A vapor-solid strategy to silica sheathed metal nanostructures and microstructures via reactions of metal chlorides with silicon

Article abstract of DOI:10.1021/jp055139g

A facile vapor-solid strategy has been developed to prepare silica-sheathed metal micro/nanostructures with controllable shapes. As examples, silica-sheathed nickel nanowires (diameter ～50 nm), microcubes (edge length 1-3 μm), nanocubes (edge length ～200 nm) with an epitaxial tail (diameter <100 nm), and 1D assembly structures of nanoparticles (particle diameter <100 nm) as well as silica-sheathed cobalt and copper micro/ nanostructures are synthesized. The possible reaction and growth mechanisms of Ni/SiO ₂ structures are discussed. The method is expected to be applied to a wider range of metals.

Full text of DOI:10.1021/jp055139g

J. Phys. Chem. B 2006, 110, 807-811
807
A Vapor-Solid Strategy to Silica Sheathed Metal Nanostructures and Microstructures via
Reactions of Metal Chlorides with Silicon
†
‡
†
,†
Jin Wang, Haoxu Zhang, Jianping Ge, and Yadong Li*
Department of Chemistry, Tsinghua UniVersity, Beijing 100084, People’s Republic of China, and
Nanoscale Physics & DeVices Laboratory, Institute of Physics, Chinese Academy of Science,
Beijing 100080, People’s Republic of China
ReceiVed: September 12, 2005; In Final Form: NoVember 4, 2005
A facile vapor-solid strategy has been developed to prepare silica-sheathed metal micro/nanostructures with
controllable shapes. As examples, silica-sheathed nickel nanowires (diameter ∼50 nm), microcubes (edge
length 1-3 µm), nanocubes (edge length ∼200 nm) with an epitaxial tail (diameter <100 nm), and 1D assembly
structures of nanoparticles (particle diameter <100 nm) as well as silica-sheathed cobalt and copper micro/
nanostructures are synthesized. The possible reaction and growth mechanisms of Ni/SiO
discussed. The method is expected to be applied to a wider range of metals.
2
structures are
1
. Introduction
assembly structures of nanoparticles were obtained. They were
coated with amorphous silica during the synthesis process.
Furthermore, the method was extended to cobalt and copper
and proven to be an effective approach to shape-controlled
synthesis of metals. To clearly describe experimental results,
we denote the as-prepared particles with different experimental
parameters (Table 1) as Sample 1, Sample 2 and so on.
In recent years, because small size could enrich materials with
versatile chemical and physical properties of novelty, micro/
nanostructures have become one of the most active fields both
in science research and practical applications. Various fabrica-
tion approaches have been developed and have provided more
1
-11
multiple and complex units and systems for investigations.
Chemical vapor deposition (CVD), which had been widely used
in preparing films, was introduced into this area and proved to
be a very effective method, characterized by many advantages
such as facility and adjustability.12-15 In particular, atmospheric
pressure chemical deposition (APCVD) is the simplest type of
CVD, which allows for high throughput and continuous opera-
tion at low cost. We have successfully prepared a series of
sulfide and silicate nanostructures with it and demonstrated that
the chlorides are ideal candidate precursors in the APCVD
methods due to their relatively high vapor pressure and adequate
reactivity.1
2
. Experimental Section
The experiments were conducted in a furnace with a
horizontal quartz tube of 2.70 cm in the inner diameter and 110
cm in length. The metal chloride powders, which had been
dehydrated via heating in a vacuum of tens of Pa (specifically,
NiCl2‚6H2O and CoCl2‚6H2O were heated at 220 °C for 15 min
and CuCl2‚2H2O was heated at 150 °C for 15 min), was put in
an alumina boat in the middle of the furnace. Another alumina
boat containing silicon powders and a (110) alumina substrate
was placed downstream away from the metal chloride. A smaller
quartz tube with one end open toward the downstream direction
6-19
During the research process of exploring the application of
chloride-precursor APCVD, we found an interesting phenom-
enon by chance that vapor NiCl2 could react with a Si wafer
and produce a dominant phase of metal Ni with the characteristic
dimension of less than a micrometer and small amounts of Ni
silicides. Ni nanoparticles are known as a catalyst in the VLS
growth of many 1D nanostructures as well as in industrial
manufacture, and their magnetic properties endow them with
(length, 20 cm; outer diameter, 2.05 cm; inner diameter, 1.90
cm) was selectively used as an accessory to hold the alumina
boats according to Table 1. The apparatus was evacuated and
filled with pure nitrogen. Then a nitrogen flow of 25 cm (STP)
3
-1
min was used as the carrier gas. The temperature of the
furnace was raised to the designated temperature from room
temperature at a rate of 10 °C/min and maintained for 60 min.
