Monolithic self-sustaining nanographene sheet grown using plasma-enhanced chemical vapor deposition

Article abstract of DOI:10.1002/pssa.200925230

We have fabricated carbon nanowalls (CNWs) composed of monolithic self-sustaining nanographene sheets standing vertically on a Si substrate, using plasma-enhanced chemical vapor deposition with a C₂F ₆/H₂ mixture. The crystallinity, evaluated by Raman spectroscopy and synchrotron X-ray surface diffraction, and the electrical properties of the CNWs were improved by introducing O₂ gas into the source gas mixture during the CNW growth process. The temperature dependence of the resistivity of the CNW films exhibited semiconductor behavior.

Full text of DOI:10.1002/pssa.200925230

Phys. Status Solidi A 207, No. 1, 139–143 (2010) / DOI 10.1002/pssa.200925230
a
p s s
www.pss-a.com
applications and materials science
Monolithic self-sustaining
nanographene sheet grown using
plasma-enhanced chemical vapor deposition
,1
Wakana Takeuchi* , Keigo Takeda1, Mineo Hiramatsu2, Yutaka Tokuda3, Hiroyuki Kano4, Shigeru Kimura5,
Osami Sakata5, Hiroo Tajiri5, and Masaru Hori1
1 Department of Electrical Engineering and Computer Science, Nagoya University, Chikusa, Nagoya 468-8603, Japan
2 Department of Electrical and Electronic Engineering, Meijo University, Tempaku, Nagoya 468-8502, Japan
3 Department of Electrical and Electronics Engineering, Aichi Institute of Technology, Yakusa, Toyota 470-0392, Japan
4 NU Eco-Engineering Co., Ltd., Kurozasa, Miyoshi, Nishikamo 470-0201, Japan
5 Japan Synchrotron Radiation Research Institute (JASRI)/SPring-8, Kouto, Sayo 679-5198, Japan
Received 7 May 2009, revised 20 August 2009, accepted 20 August 2009
Published online 18 September 2010
PACS 52.77.Dq, 61.46.–w, 78.30.Na, 81.15.Gh
Corresponding author: e-mail w_takeuchi@nagoya-u.jp
We have fabricated carbon nanowalls (CNWs) composed of diffraction, and the electrical properties of the CNWs were
monolithic self-sustaining nanographene sheets standing ver- improved by introducing O2 gas into the source gas mixture
tically on a Si substrate, using plasma-enhanced chemical vapor during the CNW growth process. The temperature dependence
deposition with a C2F6/H2 mixture. The crystallinity, evaluated of the resistivity of the CNW films exhibited semiconductor
by Raman spectroscopy and synchrotron X-ray surface behavior.
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1 Introduction Two-dimensional (2D) graphene frequency inductively coupled PECVD [4], and hot-filament
sheets, which range from a single layer to stacks of a few CVD [5]. We have fabricated CNWs by fluorocarbon
layers, have attracted great interest for use in next-generation PECVD with hydrogen (H) radical injection. This method
electronic devices. Few-layer graphene sheets have been has the potential to control the morphology of CNWs, such as
reported to have high mobilities of 10 000–15 000 cm2 Vsꢀ1 the thickness and the spacing between the walls [6, 7].
and a band gap of 0.26 eV [1, 2]. However, the primary Recently, we fabricated nitrogen-doped CNWs by injecting
method for isolating graphene, i.e., micromechanical N2 gas into the processing plasma and evaluated their
cleavage of graphite, cannot be easily scaled up for large- electrical conduction properties [8].
scale applications.
CNWs have been reported to consist of nanodomains
Recently, several reports have been published on the that are a few tens of nanometers in size [9] and individual
growth of carbon nanowalls (CNWs), which are 2D carbon CNWs were found to have many edges and defects [10].
nanostructures comprising of planar graphene layers stand- Therefore, in order to fully exploit the potential of graphene
ing almost vertically on a substrate [3–6]. CNWs can be and to develop next-generation electronic devices, reliable
described as 2D graphite nanostructures with edges and they methods are required for fabricating graphene sheets with
have a wall-like structure with an ultrahigh aspect ratio. high crystallinity. In our previous study, O2 plasma was used
Since CNWs essentially consist of graphene sheets, they to remove the carbon-related coating on the chamber wall for
are expected to have high mobilities and large sustainable the fabrication of CNWs with high reproducibility [11]. In
current densities. Therefore, CNWs are considered to contrast, the addition of small amount of CO2 or H2O to the
be one of the most promising carbon materials for nanoscale hydrocarbon plasma is effective to etch disordered carbon
electronic devices.
