Thermal metallization of silver stearate-coated nanoparticles owing to the destruction of the shell structure

Article abstract of DOI:10.1021/jp0479724

Nanoparticles composed of a silver stearate shell and a silver core became metallic bulk silver by heating at a temperature (340?°C) lower than the melting point of silver metal (960?°C). The metallization was investigated by thermogravimetric analysis, difference thermogravimetric analysis, FT-IR and UV - vis spectroscopy, electrochemical techniques, scanning electron microscopy, and measurements of the density of the nanoparticles. The nanoparticles decreased in mass by 20%, which was equivalent to the amount of the stearate shell. The low-temperature metallization resulted from the thermal decomposition of the shell of silver stearate into silver species. It involved two steps. The first, occurring at 250?°C, was the decomposition to porous silver particles through the first-order reaction with an activation energy of 111 kJ mol ^-1. The second, occurring at 340?°C, was a structure change from porous particles to silver bulk crystals.

Full text of DOI:10.1021/jp0479724

J. Phys. Chem. B 2004, 108, 15027-15032
15027
Thermal Metallization of Silver Stearate-Coated Nanoparticles Owing to the Destruction of
the Shell Structure
Nianjun Yang, Koichi Aoki,* and Hiroshi Nagasawa†
Department of Fiber Amenity Engineering, UniVersity of Fukui, 3-9-1 Bunkyo, Fukui-shi, 910-8507 Japan
ReceiVed: May 12, 2004; In Final Form: July 19, 2004
Nanoparticles composed of a silver stearate shell and a silver core became metallic bulk silver by heating at
a temperature (340 °C) lower than the melting point of silver metal (960 °C). The metallization was investigated
by thermogravimetric analysis, difference thermogravimetric analysis, FT-IR and UV-vis spectroscopy,
electrochemical techniques, scanning electron microscopy, and measurements of the density of the nanoparticles.
The nanoparticles decreased in mass by 20%, which was equivalent to the amount of the stearate shell. The
low-temperature metallization resulted from the thermal decomposition of the shell of silver stearate into
silver species. It involved two steps. The first, occurring at 250 °C, was the decomposition to porous silver
-
1
particles through the first-order reaction with an activation energy of 111 kJ mol . The second, occurring at
40 °C, was a structure change from porous particles to silver bulk crystals.
3
Introduction
improved11 with the aim of agreement with the experimental
12
13
results. Kofaman et al. and Sakai presented surface premelt-
ing models in which the particle began to melt from the surface
atoms and then abruptly fused over the inside. On the basis of
the Landau theory, they explained the nonlinear relationship
between the melting point and the inverse of the radius of par-
1
2
Heating processes of mono- and bimetallic nanoparticles
have recently been of great interest because they have the
potential to exhibit fascinating properties different from those
of bulk metals. The well-known observation of thermal
properties of metal nanoparticles is a decrease in the melting
point with a decrease in size. For example, Takagi found as
early as 1954 by the use of transmission electron microscopy
that the melting point of gold nanoparticles decreased by 300
3
1
4
ticles as a surface curvature effect. Beck et al. and Honeycutt
1
4
1
5
et al. pointed out that an assembly of small particles in the
melting region is a mixture of solidlike and liquidlike forms. A
predominant cause of the low melting points is reportedly due
to not only the thermodynamic change in the melting temper-
ature but also the enhancement of the latent heat of fusion.16
Most theories on the lower shift of melting points have been
based on the surface energy of nanoparticles caused by curvature
effects combined with interfacial phase effects. Because nano-
particles are dispersed against aggregation with the help of
surfactants, it is important to take into account chemical effects
of surfactants (e.g., desorption of surfactants from the surface
of nanoparticles or decomposition of surfactants). We have
5
°
C from the bulk melting point. Buffat and Borel observed the
dependence of the melting point of gold nanoclusters in a holey
carbon grid on particle size by means of selected-area diffraction.
