The sonochemical and microwave-assisted synthesis of nanosized YAG particles

Article abstract of DOI:10.1039/b507942j

Nanoparticles of YAG were prepared by sonochemistry and by microwave radiation. In both cases, the last stage of the preparation was annealing at a high temperature, which led to a highly aggregated product. To reduce the aggregation we have added κ-carrageenan (a sugar) to the reaction mixture. The effect of the sugar is presented and discussed herein. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2005.

Full text of DOI:10.1039/b507942j

View Article Online / Journal Homepage / Table of Contents for this issue
P A P E R
The sonochemical and microwave-assisted synthesis of nanosized
YAG particles
a
a
b
a
Yana Letichevsky, Lena Sominski, Jose Calderon Moreno and Aharon Gedanken*
a
Department of Chemistry and Kanbar Laboratory for Nanomaterials at the Bar-Ilan
University Center for Advanced Materials and Nanotechnology, Bar-Ilan University,
Ramat-Gan, 52900, Israel. E-mail: gedanken@mail.biu.ac.il; Fax: þ972-3-5351250;
Tel: þ972-3-5318315
b
`
Universitat Polite`cnica de Catalunya, Departament de FuˆUsica Aplicada, EPSC-PMTA-201,
Avendida del Canal Olı´mpic s/n, Castelldefels, 08860 Barcelona, Spain
Received (in Montpellier, France) 6th June 2005, Accepted 25th August 2005
First published as an Advance Article on the web 12th September 2005
Nanoparticles of YAG were prepared by sonochemistry and by microwave radiation. In both cases, the last
stage of the preparation was annealing at a high temperature, which led to a highly aggregated product. To
reduce the aggregation we have added k-carrageenan (a sugar) to the reaction mixture. The effect of the
sugar is presented and discussed herein.
7
complex route, metal–organic preceramic processing, copre-
8
Introduction
9
,10
cipitation methods
and the yttrium carboxylate–alumoxane
1
1
Ceramics based on the Y O –Al O formula are promising
2
materials for optical, electronic and structural applications.
The composition of 3Y O :5Al O , commonly called yttrium
route have been used to produce YAG phases. Most of these
methods suffer from being complex as well as time consuming
(long refluxing times, gelation periods of several days, etc.)
and/or there being mismatches in the solution behaviour of the
3
2
3
2
3
2
3
3 5
aluminium garnet (YAG, Y Al O12), exists in a cubic form
2
with a garnet structure. When YAG is doped with a transition
metal or a lanthanide element it becomes an important solid-
state laser material, widely used in luminescence systems, as
well as a window material for a variety of lamps. In view of its
constituents.
With the growing interest in nanotechnology, special syn-
thetic routes have been developed for the preparation of
nanoYAG particles. The nanoparticles were prepared by
1
2
1–3
high thermal stability and an extremely high creep resistance,
sol–gel methods, a citrate–nitrate gel combustion process
1
and a precipitation method.
2
YAG is a promising fiber material for the preparation of
ceramic composites. The electrical conductivity of YAG is also
reported to be lower than that of any other polycrystalline
oxide. Owing to there being such wide and diverse potential
applications for YAG-based ceramics, new routes for the
synthesis of pure and homogeneously-doped yttrium alumi-
In this paper we report on the application of ultrasound and
microwave radiation to the synthesis of nanoYAG particles.
Nanosized particles were obtained by a sonochemical reaction,
as well as by microwave dielectric heating. The products of the
sonochemical reaction varied in size, shape, structure and in
their solid phase (amorphous or crystalline), but they were
2
,3
nium garnet are highly desired.
There are three stable phases of the Y
first has the composition Y Al (YAM- monoclinic), the
second YAlO (YAP-orthorhombic or hexagonal structure)
and the third Y
synthetic methods for the synthesis of YAG nanoparticles
nanoYAG), the prepared product is amorphous. The crystal-
1
3
2
O
3
–Al
2
O
3
system. The
always of nanometre size. The microwave reaction was
conducted in a regular domestic microwave oven and did not
require special hydrothermal equipment. In most experiments
the microwave reaction was complete within a few minutes,
4
2 9
O
3
1
,4–6
3
Al
5
O
12 (YAG).
