Copyright © 2011 American Scientific Publishers
All rights reserved
Printed in the United States of America
Nanoscience and
Nanotechnology Letters
Vol. 3, 240–245, 2011
Study of Silicon Dioxide Nanowires Grown via
Rapid Thermal Annealing of Sputtered
Amorphous Carbon Films Doped with Si
Feng Ji Li1 , Sam Zhang1 ∗ , Jun Hua Kong2 , and Wa Li Zhang1
1
School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore 639798, Singapore
2
School of Materials Science and Engineering, Nanyang Technological University, Singapore 639798, Singapore
Silica nanowires are usually synthesized by means of vapor liquid solid method with metal catalyst introduced at the top which will unambiguously affect the excellent light emission properties of
silica nanowires in optoelectronic devices
and by
optical
signal to:
sensors. In this study, silicon dioxide
Delivered
Ingenta
nanowires without traces of catalyst
areTechnological
grown via rapid thermal
annealing of magnetron sputtered
Nanyang
University
amorphous carbon film doped with silicon.
These
high
density
silicon
dioxide nanowires were amorIP : 155.69.4.4
phous with a length longer than Thu,
20 m
and
a
diameter
of
30–140
nm.
Detailed morphology and
18 Aug 2011 06:03:16
microstructure analysis are conducted with field emission scanning electron microscopy and high
resolution transmission electron microscopy. Graphitization of carbon and oxidation of silicon during
rapid thermal annealing were revealed by Raman and X-ray photoelectron spectroscopy. This study
indicates that high growth rate of >6 m/min of high purity silicon dioxide nanowire is possible
simply by sputtering followed by rapid thermal annealing and an additional heating treatment.
Keywords: Magnetron Sputtering, Rapid Thermal Annealing, SiO2 Nanowire, Oxidation.
1. INTRODUCTION
With the discovery of carbon nanotubes (CNTs),1 there
has been tremendous thrust on the synthesis and characterization of one-dimensional (1-D) nanostructures such
as nanowires and nanorods.2–6 The use of 1-D nanostructures for optoelectronic applications is under intensive study. SiOx nanowires are reported to emit blue,7–9
red,9 green10 and ultraviolet9 light. The possible reasons
for red, blue, and ultraviolet light emissions could be
attributed to nonbridging oxygen hole centers, oxygen
deficient centers or neutral oxygen vacancies, and peroxy
linages formed in the nanowires, respectively.9 11 In conclusion, silica nanowires with these light emissions may
be used as effective optoelectronic devices and optical signal humid sensors. Different processing approaches will
result in different SiOx nanowire structures using different catalysts such as Ga,12 13 Pd/Au,14 Sn,15 Pt.6 16 SiOx
nanowire fabricated directly from transforming the silica layer into nanowires of ∼200 nm in diameter by
annealing at 1100 C in an Ar/methane flow has also
been demonstrated.17 Formation of SiO2 nanowires using
reactive magnetron sputtering (RMS) and rapid thermal
∗
Author to whom correspondence should be addressed.
240
Nanosci. Nanotechnol. Lett. 2011, Vol. 3, No. 2
annealing (RTA) has not been reported. Growth of SiO2
naowire by means of RTA of amorphous carbon films
in argon atmosphere is demonstrated in this study. There
are two major merits of RTA: inert gas protected annealing environment and the rapid heating rate. These are
the requirements for SiO2 nanowires to grow, whereas
impossible in normal furnace. The use of the RMS-RTA
combined technique provides a straightforward, convenient
and cost-effective way possible to grow large quantities
and areas of SiO2 nanowire directly onto silicon wafer.
This has potential applications in Si-based optoelectronic
devices.
2. EXPERIMENTAL DETAILS
2.1. Film Deposition
A thin nickel layer of about 30 nm was first deposited
on silicon N -type (100) substrates (100 mm diameter,
475 m thickness and 0.5 nm in Rq using RF E303A
magnetron sputtering system (Penta-Vacuum, Singapore).
Subsequently, silicon and graphite were co-sputtered to
cover the nickel layer to form the a-C:Si/Ni film on Si
wafer. All the sputtering targets were 100 mm in diameter and placed 70 mm away from the substrate. The
1941-4900/2011/3/240/006
doi:10.1166/nnl.2011.1163
Li et al.