6
Andres et al. found a decrease in the melting point of gold
nanoparticles by as much as 40% of that of the bulk. Schmidt
7
et al. reported that the melting point of Na139 nanoclusters 2.2
nm in diameter was 28% lower than that of bulk sodium. A
melting-point depression of 31% was found by Allen et al. for
islands of Sn nanoparticles 10 nm in diameter using scanning
8
nanocalorimetric analysis. However, an inverse phenomenon
17
was reported for tin clusters9 because their structure was
observed the significance of surfactants in silver nanoparticles
dispersed in organic solvent through electrochemical measure-
different from that of the bulk. Gold nanoparticles smaller than
17
6
ments. The silver stearate-coated nanoparticle is composed of
a core of silver atoms and a shell of silver stearate. The shell
was reduced to silver metal at -0.6 V, a more positive potential
than that used for the silver stearate molecule (at -1.5 V). The
positive potential shift indicated that the silver stearate of the
nanoparticle has higher energy than the silver stearate molecule.
Consequently, the nanoparticle is expected to fuse at a temper-
ature lower than the melting point of bulk silver. This expecta-
tion stimulates us to heat the silver nanoparticle to find the role
of the shell of silver stearate. In this paper, we report heated
states of silver nanoparticles by means of UV-vis and FT-IR
spectroscopy, scanning electron microscopy, thermogravimetric
analysis, difference thermogravimetric analysis, and electro-
chemical techniques.
2
.0 nm were reported not to vary their melting points.
Thermodynamic models1
0-15
have been proposed to account
for the dependence of the melting point on the size of the
nanoparticles. It has been commonly recognized that the surface
atoms have quite different properties from the inside atoms. The
decrease in the melting point is due to the large surface energy
10
coming from the large specific surface area. In 1909, Pawlow
was the first to derive the quantitative relationship between
the melting point and the particle size on the basis of the
classical Kelvin equation. Pawlow predicted that the melting
point had a linear relation with respect to the inverse of the
diameter of the particle. Buffat and Borel modified Pawlow’s
equation to predict the nonlinear decrease in the melting point
with the inverse of the diameter. Theory has been gradually
4
Experimental Section
*
Corresponding author. E-mail: d930099@icpc00.icpc.fukui-u.ac.jp.
Tel: +81-776-278665. Fax: +81-776-278494.
†
All of the chemicals were of analytical grade. The silver
stearate nanoparticles were synthesized with the method reported
Current address: Development Department 2, Development Center 2,
Ebara Corp.
1
0.1021/jp0479724 CCC: $27.50 © 2004 American Chemical Society
Published on Web 09/04/2004

15028 J. Phys. Chem. B, Vol. 108, No. 39, 2004
Yang et al.
Figure 2. FT-IR spectra of (a) stearate acid, (b) silver stearate, (c) the
nanoparticle, and (d) the nanoparticle heated to 340 °C for 5 min in
the air. These samples were mounted in KBr pellets.
Figure 1. Photographs of (a) the unheated nanoparticle and (b) the
nanoparticle heated to 340 °C for 5 min in the air. They were obtained
by an optical microscope.
shell, the color change may be due to not only metallization
but also the thermal decomposition of silver stearate.
elsewhere.18 Their average diameter (5 nm) and the size distri-
bution (from 1 to 9 nm) were the same as those reported pre-
To find participation in the thermal decomposition of silver
stearate in the nanoparticle, FT-IR spectra of stearic acid, the
silver stearate molecule, the nanoparticle, and the heated
nanoparticle were obtained and are shown as curves (a)-(d) in
1
7,18
viously.
The crystal-like nanoparticles were dissolved in
cyclohexane. The solution was dispersed on a glass plate or an
indium-tin oxide (ITO)-coated glass plate and dried to form
films.
-1
Figure 2, respectively. The band at 1750 cm that is attributed
UV-vis spectroscopy of the nanoparticle film was carried
out at room temperature on a UV-570 spectrometer (JASCO,
Tokyo). FT-IR spectra were recorded with 2-cm-1 resolution
on a Protege 460 FT-IR instrument (Nicolet, Japan) at room
temperature. A specimen for the analysis was prepared as a KBr
pellet of dry samples.
to the functional group of -COOH was observed only in curve
(a). A series of bands in curves (a) and (c) in the domain of
-1
2
850-2970 cm , which are attributable to the stretching of
the CH2 groups and the terminal CH3 group, were not observed
in curve (d) for the heated nanoparticle. The bands in the region
-1
-
of 1550-1620 cm for COO in curves (a)-(c) were not found
Scanning electron microscopy (SEM) was performed with
an SEM&EDX integration system (Hitachi). The samples were
prepared directly on a copper substrate coated with carbon film.
