Subscriber access provided by UNIV OF CAMBRIDGE
The Journal of Physical Chemistry C is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.
Article
The Parameter Space of Graphene CVD on Polycrystalline Cu
Piran Ravichandran Kidambi, Caterina Ducati, Bruno Dlubak, Damian Gardiner, Robert
S Weatherup, Marie-Blandine Martin, Pierre Seneor, Harry Coles, and Stephan Hofmann
J. Phys. Chem. C, Just Accepted Manuscript • Publication Date (Web): 27 Sep 2012
Downloaded from http://pubs.acs.org on September 28, 2012
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
1 
 
The Parameter Space of Graphene CVD on Polycrystalline Cu 
Piran R. Kidambi
1
, Caterina Ducati
2
, Bruno Dlubak
1
, Damian Gardiner
1
, Robert S. 
Weatherup
1
, Marie-Blandine Martin
3
, Pierre Seneor
3
, Harry Coles
1
 and Stephan Hofmann
1,
  
1
Department of Engineering, University of Cambridge, Cambridge CB3 0FA, UK 
2
Department of Materials Science and Metallurgy, University of Cambridge, Cambridge CB2 3QZ, UK 
3
Unité Mixte de Physique CNRS/Thales, 91767 Palaiseau, France and Université de Paris-Sud 11, 91405 Orsay, 
France 
 
Abstract 
A systematic study on the parameter space of graphene CVD on polycrystalline Cu foils is 
presented, aiming at a more fundamental process rationale in particular regarding the choice 
of carbon precursor and mitigation of Cu sublimation. CH4 as precursor requires H2 dilution 
and temperatures ≥1000°C to keep the Cu surface reduced and yield a high quality, complete 
monolayer graphene coverage. The H2 atmosphere etches as-grown graphene, hence 
maintaining a balanced CH4/H2 ratio is critical. Such balance is more easily achieved at low 
pressure conditions, at which however Cu sublimation reaches deleterious levels. In contrast, 
C6H6 as precursor requires no reactive diluent and consistently gives similar graphene quality 
at 100-150°C lower temperatures. The lower process temperature and more robust processing 
conditions allow the problem of Cu sublimation to be effectively addressed. Graphene 
formation is not inherently self-limited to a monolayer for any of the precursors. Rather, the 
higher the supplied carbon chemical potential the higher the likelihood of film inhomogeneity 
and primary and secondary multilayer graphene nucleation. For the latter, domain boundaries 
of the inherently polycrystalline CVD graphene offer pathways for a continued carbon supply 
to the catalyst. Graphene formation is significantly affected by the Cu crystallography, i.e. the 
evolution of microstructure and texture of the catalyst template form an integral part of the 
CVD process. 
 
Keywords: mono and few-layer graphene (M/FLG), chemical vapor deposition (CVD), 
copper (Cu), polycrystalline, methane (CH4), benzene (C6H6)       

   email: sh315@cam.ac.uk   
Page 1 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2 
 
Introduction 
Economic, large-area growth combined with viable front and back-end integration strategies 
of mono and few-layer graphene (M-/FLG) are key requirements for the commercial 
exploitation of graphene's unique properties. Chemical vapor deposition (CVD) is the most 
promising route towards M-/FLG production and integration, based on its versatility and 
success with other nanomaterials.
1–3
 While progress has been made in achieving MLG CVD 
over large areas,
4,5
 the underlying growth mechanisms have yet to be fully understood
6–8
 and 
the often narrow empirical process optimizations allow little generalization due to the vast 
CVD parameter space.
4,5,9–13
 Most current literature focuses on exposing polycrystalline Cu
4,5
 
foils to methane (CH4) at low pressures (LP) and high temperatures (≥1000°C). As-grown 
graphene can be fully continuous, but is inherently polycrystalline,
10
 with MLG domain sizes 
typically ≤ 5µm in dimension.
10–12
 Recent efforts have focused on increasing the MLG 
domain size,
12,13
 but in general the compromise made to achieve high quality CVD graphene 
is to face undesirably high levels of Cu sublimation.
14
  
Here, we focus on understanding graphene formation on polycrystalline Cu foils via a 
systematic exploration of the wider CVD parameter space, in particular regarding the choice 
of carbon precursor and mitigation of Cu sublimation, aiming at more rational process design. 
For CH4 as precursor we find that, in agreement with previous literature, 
4,5,9–12
 uniform, high 
quality MLG growth is restricted to a rather narrow CVD parameter set of LP conditions, H2 
dilution and temperatures ≥1000°C, at which Cu sublimation is at deleterious levels. The H2 
atmosphere is required to keep the Cu surface reduced, but at the same time can etch as-
grown graphene. Hence maintaining a balanced CH4/H2 ratio is critical, which makes the CH4 
based CVD process so delicate. In contrast, we find that benzene (C6H6) as precursor requires 
no reactive diluent, i.e. no delicate balance to be maintained, and consistently gives similar 
graphene quality at 100-150°C lower temperatures compared to CH4 based CVD. The lower 
process temperature and more robust processing conditions allow the problem of Cu 
sublimation to be effectively addressed. Our growth study shows that Cu catalyzed CVD 
graphene formation is not inherently self-limited to a mono-layer. Rather we find the 
nucleation density, percentage of multi-layer nuclei and film uniformity/quality to critically 
depend on CVD conditions and growth kinetics. We suggest that the domain boundaries of 
the inherently polycrystalline CVD graphene offer pathways for the precursor to reach the 
catalyst even after complete MLG coverage. Our data further emphasizes that the Cu catalyst 
Page 2 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3 
 
template is not static and that the involved kinetics of grain growth are highly process 
dependent, making this an important process step for controlled graphene CVD.  
 
