Ultrafine platinum nanoparticles in zirconia nanofilms: Preparation and thermal stability

Article abstract of DOI:10.1016/j.cplett.2008.02.008

Ultrafine Pt nanoparticles were facilely prepared in zirconia nanofilms, and possessed unexpected high thermal stability due to spatial confinement, morphological change, and lattice matching in addition to the strong bonding interaction between the surfa

Full text of DOI:10.1016/j.cplett.2008.02.008

Available online at www.sciencedirect.com
Chemical Physics Letters 454 (2008) 274–278
www.elsevier.com/locate/cplett
Ultrafine platinum nanoparticles in zirconia nanofilms: Preparation
and thermal stability
a,b,
b
b,a
a
a
a
*
Junhui He
, Emi Muto , Toyoki Kunitake , Shuxia Liu , Shen Zhang , Xiangmei Liu
a
Functional Nanomaterials Laboratory and Key Laboratory of Organic Optoelectronic Functional Materials and Molecular Engineering,
Technical Institute of Physics and Chemistry, Chinese Academy of Sciences (CAS), Zhongguancun Beiyitiao 2, Haidianqu, Beijing 100080, China
b
Frontier Research System, The Institute of Physical and Chemical Research (RIKEN), 2-1 Hirosawa, Wako, Saitama 351-0198, Japan
Received 8 November 2007; in final form 4 February 2008
Available online 8 February 2008
Abstract
Ultrafine Pt nanoparticles were facilely prepared in zirconia nanofilms, and possessed unexpected high thermal stability due to spatial
confinement, morphological change, and lattice matching in addition to the strong bonding interaction between the surface atoms of the
nanoparticle and the surrounding oxygen atoms of the ZrO matrix.
2
Ó 2008 Elsevier B.V. All rights reserved.
1
. Introduction
Nanoparticles of metals and semiconductors are attract-
Recently we developed a new ion-exchange method for
incorporation of metal ions into amorphous TiO films
2
[5]. A variety of metal ions can be introduced, and the
amount of metal ions incorporated is readily controlled
by the amount of template and ion-exchange conditions.
Incorporation of two or more metal species is also possible
by simultaneous or sequential procedures. Such films are
nanoporous, and serve as a nanoreactor for in situ synthe-
sis of nanoparticles, as indicated by formation of noble
metal nanoparticles [6], by reversible chemical transforma-
tion between metal and oxide moieties [7], and by success-
ful preparation of bimetallic nanoparticles in such ultrathin
films [8]. It is worthy of notice that ligand-free metallic
ing much attention, as they possess promise in photonic,
electronic, magnetic, and chemical applications [1,2]. Thin
films consisting of such nanoparticles are especially inter-
esting and important as new functional materials. Nano-
particles are usually synthesized in solutions by chemical,
photochemical, radiolytic and hydrothermal reactions.
Thus, effective immobilization of nanoparticles within the
matrix or on the surface of an ultrathin film is required
in most cases of practical applications, and becomes one
of the major challenges in fabrication of functional thin
films [3]. Efforts have also been dedicated to in situ prepa-
ration of nanoparticles in films of micrometer and sub-
micrometer thicknesses. However, in those methods, nano-
particle loadings are often low and, in addition, care must
be taken to avoid aggregation and precipitation of metal
salts and coalescence of metallic particles [4].
nanoparticles are stably kept in the TiO matrix. The sta-
2
bility probably arises from strong bonding interaction
between the surface atoms of the nanoparticle and the sur-
rounding oxygen atoms of the TiO matrix [8], as also dem-
2
onstrated for the stabilization of a Ti metal cluster [9]. This
crown ether effect was also found for other oxygen-rich
matrices, such as cellulose fibers and porous ZrO fibers
2
[
(
10,11]. It is expected that physicochemical properties
e.g., amorphous or crystalline) of film matrices would play
*
Corresponding author. Address: Functional Nanomaterials Labora-
tory and Key Laboratory of Organic Optoelectronic Functional Materials
and Molecular Engineering, Technical Institute of Physics and Chemistry,
Chinese Academy of Sciences (CAS), Zhongguancun Beiyitiao 2, Hai-
dianqu, Beijing 100080, China. Fax: +86 10 82543535.
an important role in the formation and stability of metal
nanoparticles [12–16]. In this communication, we reported
preparation of ultrafine Pt nanoparticles in zirconia nano-
films and their unexpected high thermal stability at
E-mail address: jhhe@mail.ipc.ac.cn (J. He).
