Large-area graphene single crystals grown by low-pressure chemical vapor deposition of methane on copper

Article abstract of DOI:10.1021/ja109793s

Graphene single crystals with dimensions of up to 0.5 mm on a side were grown by low-pressure chemical vapor deposition in copper-foil enclosures using methane as a precursor. Low-energy electron microscopy analysis showed that the large graphene domains had a single crystallographic orientation, with an occasional domain having two orientations. Raman spectroscopy revealed the graphene single crystals to be uniform monolayers with a low D-band intensity. The electron mobility of graphene films extracted from field-effect transistor measurements was found to be higher than 4000 cm² V^-1 s^-1 at room temperature.

Full text of DOI:10.1021/ja109793s

COMMUNICATION
pubs.acs.org/JACS
Large-Area Graphene Single Crystals Grown by Low-Pressure
Chemical Vapor Deposition of Methane on Copper
Xuesong Li,† Carl W. Magnuson,† Archana Venugopal,‡ Rudolf M. Tromp,§ James B. Hannon,§
Eric M. Vogel,‡ Luigi Colombo,*,|| and Rodney S. Ruoff*,†
†Department of Mechanical Engineering and the Texas Materials Institute, The University of Texas at Austin,
1 University Station C2200, Austin, Texas 78712-0292, United States
‡Department of Electrical Engineering, The University of Texas at Dallas, Richardson, Texas 75080, United States
§IBM T. J. Watson Research Center, Yorktown Heights, New York 10598, United States
Texas Instruments, 13121 TI Boulevard, Dallas, Texas 75243, United States
25 μm thick copper foil and then crimping the three remaining
ABSTRACT: Graphene single crystals with dimensions of
up to 0.5 mm on a side were grown by low-pressure chemical
vapor deposition in copper-foil enclosures using methane as
a precursor. Low-energy electron microscopy analysis
showed that the large graphene domains had a single
crystallographic orientation, with an occasional domain
having two orientations. Raman spectroscopy revealed the
graphene single crystals to be uniform monolayers with a
low D-band intensity. The electron mobility of graphene
films extracted from field-effect transistor measurements
was found to be higher than 4000 cm2 V-1 s-1 at room
temperature.
sides. The basic growth conditions were similar to those pre-
viously reported2,3 but employed slightly lower methane flow
rates and partial pressures (less than 1 sccm and 50 mTorr,
respectively). Graphene grew on both the inside and outside of
the Cu enclosure. The graphene growth on the outside showed
behavior similar to the graphene growth reported by Li et al.,3 but
there was one difference: at the lower partial pressure and flow
rate and at much longer growth times, a higher density of bilayers
and trilayers was observed. This will be reported in future work.
In contrast, the growth on the inside showed a much lower
density of nuclei followed by very large domain growth after
extended periods of time (>1 h) and a much lower density of
adlayers. At this time, the precise growth conditions inside the
enclosure are not well understood. However, we believe that the
low density of nuclei is due to the much lower partial pressure of
methane and an “improved” environment during growth; that is,
the Cu vapor is in static equilibrium, and there is potentially a
much lower pressure of unwanted species in our non-ultrahigh
vacuum system. Figure 2 shows the average domain branch
length (about half the domain size from the center of the
domain) as a function of growth time for two methane flow
rates, 0.5 and 1.3 sccm, which correspond to methane partial
pressure of 8 and 21 mTorr, respectively. During the growth
process, the hydrogen flow rate was kept constant at 2 sccm with
a partial pressure of 27 mTorr, and the chamber background
pressure was 17 mTorr. The graphene domains were very large,
as shown by scanning electron microscope (SEM) images
(Figure 3a). The domains also tended to have high “edge
roughness”. The shape of the graphene nuclei in the initial stages
of growth showed a hexagonal symmetry (Figure 3b). At first, the
graphene domains grew as six-sided polygons, and these even-
tually grew into very large graphene domains with growing edges
resembling dendrites (Figure 3c).
raphene growth by chemical vapor deposition (CVD) has
G
been receiving significant attention recently because of the
ease with which large-area films can be grown, but the growth of
large-domain or large-grain-size single crystals has not been
reported to date.1 In our earlier work, we found that growth of
graphene on Cu by CVD occurs predominantly by surface
nucleation followed by a two-dimensional growth process, but
the domain size was limited to a few tens of micrometers.2,3 The
presence of domain boundaries has been found to be detrimental
to the transport properties; the precise mechanism of the degrada-
tion still remains elusive, but what is known is that structural
defects promote surface reactions with adsorbates from the
ambient or with deposited dielectrics.3 The presence of heptagons
and pentagons in the network of hexagons has been observed
experimentally, and first-principles quantum-transport calculations
have predicted that the periodicity-breaking disorder can adversely
affect transport properties.4,5 Any of these defects can give rise to
higher surface chemical activity that would further disrupt the sp2-
bonding nature of graphene and thus impact graphene’s funda-
mental properties. Therefore, it is imperative that large single
crystals of graphene be grown to minimize the presence of defects
arising from boundaries between misoriented domains. In this
communication, we report a very low pressure CVD process that
yields graphene with domains of up to 0.5 mm in size, which is a
factor of ∼30 times larger than previously reported.3
We also employed the carbon isotope-labeling technique to
delineate the graphene growth front in order to establish the
boundaries between the growing “lobes”, the time dependence,
and the spatial dependence.2 In these experiments, the graphene
films were grown by alternating the flow of 12CH4 and 13CH4
every 10 min for a total of 90 min at a flow rate of 0.5 sccm and a
corresponding partial pressure of 8 mTorr at 1035 °C. An
The large-domain graphene growth was observed on the inside
of a copper-foil enclosure at high temperature (∼1035 °C). The
copper-foil enclosure (Figure 1a) was formed by bending a
Received: November 1, 2010
Published: February 10, 2011
r
2011 American Chemical Society
2816
dx.doi.org/10.1021/ja109793s J. Am. Chem. Soc. 2011, 133, 2816–2819
|

