Free-standing single crystal silicon nanoribbons [16]

Full text of DOI:10.1021/ja0162966

J. Am. Chem. Soc. 2001, 123, 11095-11096
Free-standing Single Crystal Silicon Nanoribbons
11095
Wensheng Shi, Hongying Peng, Ning Wang,† Chi Pui Li,
Lu Xu, Chun Sing Lee, Rafi Kalish,‡ and Shuit-Tong Lee*
Center Of Super-Diamond and
AdVanced Films (COSDAF) and the
Department of Physics and Materials
City UniVersity of Hong Kong, Hong Kong SAR, China
ReceiVed May 28, 2001
In recent years much scientific and technological excitement
was raised by the discovery of different forms of nanostructures
in many materials (nanospheres, nanotubes, nanowires, oxide-
nanobelts, as well as derivatives of the above).1,2 Silicon is one
of the most important electronic materials. Its nanoscale forms,
such as nanocrystal, porous silicon, quantum well, and nanowire,3
have stimulated great interest among scientists because of their
peculiar physical properties, such as light emission, field emission,
and quantum confinement effect. In this communication we report
on the synthesis of single-crystal silicon ribbons of nano
dimensions achieved by thermal evaporation of silicon monoxide
(SiO) without using any templates or catalysts. The nanoribbons
have a thickness of only about 10 nanometers, widths of several
hundreds of nanometers, and lengths of many micrometers. This
new type of nanostructure, distinctive in shape from nanofilms,
-wires, -tubes, -chains, and -particles, might be explored for
physical property studies and interesting applications as a new
kind of nanoblock.
Free-standing quasi-two-dimensional silicon nanoribbons have
been realized in the present work by using an oxide-assisted
growth method, which has been previously applied to the growth
of one-dimensional silicon nanostructures.4 The remarkable
growth of silicon nanoribbons was carried out in a tube furnace,
in which SiO powder as the source was placed at the center zone
where the temperature was 1150 °C. The evaporated material was
carried down the tube by a gas mixture of 5% hydrogen in argon
at a flow rate of 50 standard cubic centimeters per minute at a
total pressure of 0.5 Torr. Si nanoribbons were abundantly grown
on substrates placed down the tube where the temperature was
900 °C. Untreated alumina plates were usually used as substrates,
although ribbon growth was also observed on other substrates.
Typical growth process lasted for 2 h.
Figure 1. TEM images (TEM, Philips CM 20 TEM at 200 kV) of (A)
silicon nanoribbons collected from the substrate. Arrows 1 indicate the
edge-on position, and arrows 2 indicate silicon nanowires grown together
with nanoribbons. (B) Some rippling-edge ribbons, and (C) a smooth-
edge ribbon. The thickness of the ribbons estimated from the arrows
indicating parts in B and C are 13 and 14 nm, respectively. Scale bars:
1 µm in A, and 100 nm in B and C.
the ribbons have rippling edges (Figure 1B), and a small portion
of the ribbons has smooth edges (Figure 1C). The thickness of
the ribbons can be deduced from the ribbons imaged edge-on, as
marked by arrows in Figure 1, B and C, and were estimated to
be 10-20 nm. Support for this estimate is found in the
transparency of the ribbons to the 200 keV electrons used for the
TEM imaging (Figure 1C). It is worthwhile noting that the
thickness of the ribbons is nearly uniform from ribbon to ribbon
with different width and morphology and also is remarkably
constant within an individual ribbon. Analysis of a number of
nanoribbons with different widths reveals that the average
thickness is about 15 nm. The widths of the various ribbons vary
from 50 to 450 nm as determined from the TEM images. It can
also be observed that the widths exhibit little change over the
length of individual ribbons. Using a representative ribbon as an
example (Figure 1C), the thickness is about 14 nm, and the width
is about 370 nm. Statistically, the ratio of thickness to width varies
from 4 to 22. The rippling and curling features at the edge of
most ribbons also further confirm that the nanoribbons are a quasi-
two-dimensional structure and are distinctly different in shape
from the one-dimensional silicon nanowires with a smooth
surface.