0
009-2614/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.cplett.2008.02.008

J. He et al. / Chemical Physics Letters 454 (2008) 274–278
275
elevated temperatures up to 900 °C. The mechanism of sta-
bilization was also discussed based on structural findings
revealed by high resolution electron microscopy.
80
60
4
2
0
0
2
. Experimental
A mixture of Mg(O–Et)2 (10 mM) and Zr(O–nBu)4
(
100 mM) in 2-ethoxyethanol was used for film assembly.
0
Thin films were assembled layer-by-layer by immersing a
substrate (e.g., cleaned quartz plate) in the precursor solu-
tion at room temperature for 3 min, followed by rinsing
with toluene to remove the physisorbed species, drying
0
5
10
d / nm
15
20
Fig. 1. Relationship between the ratio (N
S T
/N ) of exposed surface atoms
(N
S
) to total atoms (N ) and the diameter (d) of metal nanoparticles.
T
with N and hydrolysis in air. Eight cycles of this procedure
2
were repeated and the film thickness was estimated to be
ca. 20 nm from quartz crystal microbalance (QCM) mass
As shown in Fig. 2a, both Pt nanoparticles and ZrO film
were clearly seen. The Pt nanoparticles are nearly spherical
2
2
+
decrease. To remove Mg ions, the thin film was immersed
in aqueous HCl of pH 4 for 20 min, rinsed with pure water
and dried with nitrogen gas. It was then treated with aque-
ous NaOH (pH 10) for 20 min, followed by rinsing and
drying with nitrogen gas. The film on quartz plate was
immersed in an aqueous solution (10 mM) of platinum
and distributed homogeneously in ZrO film. The mean
2
diameter (d) and its standard deviation (r) of Pt nanoparti-
cles were estimated to be ca. 1.4 nm and ca. 0.4 nm, respec-
tively, indicating that the particle size distribution is narrow
(Fig. 2b). A selected area electron diffraction (SAED) pat-
tern (Fig. 2c) as revealed by focusing an electron beam on
the Pt nanoparticles matches that ((111), (200), (220),
(311)) of cubic metallic platinum. On the other hand, when
(
IV) chloride anhydrous (PtCl ) for 4 h, rinsed with pure
4
water and dried by flushing nitrogen gas. It was then
exposed to H plasma treatments. H plasma treatments
2
2
were carried out on a PE-2000 Plasma Etcher (South Bay
Technology, USA). The operating pressure was regulated
at ca. 180 mTorr. The forward power was set at 10 W,
while the reflected power was optimized [6].
the electron beam was focused on a ZrO area which does
2
not contain Pt nanoparticles, a diffuse halo appeared, indi-
cating the amorphous nature of the as-prepared ZrO2
matrix. Fig. 2d and e shows high resolution TEM
(HRTEM) images of Pt nanoparticles. A regular lattice
was observed for Pt nanoparticles, and the lattice fringe
reveals a periodicity of 0.22 nm, which is attributed to the
(111) planes of cubic metallic platinum [17,18]. In contrast,
Nanoparticle-containing ZrO nanofilm was scratched
2
off from its quartz substrate in 2-ethoxyethanol and trans-
ferred to TEM grids (SiO -coated gold grid) by dispenser.
2
The fragmented films were then observed on a JEOL
JEM 2100 F/SP transmission electron microscope at
no ordered lattice structures were observed for the ZrO film
2
2
00 kV.
matrix, and the film matrix showed an amorphous, nano-
porous morphology, in agreement with the SAED results.
3
. Results and discussion
The composition of the Pt/ZrO nanofilm was estimated
2
by X-ray photoelectron spectroscopy (XPS). XPS measure-
ments were carried out on ESCALAB 250 (VG) using Al
Ka (1486.6 eV) radiation. The results gave the composition
Previous results indicated that the catalytic activity of
metal nanoparticle largely depends on the ratio (N /N )
S
T
of its exposed surface atoms (N ) to its total atoms (N )
of the Pt/ZrO nanofilm as Pt:Zr = 1:2.96. Thus, high load-
S
T
2
[
8], and thus, the size of metal nanoparticle would play
ing of platinum was realized.
an important role in determining its catalytic activity.
