Synthesis of Crystalline Silicon Tubular Nanostructures with ZnS Nanowires as Removable Templates

Article abstract of DOI:10.1002/anie.200352483

Silicon epitaxy on ZnS nanowires results in the formation of ZnS/Si core/shell nanowires; chemical removal of the zinc blende nanowire templates produces monocrystalline silicon tubular nanostructures with outer diameters of about 60-180 nm, wall thicknes

Full text of DOI:10.1002/anie.200352483

Angewandte
Chemie
Silicon Tubular Nanostructures
Synthesis of Crystalline Silicon Tubular
Nanostructures with ZnS Nanowires as
Removable Templates**
Junqing Hu,* Yoshio Bando, Zongwen Liu,
Jinhua Zhan, Dmitri Golberg, and Takashi Sekiguchi
Since the discovery of carbon nanotubes in 1991,[1] a large
number of inorganic solids with layered or nonlayered crystal
lattices have been found to form stable and energetically
viable nanotubes (composed of molecular layers) or tubular
nanostructures.[2] Nowadays, inorganic nanotubes and tubular
nanostructures with interesting properties and potential
applications constitute an important domain of the nano-
structural family. In particular, one-dimensional Si nano-
structures are of special interest due to the key role of Si in the
modern semiconductor industry. Different methods have
been developed for the fabrication of Si nanowires, for
example, the laser-ablation metal-catalytic method,[3] the
oxide-assisted method,[4] and the solution technique.[5] There
have also been reports on the synthesis of Si nanotubes, but
these tubular forms of Si are either polycrystalline or
amorphous.[6] Although the existence of stable monocrystal-
line Si nanotubes and tubular nanostructures has been
predicted theoretically,[7] they have never been observed
experimentally.
Among the many routes for synthesizing nanostructured
materials, template-based methods offer various advan-
tages.[8] Recently, monocrystalline GaNnanotubes were
prepared by “epitaxial casting”, in which hexagonal ZnO
nanowires were used as templates for epitaxial overgrowth of
thin GaNsheaths, and the ZnO nanowire templates were
subsequently removed by thermal reduction and evaporation
with formation of GaNtubular nanostructures. [9] We believed
that this template process could be exploited to prepare
monocrystalline Si tubular nanostructures. It is known that
diamondlike cubic Si and zinc blende ZnS have similar crystal
structures and very close lattice constants (ZnS: a =
0.5431 nm, Si: a = 0.5420 nm).[10] In accordance with lattice-
matching theory, effective epitaxial growth of Si on ZnS
substrates should be possible, since the growth of ZnS thin
films on Si substrates has already been demonstrated.[11] Here
we report on the growth of crystalline Si tubular nano-
structures. A novel template process generates crystalline Si
[*] Dr. J. Hu, Dr. Y. Bando, Dr. Z. Liu, Dr. J. Zhan, Dr. D. Golberg,
Dr. T. Sekiguchi
Advanced Materials Laboratory and Nanomaterials Laboratory
National Institute for Materials Science (NIMS)
Namiki 1-1, Tsukuba, Ibaraki 305-0044 (Japan)
Fax: (+81)298-51-6280
E-mail: hu.junqing@nims.go.jp
[**] This work was supported by the Japan Society for the Promotion of
Science (JSPS) Fellowship tenable at the National Institute for
Materials Science, Tsukuba, Japan.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2004, 43, 63 –66
DOI: 10.1002/anie.200352483
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
63

Communications
tubular nanostructures with outer diameters of about
diameters of about 120–180 nm, and a small fraction of the
tubes have smaller outer diameters of about 60–70 nm. The
majority of tubes have wall thicknesses of about 40–60 nm,
and a small number have thinner walls with thicknesses of
about 20 nm (Figure 2b). Some tubes have one closed end
(Figure 2a), while others are open-ended (Figure 2c). Anal-
ysis by energy-dispersive X-ray spectroscopy (EDS, Fig-
ure 2d) confirmed that the nanotubes are made of Si with a
minimal content of O. Together with the XRD data, these
results show that the obtained sample consists of crystalline Si
tubular nanostructures.