From the synthesized material of light-yellow color, two kinds
of morphology of the nanoribbons have been observed from a
low-magnification TEM overview image (Figure 1A). Most of
* Corresponding author. E-mail: apannale@cityu.edu.hk. Fax: 852-2784-
4696. Telephone: 852-2788-9606.
† On leave from Technion Israel Institute of Technology, Solid State
Institute, Technion City, Haifa 3200, Israel.
‡ Current address: Physics Department, the Hong Kong University of
Science and Technology, SAR, China.
(1) (a) Iijima, S. Nature 1991, 354, 56. (b) Hu, J.; Odom, T. W.; Lieber,
C. M. Acc. Chem. Res. 1999, 32, 435. (c) Kondo, Y.; Takayangi, K. Science
2000, 289, 606. (d) Stevens, R.; Nguyen, C.; Cassell, A.; Delzeit, L.;
Meyyappan, M.; Han. Jie Appl. Phys. Lett. 2000, 77, 3453. (e) Pan, Z. W.;
Dai, Z. R.; Wang, Z. L. Science 2001, 291, 1947. (f) Guo, L. J.; Leobandung,
E.; Chou, S. Y. Science 1997, 275, 649.
(2) (a) Wang, Z. L.; Dai, Z. R.; Gao, R. P.; Bai, Z. G.; Gole, J. L. Appl.
Phys. Lett. 2000, 77, 3349. (b) Zhang, Y.; Suenaga, K.; Colliex, C.; Iijima, S.
Science 1998, 281, 973. (c) Vigolo, B. et al. Science 2000, 290, 1331.
(3) (a) Choi, S. H.; Elliman, R. G. Appl. Phys. Lett. 1999, 75, 968. (b)
Wei, J.; Buriak, J. M.; Siuzdak, G. Nature 1999, 399, 243. (c) Balchin, G.
A.; Amirtharaj, P. M.; Silvestre, C.; Thompson P. J. Appl. Phys. 1999, 85,
2875. (d) Wang, N.; Tang, Y. H.; Zhang, Y. F.; Lee, C. S.; Bello, I.; Lee, S.
T. Chem. Phys. Lett. 1999, 299, 237.
A high-resolution TEM image (Figure 2) of a single nanoribbon
revealed that the ribbon has a crystal core nipped by amorphous
layers with atomically sharp interfaces. The fact that the distance
between atomic planes (indicated by lines in Figure 2) is identical
to that of (110) lattice planes of bulk silicon suggests that the
in-plane layers of the nanoribbon are silicon (110) facet with a
perfect atomic, defect-free, single-crystal structure grown along
(4) (a) Lee, S. T.; Wang, N.; Zhang, Y. F.; Tang, Y. H. Mater. Res. Bull.
1999, 24, 36. (b) Lee, S. T.; Zhang, Y. F.; Wang, N.; Tang, Y. H.; Bello, I.;
Lee, C. S.; Chung. Y. W. J. Mater. Res. 1999, 14, 4503. (c) Shi, W. S.; Peng,
H. Y.; Zheng, Y. F.; Wang, N.; Shang, N. G.; Pan, Z. W.; Lee, C. S.; Lee, S.
T. AdV. Mater. 2000, 12, 1343.
10.1021/ja0162966 CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/11/2001

11096 J. Am. Chem. Soc., Vol. 123, No. 44, 2001
Communications to the Editor
growth process, in which the silicon oxide shell was formed by
the reaction of SiO + SiO ) Si + SiO2, and subsequently
separated from the silicon core during growth.4 According to this
reaction, the amount of segregated silicon oxide is proportional
to the amount of the silicon at the same place. In other words,
the anisotropic growth of the nanoribbons also contributes to the
anisotropic thickness of the silicon oxide shell.