species [12, 13]. Carbon–oxygen reactions on various kinds
CNWs have been synthesized by various chemical vapor of carbon surfaces have been studied [14]. In some cases,
deposition (CVD) techniques, including microwave plasma- the trigonal bond of carbon at the edges of basal planes of the
enhanced chemical vapor deposition (PECVD) [3], radio- crystallites and defects within the basal plane act as
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
p s s
140
W. Takeuchi et al.: Monolithic self-sustaining nanographene sheet grown using PECVD
nucleation sites for undesirable nanostructured carbon. C2F6/H2 mixture without O2. These images show slightly
These nucleation sites can be eliminated by reacting with branching 2D carbon sheets standing almost vertically on the
atomic oxygen to form volatile by-products such as CO and substrate. The CNW height, which is equivalent to the CNW
CO2. Thus, using carbon–oxygen reactions would enable us film thickness, was approximately 1 mm. Figure 1b shows
to synthesize carbon materials that have larger crystallite SEM images of CNWs grown for 40min with the addition of
sizes, higher crystallite alignment, and higher purity.
O2 at a flow rate of 5 sccm. These CNWs exhibit less
In this study, CNW films were fabricated using PECVD branching than those produced without O2 so that monolithic
with a C2F6/H2 mixture. We investigated the effect of O2 gas graphene sheets were obtained; however, the growth rate was
addition to the process gas mixture of C2F6/H2 on the reduced by approximately 33%. CNW films grown with O2
structure and electrical properties of CNW films. We found had larger plane sheets with wider interspaces than those
that the crystallinity of vertically standing graphene sheets is grown without O2.
improved by introducing O2 into the plasma used for CNW Cross-sectional transmission electron microscope (TEM)
growth, and that the well-defined CNW films synthesized by imagesofCNW filmsgrownwithoutandwithO2 areshownin
PECVD exhibit semiconductor behavior.
Fig. 1c and d, respectively. Small overlapping multi-layered
graphene domains with random orientations were observed
2 Experimental CNW films were fabricated using forCNWsgrownwithoutO2 (Fig. 1c).Bycontrast, monolithic
PECVD with H radical injection with a mixture of C2F6 and self-sustaining graphene sheets larger than 200 nm in size
H2. The radical-injection PECVD system consists of a were clearly observed in the CNWs grown with O2 (Fig. 1d).
parallel-plate VHF (100 MHz) capacitively coupled plasma The insets show magnified images of the areas enclosed by
(CCP) with C2F6 and a surface-wave-excited microwave H2 squares in Fig. 1c and d; these images allow the fine structure
plasma as a remote H radical source. Details of the radical- of the CNWs to be assessed. The inset of Fig. 1c clearly shows
injection PECVD system are provided elsewhere [8, 11]. The a bent multi-layered graphene structure with a thickness of
C2F6 and H2 flow rates were maintained at 50 and 100 sccm, approximately 9 nm in the case of CNW grown without O2.
respectively. O2 was introduced into the CCP region at flow On the other hand, a highly orientated, smooth multi-layered
rates of 0–5 sccm and the total pressure was maintained at graphene structure was clearly obtained in the CNW grown
160 Pa.
with O2 (inset of Fig. 1d).
CNW grown without and with O2 gas addition were also
3 Results and discussion Figure 1a and b shows characterized by Raman spectroscopy to investigate the
cross-sectional scanning electron microscopy (SEM) images influence of O2 gas addition on the structural property of
of the CNW films, and the insets show top-view SEM images CNWs. Raman spectroscopy is a powerful tool and widely
of the same CNW films. Figure 1a shows SEM images of a applied to determine the structural properties of various forms
typical CNW film grown for 30min on a Si substrate using a of carbon nanostructures. Raman spectra for the deposits were
measured at room temperature with a triple monochromator
(Jobin Yvon, Ramanor T64000) using the 514.5 nm line of an
Ar laser in the spectral range from 1200 to 1800cmꢀ1
.