The thermal stability of the Pt/ZrO nanofilm was inves-
2
Rough estimation of N /N of Pt nanoparticle that has a
tigated by annealing it at 500 °C, 700 °C, and 900 °C. The
obtained specimen was observed by TEM, and the mean
diameter (d) and standard deviation (r) were determined
by sampling over 100 Pt nanoparticles (histograms not
shown). The effect of temperature on the particle mean
diameter and standard deviation is shown both in Fig. 3
and Table 1. After annealing at 500 °C for 5 h, only the
mean diameter of Pt nanoparticle increased slightly to ca.
2.0 nm from that (ca. 1.4 nm) of the as-prepared Pt nano-
particle. Additional annealing at 700 °C caused larger
increases in mean diameter and standard deviation to ca.
4.0 nm and ca. 1.3 nm, respectively. Further annealing at
900 °C, however, only slightly increased the mean diameter
and standard deviation to ca. 4.1 nm and ca. 1.5 nm,
respectively.
S
T
diameter of d was carried out according to the literature
8]. The N /N value was plotted against d, and the
[
S
T
obtained curve is shown in Fig. 1. When d is 20 nm,
N /N is as low as 6.55%. N /N increases slowly to ca.
S
T
S
T
3
3% with decrease of d when the size of Pt nanoparticle
is greater than 3.3 nm. It increases sharply, however, with
decrease of d when the size of Pt nanoparticle is smaller
than 3.3 nm. The reliable N /N value reaches ca. 70.4%
S
T
at a diameter of 1 nm. Clearly, it would be desirable to con-
trol the size of metal nanoparticles as small as possible for
catalytic purposes. Based on the above theoretical investi-
gation, fabrication of ZrO nanofilms containing ultrafine
2
Pt nanoparticles (ꢀ1 nm) was carried out and their thermal
stability was studied.

276
J. He et al. / Chemical Physics Letters 454 (2008) 274–278
30
25
20
15
10
5
0
d = 1.4 nm
σ = 0.4 nm
0
.0 .5 1.0 1.5 2.0 2.5 3.0
Diameter / nm
1
0 nm
Pt (111)
Pt (111)
.22 nm
0
.22 nm
0
(
200)
(
111)
(
220)
(
311)
2
nm
2 nm
Fig. 2. TEM image (a), histogram (b), SAED pattern (c) and high resolution TEM images (d) and (e) of Pt nanoparticles in ZrO
2
nanofilm.
5
4
3
2
1
0
5
4
3
2
1
tetragonal ZrO [19–22]. The lattice domain size increases
with increase of annealing temperature. Very interestingly,
2
Pt nanoparticles seem to dock on ZrO lattice domains by
2
epitaxy, and thus the lattice of Pt nanoparticles coexists
harmoniously with that of tetragonal ZrO2 domains
(Fig. 5). This is probably because of the small difference
(0.07 nm) in the (111) spacing of Pt nanoparticles
(
0.22 nm) and tetragonal ZrO domains (0.29 nm).
2
0
The unexpected high thermal stability of the Pt nano-
0
200 400 600 800 1000
particles in the zirconia nanofilm would be explained in
terms of spatial confinement, morphological change, and
lattice matching in addition to the strong bonding interac-
tion between the surface atoms of the nanoparticle and the
Annealing temperature / C
o
Fig. 3. Temperature dependence of particle mean diameter (d) and
standard deviation (s).
surrounding oxygen atoms of the ZrO matrix [8]. The par-
2
ticle transferring model is often used to account for sinter-
ing of supported metals [23]. In this model, metal
microcrystals are supposed to exist in a quasi-liquid state
when the temperature is beyond their Tamman tempera-
ture (0.4 times the metal melting point in Kelvin). Thus,
the metal microcrystals could transfer, collide, and grow
up on the surface of support over their Tamman tempera-
ture. Because of their smaller sizes, Pt nanocrystals must
have an even lower Tamman temperature than Pt micro-
crystals (544 °C). Thus, they might still exist in their solid
state and hardly grow up at below 500 °C. This specula-
tion was supported by the negligible increase of the mean
diameter and standard deviation in the temperature range
of 20–500 °C. The mean diameter and standard deviation
increased evidently faster in the range 500–900 °C than
below 500 °C. In this range, Pt nanoparticles might be in
the quasi-liquid state and began to transfer and grow.
