Preparation and characterization of oriented silica nanowires

Article abstract of DOI:10.1016/j.ssc.2003.08.021

Large-scale of oriented closely packed silica nanowire bunches have been synthesized by using large size (1-10 μm in diameter), low melting point tin droplets as catalyst on silicon wafers at 980 °C. Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) analyses show that the amorphous silica nanowires have lengths of 50-100 μm and diameters of 100-200 nm. Unlike any previous observed results using high melting point metal (such as gold and iron) as catalyst, the Sn catalyst growth exhibits many interesting phenomena. Each Sn ball can simultaneously catalyze the growth of many silica nanowires, which is quite different from the conventional vapor-liquid-solid process.

Full text of DOI:10.1016/j.ssc.2003.08.021

Solid State Communications 128 (2003) 287–290
www.elsevier.com/locate/ssc
Preparation and characterization of oriented silica nanowires
a,c
a
a
a
S.H. Suna, G.W. Menga,b, , M.G. Zhang , Y.T. Tian , T. Xie , L.D. Zhang
*
aInstitute of Solid State Physics, Chinese Academy of Sciences, P.O. Box 1129, Hefei 230031, People’s Republic of China
bDepartment of Materials Science and Engineering, Rensselaer Polytechnic Institute, 1108th Street, Troy, NY 12180, USA
cDepartment of Materials Science and Engineering, Taiyuan Heavy Machinery Institute, Taiyuan 030024, People’s Republic of China
Received 13 June 2003; accepted 26 August 2003 by T.T.M. Palstra
Abstract
Large-scale of oriented closely packed silica nanowire bunches have been synthesized by using large size (1–10 mm in
diameter), low melting point tin droplets as catalyst on silicon wafers at 980 8C. Scanning electron microscopy (SEM) and
transmission electron microscopy (TEM) analyses show that the amorphous silica nanowires have lengths of 50–100 mm and
diameters of 100–200 nm. Unlike any previous observed results using high melting point metal (such as gold and iron) as
catalyst, the Sn catalyst growth exhibits many interesting phenomena. Each Sn ball can simultaneously catalyze the growth of
many silica nanowires, which is quite different from the conventional vapor–liquid–solid process.
q 2003 Elsevier Ltd. All rights reserved.
PACS: 81.16.Hc; 68.65. 2 k
Keywords: A. Nanostructures; B. Nanofabrications; C. Scanning and transmission electron microscopy
1. Introduction
nanostructures could be explained by the vapor–liquid–
solid (VLS) mechanism [13], which was proposed by
Wagner and Ellis in 1964 for silicon whisker growth. Very
recently, Liu et al. [14] have synthesized ultra-long and
highly oriented silicon oxide nanowires via Ga as catalyst;
Pan and his co-workers push the conventional VLS
mechanism to a new range by using molten gallium as
catalyst to grow highly aligned silica nanowires [15].
In this letter, we report a simple method to synthesize
large quantities of closely packed silica nanowire bunches,
by using tin (Sn) powders as catalyst and silicon wafers as a
source material, via the extended VLS mechanism [15].
Further more, the possible growth mode corresponding to
these novel structures was proposed in this paper.
Researches on the synthesis of one-dimensional (1D)
nanostructures have flourished in recent years because of
their fundamental importance to nanotechnology [1].
Nanowires are particularly interesting as they offer the
opportunity to investigate electrical and thermal transport
process in size-confined systems, with the possibility of
providing a deep understanding of physics at the nano-scale.
Among them, silica nanowires have attracted considerable
attention because of their unique properties and promising
applications in mesoscopic research and optoelectronic
nanodevices [2–5].
It is of interest to note that silicon oxide may form some
novel morphology, such as silica ‘nanoflowers’ [6–8],
radial pattern of carbonated silica fibers [9,10], silica
nanowire ‘bundles’ and ‘nanobrushes’ arrays and silica
nanotubes [11], tree-like and tadpole-like SiOx nanostruc-
tures [12]. Most of the growth of the above-mentioned
2. Experimental
The synthesis of silica nanostructures was carried out in
a high-temperature tube furnace. Briefly, an alumina tube
was placed horizontally inside the tube furnace. An alumina
boat loaded with pure Sn powders (99.9 wt%, 0.1 g) was
*
Corresponding author. Tel.: þ86-551-559-3060; fax: þ86-551-
559-1434.
E-mail address: qwmeng@issp.ac.cn (G.W. Meng).
0038-1098/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ssc.2003.08.021