Journal of the American Chemical Society
COMMUNICATION
Figure 1. (a) Copper foil enclosure prior to insertion in the furnace.
(b) Schematic of the CVD system for graphene on copper.
Figure 3. SEM images of graphene on copper grown by CVD. (a)
Graphene domain grown at 1035 °C on Cu at an average growth rate of
∼6 μm/min. (b) Graphene nuclei formed during the initial stage of
growth. (c) High-surface-energy graphene growth front shown by the
arrow in (a).
Figure 2. Graphene growth inside the enclosure as a function of time
for two methane flow rates and partial pressures at 1035 °C.
analysis of the images in Figure 4 shows that the graphene growth
rate was higher at the tips of the lobes (∼1.2 μm/min; region 3b)
and lower between the lobes and near the end of the growth
process (0.1 μm/min; region 7b), as also seen by the spatial
coverage variation of graphene. Furthermore, graphene covered
the copper surface and closed onto itself as the growth front
advanced. The growth rate at points where the graphene joined
was lower than at the tip of the growing front. Figure 4 also shows
the shape of the growth front at different times during growth, as
delineated by the boundaries between the domains of 12C- and 13C-
based graphene. The structure of the domains, their distribution,
and the graphene growth-front structure provide further indications
that the growth was surface-mediated, as previously reported.2
These data also show that the growth was a result of surface growth
on the inside of the enclosure rather than diffusion through the Cu
from outside the enclosure followed by precipitation upon cooling.
In order to probe the domain size of the graphene, we made
spatially resolved electron diffraction measurements using low-
energy electron microscopy (LEEM).6 Using photoelectron
emission microscopy (PEEM), we located the edge of a graphene
domain (Figure 5a). We then recorded the diffraction pattern
from 2 μm areas of the surface. Diffraction from the substrate
revealed a highly faceted, rough Cu(100) surface with sharp
diffraction spots, while diffraction from the graphene was diffuse
(Figure 5b-e). The diffuse pattern was similar to that for
Figure 4. Raman map of the G bands corresponding to 12C (yellow)
and 13C (black). The numbers in the figure correspond to the relative
methane cycle numbers.
diffraction from free-standing graphene,7 perhaps indicating a
weak coupling to the rough substrate. To estimate the domain
size, we recorded the diffraction pattern as the sample was
2817
dx.doi.org/10.1021/ja109793s |J. Am. Chem. Soc. 2011, 133, 2816–2819