Experimental Methods  
Graphene synthesis is carried out in a customized cold-wall, low pressure CVD reactor 
(LPCVD, a heavily modified Aixtron BM3, base pressure ~ 5×10
-6
 mbar) and a hot-wall, 
atmospheric pressure furnace (APCVD). For LPCVD the total pressure (0.001 – 100 mbar) 
was regulated by a pressure controller at the reactor outlet. Commercial, cold-rolled Cu foils 
of different thicknesses and purities (Alfa Aesar Puratronic 99.999% purity, 25 µm and 
100µm thick; Advent Research Materials, 99.995% purity, 12 µm thick) are used as catalysts. 
A one step CVD recipe is used as benchmark process for all systems. For all CVD recipes, 
heating up and pre-annealing is carried out in H2 at 1000°C (LPCVD 4 mbar total pressure, 
heating rate ~250°C/min to 800°C followed by 50°C/min to 1000°C, APCVD heating rate ~ 
40°C/min), after which the temperature is stabilized at the chosen growth temperature. In the 
case of CH4 as the precursor, CH4 is added to the annealing gas and cooling is performed in 
pure Ar (LPCVD cooling ramp ~150°C/min to 400°C at 7 mbar, APCVD cooling rate 
~30°C/min). For C6H6 LPCVD, the exposure is to C6H6 (>99.7% purity, Sigma-Aldrich) 
without H2 and cooling is performed in vacuum.  
Samples were characterized by scanning electron microscopy (SEM, Philips XL30, 1-2 kV) 
and Raman spectroscopy (Renishaw InVia spectrometer, 532 nm excitation). For the latter 
the M-/FLG was typically transferred to SiO2(300 nm)/Si substrates, using a 
polymethylmethacrylate (PMMA) support layer and a 0.5 M aqueous solution of FeCl3 to 
etch the Cu foil. Acetone was then used to dissolve the PMMA support. Hall-bar devices 
were fabricated via e-beam lithography. Graphene layers, transferred to SiO2/Si wafer 
substrates, were etched by an O2 plasma and Au/Ti contacts evaporated on top. All electrical 
measurements were performed at room temperature. Electron backscattered diffraction 
(EBSD) experiments were performed in a FEI Helios Dual beam microscope (5-15 kV, 
current ~5.5 nA, working distance ~5-6.5 mm and sample tilt of ~ 60° with respect to the 
electron beam) with an Oxford Instruments - HKL EBSD Nordlys II detector in spot mode 
using Channel 5 software. Birefringence measurements followed the method outlined by Kim 
et al.
15
 Liquid crystals of 4-pentyl-4`-cyanobiphenyl (5CB, Merck GmbH) were drop cast 
Page 3 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4 
 
onto graphene transferred to a glass substrate. A thin cover slip was added on top and 
measurements were carried out with the as-prepared sample placed between 2 crossed 
polarizers on a rotatable stage.  
 
Results  
The process of graphene formation on metal surfaces comprises nucleation, a subsequent 
expansion of the nuclei into domains, followed by a merging of the domains into a 
continuous covalently bonded film.
16
 Here we refer to domains as regions that grow from a 
single nucleation point. Figure 1 shows SEM images of the Cu surface after short CH4/H2 
exposures (see process details in caption), i.e. the early stages of CVD before the graphene 
coverage is continuous. Clear differences can be already seen for the different CVD 
conditions. Figs. 1 a,b and 1 c,d compare low (1:10) and high (1:1) CH4/H2 ratios for LPCVD 
conditions. For the former we observe MLG domains ranging from typically ~30-40µm
2
 in 
size, with a few isolated areas (2-4µm
2
) of FLG (as seen by SEM contrast and confirmed by 
Raman spectroscopy, see below). This is largely consistent with recent literature on 
optimized CVD with CH4 on Cu foils.
4,5,9–13
 We note that, due to the pressure regulation and 
backfilling procedure, the CH4/H2 ratio for our process is initially lower at the point of CH4 
addition, and thus partly resembles two step exposures reported in literature.
17
 A higher CH4 
partial pressure (Fig. 1 c,d) leads to predominantly multilayer graphene nucleation and 
decreased sample homogeneity. This highlights that at the initial stages of growth, which we 
refer to the as primary nucleation stage, Cu is not inherently limiting graphene formation to a 
monolayer.  
Figures 1 e-j highlight that the nature of graphene nucleation and growth at the early stages is 
highly dependent on catalyst surface orientation and impurity levels. Figs. 1 e,f show that for 
the same CVD conditions the resultant graphene coverage is different on adjacent Cu facets, 
with the imaged Cu (111) surface showing less graphene coverage compared to the Cu (110) 
surface. APCVD conditions as in Figs. 1 g,h and 1 i,j result in predominantly multilayer 
nucleation pattern. The nucleation density is notably increased and the shape of nuclei 
significantly changed for a Cu foil of lower purity (Figs. 1 i,j) at otherwise identical APCVD 
conditions. The FLG nuclei preferentially decorate Cu grain boundaries and appear aligned 
along the rolling striations of the foil. We find that a variation from 25-100 µm in foil 
Page 4 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5 
 
thickness does not appear to influence the kinetics of graphene formation on that scale (Figs. 
1 c-f), but the foil thickness does influence the Cu grain growth kinetics as discussed in the 
following.  
Figure 2 shows the results of EBSD analysis marked across SEM images of Cu foil surfaces 
for various process stages and conditions, highlighting the effects of Cu recrystallization and 
grain growth. Rolling striations are a dominant feature of the as-received Cu foils, for which 
EBSD shows Cu grain sizes < 2 µm with diverse surface orientations (Fig. 2 a, Table S1). We 
find the detailed deformation texture of the used commercial cold-rolled Cu foils to vary, 
despite being advertised as the same product. Hence the starting point cannot be 
automatically assumed as constant. It should be noted that here we do not use any additional 
Cu surface treatment procedure, such as electropolishing.
18
 After annealing in H2 at 1000°C, 
the Cu grain sizes increase to ~50 µm - 500 µm for APCVD and ~50 µm - 2 mm for LPCVD 
conditions (Figs. 2 b,c), which exceed the foil thickness. Whilst for APCVD we still find a 
crystallographically diverse Cu surface (Fig. 2b), the texture after LPCVD annealing 
becomes (111) dominated (Fig. 2c). The surface topography appears rougher and stepped for 
LPCVD conditions (Fig. 2c inset, with individual steps of ~10-50 nm) compared to the 
relatively smooth surface seen for APCVD conditions. The Cu grain size and orientation 
distributions are similar before and after hydrocarbon exposure (Figs. 2 d,e), most notably a 
(111) dominated texture for LPCVD conditions is maintained and several orientations are 
seen for APCVD.A polycrystalline material has no equilibrium structure, but depending on 
processing reaches a metastable equilibrium where the total grain boundary energy is locally 
minimized. At the given conditions, recrystallization followed by normal and abnormal grain 
growth are expected
19
 and Fig. 2 is consistent with that. Our data emphasizes that the Cu 
catalyst template is not static and that the involved kinetics of grain growth are highly process 
dependent, making this an important process step for controlled graphene CVD.  
We note that most literature directed towards the optimization of uniform MLG CVD focuses 
on low pressure conditions in the mbar range
9,17
 during H2 pre-treatment and CH4/H2 
exposure, similar to our  LPCVD conditions of Figs. 1 a,b and 2e. Based on their prevalence, 
we choose these LPCVD conditions as the standard "reference" for our further parametric 
study. Our discussion below will highlight why achieving continuous MLG films based on 
APCVD is very challenging using CH4 as precursor. Figure 3 highlights the quality of 
graphene grown at our "reference" CH4 based LPCVD conditions. Fig. 3a shows an optical 
Page 5 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6 
 