Study of Silicon Dioxide Nanowires Grown via RTA of Sputtered Amorphous Carbon Films Doped with Si
deposition was carried out in 0.35 Pa chamber pressure
with 50 sccm argon (99.999%) gas flow at room temperature. The sputtering details are summarized as follows: On
nickel (99.99%) target, RF power 150 W, sputtering time
5 min, on graphite target (99.999%), DC power 990 W,
sputtering time 60 min, on silicon target (99.999%), RF
power 200 W, sputtering time 60 min. The as-deposited
a-C:Si/Ni film underwent rapid thermal annealing (RTA,
Jipelec Jetfirst rapid thermal processor) in argon ambient
at 1100 C for 180 s, at a ramping rate of 87.2 C/s, argon
flow rate of 2000 sccm and chamber pressure of higher
than 0.1 MPa.
2.2. Characterization
Silicon
a-C/Ni/Si film
Fig. 1. Cross-sectional image of the as-deposited a-C:Si/Ni film.
The thickness of the as-deposited film and the surface morVapor-phase synthesis is probably the most extensively
phology were examined by field emission scanning elecexplored approach in formation of one-dimensional nanotron microscopy (FESEM, JEOL JBM-7600F). Structure
structures such as whiskers, nanorods, and nanowires.18 In
of the film was investigated with a Renishaw Raman specDelivered
by
Ingenta
to: synthesis of the nanowire, the initial startvapor phase
troscope at a wavelength of 633 nm radiation excited with
Nanyang
Technological
University
ing reactants are gas phase species. The concentrations
a He-Ne laser. The microstructure of the as-grown film
IP : 155.69.4.4
of the gaseous reactants must be carefully regulated for
was examined by high resolution transmission electron
Thu, 18 Aug 2011
06:03:16
nanowire
synthesis to allow the nanowire growth mechmicroscope (HRTEM, JEOL 2010) operated at 200 kV
anism to predominate and suppress secondary nucleation
and X-ray photoelectron spectroscopy (XPS, Kratos) using
events. Among all vapor-based methods, the formation
a Kratos AXIS spectrometer with monochromatic Al K
mechanism of nanowire was mostly explained by the vapor
(1486.71 eV) radiation (15 kV and 10 mA) and hemiliquid solid (VLS) model in generating large quantities of
spherical electron energy analyzer at a scanning step size
nanowires with single crystalline structures.19 20 A typical
of 0.1 eV after argon ion etching to remove surface conVLS process starts with the dissolution of gaseous reactamination. A pair of sharp stainless tweezers was used
tants into nanosized liquid droplets of a catalyst metal, folto scratch off the film into a small plastic container of
lowed by nucleation and growth of single-crystalline rods
acetone. The container was then subjected to ultrasonic
and then wires. The one-dimensional growth is induced
vibration to disperse the film scraps to be scooped up with
and dictated by the liquid droplets, whose sizes remain
a copper grid for HRTEM study. Finally, energy dispersive
essentially unchanged during the entire process of wire
X-Ray spectroscopy (EDS, JEOL 2010) was used to detect
growth. Each liquid droplet serves as a virtual template to
the chemical composition of the as-grown nanowires after
strictly limit the lateral growth of an individual wire. The
heating up to 600 C in air for 10 min.
establishment of the symmetry-breaking solid–liquid interface is the key step for the one-dimensional nanocrystal
3. RESULTS AND DISCUSSION
growth in this process. However, in this study, it is impossible for the nickel layer to contact the vapor gas in the
3.1. Morphology
chamber and act as the direct catalyst because it is covered
Figure 1 shows the representative cross-sectional image
by solid a-C:Si film. The XPS depth profile also showed
of the as-deposited a-C:Si/Ni film. As seen, the film has
no Ni 2p peaks were detected as shown in Figure 3.
a thickness of about 345 nm. The low and high magnification FESEM images of the nanowires after rapid
3.2. Chemical Composition
thermal annealing are shown in Figures 2(a and b).