Then the furnace was cooled to room temperature naturally.
The location of the silicon powders and the alumina substrate,
the furnace center temperature and the amount of the used metal
chloride varied as listed in Table 1. Scheme 1 illustrates the
apparatus setting for the preparation of Sample 1 as an example.
It should be mentioned that a (100) silicon wafer is used to
replace silicon powders and the alumina substrate in the
experiment for Sample 4, and its reproduction is harder than
the others.
20
rich theoretical and practical meanings; nickel silicides are
also very important electronic materials.21 So it is attractive for
us to obtain small particles of pure Ni or its pure phase silicide
through investigation of the phenomenon. Luckily, after a series
of experiments, we succeeded to synthesize diversified micro/
nanostructures of pure Ni and found the new method we used
was rather general.
The nickel particles with well-controlled shapes from nano-
wires, microcubes, and nanocubes with an epitaxial tail to 1D
The synthesized products were characterized by scanning
electron microscopy (SEM, Sirion 200), energy-dispersive
analysis (EDS, Sirion 200), X-ray diffraction (XRD, Br u¨ ker D8
Advance X-ray diffractometer), transmission electron micro-
*
To whom correspondence should be addressed. E-mail:
ydli@tsinghua.edu.cn; Tel: 86-10-62772350, Fax: 86-10-62788765.
†
‡
Tsinghua University.
Chinese Academy of Science.
1
0.1021/jp055139g CCC: $33.50 © 2006 American Chemical Society
Published on Web 12/21/2005

808 J. Phys. Chem. B, Vol. 110, No. 2, 2006
Wang et al.
TABLE 1: Experimental Parameters of the Discussed Samples
sample
number
starting
materials
furnace center
temperature
Si powder
location
Al
2
O
3
substrate
existence of
smaller quartz tube
location
1
2
3
4
5
6
7
8
9
0.050 g NiCl
2
2
2
2
700 °C
900 °C
650 °C
900 °C
900 °C
900 °C
700 °C
850 °C
700 °C
800 °C
1.75 cm
1.50 cm
6.25 cm
1.50 cm
Yes
0
.016 g Si
0.050 g NiCl
.016 g Si
0.050 g NiCl
.016 g Si
0.100 g NiCl
6.00 cm
1.75 cm
Yes
Yes
No
0
0
11.50 cm
(silicon wafer)
6.00 cm
11.50 cm
(silicon wafer)
6.25 cm
(silicon wafer)
0.050 g NiCl
.016 g Si
0.050 g NiCl
.016 g Si
0.050 g CoCl
.016 g Si
0.050 g CoCl
.016 g Si
0.050 g CuCl
.016 g Si
0.050 g CuCl
.016 g Si
2
No
0
2
10.25 cm
1.75 cm
5.50 cm
1.75 cm
5.50 cm
10.50 cm
1.50 cm
5.75 cm
1.50 cm
5.75 cm
Yes
Yes
Yes
Yes
Yes
0
2
2
2
2
0
0
0
1
0
0
SCHEME 1: Apparatus Scheme for the Preparation of
Sample 1 as an Example
Figure 2a and its inset are the SEM and TEM images of
Sample 2, respectively. Well-defined cubes with edge length
ranging from 1 to 3 µm were deposited on the (110) Al2O3
substrate. No obvious silica sheath could be seen from the TEM
image, but the EDS result (Figure S1 in the Supporting
Information, obtained by analyzing one of the detached cubes
on the conductive tape from Sample 2) shows the existence of
a small amount of SiO2, in addition to Ni, which indicates that
a very thin layer of silica surrounds the nickel core.
