scispace - formally typeset
Search or ask a question
Journal ArticleDOI

A catalytic alloy approach for graphene on epitaxial SiC on silicon wafers

TL;DR: In this article, the synthesis of high-quality and highly uniform few-layer graphene on silicon wafers, based on solid source growth from epitaxial 3C-SiC films, is presented.
Abstract: We introduce a novel approach to the synthesis of high-quality and highly uniform few-layer graphene on silicon wafers, based on solid source growth from epitaxial 3C-SiC films. Using a Ni/Cu catalytic alloy, we obtain a transfer-free bilayer graphene directly on Si(100) wafers, at temperatures potentially compatible with conventional semiconductor processing. The graphene covers uniformly a 2″ silicon wafer, with a Raman ID/ IG band ratio as low as 0.5, indicative of a low defectivity material. The sheet resistance of the graphene is as low as 25 Ω/square, and its adhesion energy to the underlying substrate is substantially higher than transferred graphene. This work opens the avenue for the true wafer-level fabrication of microdevices comprising graphene functional layers. Specifically, we suggest that exceptional conduction qualifies this graphene as a metal replacement for MEMS and advanced on-chip interconnects with ultimate scalability.

Summary (2 min read)

Introduction

  • The authors introduce a novel approach to the synthesis of high-quality and highly uniform few-layer graphene on silicon wafers, based on solid source growth from epitaxial 3C-SiC films.
  • This work opens the avenue for the true wafer-level fabrication of microdevices comprising graphene functional layers.
  • Ever since graphene was experimentally isolated about a decade ago,1 the high temperature (1300–1700 °C).
  • This endeavor has proved more challenging than expected, in particular, because of the upper limit set by the melting temperature of silicon and the scarce availability of a defect-free and atomically smooth epitaxial SiC on Si(111) starting template.
  • 10,11 However, their catalyst-mediated process10 allows for a reduction in the optimal synthesis temperature and for a relaxation of the strict requirements on the starting template.

II. EXPERIMENTAL

  • Monocrystalline 3C-(cubic polytype) SiC films, 250 nm thick, were epitaxially grown on ,111.
  • Sample foils for transmission electron microscopy were prepared via a focused ion beam liftout technique using a FEI Strata DB235 FIB/SEM with a Ga1 ion source.
  • A total of eight measurements were taken for each sample by positioning the electrical probes on the sputtered metal contacts and sweeping the DC input current from a HP4145B parameter analyzer from 0 to 10 mA, which were then separately averaged into “vertical” and “horizontal” sheet resistance groups.

A. Mechanisms for catalytic alloy graphitization and physical analysis

  • Figure 2 shows an overview of the ID/IG band ratios, indicative of the defect density of graphene layers,16 calculated from the Raman spectra of graphene prepared with different compositions of metal catalysts on SiC (100) and SiC(111), as measured after the removal of the reacted metal layer by wet etching.
  • The error bar represents the variation obtained by measuring five different sites over the prepared samples.
  • This means that asperities larger than 200 nm and smaller than 50 nm are considerably reduced with the use of Ni/Cu.
  • The second important advance of this study is the demonstration that the Ni/Cu alloy graphitization achieves quality graphene on both SiC(111) and SiC (100) surface orientations, with a consistently better quality obtained from SiC(100), as inferred from the ID/IG trend in Fig.
  • It is reasonable to expect that the metal-induced graphitization of SiC would proceed through an amorphization of a thin top portion of the SiC film, as the Ni helps to weaken the Si–C bonds and to release the atomic C. Given the ,1 nm thin nature of the graphene, the XPS analysis is also clearly probing the underlying amorphous layer.

C. Electrical measurements

  • The authors have performed AU9sheet resistance measurements on the graphene samples prepared with the same graphitization process as the one shown in the TEM micrograph in Fig.
  • The small magnification image shows a region near the top surface of the epitaxial SiC. J. Mater.
  • Res., Vol. 30, No. 0, 2015 5 value of all measurements where conduction was along one axis of the samples (horizontal and vertical), so that each point on those curves is an average of four distinct van der Pauw measurements.
  • By averaging measurements taken on five samples fabricated with the same graphitization procedure over separate runs, the authors can conclude that the sheet resistance of their bilayer graphene prepared on a SiC(100) layer with Ni/Cu is around 24.8 X/square 6 0.7 from sample-to-sample variation.

IV. CONCLUSIONS

  • The authors have demonstrated what is, to their knowledge, the first successful approach to transfer-free, direct growth of uniform and high-quality bilayer graphene over full silicon wafers.
  • The methodology relies on the use of an epitaxial SiC layer on silicon as a carbon source for graphene, and the use of a catalyst alloy of Ni and Cu, and can be used with both SiC(111) and SiC(100) surface orientations.
  • The obtained bilayer graphene is around 0.8–0.9 nm thick, matching the electrical resistivity of bulk Au.
  • Additionally, preliminary results indicate that the adhesion of graphene to the underlying substrate could be an order of magnitude higher than the adhesion of a graphene layer grown ex situ and transferred onto a SiO2 layer on silicon.
  • Large benefit from this advance is anticipated in areas such as actuation forAU10 MEMS/NEMS and advanced onchip interconnects for micro and nanocircuits.