In all of the various
1
4,15
and within 1 h at the most.
(
lization process to form a YAG phase starts at 800 1C, and it
completely transforms into YAG below 900 1C. A certain
amount of the YAM phase co-exists with the YAG phase
between 850 1C and 900 1C, but above 900 1C, only the YAG
Experimental
All the reagents were of the highest commercially-available
purity. Yttrium oxide, aluminium nitrate nonahydrate and k-
carrageenan were purchased from Aldrich and used without
further purification. The X-ray diffraction patterns of the
products were recorded with a Bruker AXS D8 Advance
4
phase exists.
The final product of calcination and its behavior is greatly
influenced by the nature of the starting precursor powder.
Thus, it is important to understand the conversion mechanism
of precursor into crystalline YAG in order to control the
˚
Powder X-ray Diffractometer (using Cu-Ka l B 1.5418 A
radiation). Peak fitting and lattice parameter refinement were
computed using the Topas and Metric programs (Bruker
Analytical X-Ray Systems). The morphologies and micro- or
nanostructures of the synthesized products were further char-
acterized with a JSM-840 SEM. EDAX measurements were
performed on an X-ray microanalyzer (Oxford Scientific) built
on a JEOL JSM-840 SEM. High resolution SEM (HR-SEM)
micrographs were obtained using a Field Emission Hitachi
S-2300. The transmission electron micrographs (TEM) were
1
nature of the final product.
The solid-state synthesis of YAG ceramic from Al
2 3
O and
Y O powders usually requires extensive mechanical mixing
2
3
4
and lengthy heat treatments above 1600 1C. These processing
conditions do not allow facile control over micro-structure,
grain size and grain size distribution in the resulting powders or
shapes. Several wet chemical techniques were reported for the
synthesis of YAG, and methods such as the polymerized
T h i s j o u r n a l i s & T h e R o y a l S o c i e t y o f C h e m i s t r y a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 5
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 4 4 5 – 1 4 4 9
1445

View Article Online
imaged on a JEOL-JEM 100SX microscope using a 100 kV
accelerating voltage. Samples for TEM were prepared by
placing a drop of the sample suspension on a copper grid
The microwave (MW) synthesis of nanoYAG. The composi-
tion of the irradiated solution was identical to that of the
sonochemical solution, except that NaOH replaced ammonia
in creating the basic conditions. The sample was irradiated
under ambient conditions for 10 min. The temperature at the
end of the reaction was not measured. However, it is assumed
that it was about 10 1C above the boiling point due to the
superheating phenomenon.
(
400 mesh, Electron Microscopy Sciences) coated with carbon
film. The grid was then air dried. High resolution TEM
HRTEM) micrographs were taken using a JEOL 3010 with
(
a 300 kV accelerating voltage. A conventional monochrome
CCD camera with a resolution of 768 ꢀ 512 pixels was used to
digitize the images. The digital images were processed with the
Digital Micrograph software package (Gatan, Inc., Pleasan-
ton, CA, USA). FTIR spectra were recorded in KBr pellets
using an Impact 410 Nicolet spectrometer. DSC was carried
out on a Mettler Toledo TC 15 using nitrogen or argon as a
Results and discussion
The colour of the prepared and annealed materials was white.
The annealed sample underwent a series of standard charac-
terization measurements to determine its composition and
morphology. The results will be presented for both the sono-
chemical and MW products.
ꢁ
1
purging gas at a scanning rate of 5 1C min . TGA was
performed on a Mettler Toledo TGA/SDTA851 instrument
using nitrogen or argon as a purging gas, also at a scanning
ꢁ1
rate of 5 1C min
Specific surface areas were measured by the BET method at
7 K using N gas as an absorbent, after heating the sample at
20 1C for 1 h.
The C, H, N and S analyzer was made by the Thermo
.
Powder XRD studies
7
1
2
The X-ray diffraction patterns of the samples annealed at
different temperatures are shown in Fig. 1. These samples
resulted from the sonochemical reaction. Fig. 1 exhibits the
results of calcination for 1 h at 800 1C and 900 1C. Fig. 1a
shows that the ash annealed at 800 1C is amorphous, while the
sample calcined at 900 1C has a crystalline YAG phase (Fig.