60–180 nm, wall thicknesses of about 20–60 nm, and lengths
of several micrometers. Zinc blende ZnS nanowires were used
as one-dimensional templates for epitaxial growth of thin
monocrystalline Si sheaths, which led to the formation of ZnS/
Si core/shell nanowires. Crystalline Si tubular nanostructures
were formed by chemical removal of the ZnS nanowire cores.
The X-ray diffraction pattern of a sample of Si tubular
nanostructures is shown in Figure 1. The reflections can be
indexed to diamondlike cubic Si with a lattice constant of a =
5.428 Š, consistent with the standard value (JCPDS file 27-
1402: a = 5.430 Š) within experimental error.
Figure 3a shows a high-magnification TEM image of a
segment of a straight Si tubular structure with an outer
diameter and wall thickness of about 60 and 20 nm, respec-
tively. The electron diffraction (ED) pattern (Figure 3a,
inset) can be indexed to the [110] zone axis of crystalline Si,
in which the streaking along the [111] direction originates
from the stacking faults on the {111} planes, as suggested by
the streaklike contrasts on the corresponding faces in TEM
images. Figure 3b depicts the high-resolution TEM image of
this crystalline tubular nanostructure, in which the (111)
fringes with a d spacing of about 0.31 nm are aligned
perpendicular to the longitudinal direction of the tube,
suggesting a [111] growth orientation of the monocrystalline
Si tubular domains. The thin sheath of amorphous SiO2 that
was noticed on the surface was attributed to unavoidable
surface oxidation and/or surface adsorption of oxygen on the
tubes during sample processing. Figure 3c and d respectively
show a TEM image and its corresponding Si elemental map
for a thin Si tubule. Line-scanning elemental profiles across
the Si nanotube (white line in Figure 3c) representing the
spatial distribution of the atomic species are shown in
Figure 3e. The Si profile displays two peaks with a gap in
the center. Both the elemental map and spatially resolved
elemental profile show the tubular shape of the Si nano-
material.
The TEM images (Figure 2a–c) reveal that the product
consists of relatively straight tubular structures. Typically, the
lengths of the as-synthesized tubular structures can reach
several micrometers. Most of them (Figure 2a) have outer
Figure 1. XRD pattern recorded from as-synthesized Si tubular nano-
structures.
In the template-based growth of the crystalline Si tubular
nanostructures, ZnS/Si core/shell nanowires were initially
obtained, as shown in the TEM image in Figure 4a, in which
the shell is a Si thin layer, and the core is a ZnS nanowire. As
seen in Figure 4b, residual segments (nanorods) of the ZnS
nanowire template remains in an open-ended Si tubular
nanostructure after treatment of the ZnS/Si core/shell nano-
wires with HCl solution, owing to incomplete decomposition
of the ZnS nanowire, which suggests that the Si tubular
nanostructure is an extension of the thin Si sheath of the ZnS/
Si core/shell nanowire. Figure 4c shows an HRTEM image of
the interfacial domain between the Si shell and the ZnS core
of a ZnS/Si core/shell nanowire. The well-defined epitaxial
relationship ((111)Si//(111)ZnS; [111]Si//[111]ZnS) between the Si
shell and ZnS core that is clearly observed is due to the similar
crystal structure and the close lattice-matching of Si and ZnS.
Line-scanning elemental profiles across a ZnS/Si core/shell
nanowire (white line in Figure 4a) are shown in Figure 4d.