After a total of 230 min etching with etching rate of
Figures 2(c and d) show the HRTEM structures of as∼1.2 nm/min on the as-deposited a-C:Si/Ni film (total
grown nanowires. As seen from these images, the length of
etching depth ∼276 nm is smaller than thickness of asthe as-grown amorphous nanowires is longer than 20 m.
deposited a-C:Si film (∼315 nm)) and 75 min etching on
The diameter of the nanowires ranges from 30 nm to
the as-annealed film, the XPS depth profiling results were
140 nm. The dark end is not attributed to the nickel
plotted in Figure 3. Si 2p, O 1s and C 1s varied greatly
particle but to the overlapping of nanowire and the condue to RTA: before RTA, the C 1s concentration (9004 ±
sequence of defocusing. The amorphous structure of the
12 at.%) was ultra high in comparison with Si 2p (521 ±
as-grown nanowires was confirmed by the HRTEM and
04 at.%) and O 1s (475±097 at.%). After RTA, however,
haze rings in the selected area electron diffraction (SAED)
in Figure 2(d).
there was a tremendous decrease in C 1s concentration
Nanosci. Nanotechnol. Lett. 3, 240–245, 2011
241
Study of Silicon Dioxide Nanowires Grown via RTA of Sputtered Amorphous Carbon Films Doped with Si
(a)
(b)
(c)
(d)
Li et al.
Delivered by Ingenta to:
Nanyang Technological University
IP : 155.69.4.4
Thu, 18 Aug 2011 06:03:16
Fig. 2. (a) low magnification (b) high magnification FESEM surface morphology (c) low resolution and (d) high resolution HRTEM images with
selected-area electron diffraction pattern (SAED pattern) in the inset of SiO2 nanowires grown at 1100 C for 180 s.
(1833 ± 458 at.%) and sharp increase in O 1s (3409 ±
222 at.%) and Si 2p (4746 ± 349 at.%). Note that there
was no Ni 2p signal detected during the whole depth profiling, indicating that Ni didn’t move during RTA process.
Fig. 3. XPS depth profiling of the as-deposited and RTA-processed
a-C:Si/Ni film.
242
The additional oxygen may have come from the leak of
the RTA system and the residual oxygen in the reaction
chamber. Even if the chamber pressure is higher than the
outside atmosphere (101 × 105 Pa), a small amount of
oxygen could still be introduced into the chamber. In addition, the RTA system is not a high vacuum device thus
the presence of residual oxygen is unavoidable. Owing to
the presence of the additional oxygen and the high temperature at annealing, the nanowire is further oxidized and
finally the chemical characterization shows all Si4+ or SiO2
structure as shown in Figure 5.
Surface morphology of as-grown nanowires after 600 C
10 min isothermal heating treatment is shown in Figure 4.
The residual carbon was further burnt out. The EDS analysis reveals that the as-grown nanowires consist of Si and O
elements in an atomic ratio of approximately 1:2 confirms
that the chemical composition of the as-grown nanowire
is silicon and oxygen. Moreover, lots of cracks are shown
up on the nanowire surface indicating the disappearance
of residual carbon and the thermal strain during the rapid
thermal annealing process.
Nanosci. Nanotechnol. Lett. 3, 240–245, 2011
Li et al.
Study of Silicon Dioxide Nanowires Grown via RTA of Sputtered Amorphous Carbon Films Doped with Si
(a)
Cracks
(b)
Element
Weight % Atomic % Compd % Formula
Si K
O
Totals
46.74
53.26
100.00
33.33
66.67
100.00
SiO2
(c)
Fig. 4. (a) Surface morphology (b) EDS spectrum and (c) Element stoichiometry analysis of as-grown SiO2 nanowire after 600 C 10 minute
isothermal heating treatment.
Si4+ in the film. However, after RTA, all silicon bonds
have beento:
completely converted to Si4+ ,21–24 indicating silDelivered by Ingenta
The deconvoluted Si 2p XPS peaks for
the
as-deposited
icon inUniversity
a-C:Si/Ni film has been totally oxidized into silicon
Nanyang Technological
and RTA films are shown in Figure 5. Big difference
dioxide. The deconvoluted C 1s XPS peaks for the asIP : in
155.69.4.4
the Si 2p was observed between the as-deposited
deposited
and post-RTA a-C:Si/Ni films are demonstrated
Thu,and
18 postAug 2011
06:03:16
RTA films. Before RTA, there are silicon bonds Si2+ , Si3+ ,
in Figure 6(b).25 As seen from the plot, the sp3 /sp2 ratio
changes from 0.67 to 0.52 meaning there is a conversion of
sp3 C–C bonds into sp2 C C bonds during RTA process.