image of as-grown MLG transferred to a SiO2 (300nm)/Si substrate. A corresponding Raman 
spectrum in Fig. 3b with G (~1600 cm
–1
, FWHM ~ 23-25 cm
–1
), D (~1360 cm
–1
) and 2D 
(~2700 cm
–1
, FWHM ~35-37 cm
–1
 which can be fitted with a single Lorentzian function) 
peaks as well as ratios of I2D/IG ~3.5 and ID/IG ~0.05 demonstrate the high quality of the 
MLG.
3,19
  Figs. 3 c, e show Raman I2D/IG and ID/IG maps of the MLG (dimensions 50 µm × 
50 µm), respectively, along with the corresponding distribution statistics (Figs. 3 d,f). The 
maps show average values of I2D/IG ~ 3.5 and ID/IG ~ 0.05 over a large area. 6 contact Hall 
geometry devices (Fig. 3g) give sheet resistances (on SiO2 support) in the range 400-800 Ω/□ 
and mobilities in the 2000-3000 cm
2
V
-1
s
-1
 range (with a p doping of few 10
12
 cm
-2
). In order 
to characterize sample uniformity and polycrystallinity over larger areas, we validate the 
potential of a liquid crystal based polarizing optical microscopy technique as recently 
reported by Kim et al.
15
 Figs. 3 h,i show polarizing optical microscopy (POM) images of a 
empty control and our reference MLG, respectively. The POM contrast is based on sample 
interactions with a 5CB nematic liquid crystal (See Supplementary Information). Across a 1 
cm
2
 MLG area POM indicates feature sizes ranging from ~40 µm
2
 to a few hundred µm
2
, 
which is consistent with the observed nuclei sizes of ~30-40 µm
2
 in Fig. 1 and their 
subsequent merging. We note that all characterization above is done after graphene transfer, 
i.e. includes possible degradation incurred during transfer. 
 
Whereas "optimized" CVD parameters can be highly system specific, we note that the 
variation of key parameters over a wide range offers fundamental insights into the growth 
process and allows the establishment of more generic growth guidelines. Fig. 4 shows results 
of our systematic exploration of the wider CVD parameter space and focuses on the effects of 
total pressure, growth time and hydrocarbon partial pressure, where for each experiment only 
one specified parameter was varied starting from the LPCVD benchmark recipe (see Fig. 3). 
The results are presented in terms of optical images of as-grown graphene films transferred to 
SiO2 (300nm)/Si (Fig. 4 a-f) and corresponding Raman spectra (Fig. 4g). For a lower total 
pressure of 1 mbar only MLG and no FLG is observed, but for the given exposure the film 
has large holes (Fig. 4a). This is indicative of a lower graphene nucleation density and growth 
rate. We note that rather than focusing only on the carbon precursor and carbon addition, also 
competing etching processes, e.g. by H2 or H2O, have to be considered.
21,22
 We clearly 
observe that as-grown graphene on Cu is etched while annealing in a H2 atmosphere, which is 
the main reason why we do not use hydrogen during cool down, in contrast to other studies.
23
 
Page 6 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7 
 
We note that although this etching occurs in the presence of H2 (at sufficiently high partial 
pressures), it may also arise from residual water or oxygen contamination.
8
 8 mbar of total 
pressure on the other hand leads to a significant increase in FLG nucleation and film 
inhomogeneity. An analogous behavior is seen for an increase in the CH4 partial pressure 
(Fig. 4 e,f; Fig. 3a), where a 1:10 CH4:H2 results in incomplete MLG coverage and a 1:1 ratio 
shows significant multilayer coverage. In general, the lower the carbon precursor pressure the 
lower the likelihood of achieving complete MLG surface coverage. 
Figs. 4 c,d combined with the reference sample in Fig. 3a show the effect of growth time: the 
longer the growth time the more complete the graphene coverage. Extended exposures, 
however, increase the fractional multilayer coverage. Importantly we find that new graphene 
layers can nucleate after the completion of a monolayer. We refer to this as secondary 
nucleation, as compared to the primary nucleation discussed above. In this context, we note 
that in Figs. 4 b, d, f the nucleation pattern of multilayered graphene appears to follow the 
rolling striations of the Cu foil. Fig. 4g shows that the interpretation of optical contrast in 
Figs. 4 a-f is in full agreement with measured Raman spectra corresponding to mono, bilayer 
(I2D/IG~1) and multilayer (I2D/IG<1) graphene. Further, the Raman measurements in Fig. 4g 
also confirm the interpretation of SEM contrast regarding multilayer graphene primary 
nucleation for APCVD conditions in Figs. 1e-h. The bilayer and FLG seen in areas of Fig. 
4d-f show Raman spectra (Fig. 4g) corresponding to turbostratic graphene with 2D peaks that 
can be fitted with single Lorentzian peaks, whereas the APCVD conditions shows a Raman 
signature consistent with Bernal stacking.
20
  