Electron Corporation (Model Flash EA, 1112).
Ultrasonic irradiation was provided with a high intensity
ultrasonic probe (Misonix XL Sonifier, 1.13 cm diameter Ti
ꢁ2
horn, 20 kHz, 60 W cm measured calorimerically at 60%
efficiency) immersed directly into the reaction solution. The
microwave-assisted reaction was carried out in a Spectra
1
b). The sizes of the nanocrystals (at 900 1C) were calculated by
14
using the Debye–Scherrer (DS) equation and were found to
be in the range 17–20 nm.
9
00 W microwave oven, with a 2.45 GHz working frequency.
As explained above, for the MW reaction, two bases for
establishing the pH of 7–8 have been examined, i.e., ammonia
and NaOH. The results of the powder XRD analysis for the
MW annealed samples are shown in Fig. 2. The basic condi-
tions were formed by the use of NaOH. Fig. 2 shows only the
diffraction patterns for the sample annealed at 900 1C. The
reason why the 1000 1C XRD pattern is not presented is
because that, at this temperature, it is identical to that depicted
in Fig. 1b; in other words, only the YAG phase is detected.
EDAX measurements showed that there are no sodium atoms
in the final product. On the other hand, the sample annealed at
The oven was modified to include a refluxing system. In all the
experiments the microwave oven was cycled as follows: on for
2
1 s, off for 9 s, with the total power always at 900 W. This
cycling mode was chosen in order to reduce the risk of super-
heating the solvent. All reactions were conducted under air.
Synthesis
Procedure for sonochemical reaction. The two reactants,
Al(NO
3
)
3
ꢂ 9H
2 3 3
O and Y(NO ) , were mixed to form the starting
aqueous solution. Y(NO ) was prepared by dissolving 0.847 g
3
3
9
00 1C reveals two phases, YAG and YAP. Raising the
2 3
of Y O (0.0075 mol yttrium) in a small amount of a concen-
calcination temperature to 1000 1C (and annealing for 1 h)
resulted in a pure phase of YAG. The sizes of the 1000 1C
calcined nanocrystals were calculated to be in the range 19 ꢃ 2
nm. Thus, the comparison between the sonochemical and the
MW product, based on the XRD results, shows that sonication
is advantageous, already yielding the YAG phase at 900 1C,
while at that temperature, YAG and YAP phases are observed
for the MW product. Since the calculated particle sizes are
almost equal for the two products, we will attribute the
difference between them to the MW method yielding more
defects and pores than the sonochemical technique. Thus, in
the conversion to the stable YAG phase, a higher temperature
is required.
trated HNO solution. After the dissolution of the oxide, the
3
solution was evaporated, leaving only the solid nitrate. This
solid was introduced into the mixed solution of the Al and Y
nitrates. A typical precursor solution was prepared by mixing
4
.68 g of Al(NO ) ꢂ 9H O (0.0125 mol) and 0.847 g of Y(NO )
3
3
2
3 3
ꢁ3
(
0.0075 mol) in 200 cm of water. Thus, the molar ratio of
13
13
Al : Y ¼ 0.0625 : 0.0375 ¼ 1.67 : 1. An aqueous ammonia
solution was added to the sonication cell prior to the irradiation,
keeping the pH in the 7–8 range. The solution was sonicated for
1
h without cooling the irradiated solution. The temperature at
the end of the sonication reached 80 1C. The precipitate was
separated from the solution by centrifugation and was washed
with ethanol and water three times. The product was dried
under vacuum at rt. The heating rate for the conversion of the
ꢁ
1
.
prepared material into the crystalline phase was 10 1C min
Sonication in the presence of j-carrageenan (a sugar). The
highly aggregated form of the nanoYAG obtained in our
experiments led to an attempt to de-aggregate the annealed
product. For this purpose we added 0.22 g of k-carrageenan to
the above-mentioned sonicated mixture of Al(NO
Y(NO . The sugar was added prior to the addition of
ammonia. After its addition, the solution was heated to
0 1C for about 1 h to allow its dissolution, and only then
3 3
) and
3 3
)
5
was the ammonia added, keeping the pH between 7 and 8. The
sonication time and conditions were identical to the experi-
ments without k-carrageenan. Since the k-carrageenan is sen-
sitive to light, the sonication cell was wrapped in aluminium
foil.