XRD pattern (Figure 2b) definitely confirms these cubes are
fcc nickel crystals (JCPDS, No.04-0850). Something worthy of
note is that the Ni (2 0 0) peak greatly overwhelms the other
scopy (TEM, JEM-1200EX), and high-resolution transmission
electron microscopy (HRTEM, Tecnai G2 F20 S-Twin).
3. Results and Discussion
3
.1 Different Morphologies of Ni/SiO2 Structures. Figure
1a presents a typical scanning electron microscopy (SEM) image
of Sample 1. The Al2O3 substrate was covered with curly
nanowires up to tens of micrometers long, which can be
identified as pure face-centered cubic (fcc) nickel with a lattice
constant of a ) 0.3523 nm (JCPDS, No.04-0850) by the XRD
result (Figure 1b). TEM results reveal that the as-prepared
nanowires have core-shell structure with typical outer diameter
of 200 nm and core diameter of 50 nm, as the example shown
in Figure 1c. Figure 1d shows the tail part of a long nanowire.
As the nanowire goes to the end, the core becomes thinner and
thinner while the outer diameter remains unchanged. The
selected area electron diffraction (SAED) pattern recorded from
the marked region can be indexed to the same fcc phase as has
been designated by the XRD result. Meanwhile, the SAED
pattern from only the shell part does not show the feature of
crystals. So it is safe to conclude that the core is a nickel phase
and the shell is noncrystalline. Additionally, the (220) direction
of the SAED pattern is roughly parallel to the axis of the marked
core part, which suggests that the nickel wire should grow along
22
the (220) direction in good accord with the previous literature.
Further investigation with the energy-dispersive spectrum (EDS,
Figure 1e, obtained by analyzing the attached part on the
conductive tape from Sample 1) indicates that the nanowires
consist of Ni, O, and Si, and the ratio of Si to O is approximately
Figure 1. (a) Typical SEM image of Sample 1. (b) XRD pattern for
Sample 1. (c) Typical TEM image of Sample 1. (d) TEM image of the
tail part of a long silica-sheathed nickel wire (inset: the SAED pattern
recorded from the marked region). (e) EDS spectrum of Sample 1.
1:2. Combined with the previous SAED analysis, the shell can
be determined to be noncrystalline SiO2.

V-S Strategy to Silica Sheathed Metal Nanostructures
J. Phys. Chem. B, Vol. 110, No. 2, 2006 809
Figure 2. (a) SEM image of Sample 2 (inset: TEM image of a single
microcube). (b) XRD pattern for Sample 2.
Figure 4. (a) SEM image of Sample 4. (b) TEM image of 1D assembly
of nanoparticles in a nanobelt. (c) TEM image of 1D assembly of
nanoparticles or nanorods in nanotubes.
Figure 4a. The silica 1D nanostructures mainly have two
types: nanobelts and nanotubes. The former can be easily found
in both the SEM image (Figure 4a) and the TEM image (Figure
4
b), with width ranging from 300 to 500 nm and thickness of
about 20 nm. The latter could be clearly seen in the TEM image
Figure 4c) with an obvious characteristic of open ends shown
(
by the magnified part. The assemblies of several short nanorods
are also observed. The SAED pattern (the inset of Figure 4b)
recorded from the marked nanoparticle confirms the nanopar-
ticles are nickel. The EDS result from a large area of the as-
prepared 1D assembly structures demonstrates the products
contain only Ni, Si, and O. The quantitive analysis tells us the
ratio of Si to O is 1:2, while the corresponding value of Si to
Ni is 7:1, which indicates that such assemblies in which nickel
nanoparticles are sparely embedded in the silica matrix occur
only when the amount of formed SiO2 is much larger than that
of formed Ni.