Figure 5 shows the results of LPCVD in undiluted CH4 in order to assess the role of hydrogen 
dilution during growth in more detail. The post-growth Cu surface is dominated by triangular 
and other 3 lobed structures (Fig. 5 a-d), partly resembling a Sierpinski triangle like fractal 
pattern. An EBSD analysis (Fig. 2f) shows a predominant (111) texture for the processed Cu 
foil, whereby the triangles and lobed structures formed on Cu (111) facets and are not present 
on Cu (001). Attempts to transfer the structures to SiO2/Si substrates resulted in small 
discontinuous patches of graphene and residual PMMA. Raman confirms the transferred 
patches as MLG (Fig. 5f). Importantly, Raman spectra (457 nm excitation) measured directly 
on the processed Cu foil show peaks corresponding to CuO (300 and 652 cm
-1
)
24,25
and CuO2  
(217, 415, 504, 808 cm
-1
)
24,25
 for the triangular areas. The mechanisms of this self-
organization are unclear, but we suggest it arises based on the balance of three competing 
Page 7 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8 
 
processes, namely graphene formation, its etching by residual oxygen and the formation of 
copper oxide from this residual oxygen. The parallel processes of reduction and oxidation 
could make the copper oxide species extremely mobile causing them to self-align in a 
triangular fashion due to the 3 fold symmetry of Cu (111).
26,27
 Hence, our data shows that the 
presence of a hydrogen atmosphere suppresses the formation of Cu oxide from trace oxygen 
contamination during CVD. We note that the observation of triangular graphene on Cu(111) 
has been reported in recent literature,
26 
without however considering the role of oxygen. This 
highlights why CH4 as precursor requires H2 dilution and our data above emphasized the 
delicate effects of the CH4/H2 balance. 
Fig 2h shows that for LPCVD conditions at 1000°C the rate of Cu sublimation is significant 
and deleterious. Increasing the total pressure with an inert diluent can suppress the Cu 
sublimation, but, as discussed below, maintaining the CH4/H2 balance and achieving 
complete MLG coverage then becomes increasingly challenging.
28
 The exponential variation 
of vapor pressure with temperature strongly motivates a temperature reduction to mitigate the 
Cu sublimation. Figs. 6 a-d show the effect of lowering the process temperature for CH4 
based graphene CVD on Cu. Comparing graphitic films grown at 1000, 900 and 800°C, the 
optical images of the transferred films all appear homogeneous (Fig. 6 a-c). The 
corresponding Raman spectra, however, show a significant deterioration in graphene quality, 
as highlighted by the significantly increased D peak intensity. The spectrum for 900°C shows 
ratios of I2D/IG of ~2.5 and ID/IG ~0.35. Also observed is the emergence of an additional 
defect peak, referred to as the D’ peak,
29
 near the G peak at higher wave numbers. A further 
reduction in temperature to 800°C leads to I2D/IG ~1.2 and ID/IG > 2 and a D’ peak intensity 
increased to the level of the IG peak. We also note that all these films are continuous, in 
contrast to recent literature that claims that no continuous films can be obtained below 
1000°C at comparable experimental conditions.
23
 Whereas the parameters discussed in Fig. 4 
mainly influence the M/FLG ratio and coverage, growth temperature is clearly the most 
significant parameter influencing the crystalline quality of the as-grown material. Our data 
shows that with CH4 as precursor the growth temperature cannot be lowered sufficiently to 
mitigate Cu sublimation and at the same time maintain a high graphene quality.  
This raises the question of what fundamentally determines at how low a temperature can high 
quality graphene CVD be grown? We approach this question here by using benzene (C6H6) as 
an alternative carbon precursor. Fig. 6e shows the results of LPCVD, for which a simple 
Page 8 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9 
 
exposure to undiluted C6H6 was adopted. At 900°C, highly uniform MLG films of high 
quality (ID/IG ~0.06, Fig. 6e) are achieved with greatly reduced Cu sublimation compared to 
1000°C. Analogous to Fig. 6d, the graphene quality decreases with decreasing growth 
temperature (Fig. 6 e). However the MLG quality for C6H6 based CVD is better at any given 
temperature (compare Figs. 6 d and e), and unlike for CH4, graphitic material (albeit highly 
defective) nucleates at temperatures as low as 600°C for C6H6. Fig. 6f directly compares the 
measured ID/IG ratios for the two different carbon precursors. Raman maps for the C6H6 
derived MLG 900°C show a uniform ID/IG ~0.06 and I2D/IG ~ 2.5 distribution over large area 
(dimensions 50 µm × 50 µm) as seen in Figs 7a-d. POM indicates a similar grain size 
distribution and polycrystallinity for the C6H6 derived MLG films (Figs. 7 i,j) as compared to 
the reference samples for CH4/H2 based CVD (Figs. 3 h,i). Further, 6 contact Hall geometry 
devices based on graphene grown from C6H6 at 900°C give sheet resistances (on SiO2 
support) in the range 400-800 Ω/□ and mobilities in the 2000-3000 cm
2
V
-1
s
-1
 range (with a p 
doping of few 10
12
 cm
-2
). These results highlight that C6H6 enables similar graphene quality 
at 100-150°C lower temperatures. i.e. that the apparent low temperature limit is precursor 
dependent.  
The variation of CVD parameters over a wider parameter space for the benzene based process 
shows a similar general behavior compared to the CH4 as precursor (Fig. 7 e-h). In particular, 
the lower the carbon precursor partial pressure the lower the likelihood of achieving complete 
MLG surface coverage, and the higher the exposure pressure the higher the likelihood of 
multilayer nucleation and film inhomogeneity. Again, the growth rate depends on the specific 
Cu surface. However, considering that between Figs. 7 e,f and g,h the C6H6  partial pressure 
increase was ~100 fold, the increase in fractional FLG coverage and inhomogeneity is 
surprisingly little. Significantly, we note that the partial pressure range in which C6H6 yields 
MLG (~10
-4
 –10
-2 
mbar) is much wider (relative to the partial pressure used) than for the CH4 
process (~0.2 - 1.5 mbar). Hence, for the conditions used, C6H6 as precursor does not require 
H2 dilution and  enables growth at lower temperatures  with more robust processing 
conditions. 
Discussion 
 The overall CVD process for graphene growth can be discussed in the context of basic 
heterogeneous catalysis and 2D crystal growth kinetics as a multistep reaction comprising: 
Page 9 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10 
 