Electrochemical experiments were made with a model 1112
potentio/galvanostat (Huso, Kawasaki) controlled by a computer.
The reference and the counter electrodes were Ag|AgxO and
platinum wire, respectively. The potential difference between
Ag|AgxO and Ag|AgCl in 3 M NaCl was -0.055 V. The
working electrode was a platinum disk electrode or an ITO plate.
Thermogravimetric analysis (TGA) and difference thermo-
gravimetric analysis (DTA) were simultaneously applied to the
dried powder of the nanoparticle (8-12 mg) with a computer-
controlled DTG-60 apparatus (Shimazu, Kyoto). Temperature
was scanned from room temperature to 500 °C at some heating
in curve (d) either. Therefore, the increase in temperature up to
3
40 °C removes the stearate moiety from the nanoparticle. A
plausible reaction mechanism is the thermal decomposition of
silver stearate to silver metal and gases in the shell:
AgOOC(CH ) CH f Ag + gas
(1)
2
16
3
The final products of the gas should be CO2 and H2O because
heating was done in air. Because the formation free energy of
silver oxide is positive at temperatures greater than 200 °C,19
silver metal is more stable than its oxide.
Reaction 1 suggests either the case in which all of the shell
disappears in the gas or the shell is left as Ag metal. To obtain
a reasonable reaction mechanism, we weighed the nanoparticles
before and after heating. The nanoparticle decreased in mass
by 20%, for example, from 15.6 to 12.5 mg, when it was heated
to 340 °C and cooled to room temperature. We assume that the
loss of mass is ascribed only to the gasification (the second
case). Letting the number of silver stearate molecules in the
shell be nAg18, that of the silver atoms in the core be nAg, the
-
1
rates (2, 5, 10, 20, 50 °C min ). Temperature was monitored
with a thermocouple located near the chamber under a nitrogen
atmosphere.
The density of aggregation of heated nanoparticles was
3
evaluated from the mass and the excluded volume in a 10-cm
water vessel. The weighed aliquot (ca. 10 mg) of the nanopar-
ticle was sunk in water, and the increment of the meniscus was
read.
-
1
molar mass of silver stearate be MAg18 () 391 g mol ), and
-1
that of silver be MAg () 108 g mol ), we can express the mass
of the nanoparticle before and after heating as (nAg18 MAg18 +
nAg MAg) and (nAg18 + nAg) MAg, respectively. These expressions
correspond to 15.6 and 12.5 mg. Taking the ratio, we obtain
the molar fraction of silver stearate, nAg18/(nAg18+ nAg), to be
0.095. The resulting ratio, nAg/nAg18 () 9.5), is close to the
Results and Discussion
Decomposition of the Nanoparticles. The crystal-like nano-
particles heated to 340 °C for 10 minutes were aggregated into
a lump. When the nanoparticle powder was dispersed on the
ITO electrode by dropping nanoparticle-included cyclohex-
ane solution and was heated to 340 °C, it became a thin film
on the ITO. Figure 1 shows photographs of nanoparticles (a)
before and (b) after heating, obtained by an optical microscope.
The film before heating showed a rough, black surface. The
film after heating to 340 °C for several minutes in air showed
a smooth and brilliant surface (Figure 1b), as if it were a
flattened droplet like fused solder. The film cooled to room
temperature exhibited a white surface. An increase in the heating
temperature reduced the time of the color change from black to
brilliant. Because the nanoparticle involves silver stearate as a
1
7
previously reported results (9.6). Consequently, we can
demonstrate quantitatively that the shell is decomposed into
silver metal that is left on the silver core when the nanoparticles
are heated to temperatures greater than 340 °C.