Figure 3. (a) SEM image of Sample 3. (b) Higher magnificaiton SEM
image of Sample 3. (c) TEM image of Sample 3 (inset: magnified
image of the marked rectangular region). (d, e, f) SAED patterns of
the marked circle regions of Figure 3c. (g) TEM image of Sample 3.
(h, i, j) HRTEM images of the marked rectangular regions of Figure
3
g, scale bar 1 nm.
ones while in powder diffraction pattern the Ni (200) peak is
only about 50% of the strongest (111) peak in intensity. It can
be explained as follows. The well-defined cubic morphology is
characteristic of single-crystalline cubic-structured Ni crystals
bound by six {1 0 0} planes. Meanwhile, as shown by the SEM
image, almost every microcube is located with two of the six
outer facets parallel to the Al2O3 substrate (110) surface to keep
stable. Therefore, the reflection of crystal planes corresponding
to these six equivalent facets should give rise to the dominant
3
.2 Reaction and Growth Mechanism. The above section
has exhibited various micro/nanostructures assembled with
nickel and silica, but a basic question of how the two kinds of
assembly units came out from the source materials NiCl2 and
Si is still unsolved. As shown in the Experimental Section, NiCl2
and Si were placed in different ceramic boats and thus a transfer
process was needed for them to contact with each other. Over
our experimental temperature range, NiCl2 could be easily
vaporized giving the partial pressure of a few millibar at 650
(200) peak detected by XRD.
SEM images in Figure 3a and 3b portray the morphology of
°C while almost no Si vapor could be produced from the silicon
Sample 3. The typical unit is a nanocube with a long tail, up to
several microns, which turns thinner and thinner as growing
farther from the nanocube. The edge length of the nanocubes
is estimated to be 200 nm and the diameter of the tails is less
than 100 nm. The inset of Figure 3c shows that both the
nanocubes and their tails have a sheath. The SAED patterns
powders. When NiCl2 gas contacted with the Si powders, the
possible reaction occurred as follows:
2
NiCl + Si f 2Ni + SiCl
(1)
2
4
The produced fresh Ni species existed in the vapor phase and
rapidly deposited on the Al2O3 substrate, the top of the Si
powders, and the nearby ceramic boat wall. At the same time,
the simultaneously formed SiCl4 readily reacted with O2 or H2O
gas, which had leaked into the quartz tube or desorbed from
the tube wall at a high temperature (trace amount of water in
the dehydrated chlorides may also serve as a source), giving
products as shown by the equations below:
(Figure 3d, 3e and 3f) of the cube part, the neck part and the
tail part can be indexed into the same nickel fcc structure and
have a uniform orientation. As shown in Figure 3d, the (111)
direction is roughly parallel to the tail axis, indicating that the
tail grows along (111) direction. Meanwhile, the consistence
of the SAED patterns of the tail part, the neck part, and the
cube part demonstrates that the growth of the tail is an epitaxial
process with the (111) crystal plane of the cube as the epitaxial
plane. Figure 3h, 3i, and 3j are HRTEM images of the marked
regions of another cube-tail structure, clearly elucidating the
uniformly oriented (111) growth plane along the whole tail.
Assemblies of nickel nanoparticles/nanorods embedded in
silica 1D nanostructures were found in Sample 4, as shown in
SiCl + O f SiO + 2Cl
2
(2)
(3)
4
2
2
SiCl + 2H O f SiO + 4HCl
4
2
2
The nearly simultaneously formed SiO2 relative to Ni tended

810 J. Phys. Chem. B, Vol. 110, No. 2, 2006
Wang et al.
obvious silicides could be detected above 800 °C (Figure 5).