(1) gaseous precursor transport to and dissociation on the catalyst surface, (2) transport of 
carbon (species) on the surface and into/out of bulk of the catalyst, (3) graphene nucleation 
and carbon incorporation into the growing graphene layer, (4) etching of the as-formed 
graphene. Step (1) thereby comprises transport of gas reactants through the boundary layer 
above the catalyst surface and the adsorption/desorption kinetics of the catalytic reaction. 
Simultaneous to its formation, (4) graphene etching can occur depending on the composition 
of the gas atmosphere and the presence of contaminants in the CVD set-up/process. The 
thermodynamic driving force for growth, i.e. step (3), is a carbon supersaturation at the 
catalyst surface.
16
 In CVD this supersaturation is created via step (1), whereby the different 
CVD conditions can be expressed as different carbon chemical potentials. The chemical 
potential depends on temperature and partial pressures, which in turn depend on the choice of 
precursor and the reaction considered. A more reactive carbon source corresponds to carbon 
supplied at a higher chemical potential. We adopt this general framework here to qualitatively 
rationalize our findings, even though the experimental conditions might not be close to 
equilibrium, i.e. the carbon chemical potential is difficult to quantify.
30
                                            
For carbon supplied at a very high chemical potential, graphene growth is very favorable and 
can become non-specific to details of the catalyst surface. Our data here (Figs. 1,4,7) is 
consistent with such a generic behavior insofar that the higher the supplied carbon chemical 
potential the higher the likelihood of film inhomogeneity and primary and secondary 
multilayer nucleation. Assuming growth occurs isothermally during CVD exposure and not 
during cool down (see below) and that additional layers grow in contact with the catalyst, i.e. 
underneath the existing graphene,
31,32
  secondary nucleation indicates that carbon reaches the 
Cu surface even after complete MLG coverage. Isolated graphene flakes have been shown to 
be impermeable to gases,
33
 hence we suggest that the observed carbon leakage is due to the 
inherent polycrystallinity of as-grown MLG whereby the domain boundaries and other 
defects offer pathways for the precursor to reach the catalyst. Hence clearly graphene CVD 
on Cu cannot be expected to inherently (independent of conditions applied) give a self-
limiting homogeneous monolayer coverage. We emphasize that this has important 
ramifications for FLG CVD. Primary nucleation will not give a homogeneous FLG coverage 
due to the different growth rates of the layers. Secondary nucleation will require leakage 
through the covering layer(s), and the challenge is thereby to feed homogeneous growth 
Page 10 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11 
 
through inhomogeneous leakage. Further we expect the leakage to significantly decrease with 
the number of layers, so the number of layers possible is clearly limited by this method. 
The more the supplied carbon chemical potential is lowered the more specific the graphene 
formation becomes to details of the catalyst surface, nature of nucleation sites, energy costs 
associated with graphene edges and for instance additional strain energies depending on the 
lattice mismatch.
30
 Figs. 1 and 7 highlighted that graphene formation is indeed dependent on 
catalyst surface orientation and impurity levels. Improved growth on Cu(111) has been 
previously attributed to improved precursor adsorption and high diffusion of carbon species.
34
 
Compared to for instance Ni, Cu in its given state is a less active catalyst for step (1), hence 
higher temperatures are required to supply carbon at a given rate. Compared to CH4, C6H6 
represents a more reactive carbon source, which is captured in the temperature dependence of 
Fig. 6. Hence it is not surprising that we do not see any graphitic deposits at temperatures 
below 700°C for CH4, whereas for C6H6 we observe carbon film deposition at temperatures 
as low as 300°C (see Fig. 6e) on Cu for the given conditions. Below 600
o
C, the crystallinity 
of the as-grown carbon is poor, and although the defect density will again depend on the 
detailed growth kinetics (e.g. carbon arrival rate vs incorporation rate), we cannot reproduce 
the Cu-catalyzed growth of graphene at temperature of  600
o
C or below, as recently reported 
for toluene
18
(~ 600
 o
C, ID/IG~0.35) and ill-defined C6H6 exposures in hot-wall furnaces.
35
 We 
note in this context that these previous efforts have focused on lowering the temperature, but 
clearly compromised on graphene quality.
18,35
 Bearing in mind that graphene has to be 
transferred off the catalyst metal for most applications,
36
 our motivation here is a temperature 
reduction while maintaining the quality. 
Our data indicates that C6H6 as precursor does not only enable growth at lower temperatures 
but also more robust processing conditions. Some previous literature suggests that the 6-
membered ring configuration of benzene would provide an inherent advantage, that could 
help to explain our findings.
35
  However, even though the detailed nature of carbon species in 
steps (2) and (3) remains unknown, we see no evidence that the 6-membered ring 
configuration of C6H6 will be preserved at step (1) at the given conditions, as sometimes 
suggested in the literature.
35,37
 Rather we suggest that the advantage of C6H6 lies in the rate 
balance that it allows during CVD. CH4 requires H2 dilution to reduce the Cu surface (Fig. 5) 
and the required CH4/H2 balance is critical (Fig. 4). We show that a lower CH4/H2 ratio leads 
to etching, hence the CH4/H2 balance reflects a balance between carbon deposition and 
Page 11 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
12 
 
etching, i.e. between steps (3) and (4). We emphasize that this balance is highly process 
parameter dependent, which is why CH4 based CVD is more delicate to control. At APCVD 
conditions, for instance, well known CVD kinetic models predict a mass transfer limited 
regime, whereby the boundary layer in step (1) is rate limiting. This has also been discussed 
in the context of graphene APCVD.
28
 Based on our data, we suggest here that the delicate 
CH4/H2 balance shifts for pressure induced changes of the boundary layer and this is why 
achieving continuous MLG films based on APCVD is very challenging using CH4 as 
precursor. C6H6 as precursor on the other hand is more reactive and requires no reactive 
diluent and related delicate balancing. We also suggest that diluting C6H6 with a neutral gas 
such as Ar should be much more straightforward than for CH4 in terms of maintaining high 
quality graphene growth.  
The above argumentation assumes that the observed graphene formation on Cu occurs 
predominantly during the precursor exposure at isothermal conditions, rather than due to 
precipitation upon cooling.
7,8
 This assumption is supported by in-situ observations of 
isothermal graphene growth on Cu during elemental carbon deposition,
38
 however the 
importance of the contribution of carbon precipitation on cooling should also be considered. 
Based on a simplistic consideration of carbon solubility in Cu at 1000
o
C of between 0.00070 
at%
39
  and 0.028 at% 
40
, the amount of carbon dissolved in the 25µm foil corresponds to 
between 0.4 and 15.5 layers of graphene with an atomic density of 3.8×10
19
 carbon atoms m
-
2
. It should be noted that the large uncertainties here reflect the significant disparities between 
the solubility values reported in the literature.
39,40
 Realistically the diffusion lengths of carbon 
dissolution and precipitation may limit the active volume to some fraction of the foil 
thickness and should be considered when estimating whether the quantity of carbon that 
precipitates as graphene upon cooling is significant.
8
 For such a calculation to be informative, 
a validation of the solubility and diffusivity of carbon in Cu is required which lies beyond the 
scope of the present work. We note that further in-situ experiments are needed to fully clarify 
the relative importance of growth by precipitation upon cooling. 
 