Did you find this useful? Give us your feedback

Content maybe subject to copyright    Report

A catalytic alloy approach for graphene on epitaxial SiC on
silicon wafers
Author
Iacopi, Francesca, Mishra, Neeraj, Cunning, Benjamin Vaughan, Goding, Dayle, Dimitrijev,
Sima, Brock, Ryan, Dauskardt, Reinhold H, Wood, Barry, Boeckl, John
Published
2015
Journal Title
Journal of Materials Research
Version
Accepted Manuscript (AM)
DOI
https://doi.org/10.1557/jmr.2015.3
Copyright Statement
© 2015 Cambridge University Press. This is the author-manuscript version of this paper.
Reproduced in accordance with the copyright policy of the publisher. Please refer to the
journal's website for access to the definitive, published version.
Downloaded from
http://hdl.handle.net/10072/69321
Griffith Research Online
https://research-repository.griffith.edu.au

INVITED FEATURE PAPER
A catalytic alloy approach for graphene on epitaxial SiC on silicon
wafers
F. Iacopi,
a)
N. Mishra, B.V. Cunning, D. Goding, and S. Dimitrijev
Queensland Micro- and Nanotechnology Centre, Grifth University, Nathan, Queensland 4111, AustraliaAU2
R. Brock and R.H. Dauskardt
Department of Materials Science and Engineering, Stanford University, Stanford, California 94305, USA
B. Wood
Centre for Microscopy and Microanalysis, The University of Queensland, St. Lucia, Queensland 4072, Australia
J.J. Boeckl
Materials and Manufacturing Directorate, Air Force Research Laboratories, Wright-Patterson AFB, Ohio, USAAU3AU1
(Received 7 July 2014; accepted 1 January 2015)
We introduce a novel approach to the synthesis of high-quality and highly uniform few-layer
graphene on silicon wafers, based on solid source growth from epitaxial 3C-SiC lms. Using
a Ni/Cu catalytic alloy, we obtain a transfer-free bilayer graphene directly on Si(100) wafers,
at temperatures potentially compatible with conventional semiconductor processing. The graphene
covers uniformly a 20 silicon wafer, with a Raman I
D
/I
G
band ratio as low as 0.5, indicative of
a low defectivity material. The sheet resistance of the graphene is as low as 25 X/square, and its
adhesion energy to the underlying substrate is substantially higher than transferred graphene. This
work opens the avenue for the true wafer-level fabrication of microdevices comprising graphene
functional layers. Specically, we suggest that exceptional conduction qualies this graphene as
a metal replacement for MEMS and advanced on-chip interconnects with ultimate scalability.
I. INTRODUCTION
Ever since graphene was experimentally isolated about
a decade ago,
1
the high temperature (13001700 °C) Si
sublimation from crystalline silicon carbide (SiC) bulk
substrates has been extensively regarded as the cleanest
and most controlled means of obtaining quality graphene
at the wafer level.
24
However, the transfer of this
approach to its natural pseudo-substrate, i.e., epitaxial
SiC lms on silicon,
57
has lagged behind despite being
strongly driven by substantial cost (SiC wafers are about
100 times more expensive than silicon) and large-scale
fabrication arguments. This endeavor has proved more
challenging than expected, in particular, because of the
upper limit set by the melting temperature of silicon and
the scarce availability of a defect-free and atomically
smooth epitaxial SiC on Si(111) starting template.
8,9
In response to such challenges, we have recently
demonstrated an alternate approach to the wafer-level,
transfer-free uniform synthesis of graphene on silicon.
10
As for the sublimation process, our new methodology
relies on the use of epitaxial SiC on silicon as a solid
source of carbon. The main advantage of this approach is
that it enables a wafer-scale patterned synthesis of
graphene by prepatterning of the source SiC layer.
10,11
However, our catalyst-mediated process
10
allows for
a reduction in the optimal synthesis temperature and for
a relaxation of the strict requirements on the starting
template. In this study, we will review the alloy-mediated
catalytic synthesis and show that this process works
equally well on both SiC(100) and SiC(111) orientations
of the pseudo-substrate. We demonstrate an outstanding
graphene bilayer with remarkable adhesion to the sub-
strate and outstanding electrical conduction properties.
II. EXPER IMENTAL
Monocrystalline 3C-(cubic polytype) SiC lms,
250 nm thick, were epitaxially grown on ,111. and
,100. oriented 20 silicon substrates in a hot wall,
horizontal low-pressure chemical-vapor deposition reac-
tor, as described by Wang et al.,
12
yielding two types of
lms that will be referred to as SiC(111) and SiC(100),
respectively
13
[Fig. 1(a)]. Subsequently, either a single
layer of Ni or a double layer of Ni and Cu was sputtered
on the SiC samples, as illustrated in Fig. 1(b), and then
annealed in a Carbolite HT furnace for 75 min at 1100 °C
in a medium vacuum atmosphere (below 10
3
mbar) to
induce graphitization. As per our recent work,
10
the
annealing generates a highly intermixed layer, which is
then removed by immersion of the samples in a wet etch
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
Contributing Editor: Mauricio Terrones
a)
Address all correspondence to this author.
e-mail: f.iacopi@grifth.edu.au
This paper has been selected as an Invited Feature Paper.
DOI: 10.1557/jmr.2015.3
J. Mater. Res., Vol. 30, No. 0, 2015 Ó Materials Research Society 2015 1