Nevertheless, the increase was surprisingly not significant,
Table 1
Effect of annealing temperature on the mean diameter and standard
deviation of Pt nanoparticles in ZrO
2
nanofilm
Annealing temperature Mean diameter (d)
Standard deviation (r)
(nm)
(
°C)
(nm)
As-prepared
1.4
2.0
4.0
4.1
0.4
0.4
1.3
1.5
5
7
9
00
00
00
HRTEM images of the as-prepared and annealed Pt/
ZrO nanofilms are compared in Fig. 4. Identical lattice
2
fringes were observed for Pt nanoparticles, indicating the
crystalline structure of Pt nanoparticle had not changed
upon thermal annealing at different elevated temperatures.
In contrast, unlike the as-prepared Pt/ZrO nanofilm, after
2
annealing at 500 °C, a regular lattice was formed in the
ZrO matrix (Fig. 4b). The majority of the lattice fringe
compared with in other matrices such as TiO . The Pt
2
2
reveals a periodicity of 0.29 nm, which is attributed to
nanoparticles were located within the nanopores of the

J. He et al. / Chemical Physics Letters 454 (2008) 274–278
277
T denotes tetragonal
ZrO T(111)
0
.29 nm Pt (111)
0
.22 nm
ZrO T(111)
.29 nm
Pt (111)
0.22 nm
0
2
nm
ZrO T(111)
.29 nm
0
o
After 900 C
annealing
2
nm
Pt (111)
0.22 nm
Pt (111)
.22 nm
o
0
After 700 C
annealing
2
nm
o
After 500 C
annealing
2
nm
As-prepared
Annealing temperature
2
Fig. 4. HRTEM images of the Pt/ZrO nanofilm before and after annealing at elevated temperatures.
atures, these tetragonal domains, unlike TiO matrices, had
2
not grown to three dimensional crystals [24]. Thus, they
would not exclude molten Pt nanoparticles at elevated tem-
peratures, and the fusion of molten Pt nanoparticles was
suppressed.
Pt (111)
.22 nm
0
The unveiled lattice matching of Pt nanoparticles and
tetragonal ZrO domains above should also contribute to
2
the unexpected high thermal stability of the Pt nanoparti-
cles in the ZrO nanofilm. The fact that the lattice of Pt
nanoparticles coexists harmoniously with that of tetrago-
ZrO T(111)
2
2
0.29 nm
nal ZrO domains by epitaxy suggests that the tetragonal
2
ZrO domains enhanced the stability of the Pt nanoparti-
2
cles rather than excluded them from the ZrO2 matrix.
These Pt nanoparticles were, therefore, very stable in the
2
nm
ZrO nanofilm.
2
Clearly, the zirconia nanofilm effectively immobilized
and isolated Pt nanoparticles. Therefore, Pt nanoparticles
in the as-prepared zirconia nanofilm had unexpected high
thermal stability.
Fig. 5. HRTEM image that shows harmonious coexistence of Pt
nanocrystal and tetragonal ZrO domain by epitaxy. This image was
taken after annealing at 500 °C, T denotes tetragonal.
2
zirconia nanofilm [5–8,11]. The confining effect of these
nanopores on Pt nanoparticles may make them less easy
to transfer and grow. Clearly, it is difficult for Pt nanopar-
ticles to grow up beyond the confining nanopores of the
zirconia nanofilm.