Taking in consideration that the deposits on the Al2O3 substrate
had no silicides but pure nickel in any experiment listed in Table
1
, we conjectured that the silicides came from the reaction of
Ni and unreacted Si. Lee et al. reported that Ni films did not
react with the silicon wafer with a native silica layer until 800
2
4
°
C. This is in good accord with our experiment facts and
sufficiently confirmed the origin of the nickel silicides.
There are mainly two mechanisms to explain the growth of
1
D structure in the vapor phase: vapor-liquid-solid (VLS)
mechanism and vapor-solid (VS) mechanism. The characteristic
feature of VLS mechanism is the existence of a catalyst
nanoparticle at the tip of the nanowire. The absence of such
nanospheres indicates the VS mechanism should be responsible
for the growth of the silica-sheathed nickel nanowires. After
the reaction of NiCl2 and Si, Ni species were produced and
nucleated on the substrate surface when the concentration
reached a certain supersaturation. Then newly generated nickel
species were absorbed on the nuclear surface, and meanwhile
some outer nickel atoms of the nucleus were vaporized. The
two competitive processes made only the nickel atoms packing
on the growth crystal plane stable in the solid phase, and thus
led to the 1D prolongation of the crystal. At the same time, the
almost simultaneously formed SiO2 adhered to the surface of
the nickel nanowire as a sheath. Recently Zhang et al.
synthesized silica-sheathed Fe7S8 nanowires via a CVD method
and proposed that SiO2 were preferentially absorbed on some
certain Fe7S8 surfaces and induced the 1D growth of Fe7S8.25
Perhaps some similar roles of SiO2 in our method may be taken
into consideration.
Figure 5. XRD patterns for remained silicon powders at different
reaction temperatures.
Figure 6. (a) SEM image of Sample 5. (b) SEM image of Sample 6.
to adsorb onto the Ni surface and form a sheath around the Ni
particles. Reactions 2 and 3 have been comprehensively
Microcubes are also dominated by VS mechanism under the
condition of much higher supersaturation of deposited product
species. To achieve this, in addition to the relatively high
temperature, we inserted a thinner quartz tube with one closed
end into the outer quartz tube to hold the ceramic boat of NiCl2
and another one of Si powders and the substrate. Such an
accessory could restrict the gas-phase species to diffuse in a
smaller volume of space and thus keep higher concentrations.
The product (Sample 5) obtained from the experiment with the
same parameters as Sample 2 except for the absence of the
inserted thinner quartz tube are shown in the Figure 6a, in the
shape of nanorods rather than microcubes due to the lower
concentration of deposited species. With varying the location
of Si powders and the Al2O3 substrate from 6.00 to 10.50 cm,
rods with average width of 600 nm and average length of 4.5
µm mixed with a small amount of microcubes (Figure 6b) were
found in Sample 6. As farther from the furnace center, the NiCl2
vapor concentration and the reaction temperature changed lower,
which usually led to low concentrations of deposited product
species and supplied the condition for the formation of 1D rods
in Sample 6. Nevertheless, the thickness of the rods as large as
600 nm strongly indicates that the concentration of nickel species
was still rather high and could even supply the formation of
2
3
investigated, and it has been pointed that the SiO2 obtained
via them is always amorphous and nanoscaled. This is consistent
with our experiments.
The chemical stability of the deposited Ni and SiO2 is also
confirmed by the previous research.24 Lee et al. calculated the
ternary phase diagram of Ni-Si-O and found that a tie line
exists between Ni and SiO2, indicating that the two phases will
coexist without any reaction. Experimentally, they found that
the reactivity of Ni and SiO2 depends strongly on the silicon
oxide nature. The native SiO2 layer formed with incorporation
of water or moisture is stable in contact with pure Ni while the
thermal SiO2 grown by dry oxidation will react with Ni to form
NiO. The growth environment of our products in the presence
of water determined that the final deposits of Ni and SiO2 are
stable.