solution (Freckle) and sonication, to reveal a few-layer
graphene at the top surface of SiC [Fig. 1(c)]. In this
study, we have kept the Cu thickness constant to ;15
nm, while varying the Ni thickness to analyze and
compare the properties of the produced graphene.
The few-layer graphene on SiC/Si samples was char-
acterized with Raman spectroscopy with a Renishaw
inVia Raman, using a 514 nm laser. The laser spot size
was ;1 lm (50x objective), and the laser power at the
sample was about 7.5 mW. Atomic force microscopy
(AFM) was performed with a Park NX20 instrument in
a noncontact mode with a 512 512 pixel resolution
on 5 5 and 1 1 lm
2
scan areas to extract the root-
mean-square (RMS) surface roughness as well as the
power-spectral-density (PSD) of the spatial frequencies
composing the surface topography pattern. X-ray photo-
electron spectroscopy (XPS) was carried out with a Kratos
Axis ULTRA
AU4 high resolution system with monochro-
matic Al K
a
(1486.6 eV) incident radiation. Survey
(wide) scans were taken with an analyzer pass energy
of 160 eV while multiplex (narrow) high resolution scans
at 20 eV. The used aperture leads to an analysis area of
700 300 lm
2
. The base pressure in the analysis
chamber was below 1.0 10
8
Torr. Data analysis
was performed with the CasaXPS software and a Shirley
baseline with Kratos library relative sensitivity factors.
Sample foils for transmission electron microscopy
were prepared via a focused ion beam liftout technique
using a FEI Strata DB235 FIB/SEM with a Ga
1
ion
source. Prior to ion milling, the samples were protected
with a plasma sputtered gold layer of ;300 nm and then
a2lm Pt cap was deposited at 5 keV by an e-beam to
preserve the initial surface integrity. Site specic foils
were excavated from the bulk samples and thinned to
;500 nm. Subsequent Ar
1
ion milling was conducted in
a Fiscione NanoMill to remove Ga ion damage and
provide a high resolution transmission electron micro-
scope (HRTEM) foil. HRTEM characterization was
performed using a FEI Cs corrected Titan3 microscope
under 80 keV source illumination.
An initial analysis of the adhesive fracture energies of the
graphene/SiC/Si system was made by loading these samples
within a four point bend sandwiched conguration.
14,15
The
sandwiched structures containing the graphene on SiC were
prepared by sputtering an additional 500 nm Si spacer
layer, subsequently bonded with epoxy to the mirroring
silicon slab. 5 40 mm beams were diced out of the
sandwiched structures and tested. The average critical strain
energy release rates measured from such specimens yield an
indication of the adhesion energy along the debonded
pathway. Extensive detailed descriptions of this methodol-
ogy are found in the literature.
14,15
The location of the
debonding is veried through failure analysis with surface
XPS analysis on both sides of the delaminated interface.
Sheet resistance measurements were taken using four
electrical contacts in a van der Pauw conguration on the
corners of
AU51 1cm
2
samples. 150 nm of Ni were
sputtered at 100 °C through a shadow mask for forming
the contacts. A total of eight measurements were taken
for each sample by positioning the electrical probes on
the sputtered metal contacts and sweeping the DC
input current from a HP4145B parameter analyzer from
0 to 10 mA, which were then separately averaged into
vertical and horizontal sheet resistance groups. Pro-
vided the material is uniform and isotropic, the measure-
ments are expected to all converge to the same line with
only marginal variation due to probe positioning.
III. RESULTS AND DISCUSSION
A. Mechanisms for catalytic alloy graphitization
and physical analysis
Figure 2 shows an overview of the I
D
/I
G
band ratios,
indicative of the defect density of graphene layers,
16
calculated from the Raman spectra of graphene prepared
with different compositions of metal catalysts on SiC
(100) and SiC(111), as measured after the removal of the
reacted metal layer by wet etching. The error bar
represents the variation obtained by measuring ve
different sites over the prepared samples. Note that data
extracted from samples produced over more than 10
experimental runs fell within those error bars. Conrming
and extending the scope of our recent ndings for
graphene on SiC(111),
10
these data show that the use
of a Ni/Cu mixture improves dramatically the quality of
the obtained graphene on SiC(100), as compared to when
only Ni is used. In fact, the I
D
/I
G
ratio of graphene on SiC
(100) decreases from ;2 for 5 nm Ni catalyst alone
to ;0.8 when the Cu catalyst layer is added. Note that the
graphene on SiC(100) seems to be consistently of better
quality than that grown on SiC(111) under the same
conditions. Figure 3 shows an example of Raman
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
FIG. 1. Schematic of the process steps for the preparation of graphene on epitaxial 3C-SiC on silicon. The starting substrate can be either a ,100. or
,111. oriented substrate (a), on which a single- or double-thin catalyst layer of Ni alone or Ni and Cu is deposited (b). An anneal at 1100 °C leads to
the formation of a few-layer graphene covered with a metallic highly intermixed layer (c) that is subsequently removed by wet etching. (color online)
F. Iacopi et al.: A catalytic alloy approach for graphene on epitaxial SiC on silicon wafers
J. Mater. Res., Vol. 30, No. 0, 20152

spectrum of graphene grown on SiC(100) with 8 nm Ni
and 15 nm Cu. The intensity of the 2D band around
2700 cm
1
is slightly higher than that of the G band
(;1580 cm
1
), which indicates monolayer to bilayer
graphene.
16
The spectrum in Fig. 3 includes Raman shifts
down to 750 cm
1
to reveal the TO peak of the
underlying SiC(100) at 795 cm
1
. Note from Fig. 2 that
the nal graphene quality, as seen from the Raman
spectrum, appears rather insensitive to the further increase
of Ni thickness when the Ni/Cu mixture is used.
We have also compared the surface morphology of
graphene samples prepared with Ni only (8 nm) and the
Ni/Cu approach with AFM. As shown in Table I, the use
of the alloy halves the surface RMS roughness of the nal
graphene sample. This is also clearly shown by the power
spectral distributions in Fig. 4, indicating that the use of
Ni/Cu reduces dramatically the large-range (spatial fre-
quencies below 5 lm
1
) as well as the smallest-range
roughness (frequencies larger than 20 lm
1
). This means
that asperities larger than 200 nm and smaller than 50 nm
are considerably reduced with the use of Ni/Cu.
In our recent work,
10,17
we had presented a preliminary
elucidation of the mechanisms that allow the Ni/Cu alloy
to induce a better quality and more controlled graphiti-
zation as compared to Ni alone. The use of Ni only had
been in fact already reported in an earlier literature on
SiC wafers and then abandoned as yielding graphene on
the metal surface rather than on SiC, and of inferior
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
FIG. 2. Comparison of the defe ctivity of different graphenes on
SiC preparations as estimated from the ratio of the intensities of the
D over G Raman band s. T he thickness of the Cu layer, when
present, was kept constant. The error bar is calculated from the
measurements variation over ve different sites. The use of a Ni/Cu
alloy leads to a remarkable decrease in defectivity. Graphene on
SiC(100) shows consistently a slightly lower defect density than
that on SiC(111). (color online)
FIG. 3. Example of Raman spectrum (graphene D, G, and 2D bands
region) from graphene on SiC(100) prepared with ;8 nm Ni and
;15 nm Cu. Additionally to the low D over G band intensity,
indicative of a good quality graphene, note that the 2D over G
intensity ratio higher than 1 is indicative of a few-layer graphene.
The Raman shift region for 3C-SiC is also shown.
FIG. 4. Comparison of the spatial frequency spectra dening the
surface topography of different graphenes on SiC preparations (power
spectral densities), related to the AFM measurements in Table I. Note
that the graphene prepared with Ni only has substantially more large
(frequencies below 5 lm
1
) as well as very small surface asperities as
compared to the Ni/Cu preparations. (color online)
TABLE I. XXX AU6.
5 5mm
2
RMS
(nm 60.15)
1 1mm
2
RMS
(nm 60.15)
Graphene/SiC(100) {Ni only} 28 37
Graphene/SiC(100) {Ni/Cu} 15 12
Graphene/SiC(111) {Ni/Cu} 11 9
Surface roughness comparison of different graphenes on SiC prepara-
tions. Note that the graphene prepared with Ni/Cu indicates a 50%
decrease in RMS as compared to the use of Ni alone as a catalyst, and
that the graphene roughness does not seem dependent on the orientation
of the starting substrate. Note that the typical RMS roughness of the
initial SiC surface is 23nm.
9
F. Iacopi et al.: A catalytic alloy approach for graphene on epitaxial SiC on silicon wafers
J. Mater. Res., Vol. 30, No. 0, 2015 3