288
S.H. Sun et al. / Solid State Communications 128 (2003) 287–290
placed in the middle of the alumina tube, several pretreated
Si wafers were placed one by one on a lone alumina plate,
which was placed next to the alumina boat at the
downstream end of the tube. The Si wafers were pretreated
as follows. Several piece of strip-like Si (111) wafers were
ultrasonically cleaned in ethanol, acetone in turn for 15 min,
treated in piranha solution (30% H2O2/20% H2SO4), rinsed
with de-ionized water and then dried in air.
Before heating, the tube was evacuated by a mechanical
rotary pump to a base pressure of 5 £ 1022 Torr. And high-
purity argon was poured in to clean away the impurity gas
absorbed in the Si wafers, alumina boat and plate, and the
tube wall. Then, the furnace was heated to 980 8C at a rate of
50 8C/min and held at this temperature for 2 h. During the
experiment, high-purity (99.99%, H2O , or ¼ 20 ppm,
O2 , or ¼ 20 ppm) argon was kept flowing through the
tube at a rate of 100 sccm (standard cubic centimeters per
minute), and the flow pressure in the alumina tube was kept
at about 1 atm. The temperature between the tube center and
the downstream end was measured in situ by a sheathed
thermocouple, allowing us to readily measure the tempera-
ture in the silica nanowire formation region. After the
furnace was cooled naturally to room temperature, white
sponge-like products were found on the upper surface of the
Si wafer located in the temperature range of 940–950 8C.
The as-synthesized products were characterized by
scanning electron microscopy (SEM, JEOL JSM-6300),
transmission electron microscopy (TEM, JEOL 2010). The
chemical compositions were determined by X-ray photo-
emission spectra (XPS, VGESCALAB MKII) and energy-
dispersive X-ray spectroscope (EDS) attached to SEM
instrument. XPS data were collected in the constant analyzer
energy (CAE) mode at 20 eV. Mg Kaðhv ¼ 1253:6 eVÞ
radiation was employed as excitation source with an anode
voltage of 12 kV and an emission current of 20 mA. For
SEM observations, the product was pre-sputtered with a
conducting layer of noble Pt metal, and then together with
the Si wafer was directly transferred into SEM chamber.
Specimens for TEM investigations were briefly ultra-
sonicated in ethanol, and then a drop of suspension was
placed on a holey carbon copper grid.
Fig. 1. SEM images of the as-synthesized products. (a) An overview
of jellyfish-like microstructures. (b) Typical jellyfish-like micro-
structures with closely packed silica nanowires. (c) A SEM image of
cherry-like structure.
our products is likely governed by VLS mechanism. The tail
part of the jellyfish-like structure is composed of large
quantities of closely packed oriented nanowires with
diameters of 100–200 nm and lengths of 50–100 mm.
Quantitative EDS analyses of the jellyfish-like structure
show that the nanowires have the composition of Si and O
elements in atomic ratio of 1:2, closing to SiO2; no other
elements such as tin is detected; while the balls on the tops
of the nanowires are composed of Sn, Si, and O. It is very
interesting that each ball can simultaneously catalyze the
growth of many SiO2 nanowires, and the nanowires grow
out perpendicularly from the low hemisphere surface of the
ball, which is quite different from the conventional VLS
3. Results and discussion
The morphology of the as-synthesized product was
examined by SEM. A typical low-magnification SEM image
(Fig. 1(a)) reveals the presence of large quantities of
jellyfish-like microstructures with diameters of 2–10 mm
and lengths of up to 50–100 mm, exhibiting a pronounced
three dimensional feature. Each jellyfish-like structure
terminates at its top end in a large spherical ball with a
diameter comparable to that of the connected tail (Fig. 1(b)).
The morphology of the jellyfish-like structures is similar to
that of VLS-grown nanowires, suggesting that the growth of