Conclusions 
We systematically explored the parameter space of graphene CVD on polycrystalline Cu foils 
in particular regarding the choice of carbon precursor and mitigation of Cu sublimation as 
Page 12 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13 
 
required towards industrial manufacture. CH4, the currently most widely used carbon 
precursor, requires H2 dilution and high temperatures (1000°C) to keep the Cu surface 
reduced and yield high quality graphene. The H2 atmosphere etches as-grown graphene, 
hence maintaining a balanced CH4/H2 ratio is critical. Such balance is more easily achieved at 
low pressure conditions, at which however Cu sublimation is at deleterious levels. In contrast, 
C6H6 as precursor requires no reactive dilution, i.e. no delicate balance to be maintained, and 
consistently gives similar graphene quality at 100-150°C lower temperatures compared to 
CH4 based CVD. The lower process temperature and more robust processing conditions allow 
the problem of Cu sublimation to be effectively addressed. Our growth study shows that Cu is 
not inherently limiting graphene formation to a monolayer. Rather the higher the supplied 
carbon chemical potential the higher the likelihood of film inhomogeneity and primary and 
secondary multilayer nucleation. Secondary nucleation indicates that carbon reaches the Cu 
surface even after complete MLG coverage, whereby we suggest that the domain boundaries 
of the inherently polycrystalline layers offer pathways for the precursor to reach the catalyst. 
Our data further emphasizes that the Cu catalyst template is not static and that the involved 
kinetics of grain growth are highly process dependent, making this an important process step 
for controlled graphene CVD.  
Although the data presented concerns only two carbon precursors, we expect the insights 
achieved to be of general relevance for the optimization of graphene CVD and more rational 
process design. While C6H6 may not be the precursor of choice for industrial upscaling due to 
its harmful effects on health, we think that it serves as a good model precursor system to 
effectively study the effect of precursor reactivity for graphene CVD. We have preliminary 
data for xylene as carbon precursor, which is an example of a cheap and safe precursor that 
shows similar advantages as highlighted here for benzene, in particular giving MLG at 900°C 
when diluted in Ar at APCVD conditions without Cu sublimation. 
 
  
Page 13 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
14 
 
Acknowledgements 
S.H. acknowledges funding from ERC grant InsituNANO (n°279342) and from EPSRC 
(Grant Nr. EP/H047565/1). P.R.K. acknowledges funding from the Cambridge 
Commonwealth Trust and C.D. acknowledges funding from Royal Society. The authors wish 
to acknowledge Dr. Matt Cole for help with equipment modification, Amalya Kostanyan for 
help with the POM characterization, and Dr. Bernhard C. Bayer for discussions. 
 
Supporting Information 
 
More experimental details on the EBSD analysis and on the liquid crystal based polarizing 
optical microscopy (POM) technique are available as supporting information. This 
information is available free of charge via the Internet at http://pubs.acs.org. 
 
  
Page 14 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
15 
 
References  
(1)  Dresselhaus, M. S.; Dresselhaus, G.; Avouris, P. Carbon nanotubes: synthesis, 
structure, properties, and applications; Springer, 2001. 
(2)  Hiruma, K.; Yazawa, M.; Katsuyama, T.; Ogawa, K.; Haraguchi, K.; Koguchi, M.; 
Kakibayashi, H. J. Appl. Phys. 1995, 77, 447-62. 
(3)  Hofmann, S.; Sharma, R.; Ducati, C.; Du, G.; Mattevi, C.; Cepek, C.; Cantoro, M.; 
Pisana, S.; Parvez, A.; Cervantes-Sodi, F.; et al. Nano Lett. 2007, 7, 602–8. 
(4)  Li, X.; Cai, W.; An, J.; Kim, S.; Nah, J.; Yang, D.; Piner, R.; Velamakanni, A.; Jung, 
I.; Tutuc, E.; et al. Science 2009, 324, 1312–4. 
(5)  Bae, S.; Kim, H.; Lee, Y.; Xu, X.; Park, J.S.; Zheng, Y.; Balakrishnan, J.; Lei, T.; Kim, 
H. R.; Song, Y. I.; et al. Nat. Nanotechnol. 2010, 5, 574–8. 
(6)  Li, X.; Cai, W.; Colombo, L.; Ruoff, R. S. Nano Lett. 2009, 9, 4268–72. 
(7)  Weatherup, R. S.; Bayer, B. C.; Blume, R.; Ducati, C.; Baehtz, C.; Schlögl, R.; 
Hofmann, S. Nano Lett. 2011, 11, 4154–60. 
(8)  Weatherup, R. S.; Bayer, B. C.; Blume, R.; Baehtz, C.; Kidambi, P. R.; Fouquet, M.; 
Wirth, C. T.; Schlögl, R.; Hofmann, S. Chemphyschem. 2012,10,2544-49. 
(9)  Regmi, M.; Chisholm, M. F.; Eres, G. Carbon. 2012, 50, 134–141. 
(10)  Huang, P. Y.; Ruiz-Vargas, C. S.; Zande, A. M. van der; Whitney, W. S.; Levendorf, 
M. P.; Kevek, J. W.; Garg, S.; Alden, J. S.; Hustedt, C. J.; Zhu, Y.; et al. Nature 2011, 
469, 389–92. 
(11)  Kim, K.; Lee, Z.; Regan, W.; Kisielowski, C.; Crommie, M. F.; Zettl, A. ACS Nano. 
2011, 5, 2142–6. 
(12)  Petrone, N.; Dean, C. R.; Meric, I.; Zande, A. M. van der; Huang, P. Y.; Wang, L.; 
Muller, D.; Shepard, K. L.; Hone, J. Nano Lett. 2012, 12, 2751–6. 
(13)  Li, X.; Magnuson, C. W.; Venugopal, A.; Tromp, R. M.; Hannon, J. B.; Vogel, E. M.; 
Colombo, L.; Ruoff, R. S. J. Am. Chem. Soc. 2011, 133, 2816–9. 
(14)  Ismach, A.; Druzgalski, C.; Penwell, S.; Schwartzberg, A.; Zheng, M.; Javey, A.; 
Bokor, J.; Zhang, Y. Nano Lett. 2010, 10, 1542–8. 
(15)  Kim, D. W.; Kim, Y. H.; Jeong, H. S.; Jung, H.-T. Nat. Nanotechnol. 7, 29–34. 
(16)  Batzill, M. Surf. Sci. Rep. 2012, 67, 83–115. 
Page 15 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
16 
 