quality compared to the sublimation approach.
18,19
The
catalytic action of Ni is essential because it reacts with
SiC at relatively low temperature to form stable NiSi
x
,
20
and therefore allows for the release of the atomic
C needed for the synthesis of graphene. However, the
control of a uniform reaction of a thin Ni lm over a large
SiC surface is not trivial, as nickel silicidation appears
dominated by strong local driving forces leading to
typical clustering.
20,21
This is conrmed by the high
nonuniformity of graphene observed using only Ni in the
past and conrmed with Raman bands in Fig. 2 and by
the substantial topography of about 30 nm RMS shown
by our surface roughness measurements in Table I.
The addition of Cu has a twofold benecial action.
Firstly, Cu acts like an efcient medium to dilute and
helps to distribute as uniformly as possible the Ni over
the SiC surface. Note in this regard that the thermal
process for graphitization takes place very close to the
melting temperature of Cu. Secondly, the Cu is known as
a very efcient catalyst for graphitization, since free C
has an extremely low saturation in melt Cu (only a few
ppm, lower than Ni
22
), as calculated from a simple binary
CCu phase diagram. This allows for a fast precipitation
and graphitization of C released through the Ni silicida-
tion reaction, leading to an extraordinary improvement in
the uniformity and quality of the graphene obtained
through Ni/Cu.
The second important advance of this study is the
demonstration that the Ni/Cu alloy graphitization
achieves quality graphene on both SiC(111) and SiC
(100) surface orientations, with a consistently better
quality obtained from SiC(100), as inferred from the
I
D
/I
G
trend in Fig. 2. This is in contrast with the Si
sublimation process from SiC on silicon that is typically
more successful on the SiC(111) surfaces,
2326
which are
believed to be more suitable templates for graphene
because of their hexagonal atomic arrangement. We
suggest here that our graphene synthesis conditions are
close to those of a liquid phase epitaxy,
27
a process
whose benets are largely known to nanotechnology,
28,29
with a larger adatom surface mobility that can somewhat
relax the requirements on the starting template as
compared to a more conventional epitaxial process like
in the case of the sublimation. The reason why the SiC
(100) template leads in our case to an even better
graphene quality than SiC(111) could be related to the
combination of a considerably higher tensile stress and
higher surface defectivity of the latter epitaxial SiC as
discussed in our previous work.
13
SiC(111) layers indeed
show a higher density of stacking faults appearing at the
surface that, combined to a residual tensile SiC stress in
the range of 700 MPa to 1 GPa, could provide a strong
driving force around distinct preferential diffusion paths
of the Ni atoms into the SiC layer and thus a less uniform
silicidation reaction.
We have selected the sample series showing the best
(lowest) Raman I
D
/I
G
ratio in Fig. 2, prepared on SiC
(100) with 8 nm Ni and our standard Cu thickness (full
spectrum in Fig. 3) for a more detailed physical and
chemical analysis. XPS analysis shows a near-surface
composition of 50% C, 39% Si, and about 9% O, as
shown in the survey spectrum in Fig. 5(a). Other
elements found in just trace amounts are F (0.7%),
N (0.7%), and Ni (0.15%). The F and N traces are
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
FIG. 5. (a) XPS surface survey spectrum and (b) high resolution spectrum of the C1s region of a graphene on SiC(100) prepared with Ni/Cu,
corresponding to the Raman spectrum in Fig. 3. The C1s spectrum reveals a peak centered at 284.5 eV for the graphene, one 282.5 eV
corresponding to the underlying SiC, plus lower intensity peaks at higher eV, attributed to COSi, CO, and C5O bonds. (color online)
F. Iacopi et al.: A catalytic alloy approach for graphene on epitaxial SiC on silicon wafers
J. Mater. Res., Vol. 30, No. 0, 20154