After RTA, the Raman peak of the as-deposited
a-C:Si/Ni film changed into two isolated peaks at
∼1580 cm−1 (G peak) and 1350 cm−1 (D peak)
with broadened Raman cross-section, as demonstrated in
Figure 7. The G peak is the stretching vibration of any
pair of sp2 sites, either in C C chains or in aromatic
rings characterized as the fingerprint of amorphous diamond carbon films. The D peak is related to the breathing
mode of those sp2 sites only in rings, not in chains.26 The
Raman spectra of the as-deposited a-C:Si/Ni film could be
deconvoluted into two peaks termed D band and G band
with ID as the intensity of D band and IG as the intensity of G band.27 The results indicated that annealing at
1100 C resulted in an increase in ID /IG (from 0.94 to
1.29). Though the ID /IG ratio is not a direct measure of
sp3 or sp2 bonding fraction, the ID /IG ratio is inversely
proportional to sp3 /sp2 ratio.28 Thus, an increase in ID /IG
ratio implied a decrease in sp3 hybridization confirmed
by XPS characterization where the sp3 /sp2 ratio decreased
from 0.67 to 0.52 after annealing (c.f., Fig. 6(b)). The
broadening of the total Raman spectra could be attributed
to growing sp2 clusters and the conversion from sp3 to
sp2 binding configuration. The big distinction in Raman
spectra results before and after annealing has also indicated the presence of medium range order of clustering
in the annealed films on a lateral length scale exceeding
1 nm.29 In principle, there might be several significant factors involved in the broadening of the G peak, namely
Fig. 5. Si 2p peak deconvolution for (a) as-deposited and (b) 180 s
the crystallinity and disordering, the sp2 cluster size and
1100 C RTA process a-C:Si/Ni film (The corresponding bonding con2+
3+
4+
figurations are Si (∼101.6 eV), Si (∼102.5 eV), Si (∼103.4 eV)).
distribution, and the intrinsic stress of the film. Once the
3.3. Bonding Structure
Nanosci. Nanotechnol. Lett. 3, 240–245, 2011
243
Study of Silicon Dioxide Nanowires Grown via RTA of Sputtered Amorphous Carbon Films Doped with Si
Li et al.
of the D peak and the conversion of bonds from C–C to
C C. In addition, the disordering and increase of aromatic bonding also led the carbon signature peak to upshift
as evidenced in the shifting of both D and G peaks to
higher wave numbers with respect to their as-deposited
peak positions (c.f., Fig. 7).
4. CONCLUSION
Pure silicon dioxide nanowires (without traces of metallic
catalyst) can be easily fabricated via magnetron sputtering of amorphous carbon doping with silicon followed by
rapid thermal annealing. The diameter distribution of the
resultant silicon dioxide nanowire is affected by the uniformity of the sputtered. In the as-deposited state, Si has all
possible chemical states. During annealing, these states all
converted to Si4+ . Even annealed at 1100 C, the resultant
wires still remain amorphous.
Delivered by Ingenta to:
Acknowledgment:
Nanyang Technological
University This work was supported by the
Ministry of Education’s Research Grant T208A1218
IP : 155.69.4.4
ARC4/08.
Thu, 18 Aug 2011
06:03:16
References and Notes
Fig. 6. C 1s peak deconvolution for (a) as-deposited and (b) 180 s
1100 C RTA process a-C:Si/Ni film (The corresponding bonding configurations are C–C (∼285.2 eV), C = C (∼284.6 eV), C–O (∼286.5 eV)).
annealing temperature is raised to surpass a certain point,
the graphitization inclination is dramatically accelerated
and the crystallinity is apparently strengthened. In conclusion, the sudden huge increase of temperature of 1100 C
led to a further sp2 clustering and eventual graphitization.
The graphitization resulted in an increase in the intensity
Fig. 7. Raman spectra of a-C:Si/Ni films before and after 180 s
annealing.
244
1. S. Iijima and T. Ichihashi, Nature 363, 603 (1993).
2. X. Duan and C. M. Lieber, Adv. Mater. 12, 298 (2000).
3. H. Dai, E. W. Wong, Y. Z. Lu, S. Fan, and C. M. Lieber, Nature
375, 769 (1995).
4. W. Han, S. Fan, Q. Li, W. Liang, B. Gu, and D. Yu, Chem. Phys.
Lett. 265, 374 (1997).