S.H. Sun et al. / Solid State Communications 128 (2003) 287–290
mode in that one catalyst particle usually directs the growth
289
of one nanowire. Detailed study reveals that split growth
occurred in the silica nanowires, i.e. one nanowire splits into
two or more (usually not more than five) sub-branches, and
the newly formed branches sometimes also split into
subbranches. The branches and the subbranches have
similar growth direction and similar diameters to their
stems, which is the same as the splitting phenomenon
observed by Pan et al. in their gallium-catalyzed silica
nanowires [15]. In some extra case, only several SiO2
nanowires grow out from the ball, forming a cherry-like
morphology shown in Fig. 1(c), and the ball is relatively
smaller than that directing the growth of many SiO2
nanowires.
Empirical SEM observations suggest that the number of
the nanowires and the diameters of the balls connected with
the nanowires are intimately related. For instance, from a
comparison of the morphologies shown in the cross-section
SEM image of the as-synthesized product (Fig. 2), it appears
that the bigger the diameter of the ball, the more the SiO2
nanowires grow out from the ball.
The morphology and structure of the nanowires have
been characterized in further detail using TEM. The
nanowires (Fig. 3(a)) have smooth surfaces with diameters
of about 100–200 nm. Highly dispersed selected area
electron diffraction pattern (inset in Fig. 3(a)) taken from
many nanowires shows that the SiO2 nanowires are
amorphous, and the split growth of SiO2 nanowires occurs
(Fig. 3(b)), in agreement with SEM observations.
Fig. 3. TEM images of the as-synthesized products. (a) The fine
morphology of the tail part of the jellyfish-like structure, the highly
dispersed SAED ring pattern (inset) showing that the nanowires are
amorphous. (b) A more detailed TEM image showing the splitting
growth in SiO2 nanowires.
Further evidence for the formation of silica nanowires
can be obtained through XPS. The two strong peaks at 531.6
and 102.6 eV as shown in Fig. 4(a) and (b) correspond to the
binding energy of O (1s) and Si (2p) for silica, respectively.
On the basis of the conventional VLS growth mechanism
and the experimental results described above, we have
proposed a possible mechanism for the growth of the
jellyfish-like silica nanowires. During the early period of
heating, the hot Sn vapor generated from Sn powders rapidly
condenses into small clusters as the Sn species become cool
due to the collision with the buffer gas (argon, oxygen that
remained in the reaction chamber). The formed Sn clusters
are transferred to the downstream end of the alumina tube by
the carrier gas and then deposit onto the surface of silicon
wafers to grow into small liquid Sn droplets. The diameters
of Sn droplets increase with the reaction time through
continuously accepting the upcoming Sn clusters. The Si
wafer was first dissolved in molten Sn at elevated
temperature, i.e. the hot liquid Sn droplets etch the silicon
wafer to form liquid Sn–Si alloy droplets. The solubility of
Si in Sn at 950 8C is ,4 at.% [16]. The Si in the Sn–Si alloy
evaporates into gas species to create a dense vapor of silicon
species around the silicon wafer region, which is similar to
that happened in Ga-catalyst growth of aligned silica
nanowires [15]. In the meanwhile, the liquid alloy droplet
could also absorb Si from the vapor. Then, the dissolved
silicon was oxidized by the oxygen (the source of oxygen
will be discussed latter). Since silicon oxide is not soluble in
tin, it had to precipitate out of the alloy droplet in a form of
silicon oxide nanoparticles [15]. The growth of silica
nanowires can be broken down to four steps:
Fig. 2. Cross-section SEM image of the as-synthesized product.