Citations
More filters
DOI
Claudia Backes1, Claudia Backes2, Amr M. Abdelkader3, Concepción Alonso4, Amandine Andrieux-Ledier5, Raul Arenal6, Raul Arenal7, Jon Azpeitia6, Nilanthy Balakrishnan8, Luca Banszerus9, Julien Barjon5, Ruben Bartali10, Sebastiano Bellani11, Claire Berger12, Claire Berger13, Reinhard Berger14, M.M. Bernal Ortega15, Carlo Bernard16, Peter H. Beton8, André Beyer17, Alberto Bianco18, Peter Bøggild19, Francesco Bonaccorso11, Gabriela Borin Barin20, Cristina Botas, Rebeca A. Bueno6, Daniel Carriazo21, Andres Castellanos-Gomez6, Meganne Christian, Artur Ciesielski18, Tymoteusz Ciuk, Matthew T. Cole, Jonathan N. Coleman1, Camilla Coletti11, Luigi Crema10, Huanyao Cun16, Daniela Dasler22, Domenico De Fazio3, Noel Díez, Simon Drieschner23, Georg S. Duesberg24, Roman Fasel20, Roman Fasel25, Xinliang Feng14, Alberto Fina15, Stiven Forti11, Costas Galiotis26, Costas Galiotis27, Giovanni Garberoglio28, Jorge M. Garcia6, Jose A. Garrido, Marco Gibertini29, Armin Gölzhäuser17, Julio Gómez, Thomas Greber16, Frank Hauke22, Adrian Hemmi16, Irene Hernández-Rodríguez6, Andreas Hirsch22, Stephen A. Hodge3, Yves Huttel6, Peter Uhd Jepsen19, I. Jimenez6, Ute Kaiser30, Tommi Kaplas31, HoKwon Kim29, Andras Kis29, Konstantinos Papagelis27, Konstantinos Papagelis32, Kostas Kostarelos33, Aleksandra Krajewska34, Kangho Lee24, Changfeng Li35, Harri Lipsanen35, Andrea Liscio, Martin R. Lohe14, Annick Loiseau5, Lucia Lombardi3, María Francisca López6, Oliver Martin22, Cristina Martín36, Lidia Martínez6, José A. Martín-Gago6, José I. Martínez6, Nicola Marzari29, Alvaro Mayoral37, Alvaro Mayoral7, John B. McManus1, Manuela Melucci, Javier Méndez6, Cesar Merino, Pablo Merino6, Andreas Meyer22, Elisa Miniussi16, Vaidotas Miseikis11, Neeraj Mishra11, Vittorio Morandi, Carmen Munuera6, Roberto Muñoz6, Hugo Nolan1, Luca Ortolani, A. K. Ott3, A. K. Ott38, Irene Palacio6, Vincenzo Palermo39, John Parthenios27, Iwona Pasternak40, Amalia Patanè8, Maurizio Prato21, Maurizio Prato41, Henri Prevost5, Vladimir Prudkovskiy13, Nicola M. Pugno42, Nicola M. Pugno43, Nicola M. Pugno44, Teófilo Rojo45, Antonio Rossi11, Pascal Ruffieux20, Paolo Samorì18, Léonard Schué5, Eki J. Setijadi10, Thomas Seyller46, Giorgio Speranza10, Christoph Stampfer9, I. Stenger5, Wlodek Strupinski40, Yuri Svirko31, Simone Taioli28, Simone Taioli47, Kenneth B. K. Teo, Matteo Testi10, Flavia Tomarchio3, Mauro Tortello15, Emanuele Treossi, Andrey Turchanin48, Ester Vázquez36, Elvira Villaro, Patrick Rebsdorf Whelan19, Zhenyuan Xia39, Rositza Yakimova, Sheng Yang14, G. Reza Yazdi, Chanyoung Yim24, Duhee Yoon3, Xianghui Zhang17, Xiaodong Zhuang14, Luigi Colombo49, Andrea C. Ferrari3, Mar García-Hernández6 
Trinity College, Dublin1, Heidelberg University2, University of Cambridge3, Autonomous University of Madrid4, Université Paris-Saclay5, Spanish National Research Council6, University of Zaragoza7, University of Nottingham8, RWTH Aachen University9, Kessler Foundation10, Istituto Italiano di Tecnologia11, Georgia Institute of Technology12, University of Grenoble13, Dresden University of Technology14, Polytechnic University of Turin15, University of Zurich16, Bielefeld University17, University of Strasbourg18, Technical University of Denmark19, Swiss Federal Laboratories for Materials Science and Technology20, Ikerbasque21, University of Erlangen-Nuremberg22, Technische Universität München23, Bundeswehr University Munich24, University of Bern25, University of Patras26, Foundation for Research & Technology – Hellas27, Center for Theoretical Studies, University of Miami28, École Polytechnique Fédérale de Lausanne29, University of Ulm30, University of Eastern Finland31, Aristotle University of Thessaloniki32, University of Manchester33, Polish Academy of Sciences34, Aalto University35, University of Castilla–La Mancha36, ShanghaiTech University37, University of Exeter38, Chalmers University of Technology39, Warsaw University of Technology40, University of Trieste41, Queen Mary University of London42, University of Trento43, Instituto Politécnico Nacional44, University of the Basque Country45, Chemnitz University of Technology46, Charles University in Prague47, University of Jena48, University of Texas at Dallas49
29 Jan 2020
TL;DR: In this article, the authors present an overview of the main techniques for production and processing of graphene and related materials (GRMs), as well as the key characterization procedures, adopting a 'hands-on' approach, providing practical details and procedures as derived from literature and from the authors' experience, in order to enable the reader to reproduce the results.
Abstract: © 2020 The Author(s). We present an overview of the main techniques for production and processing of graphene and related materials (GRMs), as well as the key characterization procedures. We adopt a 'hands-on' approach, providing practical details and procedures as derived from literature as well as from the authors' experience, in order to enable the reader to reproduce the results. Section I is devoted to 'bottom up' approaches, whereby individual constituents are pieced together into more complex structures. We consider graphene nanoribbons (GNRs) produced either by solution processing or by on-surface synthesis in ultra high vacuum (UHV), as well carbon nanomembranes (CNM). Production of a variety of GNRs with tailored band gaps and edge shapes is now possible. CNMs can be tuned in terms of porosity, crystallinity and electronic behaviour. Section II covers 'top down' techniques. These rely on breaking down of a layered precursor, in the graphene case usually natural crystals like graphite or artificially synthesized materials, such as highly oriented pyrolythic graphite, monolayers or few layers (FL) flakes. The main focus of this section is on various exfoliation techniques in a liquid media, either intercalation or liquid phase exfoliation (LPE). The choice of precursor, exfoliation method, medium as well as the control of parameters such as time or temperature are crucial. A definite choice of parameters and conditions yields a particular material with specific properties that makes it more suitable for a targeted application. We cover protocols for the graphitic precursors to graphene oxide (GO). This is an important material for a range of applications in biomedicine, energy storage, nanocomposites, etc. Hummers' and modified Hummers' methods are used to make GO that subsequently can be reduced to obtain reduced graphene oxide (RGO) with a variety of strategies. GO flakes are also employed to prepare three-dimensional (3d) low density