The profile of Si shows two peaks, on the right- and left-hand
sides, with lower intensity in the center, whereas the profiles
of Zn and S both show a broad peak at the center; this verifies
the core/shell nature of the ZnS/Si nanowires. It is noteworthy
that polycrystalline Si nanotubes and microtubes were
Figure 2. a) TEMimage of Si tubular nanostructures. Scale bar:
200 nm. b) TEMimage displaying an Si nanotube with a thin wall.
c) TEMimage displaying the open end of a Si nanotube. Scale bars
in b) and c): 100 nm. d) EDS spectrum taken from an individual Si
tubular nanostructure.
64
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 63 –66

Angewandte
Chemie
Figure 4. a) TEMimage of a segment of a ZnS/Si core/shell nanowire.
b) TEMimage showing the residual segments of the ZnS nanowire
template in the open-ended Si nanotube after treatment of the ZnS/Si
core/shell nanowires with HCl solution. Scale bars in a) and b):
100 nm. c) HRTEMimage taken from the interfacial domain between
the Si shell and ZnS core of a ZnS/Si core/shell nanowire, revealing an
epitaxial relationship ((111)Si//(111)ZnS; [111]Si//[111]ZnS) between the
Si shell and the ZnS core. Scale bar: 2 nm. d) Line-scanning (indicated
by a white line in a)) elemental mapping displaying Si, Zn, and S spa-
tial elemental distribution profiles across the ZnS/Si core/shell nano-
wire.
obtained from similar synthetic processes in the absence of
ZnS nanowire templates. This implies that the ZnS nanowires
formed during the first stage of synthesis play the key role in
producing monocrystalline Si tubular structures in the two-
stage ZnS-nanowire-templated process.
Figure 5 shows a room-temperature cathodoluminescence
(CL) spectrum of the Si tubular nanostructures. One broad
and weak emission peak at about 450 nm was detected.
Compared with the photoluminescence behavior of bulk Si
nanowires (emission band centered at ꢀ 624 nm (N2 carrier
gas) or ꢀ 783 nm (Ar carrier gas)),[12] the peak position of the
Si tubular nanostructures is shifted dramatically to lower
wavelengths; this may be attributable to some intrinsic point
defects and impurities in these Si nanostructures.
Figure 3. a) High-magnification TEMimage depicting a segment of a
Si nanotube, and an ED pattern (inset) corresponding to the [110]
zone axis of a Si single crystal. Scale bar: 20 nm. b) HRTEMimage of
a Si nanotubular structure, in which the (111) (d111 =0.31 nm) fringes
are aligned perpendicular to the axis direction of the tube, which sug-
gests a [111] growth direction of this tubular nanostructure. Scale bar:
2 nm. c) TEMimage of a segment of a Si nanotube. d) Si elemental
map as further evidence for a tubular structure. Scale bars in c) and
d): 100 nm. e) Line-scanning (indicated by a line in c)) elemental map-
ping highlighting the spatial Si elemental profile across the tube.
Monocrystalline Si tubular nanostructures are distinctly
different from previously reported Si nanowires and other Si-
based nanostructures,[13] due to the characteristic internal
Angew. Chem. Int. Ed. 2004, 43, 63 –66
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
65

Communications
Keywords: crystal growth · epitaxy · nanostructures · nanotubes ·
.
silicon
[1] S. Iijima, Nature 1991, 354, 56.
[2] See, for example a) R. Tenne, L. Margulis, M. Genut, G. Hodes,
Nature 1992, 360, 444; b) E. J. M. Hamilton, S. E. Dolan, C. M.
Mann, H. O. Colijn, C. A. Mcdonald, S. G. Shore, Science 1993,
260, 659; c) Y. Feldman, E. Wasserman, D. J. Srolovitz, R. Tenne,
Science 1995, 267, 222; d) N. G. Chopra, R. J. Luyken, K.
Cherrey, V. H. Crespi, M. L. Cohen, S. G. Louie, A. Zettl,
Science 1995, 269, 966; e) P. Hoyer, Langmuir, 1996, 12, 1411;
f) R. Tenne, M. Homyonfer, Y. Feldman, Chem. Mater. 1998, 10,
3225; g) L. Pu, X. M. Bao, J. P. Zou, D. Feng, Angew. Chem.