Evidently, the growth of Si nanoribbons differs markedly from
that of silicon nanowires, and the conditions for nanoribbon
growth are quite restrictive. Where nanoribbons were produced,
a few nanowires were also observed (indicated by arrows in Figure
1A). However, in the same experiment at the higher-temperature
zone (about 950 °C) only nanowires and no ribbons were
observed. If the temperature of the furnace and the pressure of
the system (up to 1250 °C and 400 Torr, respectively) was
increased, only wires and no ribbons could be observed in the
deposited product. It has been reported that the growth of micro-
sized ribbons (size in micrometer to millimeter) via metal-
catalyzed VLS method could be interpreted by a twin-plane
growth mechanism, according to which the growth of the
microribbons was controlled by a twin plane parallel to the flat
surface of the microribbons.7 However, for Si nanoribbons, it is
impossible for the existence of (110) twin plane parallel to the
flat face of the nanoribbons according to crystallographic
knowledge. Indeed, no twins were observed in the present
nanoribbons. Therefore, the nanoribbon growth cannot be ex-
plained by the twin-plane growth mechanism. Although the
precise growth mechanism of the nanoribbons is not yet clear, it
is likely that the anisotropy in growth kinetics along different
crystallographic directions due to the particular growth conditions
is responsible for the peculiar growth reported here.
The growth conditions for achieving Si nanoribbons need to
be optimized, and the growth mechanism requires further study.
In particular, once the sideways growth can be controlled, it may
be possible to synthesize, by this method, 2D extremely thin
single-crystal Si nanosheets. The discovery of the existence of
well-ordered, defect-free, and extremely thin narrow single-crystal
silicon nanoribbons described here not only extends the under-
standing of crystal growth in nanoscale but is also scientifically
interesting and technologically promising. The proximity of
atomic planes in nano dimension in nanoribbons may offer
unprecedented opportunities for exploring novel devices.
Figure 2. High-resolution TEM (Philips CM200 FEG TEM, at 200 kV)
image of an individual silicon nanoribbon. Scale bar: 10 nm.
the 111 direction. This direction is different from the predomi-
nant 112 and 110 direction of silicon nanowires synthesized
similarly by the oxide-assisted method.5 However, it is the same
as the growth direction of Si nanowires synthesized by the metal-
catalyzed vapor-liquid-solid (VLS) method.6 From the high-
resolution lattice image shown in Figure 2, we can also conclude
that the wide part of the ribbon is along the 112 direction. The
chemical components of the amorphous edges of the ribbon
consist of silicon oxide (SiOx) as determined by using an electron
energy loss spectrometer and an energy-dispersive X-ray spec-
trometer attached to the TEM. The width of the amorphous edge
is about 10 nm. Analysis of a number of nanoribbons with
different widths shows that the width of the oxide edges varies
from 3 to 25 nm, which are similar to the thickness of the
amorphous silicon oxide shell of the nanowires synthesized by
the oxide-assisted growth.4 The width of these oxide edges is
larger than the typical thickness of a native silicon oxide layer
on a silicon wafer and was found to be related to the width of
the ribbons. Of course, there are also oxides coated on the flat
surfaces of the silicon ribbons. Taking account of the thickness
of the ribbons observed in the above TEM investigations, we
concluded that the thickness of the oxide layer covering the flat
surfaces of the ribbons is much less than the width of the oxide
edges; that is, the thickness of the oxide layers is anisotropic.
This result may be understood in terms of the oxide-assisted
Acknowledgment. This work was supported by the Research Grants
Council of Hong Kong SAR (Project no. 9040637).
JA0162966
(5) Wang, N.; Tang, Y. H.; Zhang, Y. F.; Lee, C. S.; Lee, S. T. Phys. ReV.
B 1998, 58, R16024.
(6) Morales, A. M.; Lieber, C. M. Science 1998, 279, 208.
(7) Wagner, R. S.; Treuting, R. G. J. Appl. Phys. 1961, 32, 2490.