290
S.H. Sun et al. / Solid State Communications 128 (2003) 287–290
4. Conclusions
Closely packed SiO2 nanowire bunches have been
produced in high yield on Si wafer using Sn as catalyst.
The Sn-catalyst growth shows many interesting new
phenomena, which are different from any previously
achieved results using Au, Fe, and Ni as catalysts. The use
of Sn as catalyst provides a new way for nanowire growth
and extends the conventional VLS mechanism to a broader
range. In addition, the method reported here may be
exploited to produce nanowires of other materials by
using Sn as catalyst.
Acknowledgements
We thank Professor Yongqiang Mao and Yong Qin for
their technical assistance. This work was supported by the
Ministry of Science and Technology of China (Grant No.
G1999064501) and the National Natural Science Foun-
dation of China (Grant No. 19974055 and 50271068).
References
[1] B.I. Yakobson, R.E. Smalley, Am. Sci. 85 (1997) 324.
[2] S. Manm, G.A. Ozin, Nature 382 (1996) 313.
[3] W. Wesh, N.I. Meth, Phys. Rev. B 116 (1996) 305.
[4] A. Katz, M.E. Davis, Nature 403 (2000) 286.
[5] C.T. Kresge, M.W. Leonowiez, W.J. Roth, J.C. Vartvli, J.S.
Boch, Nature 359 (1999) 710.
Fig. 4. XPS spectra of the silica sample. (a) O1s binding energy
spectrum; (b) Si2p binding energy spectrum.
† The formation of silicon oxide nanoparticles on the Sn–
Si alloy surface due to the low solubility of silicon oxide
in Sn.
[6] Y.Q. Zhu, W.B. Hu, W.K. Hsu, M. Terrones, N. Grobert, T.
Karali, H. Terrones, J.P. Hare, P.D. Townsend, H.W. Kroto,
D.R. Mwalton, Adv. Mater. 11 (1999) 844.
† These nanoparticles act as the nucleation sites, initiating
[7] Y.Q. Zhu, W.K. Hsu, M. Terrones, N. Grobert, W.B. Hu, J.P.
Hare, H.W. Kroto, D.R.M. Walton, Chem. Mater. 11 (1999)
2709.
the growth of silica nanowires.
† The alloy droplets continuously absorbing Si species
from the vapor.
[8] Y.Q. Zhu, W.H. Hsu, M. Terrones, N. Grobert, H. Terrones,
J.P. Hare, H.W. Kroto, D.R.M. Walton, J. Mater. Chem. 8
(1998) 1859.
† These nanowires may subsequently grow toward each
other, and then coalesce into one, forming the branched
nanostructures in the end [7].
[9] Z.J. Zhang, G. Ramanath, P.M. Ajiayan, D. Goldberg, Y.
Bande, Adv. Mater. 13 (2001) 197.
The source of oxygen that contributed to the formation of
silica nanowires may have several origins. The most likely
source of oxygen may come from the low content of H2O
(,20 ppm) and O2 (,20 ppm) in the carrier gas of Ar,
which can supply a constant oxygen source during the
growth process of silica nanowires. The residual oxygen
may also be a source, as the base pressure (5 £ 1022 Torr) of
the vacuum system was relatively high. The leakage of
reaction system can be ignored under the pressure of 1 atm.
More work is underway to better understand the growth
mechanism and to prepare nanowires with different
diameters and/or different materials.
[10] Z.J. Zhang, P.M. Ajiayan, G. Ramanath, Appl. Phys. Lett. 78
(2001) 3794.
[11] Z.L. Wang, R.P. Gao, J.L. Gole, J.D. Stout, Adv. Mater. 12
(2000) 1938.
[12] Y.J. Chen, J.B. Li, J.H. Dai, Chem. Phys. Lett. 344 (2001) 450.
[13] R.S. Wagner, W.C. Ellis, Appl. Phys. Lett. 4 (1964) 89.
[14] B. Zheng, Y.Y. Wu, P.D. Yang, J. Liu, Adv. Mater. 14 (2002)
122.
[15] Z.W. Pan, Z.R. Dai, C. Ma, Z.L. Wang, J. Am. Chem. Soc. 124
(2002) 1817.
[16] T.B. Massalski, H. Okamoto, P.R. Subramanian, L. Kacpro-
zak, Binary Alloy Phase Diagrams, 2nd ed., ASM Int,
Materials Park, OH, 1992.