(17)  Li, X.; Magnuson, C. W.; Venugopal, A.; An, J.; Suk, J. W.; Han, B.; Borysiak, M.; 
Cai, W.; Velamakanni, A.; Zhu, Y.; et al. Nano Lett. 2010, 10, 4328–34. 
(18)  Zhang, B.; Lee, W. H.; Piner, R.; Kholmanov, I.; Wu, Y.; Li, H.; Ji, H.; Ruoff, R. S. 
ACS Nano. 2012,6,2471-76. 
(19)  Humphreys, F. J.; Hatherly, M. Recrystallization and Related Annealing Phenomena  
Elsevier, 2004. 
(20)  Ferrari, A. C.; Meyer, J. C.; Scardaci, V.; Casiraghi, C.; Lazzeri, M.; Mauri, F.; 
Piscanec, S.; Jiang, D.; Novoselov, K. S.; Roth, S.; et al. Phys. Rev. Lett. 97, 187401. 
(21)  Zhang, Y.; Li, Z.; Kim, P.; Zhang, L.; Zhou, C. ACS Nano. 2012, 6, 126–32. 
(22)  Vlassiouk, I.; Regmi, M.; Fulvio, P.; Dai, S.; Datskos, P.; Eres, G.; Smirnov, S. ACS 
Nano. 2011, 5, 6069–76. 
(23)  Kim, H.; Mattevi, C.; Calvo, M. R.; Oberg, J. C.; Artiglia, L.; Agnoli, S.; Hirjibehedin, 
C. F.; Chhowalla, M.; Saiz, E. ACS Nano. 2012, 6, 3614–23. 
(24)  Rackauskas, S.; Nasibulin, A. G.; Jiang, H.; Tian, Y.; Kleshch, V. I.; Sainio, J.; 
Obraztsova, E. D.; Bokova, S. N.; Obraztsov, A. N.; Kauppinen, E. I. Nanotechnology 
2009, 20, 165603. 
(25)  Hamilton, J. C. J. Electrochem. Soc. 1986, 133, 739. 
(26)  Liu, J.; Wu, J.; Edwards, C. M.; Berrie, C. L.; Moore, D.; Chen, Z.; Maroni, V. A.; 
Paranthaman, M. P.; Goyal, A. Adv. Funct. Mater. 2011, 21, 3868–3874. 
(27)  Wiame, F.; Maurice, V.; Marcus, P. Surf. Sci. 601, 1193–1204. 
(28)  Bhaviripudi, S.; Jia, X.; Dresselhaus, M. S.; Kong, J. Nano Lett. 10, 4128–4133. 
(29)  Cançado, L. G.; Jorio, A; Ferreira, E. H. M.; Stavale, F.; Achete, C. A; Capaz, R. B.; 
Moutinho, M. V. O.; Lombardo, A; Kulmala, T. S.; Ferrari, A. C. Nano Lett. 2011, 11, 
3190–6. 
(30)  Saadi, S.; Abild-Pedersen, F.; Helveg, S.; Sehested, J.; Hinnemann, B.; Appel, C. C.; 
Nørskov, J. K. J. Phys. Chem. C 2010, 114, 11221–27. 
(31)  Nie, S.; Wu, W.; Xing, S.; Yu, Q.; Mccarty, K. F.; arXiv:1202.1031v2. 
(32)  Nie, S.; Walter, A. L.; Bartelt, N. C.; Starodub, E.; Bostwick, A.; Rotenberg, E.; 
McCarty, K. F. ACS Nano. 2011, 5, 2298–306. 
(33)  Bunch, J. S.; Verbridge, S. S.; Alden, J. S.; Zande, A. M. van der; Parpia, J. M.; 
Craighead, H. G.; McEuen, P. L. Nano Lett. 2008, 8, 2458–62. 
Page 16 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
17 
 
(34)  Wood, J. D.; Schmucker, S. W.; Lyons, A. S.; Pop, E.; Lyding, J. W. Nano. Lett. 2011, 
9, 4547-54.  
(35)  Li, Z.; Wu, P.; Wang, C.; Fan, X.; Zhang, W.; Zhai, X.; Zeng, C.; Li, Z.; Yang, J.; 
Hou, J. ACS Nano. 2011, 5, 3385–90. 
(36)  Kidambi, P. R.; Bayer, B. C.; Weatherup, R. S.; Ochs, R.; Ducati, C.; Szabó, D. V.; 
Hofmann, S. Phys. Status Solidi (RRL). 2011, 9, 341–343. 
(37)  Yang, Y.; Hu, Z.; Tian, Y. J.; L, Y. N.; Wang, X. Z.; Chen, Y. Nanotechnology. 2003, 
14, 733–737. 
(38)  Wofford, J. M.; Nie, S.; McCarty, K. F.; Bartelt, N. C.; Dubon, O. D. Nano Lett. 2010, 
10, 4890–96. 
(39)  López, G. A.; Mittemeijer, E. J. Scr. Mater. 2004, 51, 1–5. 
(40)  Mclellan, R. B. Scripta Metallurgica 1969, 3, 389–391.  
 