Fig. 1 The XRD patterns of the sonochemically-prepared powder
calcined at (a) 800 1C and (b) 900 1C, for 1 h.
1
446
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 4 4 5 – 1 4 4 9

View Article Online
Fig. 2 The XRD patterns of the MW-prepared nanoYAG calcined at
900 1C. The basic conditions were established by the use of NaOH.
The MW reaction was also conducted using ammonia for
determining the pH. The XRD patterns of the product an-
nealed at 900 1C (Fig. 3) indicate that an amorphous material is
obtained. The failure to obtain a YAG phase led us to replace
the ammonia with NaOH. We can interpret these results in the
following way: it is possible that ammonia easily evaporates
upon MW irradiation and therefore the pH varies accordingly.
Another possible explanation attributes this behavior to
NaOH being a stronger electrolyte and better at absorbing
the MW radiation. Thus, the amorphous product obtained is
composed of smaller particles and its crystallization to the
stable phase can be carried out at lower temperatures. Ammo-
nia, on the other hand, is a weaker electrolyte that does not
absorb the radiation as well, and thus fewer hot spots are
formed in the liquid and a lower temperature develops. The
crystallization temperature will therefore be higher than with
NaOH.
In addition to the sonochemical and MW reactions, two
control experiments were also conducted. In the first, a room
temperature reaction was undertaken using the same concen-
trations of reactants as in the sonochemical and MW reactions,
but without applying ultrasound or MW radiation. The second
control experiment was carried out sonochemically and with
MW at 100 1C. The two reactions were performed for 1 h. In
both cases the annealed sample contained an impurity phase.
Diffraction peaks at 2y ¼ 25.25, 26.8 and 44.81 were detected.
Fig. 4 FTIR spectrum of the sonochemically-prepared nanoYAG, (a)
the prepared material, (b) sample after sintering.
stretching vibration of the water. The peaks in the 1100–1700
ꢁ1
cm region can be assigned to the vibrations of the nitrate
group. Fig. 4b depicts the spectrum of the heat-treated sample.
It shows only metal–oxygen vibrations which, in accordance
with earlier reports, are characteristic of Y–O and Al–O
2
ꢁ1
ꢁ1
stretching frequencies. The 3500 cm and 1100–1700 cm
vibrational peaks have disappeared due to the removal of
water and nitrate moieties. In short, the comparison of the
current FTIR spectra with those obtained by other techniques
shows that the sonochemically-obtained product has an iden-
tical IR spectrum to the previous data.
To learn more about the content and purity of the prepared
material, TGA measurements were carried out. The TGA data
indicated that a weight loss of 39% happens between room
temperature and 900 1C. The major part of the weight loss
occurs in the temperature range of 100–500 1C; this mass loss
FTIR studies of the sonochemical reaction
2
being due to the evolution of the water, NO and NO . The
latter gases are evolved due to the decomposition of nitrate
4
ions present in the ash.
Fig. 4a shows the FTIR spectrum of the sonochemically-
prepared product. The spectrum illustrates a broad absorption
ꢁ
1
band around 3500 cm , which is assigned to the O–H
TEM and HRTEM results of the sonochemical reaction
The TEM image of the sonochemical nanoYAG sample heated
at 900 1C is illustrated in Fig. 5. The picture shows an
aggregated network of small nanocrystals. The size of the
nanocrystallites varies from 20–25 nm and is in good agree-
ment with the XRD results.
A comparison between the two pictures (Fig. 5 and Fig. 6)
reveals that, in both cases, almost identically-sized particles
form a network of aggregated particles. The nanometric char-
acter of the products is clear in both syntheses, but the
necessity to anneal the product leads to the formation of
agglomerates.
The HRTEM micrograph of sonochemically-synthesized
nanoYAG crystals is depicted in Fig. 7. The HRTEM lattice
micrograph shows nanoYAG particles which have different
crystallographic orientations. The lattice fringes are distinct
and the distance between adjacent fringes in the particle’s core
is 0.384 nm, which correlates well with the (321) planes of the
body-centered cubic YAG phase.
