
6. J. M. Robertson and M. W. van Tol, ‘‘Epitaxially Grown Monocrystalline
Garnet Cathode-Ray Tube Phosphor Screens,’’ Appl. Phys. Lett., 37 [5] 471–
472 (1980).
7. Y. S. Lin and R. S. Liu, ‘‘Chemical Substitution Effects of Tb
31
in YAG:Ce
Phosphors and Enhancement of Their Emission Intensity Using Flux Com-
bination,’’ J. Lumin., 122–123 580–582 (2007).
8. A. Leleckaitea, A. Kareivaa, H. Bettentrupb, T. Ju
¨
stelb, and H.-J. Meyerc,
‘‘Sol–Gel Preparation and Characterization of Codoped Yttrium Aluminium
Garnet Powders,’’ Z. Anorg. Allg. Chem., 631 [21] 2987–2993 (2005).
9. D. Jia, Y. Wang, X. Guo, K. Li, Y. K. Zou, and W. Jiac, ‘‘Synthesis and
Characterization of YAG:Ce
31
LED Nanophosphors,’’ J. Electrochem. Soc.,
154 [1] J1–J4 (2007).
www.ceramics.org/ACTParticle Size Effect on the Packaging Performance of YAG:Ce Phosphors in White LEDs 469