Journal of the American Chemical Society
COMMUNICATION
Figure 6. (a) D-band and (b) G-band Raman maps of graphene within
the domain and at the edges of the growing domain and (c) Raman
spectra of large-domain graphene within the bulk of the film and along
the dendrite. The FWHM of the 2D band of the dendrite was slightly
smaller than that of the bulk, and the ratio of the intensities of the 2D and
G bands was larger for the dendritic region than the bulk, suggesting a
lower carrier concentration for the dendritic region within the “bulk” of
the domainand thetip of the dendrite. The 2D peak FWHM was 38 cm-1
for the bulk and 32 cm-1 for the dendrite, whereas the G-band FWHM
was ∼23 cm-1 for both regions.
Figure 5. (a) PEEM image recorded near a graphene domain. The black
line indicates the path from which selected-area diffraction patterns were
recorded. The graphene is bright, and the surrounding Cu foil is dark.
(b-e) Electron diffraction patterns (33 eV) recorded from 2 μm areas of
the graphene. Each pattern is labeled by the position along the black line
in (a) at which the pattern was recorded. The diffraction spots are
indicated in (b). (f) PEEM image recorded at the end of the scan.
the 2D band to the G band suggests that the carrier concentration
was different for the two regions.9 This variation will have to be
investigated further.
The quality of the large-area-domain films was also evaluated by
measuring the transport properties of the graphene films transferred
onto silicon dioxide grown on Si wafers. Field-effect transistors were
fabricated using nickel for the source and drain contacts and the
highly doped Si substrate as the back-gate contact. The resistance
was measured at room temperature as a function of back-gate
voltage, and the mobility was extracted using the methodology
introduced by Kim et al.10 The mobility for these large-domain films
was found to be greater than 4000 cm2 V-1 s-1, which is reasonably
high but not as high as the highest value for exfoliated films, thus
suggesting that the films still need improvement.
scanned under the beam along a line at a constant speed of
15 μm/s, and Figure 5 shows the PEEM images and diffraction
patterns from selected points along the line scan. Each pattern is
labeled by the position from which it was recorded.
While there were slight continuous rotations of the pattern due
to waviness in the foil, there were no discontinuous changes in the
orientation of the diffraction spots, suggesting that along this scan
of over 400 μm, there were no rotational domain boundaries. We
performed similar measurements on a number of domains. On
occasion, we observed large (>50 μm) domains with a 30 degree
relative rotation of the graphene lattice, but in most of the scans,
no rotational domain boundaries were observed.
’ AUTHOR INFORMATION
Corresponding Author
colombo@ti.com; r.ruoff@mail.utexas.edu
After growth, the graphene films were transferred to SiO2/Si
substrates as described by Li et al.8 in order to analyze the films by
Raman spectroscopy and perform electrical measurements.
Figure 6 shows Raman maps of the D band (Figure 6a) and G
band (Figure 6b) and Raman spectra recorded at two different
regions of the domain (Figure 6c), one within the film and the
other close to the dendrite edge. The spectra show that the
growing material was indeed graphene, with a low D band
intensity across the domain and the presence of graphene only.
The full width at half-maximum (FWHM) of the G band was about
the same for the two regions (23 cm-1), and the intensity ratio of
’ ACKNOWLEDGMENT
The authors appreciate the support from the Nanoelectronic
Research Initiative—SouthWest Academy of Nanoelectronics
(NRI-SWAN), NSF Grant 1006350, and the Office of Naval
Research.
’ REFERENCES
(1) Li, X.; Cai, W.; An, J. H.; Kim, S.; Nah, J.; Yang, D.; Piner, R.;
Velamakanni, A.; Jung, I.; Tutuc, E.; Banerjee, S. K.; Colombo, L.; Ruoff,
R. S. Science 2009, 324, 1312–1314.
2818
dx.doi.org/10.1021/ja109793s |J. Am. Chem. Soc. 2011, 133, 2816–2819

Journal of the American Chemical Society
COMMUNICATION
(2) Li, X.; Cai, W.; Colombo, L.; Ruoff, R. S. Nano Lett. 2009, 9,
4268–4272.
(3) Li, X.; Magnuson, C. W.; Venugopal, A.; An, J.; Suk, J. W.; Han,
B.; Borysiak, M.; Cai, W.; Velamakanni, A.; Zhu, Y.; Fu, L.; Vogel, E. M.;
Voelkl, E.; Colombo, L.; Ruoff, R. S. Nano Lett. 2010, 10, 4328–34.
(4) Huang, P. Y.; Ruiz-Vargas, C. S.; Zande, A. M.; Whitney, W. S.;
Garg, S.; Alden, J. S.; Hustedt, C. J.; Zhu, Y.; Park, J.; McEuen, P. L.;
Muller, D. A. Nature 2011, 469, 389–392.
(5) Yazyev, O. V.; Louie, S. G. Nat. Mater. 2010, 9, 806–809.
(6) Bauer, E. Rep. Prog. Phys. 1994, 57, 895–938.
(7) Knox, K. R.; Wang, S.; Morgante, A.; Cvetko, D.; Locatelli, A.;
Mentes, T. O.; Nino, M. A.; Kim, P.; Osgood, R. M. Phys. Rev. B 2008,
78, No. 201408.
(8) Li, X.; Zhu, Y.; Cai, W.; Borysiak, M.; Han, B.; Chen, D.; Piner,
R. D.; Colombo, L.; Ruoff, R. S. Nano Lett. 2009, 9, 4359–4363.
(9) Casiraghi, C.; Pisana, S.; Novoselov, K. S.; Geim, A. K.; Ferrari,
A. C. Appl. Phys. Lett. 2007, 91, No. 233108.
(10) Kim, S.; Nah, J.; Jo, I.; Shahrjerdi, D.; Colombo, L.; Yao, Z.;
Tutuc, E.; Banerjee, S. K. Appl. Phys. Lett. 2009, 94, No. 062107.
2819
dx.doi.org/10.1021/ja109793s |J. Am. Chem. Soc. 2011, 133, 2816–2819