Figure 3 shows the cyclic voltammogram for the heated film
in aqueous solutions (curve (a)). A clear anodic peak at 0.15 V
appeared only in the first scan when the scan started at a
potential less than 0.0 V. Thus, the anodic reaction is irrevers-
ible. The peak current increased with an increase in the number
of dispersed nanoparticles on the ITO. As a control experiment,
voltammetry of an aqueous AgNO3 solution was carried out at
+
a platinum electrode, and its wave is shown as curve (c). Ag

Thermal Metallization of Coated Nanoparticles
J. Phys. Chem. B, Vol. 108, No. 39, 2004 15029
Figure 5. TGA and DAT curves for the nanoparticle from 25 to 500
Figure 3. Cyclic voltammograms of (a) the nanoparticle film heated
to 340 °C on the ITO electrode, (b) the unheated nanoparticle film, (c)
-
1
°
C at a heating rate of 10 °C min under a nitrogen atmosphere.
0
.01 M silver nitrate at the Pt electrode, and (d) the silver wire electrode
participation in the shell of silver stearate more accurately for
in a 0.1 M KNO solution. They were obtained at the scan rate of 0.1
V s . The current scales in curves (c) and (d) are /
original scales, respectively.
3
1
7
-
1
1
1
the previous observation that the core and the shell are
composed of silver and silver stearate, respectively. The charge
and the mass of the nanoparticle are expressed by the participa-
tion in the core and the shell:
8
and /100 of the
q ) F(nAg + nAg18
)
(2)
(3)
m ) n M + nAg18MAg18
Ag Ag
q
Then /m is given by
F(n /n
+ 1)
q
Ag Ag18
)
m
(4)
(n /nAg18) MAg + MAg18
Ag
Figure 4. Dependence of the charge for the nanoparticle film heated
to over 340 °C obtained from the anodic wave at 0.15 V on the mass
of the nanoparticle film.
By using nAg/nAg18 ) 9.5 obtained from the mass change by
-
1
heating, the value on the right-hand side in eq 4 is 716 C g .
This value is smaller than the slope, q/m, by only 5%, as shown
in Figure 4. This is because of the involvement of not only the
capacitive component in q but also insufficiently positive
potential to complete the dissolution of all of the silver metal,
as was shown in difference curves (a) and (d) in Figure 3. All
of these observations demonstrate that the shell of the nano-
particle is decomposed into pure silver when the heating
temperature is greater than 340 °C.
begins to be deposited for E < -0.2 V at the negative scan
and continues to be deposited for E < -0.03 V at the positive
scan. The deposited silver is dissolved at 0.1 V. This dissolution
wave is similar to the anodic wave in curve (a). Consequently,
the heated products should be mainly silver metal, as predicted.
We found that the shape and the potential of the anodic wave
-1
were independent of the heating rate (2-50 °C min ) and the
heating time (30-1800 s). The anodic wave of the heated
nanoparticle was broadened and more shifted by 80 mV than
the peak of curve (a). The broadness and the shift suggest the
involvement of impure silver compounds owing to the decom-
position of silver stearate or the difference in crystalline structure
of silver metal. The FT-IR spectrum (curve (d) in Figure 2)
shows no evidence of any organic impurity. A silver wire, whose
crystalline structure may be different from that of the electro-
chemically deposited silver and the heated nanoparticle, showed
a broad anodic dissolution wave (curve (d) of Figure 3).
Therefore, the broadness and the shift in curve (a) are ascribed
to the difference in the crystalline structure. The unheated film
exhibited no voltammetric wave (curve (b)). This is consistent
with the voltammograms of the nanoparticle-dissolved solu-
tion.17 The absence of any wave is due to the protection of the
Nonisothermal Kinetics of the Decomposition. To inves-
tigate the kinetics of the thermal decomposition of the shell,
thermogravimetric analysis (TGA) and differential thermogravi-
metric analysis (DTA) were performed simultaneously, and their
curves are shown in Figure 5. The TGA curve showed that the
mass started to decrease at 230°C and reached the other state
passing through a transition at 330 °C. The ratio of the mass
decrease was 20% when the temperature increased to 500 °C.
This ratio was consistent with the loss of mass by means of the
long-time heating at 340 °C described in the previous section.
The DTA curve showed two exothermic peaks at 306 and 380
°
C. The former peak may be attributed to the decomposition of
the shell, according to reaction 1. The latter peak may be caused
by a phase transition of grain silver metal into silver bulk
crystals, as will be mentioned from UV spectra and the SEM
images in a later section.