structures, such as sponges, foams, hydro- or aerogels. The assembly of flakes into 3d structures can provide improved mechanical properties. Aerogels with a highly open structure, with interconnected hierarchical pores, can enhance the accessibility to the whole surface area, as relevant for a number of applications, such as energy storage. The main recipes to yield graphite intercalation compounds (GICs) are also discussed. GICs are suitable precursors for covalent functionalization of graphene, but can also be used for the synthesis of uncharged graphene in solution. Degradation of the molecules intercalated in GICs can be triggered by high temperature treatment or microwave irradiation, creating a gas pressure surge in graphite and exfoliation. Electrochemical exfoliation by applying a voltage in an electrolyte to a graphite electrode can be tuned by varying precursors, electrolytes and potential. Graphite electrodes can be either negatively or positively intercalated to obtain GICs that are subsequently exfoliated. We also discuss the materials that can be amenable to exfoliation, by employing a theoretical data-mining approach. The exfoliation of LMs usually results in a heterogeneous dispersion of flakes with different lateral size and thickness. This is a critical bottleneck for applications, and hinders the full exploitation of GRMs produced by solution processing. The establishment of procedures to control the morphological properties of exfoliated GRMs, which also need to be industrially scalable, is one of the key needs. Section III deals with the processing of flakes. (Ultra)centrifugation techniques have thus far been the most investigated to sort GRMs following ultrasonication, shear mixing, ball milling, microfluidization, and wet-jet milling. It allows sorting by size and thickness. Inks formulated from GRM dispersions can be printed using a number of processes, from inkjet to screen printing. Each technique has specific rheological requirements, as well as geometrical constraints. The solvent choice is critical, not only for the GRM stability, but also in terms of optimizing printing on different substrates, such as glass, Si, plastic, paper, etc, all with different surface energies. Chemical modifications of such substrates is also a key step. Sections IV-VII are devoted to the growth of GRMs on various substrates and their processing after growth to place them on the surface of choice for specific applications. The substrate for graphene growth is a key determinant of the nature and quality of the resultant film. The lattice mismatch between graphene and substrate influences the resulting crystallinity. Growth on insulators, such as SiO2, typically results in films with small crystallites, whereas growth on the close-packed surfaces of metals yields highly crystalline films. Section IV outlines the growth of graphene on SiC substrates. This satisfies the requirements for electronic applications, with well-defined graphene-substrate interface, low trapped impurities and no need for transfer. It also allows graphene structures and devices to be measured directly on the growth substrate. The flatness of the substrate results in graphene with minimal strain and ripples on large areas, allowing spectroscopies and surface science to be performed. We also discuss the surface engineering by intercalation of the resulting graphene, its integration with Si-wafers and the production of nanostructures with the desired shape, with no need for patterning. Section V deals with chemical vapour deposition (CVD) onto various transition metals and on insulators. Growth on Ni results in graphitized polycrystalline films. While the thickness of these films can be optimized by controlling the deposition parameters, such as the type of hydrocarbon precursor and temperature, it is difficult to attain single layer graphene (SLG) across large areas, owing to the simultaneous nucleation/growth and solution/precipitation mechanisms. The differing characteristics of polycrystalline Ni films facilitate the growth of graphitic layers at different rates, resulting in regions with differing numbers of graphitic layers. High-quality films can be grown on Cu. Cu is available in a variety of shapes and forms, such as foils, bulks, foams, thin films on other materials and powders, making it attractive for industrial production of large area graphene films. The push to use CVD graphene in applications has also triggered a research line for the direct growth on insulators. The quality of the resulting films is lower than possible to date on metals, but enough, in terms of transmittance and resistivity, for many applications as described in section V. Transfer technologies are the focus of section VI. CVD synthesis of graphene on metals and bottom up molecular approaches require SLG to be transferred to the final target substrates. To have technological impact, the advances in production of high-quality large-area CVD graphene must be commensurate with those on transfer and placement on the final substrates. This is a prerequisite for most applications, such as touch panels, anticorrosion coatings, transparent electrodes and gas sensors etc. New strategies have improved the transferred graphene quality, making CVD graphene a feasible option for CMOS foundries. Methods based on complete etching of the metal substrate in suitable etchants, typically iron chloride, ammonium persulfate, or hydrogen chloride although reliable, are time- and resourceconsuming, with damage to graphene and production of metal and etchant residues. Electrochemical delamination in a low-concentration aqueous solution is an alternative. In this case metallic substrates can be reused. Dry transfer is less detrimental for the SLG quality, enabling a deterministic transfer. There is a large range of layered materials (LMs) beyond graphite. Only few of them have been already exfoliated and fully characterized. Section VII deals with the growth of some of these materials. Amongst them, h-BN, transition metal tri- and di-chalcogenides are of paramount importance. The growth of h-BN is at present considered essential for the development of graphene in (opto) electronic applications, as h-BN is ideal as capping layer or substrate. The interesting optical and electronic properties of TMDs also require the development of scalable methods for their production. Large scale growth using chemical/physical vapour deposition or thermal assisted conversion has been thus far limited to a small set, such as h-BN or some TMDs. Heterostructures could also be directly grown.

330 citations


Cites background from "A catalytic alloy approach for grap..."

  • ...Catalytic growth of SiC on Si with a NiCu coating [742] allows growth of graphene on predefined locations....

    [...]

  • ...Growing an epitaxial AlN layer on Si prior to SiC growth significantly reduces Si out-diffusion and helps grow higher quality graphene [738], as well as and interface NiCu layer [742]....

    [...]

Journal ArticleDOI
TL;DR: In this article, an ultrathin epitaxial graphite graphite (NPEG) was grown by thermal decomposition on the (0001) surface of 6H-SiC and characterized by surface-science techniques.
Abstract: We have produced ultrathin epitaxial graphite films which show remarkable 2D electron gas (2DEG) behavior. The films, composed of typically 3 graphene sheets, were grown by thermal decomposition on the (0001) surface of 6H-SiC, and characterized by surface-science techniques. The low-temperature conductance spans a range of localization regimes according to the structural state (square resistance 1.5 kOhm to 225 kOhm at 4 K, with positive magnetoconductance). Low resistance samples show characteristics of weak-localization in two dimensions, from which we estimate elastic and inelastic mean free paths. At low field, the Hall resistance is linear up to 4.5 T, which is well-explained by n-type carriers of density 10^{12} cm^{-2} per graphene sheet. The most highly-ordered sample exhibits Shubnikov - de Haas oscillations which correspond to nonlinearities observed in the Hall resistance, indicating a potential new quantum Hall system. We show that the high-mobility films can be patterned via conventional lithographic techniques, and we demonstrate modulation of the film conductance using a top-gate electrode. These key elements suggest electronic device applications based on nano-patterned epitaxial graphene (NPEG), with the potential for large-scale integration.

290 citations

Journal ArticleDOI
TL;DR: In this article, the authors review the enormous scientific and technological advances achieved in terms of epitaxial growth of graphene from thermal decomposition of bulk silicon carbide and fine control of the graphene electronic properties through intercalation.
Abstract: Graphene has been widely heralded over the last decade as one of the most promising nanomaterials for integrated, miniaturized applications spanning from nanoelectronics, interconnections, thermal management, sensing, to optoelectronics. Graphene grown on silicon carbide is currently the most likely candidate to fulfill this promise. As a matter of fact, the capability to synthesize high-quality graphene over large areas using processes and substrates compatible as much as possible with the well-established semiconductor manufacturing technologies is one crucial requirement. We review here, the enormous scientific and technological advances achieved in terms of epitaxial growth of graphene from thermal decomposition of bulk silicon carbide and the fine control of the graphene electronic properties through intercalation. Finally, we discuss perspectives on epitaxial graphene growth from silicon carbide on silicon, a particularly challenging area that could result in maximum benefit for the integration of graphene with silicon technologies.

182 citations

Journal ArticleDOI
TL;DR: In this article, the distinctive effects of nonequilibrium reactive chemistries and how these effects can aid and advance the integration of sustainable chemistry into each stage of nanotech product life are discussed.
Abstract: Sustainable societal and economic development relies on novel nanotechnologies that offer maximum efficiency at minimal environmental cost. Yet, it is very challenging to apply green chemistry approaches across the entire life cycle of nanotech products, from design and nanomaterial synthesis to utilization and disposal. Recently, novel, efficient methods based on nonequilibrium reactive plasma chemistries that minimize the process steps and dramatically reduce the use of expensive and hazardous reagents have been applied to low-cost natural and waste sources to produce value-added nanomaterials with a wide range of applications. This review discusses the distinctive effects of nonequilibrium reactive chemistries and how these effects can aid and advance the integration of sustainable chemistry into each stage of nanotech product life. Examples of the use of enabling plasma-based technologies in sustainable production and degradation of nanotech products are discussed—from selection of precursors derived ...

162 citations

01 Jan 2016
TL;DR: This review discusses the distinctive effects of nonequilibrium reactive chemistries and how these effects can aid and advance the integration of sustainable chemistry into each stage of nanotech product life.
Abstract: Sustainable societal and economic development relies on novel nanotechnologies that offer maximum efficiency at minimal environmental cost. Yet, it is very challenging to apply green chemistry approaches across the entire life cycle of nanotech products, from design and nanomaterial synthesis to utilization and disposal. Recently, novel, efficient methods based on nonequilibrium reactive plasma chemistries that minimize the process steps and dramatically reduce the use of expensive and hazardous reagents have been applied to low-cost natural and waste sources to produce value-added nanomaterials with a wide range of applications. This review discusses the distinctive effects of nonequilibrium reactive chemistries and how these effects can aid and advance the integration of sustainable chemistry into each stage of nanotech product life. Examples of the use of enabling plasma-based technologies in sustainable production and degradation of nanotech products are discussed—from selection of precursors derived from natural resources and their conversion into functional building units, to methods for green synthesis of useful naturally degradable carbon-based nanomaterials, to device operation and eventual disintegration into naturally degradable yet potentially reusable byproducts.

137 citations

References
More filters
Journal ArticleDOI
22 Oct 2004-Science
TL;DR: Monocrystalline graphitic films are found to be a two-dimensional semimetal with a tiny overlap between valence and conductance bands and they exhibit a strong ambipolar electric field effect.
Abstract: We describe monocrystalline graphitic films, which are a few atoms thick but are nonetheless stable under ambient conditions, metallic, and of remarkably high quality. The films are found to be a two-dimensional semimetal with a tiny overlap between valence and conductance bands, and they exhibit a strong ambipolar electric field effect such that electrons and holes in concentrations up to 10 13 per square centimeter and with room-temperature mobilities of ∼10,000 square centimeters per volt-second can be induced by applying gate voltage.

55,532 citations

Journal ArticleDOI
TL;DR: The roll-to-roll production and wet-chemical doping of predominantly monolayer 30-inch graphene films grown by chemical vapour deposition onto flexible copper substrates are reported, showing high quality and sheet resistances superior to commercial transparent electrodes such as indium tin oxides.
Abstract: The outstanding electrical, mechanical and chemical properties of graphene make it attractive for applications in flexible electronics. However, efforts to make transparent conducting films from graphene have been hampered by the lack of efficient methods for the synthesis, transfer and doping of graphene at the scale and quality required for applications. Here, we report the roll-to-roll production and wet-chemical doping of predominantly monolayer 30-inch graphene films grown by chemical vapour deposition onto flexible copper substrates. The films have sheet resistances as low as approximately 125 ohms square(-1) with 97.4% optical transmittance, and exhibit the half-integer quantum Hall effect, indicating their high quality. We further use layer-by-layer stacking to fabricate a doped four-layer film and measure its sheet resistance at values as low as approximately 30 ohms square(-1) at approximately 90% transparency, which is superior to commercial transparent electrodes such as indium tin oxides. Graphene electrodes were incorporated into a fully functional touch-screen panel device capable of withstanding high strain.