Fig. 3 The XRD patterns of the MW-prepared powder assisted by the
use of ammonia and calcined at 900 1C.
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 4 4 5 – 1 4 4 9
1447

View Article Online
Fig. 5 TEM image of the nanoYAG product of the sonochemical
process after annealing at 900 1C.
SEM measurements of the sonochemical products
The SEM micrographs of the nanoYAG crystals obtained
sonochemically and calcined at 900 1C are shown in Fig. 8.
Fig. 8a shows the high degree of aggregation that the na-
noYAG crystallites undergo upon calcination. Fig. 8b how-
ever, shows that the addition of k-carrageenan helps in
breaking up the agglomerates. The reason for the use of
Fig. 7 HRTEM image of nanoYAG. The scale of the bar is 5 nm.
process and the partial sintering of the particles are not
completed, and a nanometric nature is maintained by the
sugar-treated particles.
3
1
31
k-carrageenan is to separate the Al ions from the Y ions
already in the solution, so that when the sugar is adsorbed into
the nanoYAG particles, it prevents their aggregation. The
sugar is subsequently burnt off from the nanoYAG during
the calcination process. A C, H, N and S analysis of the
annealed product revealed 0.023% of carbon and 0.065% of
hydrogen. These results show that the sugar is completely
removed in the heating process. The particle sizes measured
for the products obtained after the calcination of the sugar-
added product are 22 ꢃ 3 nm.
BET measurements
The nanosize nature of the products obtained by MW and
sonochemical syntheses was also examined by surface area
measurements. There is a big difference in the measured surface
area of products obtained by the two procedures. The BET
To learn more about the nature of the products, and to show
the difference in the aggregation of the nanoYAG and the
nanoYAG resulting from the sugar treatment, high magnifica-
tion SEM measurements were performed. Both of the micro-
graphs in Fig. 9 show that the particles were sintered by the
calcination process. It is also seen from both pictures that
rounded particles (close to spherical) are glued to each other.
The size of the nanoYAG particles in Fig. 9a is 1 mm, while the
size of the sugar-treated nanoYAG particles in Fig. 9b is
5
2
0 nm. Since the TEM and the XRD results yielded only a
5 nm particle size, we propose that the SEM is observing an
agglomerate of two particles. This shows that the annealing
Fig. 6 TEM image of the nanoYAG product obtained by a MW
process annealed at 1000 1C.
Fig. 8 SEM images of (a) nanoYAG and (b) nanoYAG þ k-carra-
geenan. Scale bars ¼ 100 mm. Samples calcined at 900 1C.
1
448
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 4 4 5 – 1 4 4 9

View Article Online
Conclusions
This paper reports two new techniques for the preparation of
nanoYAG particles, i.e., sonochemical and microwave dielec-
tric heating. The advantages of the described processes are
their simplicity, their short reaction times and the cheapness of
the preparation of the YAG structure. The particles were
prepared in nanosizes, and identical particle sizes were mea-
sured for the annealed nanoYAG products obtained by both
methods. In addition to our attempts, we also repeated the
experiment reported in ref. 12. The annealed product was not
pure and also contained the YAP phase. Furthermore, the
measured surface area was much smaller than that originally
1
8
reported. According to Wang, the secret lies in the volume of
the 5 M ammonia solution used, which has to be large, and is
not given in ref. 12. The control experiments, in which sono-
chemical and MW techniques were not used, led also to the
formation of nanoYAG particles, however, they were accom-
panied by impurities. The absence of impurity phases in the
sonochemical and MW experiments might be due to the
creation of hot spots in both cases. Thus, although the
temperature in the solvent is moderate (100 1C), higher tem-
peratures are created locally in the solvent. Finally, neither the
sonochemical nor the MW techniques are major players in the
precipitation of the prepared material, which occurs without
their application. We attribute the reduction in particle size,
compared with the unassisted reaction, to these two methods.
Thus, upon annealing, a pure product is obtained, while when
the unassisted product is being heated, the YAG contains an
impurity phase.
Fig. 9 SEM images of (a) nanoYAG (sonochemical), scale bar ¼ 5
mm and (b) nanoYAG (sonochemical) treated with k-carrageenan,
scale bar ¼ 375 nm. Both samples were calcined at 900 1C.