In addition, we analyzed the Si powders with XRD after the
experiment and detected nickel silicides in addition to nickel
and excess silicon in some experiments, such as those for
preparing Sample 2 and Sample 4. Correlating the appearance
of nickel silicide with the temperature of Si powders during
the temperature maintaining period of the furnace, we found
that no nickel silicides appeared from 650 °C to 750 °C while
SCHEME 2: Scheme of the Ni/SiO2 Structure Formation Processes

V-S Strategy to Silica Sheathed Metal Nanostructures
J. Phys. Chem. B, Vol. 110, No. 2, 2006 811
the Supporting Information) if the experimental parameters were
properly changed. The above experimental data claim that this
simple method has certain generality in the controlled synthesis
of various silica-sheathed metal particles and is possibly
applicable to other metals.
4
. Conclusion
Various silica-sheathed nickel micro/nanostructures, from
Figure 7. (a) SEM image of Sample 7. (b) TEM image of Sample 7.
nanowires, microcubes, and nanocubes, with an epitaxial tail
to 1D assembly structures of nanoparticles, have been prepared
via a simple vapor-solid strategy. The growth and reaction
mechanisms were investigated. Furthermore, the method proved
to be successful for the shape-controlled synthesis of the
neighboring elements of nickel in the periodic table (cobalt and
copper) and is possibly effective to some other metals.
Acknowledgment. This work was supported by NSFC
(50372030, 20130030), the Foundation for the Author of
National Excellent Doctoral Dissertation of P. R. China, and
the state key project of fundamental research for nanomaterials
and nanostructures (2003CB716901). We thank Professor J.
Acker, Department of Physical Chemistry, Freiberg University
of Mining and Technology, Germany for the helpful discussions.
Figure 8. (a) SEM images of Sample 9 (inset: TEM image of Sample
9
).
Supporting Information Available: Additional spectra and
images of experimental samples. This material is available free
of charge via the Internet at http://pubs.acs.org.
mirocubes due to fluctuations. Thus we can conclude that the
supersaturation degree of the deposited species has played a
very important role in determination of the product morphology.
The cube-tail structure should undergo first 3D growth and
subsequent 1D epitaxy along the (111) direction, as has been
determined in the preceding section. The assembly of nickel
nanoparticles/nanorods occurs on the condition that SiO2 reaches
a relatively high concentration to support a spontaneous 1D
growth. As a summary, an illustration of the formation process
of the four types of Ni/SiO2 structures is given in Scheme 2.
References and Notes
(1) Kroto, H. W.; Heath, J. R.; O’Brien, S. C.; Curl, R. F.; Smalley,
R. E. Nature 1985, 318, 162.
(2) Iijima, S. Nature 1991, 354, 56.
(
(
(
3) Alivisatos, A. P. Science 1996, 271, 933.
4) Morales, A. M.; Lieber, C. M. Science 1998, 279, 208.
5) Pan, Z. W.; Dai, Z. R.; Wang, Z. L. Science 2001, 291, 1947.
(6) Gudiksen, M. S.; Lauhon, L. J.; Wang, J.; Smith, D. C.; Lieber, C.
M. Nature 2002, 415, 617.
(7) Xia, Y. N.; Rogers, J. A.; Paul, K. E.; Whitesides, G. M. Chem.
3
.3 Synthesis of Silica-Sheathed Cobalt and Copper
ReV. 1999, 99, 1823.
Particles. Furthermore, we tried to apply this method to Co
and Cu, the neighboring elements of Ni, via the simple
replacement of NiCl2 by CoCl2 or CuCl2. Figure 7 and Figure
S2 (in the Supporting Information) present a group of charac-
terization results of Sample 7 (see Table 1 for the detailed
experimental parameters). SEM (Figure 7a) and TEM (Figure
(8) Wang, X.; Zhuang, J.; Peng, Q.; Li, Y. D. Nature 2005, 437, 122.