Figure 2a shows the Raman spectra of the CNW films grown
with different O2 flow rates. The Raman spectra of all the
CNW films have a G-band peak at 1584cmꢀ1 indicating the
formation of a graphitized structure, and a D-band peak
around at 1350cmꢀ1 corresponding to the disorder-induced
phonon mode. Usually, the G-band peak is accompanied by a
shoulder peak at 1620cmꢀ1 (D0-band). The D0-band is
associated with finite-size graphite crystals [10]. The strong
D-band and D0-band peaks suggest a more nanocrystalline
structure and the presence of graphene edges and defects,
which are features of CNWs. Figure 2b shows the ratio of the
peak intensity of the D-band to that of the G-band (ID/IG). The
ID/IG decreased with an increase in the O2 flow rate, indicating
that the O2 addition improved the crystallinity of CNWs by
suppressing the second nucleation, resulting in the improve-
ment of the crystallinity of CNW films.
Figure 1 Cross-sectionalSEMimagesof(a)CNWfilmgrownona
Si substrate using C2F6/H2 and (b) CNW film grown using C2F6/H2
with additional O2 gas, together with SEM top-view images of
identical CNW films as insets. TEM images of (c) CNWs grown
without any additional gas and (d) CNWs grown with additional O2,
together with magnified images of square areas as insets, showing
layered graphene structures of CNWs.
The crystallinity of CNWs was analyzed using synchro-
tron X-ray surface diffraction at grazing incidence and exit at
the beamline BL13XU of SPring-8 [15]. The X-ray beam was
incident on the CNW film samples at glancing angles of 0.058
and 0.38 relative to the substrate surface; these incident angles
were, respectively, used to evaluate the crystallinity of the
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.pss-a.com

Original
Paper
Phys. Status Solidi A 207, No. 1 (2010)
141
d002 was determined from the 002 peak by applying Bragg’s
law with a wavelength of l ¼ 0.1003nm. It was found to be
0.342 nm for all samples, which is slightly larger than that of
bulk graphite (0.335nm), but is the same as that of the layered
carbon fluorides, the F atoms of which are reduced by the
abstractreactionwithhydrogen[16]. The interlayer spacingin
the present study may be large due the presence of F atoms,
although their concentration in the CNWs grown with O2 was
very low. The average thickness of the CNWs was determined
from the mean crystallite size (Lc), which is calculated from
the 002 peak width using the Scherrer equation [17]. The
values of Lc for CNWs grown without and with O2 were
estimated to be 9.4 and 13.0nm, respectively; these values
arealmostthesameasthewall thicknessesevaluatedbyTEM.
We evaluated the degree of vertical structure of the
CNWs using the ratio of the 002 peak intensity ratio to the
100/101 integrated intensity ratio (I002/I100/101). The values
of I002/I100/101 for CNWs grown without and with O2 were
found to be 4.8 and 10.4, respectively. The I002/I100/101 ratio
of CNWs grown with O2 was twice that of CNWs grown
without O2, indicating that orientation of the vertically
standing graphene sheets was improved by introducing O2
into the plasma for CNW growth with PECVD with a
C2F6/H2 mixture. A weak broad peak at 2u ¼ 22.88 was
observed only for CNWs grown without O2 when the glanc-
ing angle was 0.38. This peak was not observed at the
glancing angle of 0.058, suggesting it is related to the inter-
face phase near the substrate. This peak is attributed to b-SiC
(111), which has an amorphous structure and is formed
between the CNWs and the Si substrate. This interface peak
Figure 2 (a) Raman spectra of CNW film without and with O2
addition. (b) The peak intensity ratio of D-band to G-band as a
function of O2 gas addition.
vertically standing CNWs near the surface region, and to was not observed for the CNWs grown with O2.
evaluate the CNWs, including the interface between the A quartz substrate was used to evaluate the electrical
CNWs and the Si substrate. Figure 3 shows the diffraction properties of the CNW films. After synthesizing the CNW
patterns of CNW film samples recorded at the glancing angle films on a quartz substrate, four aluminum (Al) contacts were
of 0.38. An intense 002 Bragg peak, the plane of which is symmetrically positioned on the CNW film for Hall
normaltothesubstrate, isat2u ¼ 16.98andtherearealsoweak measurements by the van der Pauw method [18]. The Hall
100/101, 004, and 110 Bragg peaks. The interlayer spacing coefficients of the CNW film samples grown with and
without O2 were measured using a resistivity/Hall measure-
ment system. In this measurement, it was assumed that the
CNW film is a plane membrane from a macroscopic
standpoint and that the current flows uniformly on the
surface between the contacts. Before performing the Hall
measurement, we confirmed that the Al electrodes are ohmic
Figure 3 SR X-ray diffraction patterns of CNW films grown with
and without O2 addition; various conditions measured at beam line
BL13XUofSPring-8.Theinsetisamagnifiedprofileinthe2uregion Figure 4 Variation in resistivities of CNW films as a function of
of 20–508.
temperature.
www.pss-a.com
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
p s s
142
W. Takeuchi et al.: Monolithic self-sustaining nanographene sheet grown using PECVD
the carrier concentration and mobility as well as the band gap
inside the individual monolithic CNW sheets are of great
interest for realizing nanographene devices. We are currently
evaluating the electrical properties of individual CNW
sheets.