Figure 6 shows the TGA curves at several heating rates. The
transition temperature increased with an increase in the heating
rate. The temperature at the peak of the DTA curves (not shown)
also shifted toward higher temperature as the heating rate
increased. The shift suggests a delay in the decomposition due
to slow kinetics.
17
redox reaction by the silver stearate shell.
To confirm the heat-generated species on the ITO surface,
we obtained a relation between the electrochemical charge, q,
of the heated nanoparticle and the dry mass, m, of the mounted
nanoparticle before heating. The charge was evaluated from the
area of the anodic wave at 0.15 V. Figure 4 shows the variation
of q with m, indicating proportionality. If the nanoparticles were
It is not easy to analyze a TGA curve accurately because of
some instrumental limitations such as heat transport, location
of the thermometer, uniformity of the temperature, and heat
supplied by exothermic or endothermic reactions. However,
kinetics can be roughly estimated from the TGA curves if the
q
to be made of only silver metal, then the slope, /m, should be
q
equal to F/MAg, where F is the Faraday constant. Values of /m
-
1
in Figure 4 and F/MAg () 96 485/108 C g ) are 755 and 893
-1
C g , respectively. The difference may be due to the assumption
of only silver as the composition. We consider now the
2
0
reaction order is known. The decomposition of the shell is

15030 J. Phys. Chem. B, Vol. 108, No. 39, 2004
Yang et al.
Figure 6. TGA curves at heating rates of (a) 2, (b) 5, (c) 10, (d) 20,
-
1
Figure 8. Temperature dependence of the densities of the nanoparticles,
heated to several temperatures.
and (e) 50 °C min
.
rate, then we obtained the activation energy U ) 111 (
-1
2
0 kJ mol . The TGA measurement of molecular silver stearate
was also made as a control experiment, and its logarithm plot
is inserted into Figure 7. The activation energy U ) 108 ( 14
-
1
kJ mol was close to the value for the nanoparticles. Conse-
quently, the thermal variation of the nanoparticles at 340 °C is
caused by the decomposition of silver stearate to silver in the
shell.
Process of the Decomposition of the Nanoparticles. The
density of the nanoparticles heated to several temperatures was
evaluated by their weight and volume at room temperature. The
volume was evaluated by the excluded volume of water when
nanoparticles were sunk in water. Because a lump of nanopar-
ticles has a hydrophobic surface, it floated on the water surface.
It sank in water, however, after vigorous mixing for several
minutes. Figure 8 shows the variation of the density with heating
Figure 7. Plots of the left side of eq 8 against T- for curves
(
stearate (f).
1
a)-(e) in Figure 6 and for the TGA curve of the monomer of silver
assumed to follow first-order kinetics, as in reaction 1. Letting
x be the moles of silver stearate molecules in the shell, we can
write the Arrhenius-type kinetic equation as
-
3
temperature. The density remains constant (9.3 g cm ) at
dx
dt
-Ae-(U/RT)x
temperatures lower than 200 °C. It decreased from 9.3 to 9.0 g
)
(5)
-
3
-3
cm for 200 < T < 340 °C and then increased to 10.5 g cm
for T > 340 °C. This increase was observed even at 250 °C
when the heating time was longer than 20 min.
where A is the frequency factor and U is the activation energy.
Applying the linear heating scan condition, T ) T0 + Vt, to eq
We shall explain the variation of the density from 9.5 to 9.2
5
and integrating the resulting equation leads to
-3
g cm quantitatively in terms of the decomposition of the shell.
The density of a sphere with mass m and radius r is proportional
AU U/RT e-
z
x
x0
-3
to mr . The diameters of the nanoparticle and its core are 5.3
ln
)
∫
dz
2
(6)
U/RT0
17
Rν
z
and 5 nm, respectively, as estimated previously. The density
-3
of the nanoparticle is proportional to m(5.3) . The gas-removed
where T0 and x0 are the initial temperature and the initial number
of moles, respectively. If we assume that the variation of e is
much larger than that of z and that T0 > T then eq 6 becomes
nanoparticle may be close to the core in size. Therefore, the
-
z
-3
density is proportional to 0.8m(5) when considering the 20%
-
2
mass loss. Taking the ratio of the two densities yields 0.95,
which is close to the value (0.97 ) 9.0/9.3) obtained from the
density in Figure 8. The product by the decomposition has a
-
ln(x/x0)
AR
VU
e-(U/RT)
)
(7)
2
(
)
-3
-3
lower density (9.0 g cm ) than the silver crystals (10.5 g cm ),
and hence it should have porous structure. Pores are necessarily
produced because reaction 1 involves gas formation. The
T
or
-
3
increase in the density from 9.0 to 10.5 g cm over 340 °C
x0
AR
U
RT
may be due to the structural change from the porous structure
ln ln
- 2 ln T ) ln
-
(8)
(9)
[
( )
]
( )
-3
x
VT
to silver bulk crystals (10.5 g cm ) associated with an
exothermic reaction, as shown in the DTA curve (Figure 5).
The process of the decomposition was also proved by the
SEM images and UV spectra of the nanoparticle before and
after heating. Figure 9 gave a set of SEM photographs of the
same nanoparticles heated successively to 340 (a), 400 (b), and
Values of x/x0 were obtained from TGA curves using
m - m
f
x
t
)
x0 m - m
0
f
6
00 °C (c). The particles heated to 340 °C were aggregated to
where m0, mt, and mf are the initial mass, the mass at t, and
form secondly sized particles 0.8 µm in diameter, exhibiting
porosity. The particles heated to 400 °C further aggregated into
a solid with a smooth surface. This aggregation occurred when
the nanoparticles were heated for a long time at any temperature
in the domain from 250 to 400 °C.
Figure 10 shows the UV spectra of the unheated nanoparticle
and the nanoparticle heated to 340 and 400 °C. The band of
the dried film at 457 nm in the aqueous solution (curve (b))
the final mass of the samples, respectively. Plots of
2
-1
ln[(ln(x0/x))/T ] versus T are shown in Figure 7 at several
values of the scan rate. They fall on a straight line over the
domain for the conditions at each heating scan rate. The
deviation from the line at high temperature may be ascribed to
the instrumental limitations of TGA and the assumptions in the
derivation. If we draw a straight line on the plot for each scan

Thermal Metallization of Coated Nanoparticles
J. Phys. Chem. B, Vol. 108, No. 39, 2004 15031
actual melting of silver metal. The kinetics was first order with
respect to silver stearate with an activation energy of 111 kJ
-1
mol . This metallization can be discriminated against the low-
temperature fusion of nanoparticles owing to large surface
energy on the nanometer-scaled curvature.
The metallization of the nanoparticle occurred in two steps,
one being the decomposition of silver stearate to porous silver
retaining the nanostructure for 200 < T < 340 °C and the second
being the fused and sintered nanostructure to bulk silver crystal
at temperatures greater than 340 °C. The first step was evidenced
by FTIR, voltammetry, and the density measurement, and the
second step was confirmed by UV spectroscopy, the density
measurement, and the SEM images.
References and Notes
(1) (a) Dick, K.; Dhanasekaran, T.; Zhang, Z. Y.; Meisel, D. J. Am.
Chem. Soc. 2002, 124, 2312. (b) Link, S.; Wang, Z. L.; El-Sayed, M. A. J.
Phys. Chem. B 2000, 104, 7867. (c) Gao, S. M.; Xie, Y.; Zhu, L. Y.; Tian,
X. B. Inorg. Chem. 2003, 42, 5442. (d) Martin, J. E.; Odinek, J.; Wilcoxon,
J. P.; Anderson, R. A.; Provencio, P. J. Phys. Chem. B 2003, 107, 430. (e)
Bachels, T.; Guntherodt, H. J.; Schafer, Phys. ReV. Lett. 2000, 85, 1250.
(
f) Wuelfing, W. P.; Zamborimi, F. P.; Templeton, A. C.; Wen, X. G.;
Yoon, H.; Murray, R. W. Chem. Mater. 2001, 13, 87. (g) Luo, J.; Jones, V.
W.; Maye, M. M.; Han, L.; Kariuki, N. N.; Zhong, C. J. J. Am. Chem. Soc.
2
002, 124, 13988. (h) Luo, J.; Maye, M. M.; Han, L.; Kariuki, N. N.; Jones,
V. W.; Lin, Y. H.; Engelhard, M. H.; Zhong, C. J. Langmuir 2004, 20,
4254.
(
2) (a) Sun, S.; Murray, C. B.; Weller, D.; Folks, L.; Moser, A. Science
2
000, 287, 1989. (b) Dai, Z.; Sun, S. Wang, Z. L. Nano Lett. 2001, 1, 443.
(
c) Humann, H.; Woods, S. I.; Sun, S. Nano Lett. 2003, 3, 1643. (d)
Ashavani, K.; Chinmay, D.; Murali, S. Appl. Phys. Lett. 2001, 79, 3314.
e) Chinmay, D.; Anamika, G.; Murali, S. Nano Lett. 2002, 2, 365. (f)
Harrell, J. W.; Wang, S.; Nikles, D. E.; Chen, M. Appl. Phys. Lett. 2001,
9, 4393. (g) Weller, D.; Moster, A.; Folkes, L.; Best, M. E.; Le, W.; Toney,
M. F.; Shcweickert, M.; Thiele, J. L.; Doerner, M. F. IEEE Trans. Magn.
Figure 9. SEM images of the same nanoparticle heated succeedingly
to (a) 340, (b) 400, and (c) 600 °C.
(
7
2
000, 36, 10. (h) Kang, S. S.; Harrell, J. W.; Nikles, D. E. Nano Lett. 2002,
2, 1033.
(
3) (a) Takami, A.; Kurita, H.; Koda, S. J. Phys. Chem. B 1999, 103,
1
226. (b) Chang, S.; Shih, C.; Chen, C.; Lai, W.; Wang, C. R. C. Langmuir
1999, 15, 701. (c) Link, S.; Burda, C.; Nikoobakht, B.; El-Sayed, M. A. J.
Phys. Chem. B 2000, 104, 6152. (d) Fujiwara, H.; Yanagida, S.; Kamat, P.
V. J. Phys. Chem. B 1999, 103, 2589. (e) Link, S.; Wang, Z. L.; El-Sayed,
M. A. J. Phys. Chem. B 1999, 103, 3529. (f) Shipwasy, A. N.; Katz, E.;
Willner, I. ChemPhysChem 2000, 1, 18.
(4) Takagi, M. J. Phys. Soc. Jpn. 1954, 9, 359.
(
5) Buffat, Ph.; Borel, J. P. Phys ReV. A 1976, 130, 2287.
Figure 10. UV spectra of (a) the nanoparticle in the cyclohexane
solution and (b) the unheated nanoparticle film on the ITO surface and
the nanoparticle film heated to (c) 340 °C and (d) at 400 °C for 5 min
in air.
(6) Castro, T.; Reifenberger, R.; Choi, E.; Andres, P. Phys. ReV. B
990, 42, 8548.
1
(7) Schmidt, M.; Kusche, R.; Kromueller, W.; Von Issendorf, B.;
Haberland, H. Phys. ReV. Lett. 1996, 79, 99.
8) Lai, S.; Guo, J. Y.; Petrova, V.; Ramamath, G.; Allen, L. H. Phys.
ReV. Lett. 1996, 77, 99.
9) Shvartotsburg, A. A.; Jarrold, M. F. Phys. ReV. Lett. 2000, 85, 2530.
(
was shifted from the band at 421 nm for nanoparticle-dissolved
1
7
cyclohexane (curve (a)), which has been attributed to Mie
scattering. The band shift is due to a solvent effect. The
nanoparticle heated to 250 °C shows almost the same spectrum
(
(10) Pawlow, P. Z. Phys. Chem. 1909, 65, 545.
(11) (a) Reiss, H.; Wilson, I. B. J. Colloid. Sci. 1948, 3, 551. (b) Borel,
(curve c)) as the film without heating. Therefore, the film ought
J.-P. Surf. Sci. 1981, 106, 1. (c) Ross, J.; Andress, R. P. Surf. Sci. 1981,
06, 11. (c) Shimizu, Y.; Sawada, S.; Ikeda, K. S. Eur. Phys. J. D 1998, 4,
365. (d) Peters, K. F.; Cohen, J. B.; Chung, Y.-W. Phys. ReV. B 1998, 57,
3430. (e) Sheng, H. W.; Lu, K.; Ma, E. Nanostruct. Mater. 1998, 10, 865.
f) Cleveland, C. L.; Luedtke, W. D.; Landman, U. Phys. ReV. B 1999, 60,
065. (g) Maye, M. M.; Zheng, W.; Leibowitz, F. L.; Ly, N. K.; Zhong, C.