(
9) Hu, J. Q.; Ma, X. L.; Shang, N. G.; Xie, Z. Y.; Wong, N. B.; Lee,
C. S.; Lee, S. T. J. Phys. Chem. B 2002, 106, 3823.
10) Yu, D. P.; Hang, Q. L.; Ding, Y.; Zhang, H. Z.; Bai, Z. G.; Wang,
(
J. J.; Zou, Y. H.; Qian, W.; Xiong, G. C.; Feng, S. Q. Appl. Phys. Lett.
1998, 73, 3076.
(
11) Raman, N. K.; Anderson, M. T.; Brinker, C. J. Chem. Mater. 1996,
, 1682.
12) Cassell, A. M.; Raymakers, J. A.; Kong, J.; Dai, H. J. J. Phys.
Chem. B 1999, 103, 6484.
8
7
b) images show that silica-sheathed nanoparticles with a
(
diameter of 300-400 nm and nanowires with outer diameter
of ∼100 nm and inner diameter of ∼80 nm were obtained. The
XRD pattern (Figure S2a in the Supporting Information)
demonstrates that the only crystal phase in the deposits is fcc
cobalt (JCPDS, No. 15-0806). The EDS results (Figure S2b for
the nanoparticles and Figure S2c for the nanowires in the
Supporting Information) indicate that both of them consist of
Co, Si, and O. With a variation of the experimental parameters
as made for Sample 8, thick whiskers covered with a thin silica
layer (see Figure S3, S4, and S5 in the Supporting Information)
were formed.
Similarly, in the case of the copper element, curly sheathed
nanowires were found in Sample 9 as shown in Figure 8. The
existence of fcc Cu (JCPDS, No. 04-0836) and noncrystalline
SiO2 is confirmed by the XRD pattern (Figure S6a in the
Supporting Information) and the EDS result (Figure S6b in the
Supporting Information). The morphology could also change
into other kinds such as Sample 10 (Figure S7, S8 and S9 in
(
(
13) Ma, R.; Bando, Y.; Sato, T. Chem. Phys. Lett. 2001, 337, 61.
14) Hu, J. Q.; Bando, Y.; Zhan, J. H.; Golberg, D. Angew. Chem., Int.
Ed. 2004, 43, 4606.
(15) Hu, J. Q.; Bando, Y.; Zhan, J. H.; Xu, F. F.; Sekiguchi, T.; Golberg,
D. AdV. Mater. 2004, 16, 1465.
(
16) Ge, J. P.; Li, Y. D. AdV. Funct. Mater. 2004, 14, 157.
(17) Ge, J. P.; Wang, J.; Zhang, H. X.; Li, Y. D. Chem. Eur. J. 2004,
1
0, 3525.
(18) Ge, J. P.; Wang, J.; Zhang, H. X.; Wang, X.; Peng, Q.; Li, Y. D.
AdV. Funct. Mater. 2004, 15, 303.
(
19) Zhang, H. X.; Ge, J. P.; Li, Y. D. Chem. Vap. Deposition 2005,
1
1, 147.
(20) Fu, Y. B.; Zhang, L. D. J. Nanosci. Nanotechnol. 2005, 5, 1113.
(
21) Xu, D. X.; Das, S. R.; Peters, C. J.; Erickson, L. E. Thin Solid
Film 1998, 326, 143.
22) Jina, C. G.; Liu, W. F.; Jia, C.; Xiang, X. Q.; Cai, W. L.; Yao, L.
(
Z.; Li, X. G. J. Cryst. Growth 2003, 258, 337.
(23) Binnewies, M.; Jug, K. Eur. J. Inorg. Chem. 2000, 2000, 1127.
(
24) Lee, P. S.; Mangelinck, D.; Pey, K. L.; Ding, J.; Dai, J. Y.; Ho, C.
S.; See, A. Microelectron. Eng. 2000, 51, 583.
25) Zhang, H. X.; Ge, J. P.; Wang, J.; Wang, Z.; Yu, D. P.; Li, Y. D.
J. Phys. Chem. B 2005, 109, 11585.
(