Acknowledgements
We thank Mrs. Riki Harpeness for helping us in the course of
this research. A. G. thanks the Yeshaya Horowitz Association
for its financial assistance.
surface area of the microwave product after calcination at
2
ꢁ1
1
was increased from 10 min to 30 min, an increase in the surface
000 1C was 24 ꢃ 2 m g . When the microwave reaction time
References
1
2
3
4
X. Guo, P. S. Devi, B. G. Ravi, J. B. Parise, S. Sampath and J. C.
Hanson, J. Mater. Chem., 2004, 14, 1288.
M. Veith, S. Mathur, A. Kareiva, M. Jilavi, M. Zimmer and
V. Huch, J. Mater. Chem., 1999, 9, 3069.
B. J. Chung, J. Y. Park and S. M. Sim, J. Ceram. Proc. Res., 2003,
2
ꢁ1
area to 42 ꢃ 4 m g was measured. It should be noted that the
particle size for the annealed products of the 10 and 30 min
reaction times were the same (within experimental error). It
1
6
seems that a digestive ripening process led to repeated cycles
of dissolution and precipitation that did not dramatically
change the particle size but contributed to the porous nature
of the product. Thus, it is concluded that the annealing process
did not change considerably the porous nature of the prepared
materials. It is suggested that 10 min is sufficient to obtain the
product. In the next 20 min, the reflux that the solution
undergoes leads to a digestive ripening process that is reflected
in a higher surface area. Although the exact mechanism is not
known, it has been suggested that the digestive ripening
involves the dissolution and reprecipitation of the particles.
4
(3), 145.
S. Roy, L. Wang, W. Sigmund and F. Aldinger, Mater. Lett.,
999, 39, 138.
1
5
6
J. W. Nowok and J. P. Kay, J. Mater. Sci. Lett., 2001, 20, 1331.
M. B. Kakade, S. Ramanathan and P. V. Ravindran, J. Alloys
Compd., 2003, 350, 123.
7
8
9
B. H. King and J. W. Halloran, J. Am. Ceram. Soc., 1995, 78,
141.
Y. Liu, Z.-F. Zhang, B. King, J. Halloran and R. M. Laine, J. Am.
Ceram. Soc., 1996, 79, 385.
A. M. George, N. C. Mishra, M. S. Nagar and N. C. Jayadevan,
J. Therm. Anal., 1996, 47, 1701.
2
1
7
In fact, in a recent MW reaction, Gao et al. reduced silver
ions in a MW hydrothermal reaction for 3 h. The reduction of
silver ions in ethylene glycol usually takes 1–5 min. For Stoeva
10 M. Yada, M. Ohya, M. Machida and T. Kijima, Chem. Commun.,
998, 1941.
1
1
1
C. J. Harlan, A. Kareiva, D. B. MacQueen, R. Cook and A. R.
Barron, Adv. Mater., 1997, 9, 68.
H. Wang, L. Gao and K. Niihara, Mater. Sci. Eng. A, 2000, 288,
1
6
et al. , leaving the reaction to reflux for a long time yielded
highly organized nanostructures with a narrow size distribu-
tion that also resulted in digestive ripening. On the other hand,
the surface area of the sonochemical product after calcination
1
2
1
.
13 A. Gedanken, Ultrason. Sonochem., 2004, 11, 47.
14 R. Harpeness, A. Gedanken, A. M. Weiss and M. A. Slifkin,
J. Mater. Chem., 2003, 13, 2603.
15 R. Harpeness and A. Gedanken, Langmuir, 2004, 20, 3431.
16 S. Stoeva, K. J. Klabunde, C. M. Sorensen and L. Dragieva,
2
ꢁ1
at 9001 is very small, only 5 m g , while the nanoYAG with
k-carrageenan sample, after calcination, reaches a value of
2
1 m g . This is in accordance with the SEM pictures,
ꢁ1
2
demonstrating much smaller particles for the nanoYAG–
J. Am. Chem. Soc., 2002, 124(10), 2305.
F. Gao, Q. Lu and S. Komarneni, Chem. Mater., 2005, 17, 856.
1
18 Hongzhi Wang, private communication.
7
k-carrageenan sample.
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 4 4 5 – 1 4 4 9
1449