It was found previously that significant growth of three
dimensional nanocrystals of matrix would largely affect the
stability of imbedded metal nanoparticles, i.e., the three
dimensional nanocrystals of matrix would exclude molten
metal nanoparticles, causing their fusion and significant
growth. One of such cases was found in titania matrices
4. Conclusions
In summary, the relationship between the ratio of
exposed surface atoms to total atoms and the diameter of
metal nanoparticles was investigated by theoretical calcula-
tions. Based on these results, we carried out and succeeded
in fabrication of zirconia nanofilms containing ultrafine Pt
nanoparticles. The Pt nanoparticles showed unexpected
high thermal stability, which was discussed in terms of spa-
tial confinement, morphological change, and lattice match-
ing in addition to the strong bonding interaction between
the surface atoms of the nanoparticle and the surrounding
[
11]. Although HRTEM revealed the appearance and
expansion of tetragonal ZrO domains at elevated temper-
oxygen atoms of the ZrO matrix [8]. The zirconia matrix,
2
2

278
J. He et al. / Chemical Physics Letters 454 (2008) 274–278
however, experienced a morphological change from amor-
[7] J. He, I. Ichinose, S. Fujikawa, T. Kunitake, A. Nakao, Chem.
Commun. (2002) 1910.
phous nature to tetragonal ZrO domains, and the expan-
2
[8] J. He, I. Ichinose, T. Kunitake, A. Nakao, Y. Shiraishi, N. Toshima,
sion of these tetragonal domains within the matrix. Such
J. Am. Chem. Soc. 125 (2003) 11034.
highly stable Pt/ZrO composites are expected to have wide
2
[9] R. Franke, J. Rothe, J. Pollmann, J. Hormes, H. Boennemann,
W. Brijoux, Th. Hindenburg, J. Am. Chem. Soc. 118 (1996) 12090.
10] J. He, T. Kunitake, A. Nakao, Chem. Mater. 15 (2003) 4401.
11] J. He, T. Kunitake, Chem. Mater. 16 (2004) 2656.
applications, especially in catalysis.
[
[
[
Acknowledgments
12] K. Akamatsu, S. Ikeda, H. Nawafune, S. Deki, Chem. Mater. 15
(
2003) 2488.
This work was supported by Toyota Central Laborato-
ries, the National Natural Science Foundation of China
[
13] J.I. Abes, R.E. Cohen, C.A. Ross, Chem. Mater. 15 (2003) 1125.
[14] S. Joly et al., Langmuir 16 (2000) 1354.
(
Grant No. 20471065), and ‘Hundred Talents Program’
[15] J. Dai, M.L. Bruening, Nano Lett. 2 (2002) 497.
[16] S. Kidambi, J. Dai, J. Li, M.L. Bruening, J. Am. Chem. Soc. 126
of CAS.
(
2004) 2658.
[
[
[
17] J. He, T. Kunitake, A. Nakao, Chem. Commun. (2004) 410.
18] JCPDS Card 88-2343.
19] JCPDS Card 50-1089.
References
[
[
1] G. Schmid, Chem. Rev. 92 (1992) 1709.
[20] A.D. Polli, T. Wagner, A. Fischer, G. Weinberg, F.C. Jentoft, R.
Schl o¨ gl, M. R u¨ hle, Thin Solid Films 379 (2000) 122.
[21] S. Shukla, S. Seal, J. Phys. Chem. B 108 (2004) 3395.
[22] M. Atik, C.R. Kha, P. De Lima Neto, L.A. Avaca, M.A. Aegerter,
J. Zarzycki, J. Mater. Sci. Lett. 14 (1995) 178.
[23] R. Hughes, Deactivation of Catalysts (F. Ding, N. Yuan, Trans.),
first edn., Avon Books, Beijing, 1990, p. 63.
[24] V. Richard, S. Onoue, A. Nakao, T. Kunitake, Nat. Mater. 5 (2006)
494.
2] M.P. Pileni, in: D.L. Feldheim, C.A. Foss Jr. (Eds.), Metal Nano-
particles: Synthesis Characterization and Applications, Marcel Dek-
ker, New York, 2001, p. 207.
[
[
[
[
3] J. He, T. Kunitake, in: M.P. Pileni (Ed.), Nanocrystals Forming
Mesoscopic Structures, Wiley-VCH, Weinheim, 2005, p. 91.
4] I.-W. Shim, S. Choi, W.-T. Noh, J. Kwon, J.Y. Cho, D.Y. Chae,
K.-S. Kim, Bull. Korean Chem. Soc. 22 (2001) 772.
5] J. He, I. Ichinose, S. Fujikawa, T. Kunitake, A. Nakao, Chem. Mater.
14 (2002) 3493.
6] J. He, I. Ichinose, T. Kunitake, A. Nakao, Langmuir 18 (2002) 10005.