7,709 citations


"A catalytic alloy approach for grap..." refers result in this paper

  • ...In fact, this value is comparable with the best sheet resistance reported so far for graphene by Bae et al.(31) In Fig....

    [...]

  • ...In fact, this value is comparable with the best sheet resistance reported so far for graphene by Bae et al.31 In Fig....

    [...]

Journal ArticleDOI
TL;DR: The state of the art, future directions and open questions in Raman spectroscopy of graphene are reviewed, and essential physical processes whose importance has only recently been recognized are described.
Abstract: Raman spectroscopy is an integral part of graphene research. It is used to determine the number and orientation of layers, the quality and types of edge, and the effects of perturbations, such as electric and magnetic fields, strain, doping, disorder and functional groups. This, in turn, provides insight into all sp(2)-bonded carbon allotropes, because graphene is their fundamental building block. Here we review the state of the art, future directions and open questions in Raman spectroscopy of graphene. We describe essential physical processes whose importance has only recently been recognized, such as the various types of resonance at play, and the role of quantum interference. We update all basic concepts and notations, and propose a terminology that is able to describe any result in literature. We finally highlight the potential of Raman spectroscopy for layered materials other than graphene.

5,673 citations


"A catalytic alloy approach for grap..." refers background in this paper

  • ...The intensity of the 2D band around 2700 cm 1 is slightly higher than that of the G band (;1580 cm (1)), which indicates monolayer to bilayer graphene.(16) The spectrum in Fig....

    [...]

Journal ArticleDOI
TL;DR: In this paper, ultrathin epitaxial graphite films were grown by thermal decomposition on the (0001) surface of 6H−SiC, and characterized by surface science techniques.
Abstract: We have produced ultrathin epitaxial graphite films which show remarkable 2D electron gas (2DEG) behavior. The films, composed of typically three graphene sheets, were grown by thermal decomposition on the (0001) surface of 6H−SiC, and characterized by surface science techniques. The low-temperature conductance spans a range of localization regimes according to the structural state (square resistance 1.5 kΩ to 225 kΩ at 4 K, with positive magnetoconductance). Low-resistance samples show characteristics of weak localization in two dimensions, from which we estimate elastic and inelastic mean free paths. At low field, the Hall resistance is linear up to 4.5 T, which is well-explained by n-type carriers of density 1012 cm-2 per graphene sheet. The most highly ordered sample exhibits Shubnikov−de Haas oscillations that correspond to nonlinearities observed in the Hall resistance, indicating a potential new quantum Hall system. We show that the high-mobility films can be patterned via conventional lithographic...

3,315 citations

Related Papers (5)
Frequently Asked Questions (15)
Q1. What are the contributions mentioned in the paper "A catalytic alloy approach for graphene on epitaxial sic on silicon wafers" ?

In this paper, the first successful approach to transfer-free, direct growth of uniform and high-quality bilayer graphene over full silicon wafers was demonstrated. 

Sheet resistance measurements were taken using four electrical contacts in a van der Pauw configuration on the corners of AU51 1 cm2 samples. 

The reason why the SiC (100) template leads in their case to an even better graphene quality than SiC(111) could be related to the combination of a considerably higher tensile stress and higher surface defectivity of the latter epitaxial SiC as discussed in their previous work. 

The base pressure in the analysis chamber was below 1.0 10 8 Torr. Data analysis was performed with the CasaXPS software and a Shirley baseline with Kratos library relative sensitivity factors. 

Subsequent Ar1 ion milling was conducted in a Fiscione NanoMill™ to remove Ga ion damage and provide a high resolution transmission electron microscope (HRTEM) foil. 

This sheet resistance corresponds to a resistivity of about 2 10 8 X m for the graphene, as low as bulk Au metal; (b) sheet resistance measured on the ;23 nm Ni/Cu alloy as a reference. 

Note that the graphene prepared with Ni/Cu indicates a 50% decrease in RMS as compared to the use of Ni alone as a catalyst, and that the graphene roughness does not seem dependent on the orientation of the starting substrate. 

This allows for a fast precipitation and graphitization of C released through the Ni silicidation reaction, leading to an extraordinary improvement in the uniformity and quality of the graphene obtained through Ni/Cu. 

By averaging measurements taken on five samples fabricated with the same graphitization procedure over separate runs, the authors can conclude that the sheet resistance of their bilayer graphene prepared on a SiC(100) layer with Ni/Cu is around 24.8 X/square 6 0.7 from sample-to-sample variation. 

3. Example of Raman spectrum (graphene D, G, and 2D bands region) from graphene on SiC(100) prepared with ;8 nm Ni and ;15 nm Cu. Additionally to the low D over G band intensity, indicative of a good quality graphene, note that the 2D over G intensity ratio higher than 1 is indicative of a few-layer graphene. 

preliminary results indicate that the adhesion of graphene to the underlying substrate could be an order of magnitude higher than the adhesion of a graphene layer grown ex situ and transferred onto a SiO2 layer on silicon. 

for benchmarking purposes, it is meaningful to translate the sheet resistance of the bilayer graphene into a corresponding resistivity value by using the 0.9 nm thickness revealed by TEM. 

Note that as TEM observation is extremely challenging, because of shadowing induced by the topography of the sample surface (;15 nm RMS, Table I) and the necessity for 20–30 nm thick metal layer deposition on the top of thesample (and thus of the graphene) for TEM preparation, it is not surprising that the nanolayer is not clearly visible along the whole cross-sectional image. 

In fact, the ID/IG ratio of graphene on SiC (100) decreases from ;2 for 5 nm Ni catalyst alone to;0.8 when the Cu catalyst layer is added. 

these preliminary measurements show that it is unlikely that the tested samples in Fig. 7(a) contained any interface with adhesion energies as low as 0.45 J/m2, which is the adhesion value reported in the literature for a monolayer graphene transferred onto a SiO2 layer. 

Trending Questions (1)
What is the principle behind the graphene computer chip?

Specifically, we suggest that exceptional conduction qualifies this graphene as a metal replacement for MEMS and advanced on-chip interconnects with ultimate scalability.