5. A. M. Morales and C. M. Lieber, Science 279, 208 (1998).
6. Y.-S. Lai, J.-L. Wang, S.-C. Liou, and C.-H. Tu, Appl. Phys. A:
Mater. Sci. Process. 94, 357 (2009).
7. D. P. Yu, Q. L. Hang, Y. Ding, H. Z. Zhang, Z. G. Bai, J. J. Wang,
Y. H. Zou, W. Qian, G. C. Xiong, and S. Q. Feng, Appl. Phys. Lett.
73, 3076 (1998).
8. Y. Hao, G. Meng, C. Ye, and L. Zhang, Appl. Phys. Lett. 87, 1
(2005).
9. N. G. Shang, U. Vetter, I. Gerhards, H. Hofsäss, C. Ronning, and
M. Seibt, Nanotechnology 17, 3215 (2006).
10. J. Q. Hu, Y. Jiang, X. M. Meng, C. S. Lee, and S. T. Le, Chem.
Phys. Lett. 367, 339 (2003).
11. Y. W. Wang, C. H. Liang, G. W. Meng, X. S. Peng, and L. D. Zhang,
J. Mater. Chem. 12, 651 (2002).
12. Z. W. Pan, Z. R. Dai, C. Ma, and Z. L. Wang, J. Am. Chem. Soc.
124, 1817 (2002).
13. Z. Pan, S. Dai, D. B. Beach, and D. H. Lowndes, Nano Lett. 3, 1279
(2003).
14. P. Wu, X. Zou, L. Chi, Q. Li, and T. Xiao, Nanotechnology
18, 125601 (2007).
15. S. H. Sun, G. W. Meng, M. G. Zhang, Y. T. Tian, T. Xie, and L. D.
Zhang, Solid State Commun. 128, 287 (2003).
16. C.-Y. Wang, L.-H. Chan, D.-Q. Xiao, T.-C. Lin, and H.-C. Shih,
J. Vac. Sci. Technol. B: Microelectronics and Nanometer Structures
24, 613 (2006).
17. Z. Zhang, B. Q. Wei, and P. M. Ajayan, J. Phys.: Condens. Matter
14, L511 (2002).
18. M. Law, J. Goldberger, and P. Yang, Annu. Rev. Mater. Res. 34, 83
(2004).
19. R. S. Wagner and W. C. Ellis, Appl. Phys. Lett. 4, 89 (1964).
Nanosci. Nanotechnol. Lett. 3, 240–245, 2011
Li et al.
Study of Silicon Dioxide Nanowires Grown via RTA of Sputtered Amorphous Carbon Films Doped with Si
20. E. I. Givargizov, J. Cryst. Growth 31, 20 (1975).
21. W. Zhang, S. Zhang, Y. Liu, and T. Chen, J. Cryst. Growth 311, 1296
(2009).
22. Y. Liu, T. P. Chen, L. Ding, S. Zhang, Y. Q. Fu, and S. Fung, J. Appl.
Phys. 100, 096111-1 (2006).
23. W. L. Zhang, S. Zhang, M. Yang, and T. P. Chen, J. Phys. Chem. C
114, 2414 (2010).
24. W. L. Zhang, S. Zhang, M. Yang, Z. Liu, Z. Cen, T. Chen, and
D. Liu, Vacuum 84, 1043 (2010).
25. P. Mérel, M. Tabbal, M. Chaker, S. Moisa, and J. Margot, Appl. Surf.
Sci. 136, 105 (1998).
26. G. Irmer and A. Dorner-Reisel, Adv. Eng. Mater. 7, 694 (2005).
27. S. Zhang, X. Lam Bui, X. T. Zeng, and X. Li, Thin Solid Films
482, 138 (2005).
28. S. Zhang, X. T. Zeng, H. Xie, and P. Hing, Surf. Coat. Technol.
123, 256 (2000).
29. J. Zhu, J. Han, X. Han, S. Meng, Z. Jia, J. Yang, and K. Cheng,
Scripta Mater. 55, 1167 (2006).
Received: 20 July 2010. Accepted: 20 November 2010.
Delivered by Ingenta to:
Nanyang Technological University
IP : 155.69.4.4
Thu, 18 Aug 2011 06:03:16
Nanosci. Nanotechnol. Lett. 3, 240–245, 2011
245