  
Page 17 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
18 
 
Figure Captions 
 
Figure 1. SEM images at different magnification showing graphene nuclei on Cu before 
merging to form a continuous film a), b) Monolayer nuclei LPCVD 4 mbar, 1000
o
C, (1:10) 
CH4 : H2 for 25 min before merging to form a continuous film on 99.999% pure 25 µm foil 
c),d) multilayer nuclei LPCVD 4 mbar, 1000
o
C (1:1) CH4 : H2 for 5 min on 99.999% pure 25 
µm foil e),f) multilayer nuclei LPCVD 4 mbar, 1000
o
C (1:1) CH4 : H2 for 5 min on 99.999% 
pure 100 µm foil with the underlying Cu grain orientation measured by EBSD g),h) 
multilayer nuclei APCVD 1000
o
C (1:25) CH4 : H2 for 5 min on 99.999% pure 25 µm foil i),j) 
multilayer nuclei APCVD 1000
o
C (1:25) CH4 : H2 for 5 min on 99.995% pure 12 µm Cu foil. 
 
Figure 2. SEM images with EBSD in spot mode showing the evolution of the Cu catalyst for 
different growth conditions.  a) as received 99.999% pure 25 µm foil, b) after APCVD anneal 
for 10 min at 1000
o
C, c) after LPCVD anneal for 10 min at 1000
o
C min at 4 mbar, d) after 
APCVD growth at 1000
o
C (1:25) CH4 : H2 for 5 min e) after LPCVD growth at 1000
o
C, 4 
mbar (1:5) CH4 : H2 for 30 min, f) after LPCVD growth 1000
o
C, 4 mbar (1:0) CH4 : H2 for 30 
min. Optical images of the furnace tube g) after processing at APCVD and h) 4 mbar and 
1000
o
C respectively.   
 
Figure 3. Characterization of large area MLG film grown in LPCVD 4 mbar, 1000
o
C, (1:5) 
CH4 : H2 for 30 min on 99.999% pure 25 µm foil a) Optical image b) Raman spectrum 
confirms the presence of MLG. Raman map of large area MLG c) I2D/IG d) ID/IG and e) and f) 
show the corresponding distribution statistics. g) 6 contacts Hall geometry devices. h) Control 
Polarizing microscopy (POM) image in a region of the sample containing no graphene. i)  
POM image, for liquid crystal (LC) over graphene sample.  
 
Figure 4. Optical image of graphene film grown on 99.999% pure 25 µm foil by LPCVD and 
transferred to SiO2(300 nm)/Si a) 1 mbar and b) 8 mbar at 1000
o
C, (1:5) CH4 : H2 for 30 min. 
c) 10 min and d) 60 min at 4 mbar, 1000
o
C, (1:5) CH4 : H2, e) (1:10) and f) (1:1) at  CH4 : H2, 
4 mbar, 1000
o
C, for 30 min. g) Raman spectra for mono, bi and multilayer seen in Fig 4a-f 
marked with the corresponding color of the circle and APCVD FLG graphene seen in Fig 1g-
j. 
Page 18 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
19 
 
 
Figure 5. LPCVD growth in the absence of H2 at 4 mbar, CH4:H2 (1:0), 1000
o
C for 30 min 
a),b) optical images of Cu foil post growth c),d) SEM images of Cu foil post growth e) 
optical images post transfer to SiO2(300nm)/Si wafer f) Raman spectra measured on the 
triangular structure on the Cu foil (red, 457nm laser) and post transfer to SiO2(300nm)/Si 
wafer (blue). 
 
Figure 6. Optical images of graphitic film grown on Cu at 4mbar, 1:5 CH4 : H2,30 min at a) 
1000, b) 900 and c) 800
o
C and d) shows the corresponding Raman spectra marked with the 
respective color. e) Shows Raman spectra measured on graphitic films grown with C6H6 at 
different temperatures post transfer to a SiO2(300nm)/Si wafer. f) ID/IG ratios for graphitic 
films grown from CH4 (squares) and C6H6 (circles). Optical images of a graphitic film grown 
on Cu at with C6H6 at g) 900, h) 800 and i) 300
o
C post transfer to a SiO2(300nm)/Si wafer. 
 
Figure 7. Raman map for of large area MLG films grown from C6H6 at LPCVD conditions at 
900
o
C on 99.999% pure 25 µm foil a) I2D/IG maps b) ID/IG maps. c) and d) show the 
corresponding distribution statistics. SEM images at different magnifications showing 
graphene nuclei on Cu from C6H6 at LPCVD conditions at 900
o
C on 99.999% pure 25 µm 
foil for e),f) at low exposure conditions with C6H6 partial pressure ~ 10
-4
 mbar for 5 min and 
g),h) at high exposure conditions with C6H6 partial pressure ~10
-2
 mbar for 5 min. i),j) h) 
POM images for liquid crystal (LC) over C6H6 derived graphene sample at different rotations 
highlighting the polycrystallinity of the MLG.  
 
 
 
  
Page 19 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
20 
 
Figures 
 
 
Figure 1 Kidambi et al. 
  
Page 20 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
21 
 
 
Figure 2 Kidambi et al. 
Page 21 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
22 
 
 
Figure 3 Kidambi et al. 
Page 22 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
23 
 
  
Figure 4 Kidambi et al. 
  
 
  
Page 23 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
24 
 
 
Figure 5 Kidambi et al. 
Page 24 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
25 
 
  
Figure 6 Kidambi et al.    
Page 25 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
26 
 
 
 
Figure 7 Kidambi et al.     
Page 26 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
27 
 
Table of Contents Graphic 
 
 
 
 
 
Page 27 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
  
 
 
 
70x114mm (300 x 300 DPI)  
 
 
Page 28 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
  
 
 
 
70x110mm (300 x 300 DPI)  
 
 
Page 29 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
  
 
 
 
70x123mm (300 x 300 DPI)  
 
 
Page 30 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
  
 
 
 
60x141mm (300 x 300 DPI)  
 
 
Page 31 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
  
 
 
 
70x80mm (300 x 300 DPI)  
 
 
Page 32 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
  
 
 
 
119x137mm (300 x 300 DPI)  
 
 
Page 33 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
  
 
 
 
80x146mm (300 x 300 DPI)  
 
 
Page 34 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
  
 
 
 
570x180mm (72 x 72 DPI)  
 
 
Page 35 of 35
ACS Paragon Plus Environment
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60