2001, 113, 1538; Angew. Chem. Int. Ed. 2001, 40, 1490; h) E. J.
Mele, P. Kral, Phys. Rev. Lett. 2002, 88, 056803-1; i) M. Nath,
C. N. R. Rao, Angew. Chem. 2002, 41, 3601; Angew. Chem. Int.
Ed. 2002, 41, 3451; j) J. H. Jung, H. Kobayashi, K. J. C. van -
Bommel, S. Shinkai, T. Shimizu, Chem. Mater. 2002, 14, 1445.
[3] a) A. M. Morales, C. M. Lieber, Science 1998, 279, 208; b) Y. Cui,
L. J. Lauhon, M. S. Gudiksen, J. F. Wang, C. M. Lieber, Appl.
Phys. Lett. 2001, 78, 2214.
[4] a) N. Wang, Y. H. Tang, Y. F. Zhang, C. S. Lee, S. T. Lee, Phys.
Rev. B 1998, 58, R16024; b) S. T. Lee, Y. F. Zhang, N. Wang,
Y. H. Tang, I. Bello, C. S. Lee, Y. W. Chung, J. Mater. Res. 1999,
14, 4503; c) W. S. Shi, H. Y. Peng, L. Xu, N. Wang, Y. H. Tang,
S. T. Lee, Adv. Mater. 2000, 12, 1927; d) J. L. Gole, J. D. Stout,
W. L. Rauch, Z. L. Wang, Appl. Phys. Lett. 2000, 76, 2346.
[5] J. D. Holmes, K. P. Johnston, R. C. Doty, B. A. Korgel, Science
2000, 287, 1471.
[6] a) J. Sha, J. J. Niu, X. Y. Ma, J. Xu, X. B. Zhang, Q. Yang, D. R.
Yang, Adv. Mater. 2002, 14, 1219; b) S. Y. Jeong, J. Y. Kim, H. D.
Yang, B. N. Yoon, S.-H. Choi, H. K. Kang, C. W. Yang, Y. H. Lee,
Adv. Mater. 2003, 15, 1172.
[7] a) S. B. Fagan, R. J. Baierle, R. Mota, A. J. R. da Silva, A. Fazzio,
Phys. Rev. B 2000, 61, 9994; b) G. Seifert, T. Köhler, H. M.
Urbassek, E. Hernꢀndez, T. Frauenheim, Phys. Rev. B 2001, 63,
193409; c) A. S. Barnard, S. P. Russo, J. Phys. Chem. B 2003, 107,
7577.
[8] See, for example a) W. Liang, C. R. Martin, J. Am. Chem. Soc.
1990, 112, 9666; b) C. J. Brumlik, C. R. Martin, J. Am. Chem.
Soc. 1991, 113, 3174; c) R. Parthasarathy, C. R. Martin, Nature
1994, 369, 298; d) C. R. Martin, Science 1994, 266, 1961; e) M.
Nishizawa, V. P. Menon, C. R. Martin, Science 1995, 268, 700;
f) P. M. Ajayan, O. Stephan, P. Redlich, C. Colliex, Nature 1995,
375, 564; g) B. B. Lakshmi, P. K. Dorhout, C. R. Martin, Chem.
Mater. 1997, 9, 857; h) J. C. Hulteen, C. R. Martin, J. Mater.
Chem. 1997, 7, 1075.
Figure 5. Room-temperature CL spectrum of the as-synthesized Si
tubular nanostructures.
cavity peculiar to a tubular shape. This cavity could be filled
with different semiconducting materials with various band
gaps to give interesting electrical and optical nanodevices, for
example, nanoscale lasers and/or light emitting devices with
tunable wavelengths. In addition, theoretical predictions have
indicated that Si nanotubes (or tubular nanostructures) have a
semiconducting band gap, which in contrast to carbon nano-
tubes, is independent of the tube diameter or chirality.[7b] We
believe that the present discovery of Si crystalline tubular
nanostructures will promote further experimental studies on
their physical properties and applications.