CNW films grown with additional O2 were characterized
by secondary ion mass spectrometry (SIMS) to investigate the
effect of O2 addition to the C2F6/H2 plasma on the atomic
composition of CNWs. The depth distributions of the relative
atomic composition ratios [O]/[C] and [F]/[C] were measured
using a 10 keV Csþ primary beam for sputtering. The relative
atomic compositionratios [O]/[C] inthe CNWs grown with and
without O2 were found to be almost the same at 1.6 ꢁ 10ꢀ4
based on the SIMS depth profile (data not shown), suggesting
that O atoms were not incorporated in the CNWs grown with
Figure 5 Carrier concentration of CNW films as a function of the
reciprocal of the temperature. I and II correspond to the intrinsic
region and the saturation region, respectively.
O2. Onthe other hand, the relative atomic composition ratio [F]/
[C] in the CNWs grown with O2 was 1.7 ꢁ 10ꢀ5, while it was
7.1 ꢁ 10ꢀ4 in the CNWs grown without O2, indicating that O
atoms scavenge F atoms by forming volatile COF2 molecules
contacts for the CNW film. Figure 4 shows the resistivity during the growth of CNWs [19]. The electrical conduction of
variation of the CNW films as a function of temperature, graphite with a fluorine junction (CF)n was found to be
which was derived from the Hall measurement. The disturbed since the excess charge induced by the F atoms
resistivity of CNW film was reduced by approximately reduces the graphite p density [20]. The bent multi-layered
30% as a result of O2 addition during the CNW growth graphene structure shown in Fig. 1a and c include many defects
process. The resistivity of all the CNW films decreased with as well as impurities such as residual F atoms. These residual
an increase in the measured temperature, indicating the F atoms act as impurities and F-induced defects in CNWs
semiconductor behavior of the CNW films.
will affect electrical conduction. As a result of O2 addition
Figure 5 shows the carrier concentration of CNW films to the C2F6/H2 plasma, the F atoms in CNWs were effectively
as a function of the reciprocal of the temperature. The carrier eliminated, while the O atoms were not incorporated in
concentration of the CNW films showed the flat profile at low the CNW film. This suggests that oxygen etch F atoms and
temperatures up to 100 K (1000/T ¼ 10). As the temperature small graphitic fragments, thereby contributing to the
was increased up to room temperature (1000/T: lower than higher graphitization and improving the crystallinity and
10), the carrier concentration drastically increased. The electrical conduction to form highly oriented monolithic
regions I and II in Fig. 5 indicate the intrinsic and the graphene sheets.
saturation regions, respectively. The carrier concentration
behavior of CNW film grown without O2 addition showed
4 Conclusions We synthesized CNW films composed
almost the same tendency as that of the CNW grown with O2 of monolithic self-sustaining nanographene sheets vertically
addition. Thus, the carrier generation mechanism for CNW standing on the substrate. Synchrotron X-ray surface
films grown with and without O2 is considered to be the diffraction measurements revealed that the crystallinity of
same. However, the carrier concentration of CNW film vertical graphene sheets was improved by introducing O2
grown with O2 was slightly higher than that of CNW film into the plasma used for CNW growth. The resistivity of the
grown without additional gas. The band gap can be obtained CNW film decreased with an increase in the temperature,
from the slope of the intrinsic range I in Fig. 5. Thus, the band indicating that the CNW films exhibit semiconductor
gap of CNW film was estimated from the region I at 300 K, behavior. The current results demonstrate that CNWs
which was approximately 80 meV for both CNW films consisting of vertical, monolithic self-sustaining nano-
grown without and with O2 addition. However, the intrinsic graphene sheets have great potential for application in
region was not clearly observed in Fig. 5, because the next-generation electronic devices.
temperature range for the present Hall measurement system
was limited up to room temperature. The slope in region I for
Acknowledgements The synchrotron radiation experi-
ments were conducted on the BL13XU of SPring-8 under the
the measured curve in Fig. 5 is still on the rise at 300 K.