J. Langmuir 2000, 16, 490. (h) Mayer, M. M.; Zhong, C. J. Mater. Chem.
000, 10, 1895. (i) Zhao, S. J.; Wang, D. Y.; Cheng, D. Y.; Ye, H. Q. J.
Phys. Chem. B 2001, 105, 12857.
12) Kofman, R.; Cheyssac, P.; Aouaj, A.; Lereah, Y.; Deutscher, G.;
B.-David, T.; Penisson, J. M.; Bourret, A. Surf. Sci. 1994, 303, 231.
13) (a) Sakai, H. Surf. Sci. 1996, 348, 387. (b) Sakai, H. Surf. Sci.
996, 351, 285.
14) Beck, T. L.; Jellinek, J.; Berry, R. S. J. Chem. Phys. 1987, 87,
45.
to have the nanostructure for Mie scattering. It should necessarily
be porous, as has been demonstrated not only with the lower
density in Figure 8 (250 < T < 340 °C) but also with the SEM
images in Figure 9. The nanoparticle heated to over 340 °C
1
1
(
5
(curve (d)) shows no adsorption. The drastic variation in the
2
disappearance of the absorption band may be due to changes
in the porous silver by Mie scattering into silver bulk crystals.
(
Conclusions
(
1
The silver stearate-coated nanoparticle was metallized into
bulk silver at 340 °C, which is much lower than the melting
point of bulk silver (960 °C). The low-temperature metallization
of the nanoparticle was due to the decomposition of silver
stearate in the shell into silver metal and gas rather than the
(
5
(15) Honeycutt, J. D.; Andreson, H. C. J. Phys. Chem. 1987, 91, 4950.
16) (a) Berry, R. S.; Iellinek, J.; Natanson, G. Phys. ReV. A 1987, 91,
(
4950. (b) Resis, H.; Mirabel, P.; Whetten, R. L. J. Phys. Chem. 1988, 92,

15032 J. Phys. Chem. B, Vol. 108, No. 39, 2004
Yang et al.
7
241. (c) Wales, D. J.; Berry, R. S. J. Chem. Phys. 1990, 92, 4473. (d)
T.; Yamaguchi, T.; Takiguchi, H.; Nagasawa, H.; Nakamoto, M.; Yase, K.
Mol. Cryst. Liq. Cryst. 1998, 322, 173. (d) Abe, K.; Hanada, T.; Yoshida,
Y.; Tanigaki, N.; Takiguchi, H.; Nagasawa, H.; Nakamoto, M.; Yamaguchi,
T.; Yase, K. Thin Solid Films 1998, 327, 524.
(19) Atkins, P. W. Physical Chemistry, 6th ed.; Oxford University Press,
Oxford, England, 1998; pp 9-229.
(20) (a) Chrissafis, K. Thermochim. Acta 2004, 411, 7. (b) Srikanth, S.;
Chakravortty, M. 2001, 370, 141. (c) Sonibare, O. O.; Egashira, R.; Adeosu,
T. A. Thermochim. Acta 2003, 405, 195.
Wales, D. J.; Berry, R. S. J. Chem. Phys. 1990, 92, 4283. (e) Mirabel, P.;
Reiss, H,; Bowles. R. K. Chem. Phys. 2000, 113, 8200.
(
17) Aoki, K.; Chen, J.; Yang, N. J.; Nagasawa, H. Langmuir 2003, 19,
904.
18) (a) Nagasawa, H.; Maruyama, M.; Komatsu, T.; Isoda, S.; Koba-
9
(
yashi, T. Phys. Status Solidi A 2002, 191, 67. (b) Kuwajima, S.; Okada,
Y.; Yoshida, Y.; Abe, K.; Tanigaki, N.; Yamaguchi, T.; Nagasawa, H.;
Sakurai, K.; Yase, K. Colloids Surf., A 2002, 197, 1. (c) Abe, K.; Hanada,