Experimental Section
ZnS/Si core/shell nanowires were grown by thermal evaporation of
ZnS and SiO powders under controlled conditions in a vertical
induction furnace, as described in detail elsewhere.[14] Briefly, the
furnace consisted of a fused-quartz tube and an inductively heated
cylinder made of high-purity graphite coated with a thermally
insulating layer of carbon fiber. A graphite crucible containing ZnS
powder was placed in the central cylindrical zone. After evacuation of
the quartz tube to about 0.2 Torr, a stream of pure Ar was passsed
through the tube at a flow rate of 120 cm3 sÀ1 and ambient pressure,
and evaporation was conducted at 12008C (measured by an optical
pyrometer with an estimated accuracy of Æ 108C). Numerous ZnS
nanowires with diameters of 20–60 nm had grown on the carbon-fiber
layer after 1.5 h of evaporation. They were directly used as templates
for the growth of ZnS/Si core/shell nanowires in the following stage of
synthesis: The crucible containing SiO powder was rapidly heated to
14508C and maintained at this temperature for 1 h (the pressure in
the tube and Ar flow rate in this stage were kept unchanged). The
disproportionation of SiO led to the formation of Si vapor. A thin
layer of Si grew by condensation of the Si vapor on the ZnS nanowire
templates, and ZnS/Si core/shell nanowires formed. Finally, the ZnS
nanowire templates were removed from the ZnS/Si core/shell nano-
wires with HCl solution. This final step resulted in the formation of
crystalline Si tubular nanostructures. The collected products were
characterized and analyzed by X-ray diffraction (XRD, RINT 2200),
transmission electron microscopy (TEM, JEM-3000F and JEM-
3100FFF), and energy-dispersive X-ray spectroscopy (EDS). Cath-
odoluminescence (CL) spectroscopy was carried out with a high-
spatial-resolution, low-energy CL (HRLE-CL) system and a thermal
field-emission scanning electron microscope (TFE-SEM, Hita-
chi S4200).
[9] J. Goldberger, R. R. He, Y. F. Zhang, S. Lee, H. Q. Yan, H.-J.
Choi, P. D. Yang, Nature 2003, 422, 599.
[10] S. M. Sze, Physics of Semiconductor Devices, Wiley-Interscience,
New York, 1981.
[11] a) L. T. Romano, R. D. Bringans, X. Zhou, W. P. Kirk, Phys. B
1995, 52, 11201; b) X. H. Zhou, S. Jiang, W. P. Kirk, J. Appl.
Phys. 1997, 82, 2251; c) N. H. Tran, A. J. Hartmann, R. N. Lamb,
J. Phys. Chem. B 2000, 104, 1150.
[12] Y. F. Zhang, Y. H. Tang, H. Y. Peng, N. Wang, C. S. Lee, I. Bello,
S. T. Lee, Appl. Phys. Lett. 1999, 75, 1842.
[13] W. S. Shi, H. Y. Peng, N. Wang, C. P. Li, L. Xu, C. S. Lee, R.
Kalish, S. T. Lee, J. Am. Chem. Soc. 2001, 123, 11095.
[14] a) D. Golberg, Y. Bando, L. Bourgeois, K. Kurashima, T. Sato,
Carbon 2000, 38, 2017; b) D. Golberg, Y. Bando, W. Q. Han, K.
Kurashima, T. Sato, Chem. Phys. Lett. 1999, 308, 337; c) W. Q.
Han, Y. Bando, K. Kurashima, T. Sato, Appl. Phys. Lett. 1998, 73,
3085.
Received: July 28, 2003
Revised: September 12, 2003 [Z52483]
66
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 63 –66