Priority Nanotechnology Support Program administrated by JASRI
Therefore, the band gap of CNW film was expected larger
than 80 meV.
(Proposal No. 2007B1752). Part of this work was supported by a
Grant-in-Aid for Scientific Research in a Priority Area (Nos.
18063019 and 18063011) from the Japanese Ministry of Education,
The resistivity and carrier concentration derived in this
study reflect the electrical property of the bulk CNW film
Culture, Sports, Science, andTechnology. Theauthors wouldliketo
comprising the webof nanographene sheetswith interspaces.
These values could be useful for the design and evaluation of
electronic devices using bulk CNW films. On the other hand,
thank K. Nakatsu and J. Yamagata for the measurements and a
fruitful discussion. This study was partly supported by Sumika
Chemical Analysis Service, Ltd.
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.pss-a.com

Original
Paper
Phys. Status Solidi A 207, No. 1 (2010)
143
References
[11] S. Kondo, M. Hori, K. Yamakawa, S. Den, H. Kano,
and M. Hiramatsu, J. Vac. Sci. Technol. B 26, 1294
(2008).
[12] H. Chatei, M. Belmahi, M. B. Assouar, L. Le Brizoual, P.
Bourson, and J. Bougdira, Diamond Relat. Mater. 15, 1041
(2006).
[13] M. Hiramatsu, M. Inayoshi, K. Yamada, E. Mizuno, M.
Nawata, M. Ikeda, M. Hori, and T. Goto, Rev. Sci. Instrum.
67, 2360 (1996).
[14] P. L. Walker, Jr., R. L. Taylor, and J. M. Ranish, Carbon 29,
411 (1991).
[15] S. Goto, N. Ikeda, K. Inoue, H. Kimura, and M. Yabashi,
SPring-8 Beamline Handbook, Ver. 3 (Japan Synchrotron
Radiation Research Institute, Hyogo, Japan, 2004),
p. 48.
[16] Y. Sato, H. Watano, R. Hagiwara, and Y. Ito, Carbon 44, 664
(2006).
[1] K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y.
Zhang, S. V. Dubonos, I. V. Grigorieva, and A. A. Firsov,
Science 306, 666 (2004).
[2] S. Y. Zhou, G.-H. Gweon, A. V. Fedorov, P. N. First, W. A.
de Heer, D.-H. Lee, F. Guinea, A. H. Castro Neto, and A.
Lanzra, Nature Mater. 6, 770 (2007).
[3] Y. H. Wu, P. W. Oiao, T. C. Chong, and Z. X. Shen, Adv.
Mater. 14, 64 (2002).
[4] J. J. Wang, M. Y. Zhu, R. A. Outlaw, X. Zhao, D. M. Manos,
B. C. Holloway, and V. P. Mammana, Appl. Phys. Lett. 85,
1265 (2004).
[5] N. G. Shang, F. C. K. Au, X. M. Meng, C. S. Lee, I. Bello, and
S. T. Lee, Chem. Phys. Lett. 358, 187 (2002).
[6] M. Hiramatsu, K. Shiji, H. Amano, and M. Hori, Appl. Phys.
Lett. 84, 4708 (2004).
[7] M. Hiramatsu and M. Hori, Jpn. J. Appl. Phys. 45, 5522 (2006).
[8] W. Takeuchi, M. Ura, M. Hiramatsu, Y. Tokuda, H. Kano,
and M. Hori, Appl. Phys. Lett. 92, 213103 (2008).
[9] K. Kobayashi, M. Tanimura, H. Nakai, A. Yoshimura, H.
Yoshimura, K. Kojima, and M. Tachibana, J. Appl. Phys.
101, 094306 (2007).
[10] S. Kurita, A. Yoshimura, H. Kawamoto, T. Uchida, K.
Kojima, M. Tachibana, P. Molina-Morales, and H. Nakai,
J. Appl. Phys. 97, 104320 (2005).
[17] A. Sharama, T. Kyotani, and A. Tomita, Carbon 38, 1977
(2000).
[18] L. J. van der Pauw, Philips Tech. Rev. 20, 220 (1958).
[19] H.-P. Hsueh, R. T. McGrath, B. J. Felker, B. S. Langan, J. G.
Langan, and E. J. Karwacki, J. Vac. Sci. Technol. B 19, 1346
(2001).
[20] J.-C. Charlier, X. Gonze, and J.-P. Michenaud, Phys. Rev. B
47, 16162 (1993).
www.pss-a.com
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim