Contents lists available at ScienceDirect
Diamond & Related Materials
journal homepage: www.elsevier.com/locate/diamond
Study of graphene layer growth on dielectric substrate in microwave plasma
torch at atmospheric pressure
Ondřej Jašek
⁎
, Jozef Toman, Jana Jurmanová, Miroslav Šnírer, Vít Kudrle, Vilma Buršíková
Department of Physical Electronics, Faculty of Science, Masaryk University, Kotlářská 2, 611 37 Brno, Czech Republic
A R T I C L E I N F O
Keywords:
Graphene
Microwave plasma
Ethanol
Dielectric substrate
A B S T R A C T
The initial stage of graphene layer deposition on silicon oxide substrate by ethanol decomposition in dualchannel
microwave plasma torch at atmospheric pressure was studied in dependence on precursor flow rate and
delivered microwave power. Depending on ethanol flow rate and substrate temperature, horizontally or vertically
aligned graphene nanosheets with various density could be prepared directly on dielectric substrate. In the
regime with high microwave power, above 400 W, mixture of amorphous carbon particles and graphene sheets
was deposited on the substrate. Prepared layers were analyzed by scanning electron microscopy (SEM), Raman
spectroscopy and X-ray photoelectron spectroscopy (XPS). The microwave plasma diagnostics was carried out
using optical emission spectroscopy (OES). The sample analysis showed increasing density of horizontally
aligned carbon nanosheets with increasing ethanol flow rate and their delamination and transition into vertically
aligned graphene sheets with increasing substrate temperature. The Raman spectroscopy analysis of layers
showed presence of D (1345 cm−1
), G (1585 cm−1
) and 2D (2685 cm−1
) peaks with 2D/G ratio of 1.59 and full
width at half maximum (FWHM) of 2D peak was 42 cm−1
, corresponding to few layer graphene structure. In
case of amorphous nanoparticles deposition, the D* peak at 1210 cm−1
and D** at 1500 sccm−1
was observed in
Raman spectra with D/G ratio of 1.19 and C1s XPS spectra of carbon contained 20.4 at.% of sp3
carbon phase in
comparison to 8.3 at.% in case of graphene nanosheets layer. High D/G ratio, up to 3.5, and low intensity 2D
band was characteristic for vertically aligned graphene nanosheets layers. The possibility to influence density
and size of graphene nanosheets on substrate represents promising alternative for future deposition of graphene
on arbitrary substrate.
1. Introduction
Graphene belongs among the most promising materials in both,
fundamental and applied science [1]. Today, exfoliation from graphite
crystals and chemical vapor deposition belong to most widely used
methods to prepare graphene layers on the substrate [2]. These
methods produce high quality layers in limited amounts or require
elaborate transfer techniques to dielectric substrates. Alternative approach
is controlled decomposition of organic precursors using plasma
enhanced chemical vapor deposition (PECVD) [3]. Such approach enables
synthesis of good quality material in reasonable amounts. It is of
great interest to grow graphene layers directly on dielectric substrates.
There are two main approaches to achieve growth of graphene on the
dielectric substrate. First approach uses solid carbon source deposited
onto a desired dielectric substrate which can be transformed into graphene
films by high temperature annealing. In order to accelerate the
decomposition of solid carbon sources, capping metal layers such as Cu
or Ni deposited on carbon sources can be used [4]. Second approach is
direct CVD of graphene on dielectric substrate with introduction of
metal vapors or use of plasma enhanced CVD process. While there has
been some success in these methods, the graphene nucleation density,
the growth rate on dielectric substrates as compared to metallic substrates
and the size of graphene single crystal domains is usually small
(about few micrometers) [5]. There was also reported growth of graphene
directly on dielectric substrate by PECVD where the decomposition
of carbon precursor enabled growth at lower temperature,
however, until now this method resulted only in limited success
(450–500 °C) and small domain sizes of graphene [6,7]. Another approach
to grow graphene on dielectric substrate without metallic catalyst
is so called vertically aligned few layer graphene (FLG) or carbon
nanowalls. Such a growth was first reported by Malesevic et al. for
synthesis of FLG from mixture of methane and hydrogen in microwave
PECVD [8]. Overall, wide range of plasma sources was used for
synthesis of FLG including direct current (DC), radio-frequency (RF)
https://doi.org/10.1016/j.diamond.2020.107798
Received 1 November 2019; Received in revised form 3 March 2020; Accepted 3 March 2020
⁎
Corresponding author.
E-mail address: jasek@physics.muni.cz (O. Jašek).
Diamond & Related Materials 105 (2020) 107798
Available online 05 March 2020
0925-9635/ © 2020 Elsevier B.V. All rights reserved.
T
capacitive and inductively coupled to microwave (MW) power sources,
and naturally, PECVD has become main method for FLG synthesis [9].
Typical growth pressure range is several Pascals for capacitive coupled
systems to 103
–104
Pa for DC and microwave powered setups. The
growth temperature was limited by surface reaction kinetics and varies
from 750 to 1300 K and all experiments, except MW setups, used external
resistive heating for the elevation of substrate temperature.
Among substrates used for deposition were silicon, quartz, metals or
their combination and precursors were mixture of H2 with hydrocarbon
(CH4, C2H2) or fluorocarbons, such as CF4, CHF3 or C2F6. To achieve
better quality of the layer combination of oxygen containing precursors
such as CO2 and CO together with H2 and CH4 were used. It is worth
noting that the systems used for growth of FLG are almost always low
pressure systems and experiments at atmospheric pressure were only
partially successful and resulted in mixture of various forms of carbon
allotropes. Growth using atmospheric pressure DC normal glow discharge
in Ar/CH4 and Ar/H2/ethanol was reported on Si, Cu and
stainless steel substrate and studied by Bo et al. [10] and applied as gas
sensor by Yu et al. [11]. Meško et al. [12] reported growth on Ni using
atmospheric pressure DC glow PECVD in mixture of Ar/H2/ethanol or
hexane and reported growth rates as high as 100 nm per minute at
1000 K. This approach was in detail reviewed by Bo et al. [9] and recently
by Vasell et al. [13] and Santhosh et al. [14]. Alternative method
for the growth of graphene was developed by Dato et al. [15] concerning
microwave plasma decomposition of ethanol or dimethyl ether
for synthesis of graphene nanosheets in the gas phase. This method was
further refined by Tatarova et al. [16] and Tsyganov et al. [17] also
developed a theoretical model of the synthesis process and could simulate
and experimentally map the particle and thermal fluxes in their
plasma reactor. Melero et al. also showed that decomposition of ethanol
in TIAGO torch at atmospheric pressure could be used for synthesis of
carbon nanotubes and graphene nanosheets without use of catalyst
[18,19] and efficient hydrogen generation [20]. This approach was
recently extended to N-doping and the use of methane as precursor by
Bundaleska et al. [21,22]. In our work we have showed that dual
channel microwave plasma torch could be used for direct growth of
carbon nanotubes and nanofibers on the substrate [23] and that these
could be used as sensors [24] with comparable performance to graphene
oxide [24,25]. In comparison with hydrocarbon precursors, the
use of ethanol enables robust system for nucleation and assembly of
graphene layer without the need to add another reactive gas, such as
H2, into deposition mixture. This is caused by ethanol structure of CH3CH2OH,
which is at high temperature, above 2000 K, decomposed
predominantly into CO and H2 molecules and C2 species responsible for
graphene nucleation and growth. This way simple, single precursor,
system can be used for bottom-up synthesis of graphene with various
properties. In this work we study formation of carbon based layers on
dielectric substrate by decomposition of ethanol in dual-channel microwave
plasma torch at atmospheric pressure. We study formation of
basic building blocks directly on dielectric substrate (Si/SiO2) in dependence
on ethanol precursor flow rate, delivered microwave power
and substrate temperature.
2. Materials and methods
The graphene nanosheets were synthesized by ethanol decomposition
in dual-channel microwave plasma torch at atmospheric pressure.
The microwave discharge was ignited inside reactor formed by quartz
tube (80 mm diameter, 200 mm length) terminated by dural flanges.
The discharge electrode was hollow carbon nozzle with central channel
used for introduction of working gas - argon Qc (300–500 sccm) and
subsequent ignition of plasma. The secondary channel (annulus with
outer radius 8.4 mm and inner radius 7.7 mm) was used for introduction
of carrying gas - argon Qs (35–360 sccm) with precursor (ethanol 2
to 19 mg/min) vapors into the plasma environment. Deposition time
was 120 s. The layers were deposited on 10 × 10 mm one-side polished
Si (100) P type, 525 mm thickness wafers from ON Semiconductor s.r.o.
(Czech Republic) with 92 nm thermally oxidized SiO2 layer and fixed in
the holder during the deposition Fig. 1. More details about the experimental
setup can be found in [26]. Raman spectroscopy was carried
out using HORIBA LabRAM HR Evolution system with 532 nm laser,
using 100× objective and 25% ND filter in the range from 1000 to
3200 cm−1
. Samples were imaged with TESCAN scanning electron
microscope (SEM) MIRA3 with Schottky field emission electron gun
equipped with secondary electron (SE) and back-scattered electron
(BSE) detectors as well as Oxford Instruments EDX analyser. Transmission
electron microscopy (TEM) was carried out using JEOL JEM-
2100F and FEI Tecai F20 microscope. ImageJ Fiji software was used for
determination of particle size and fast Fourier transform (FFT) image
analysis. Particle size analysis was carried out in following sequence:
image scale calibration, denoise by median filter, adaptive threshold
procedure was applied for particle selection and analyze particles
dialog was used for particle size determination excluding particles on
image edges (area, median ferret diameter (particle size) i.e. the longest
Fig. 1. Detail of carbon plasma nozzle and deposition process of graphene layer on the left and photographs of the discharge for Qc 500 sccm and a) Qs 70 sccm and
PMW 105 W, b) Qs 70 sccm and PMW 175 W, c) Qs 140 sccm and PMW 175 W and d) Qc 360 sccm, Qs 300 sccm and PMW 420 W.
O. Jašek, et al. Diamond & Related Materials 105 (2020) 107798
2
distance between any two points along the selection boundary, also
known as maximum caliper). We use median value for the particle size
because the particle size distribution was systematically skewed to
lower values i.e. there were more smaller nanoparticles than bigger
ones on the substrate. The standard deviation of the particle size distribution
was large, 50% and more of the mean value, due to the broad
range of size and irregular shape of the nanosheets. Atomic force microscopy
(AFM) was carried out using atomic force microscope
NTEGRA Prima NT-MDT in semi-contact mode. Substrate temperature
was measured by K thermocouple with compensation line JUMO
901250/32-1043-1.5-300-48-2500. Optical emission spectra were recorded
using a Jobin-Yvon Triax 550 spectrometer (1200 g mm−1
,
3600 g mm−1
) with an LN2 cooled CCD detector and an Avantes ULS3648TEC-USB2
overview spectrometer with UA grating. The rotational
temperature was estimated by fitting experimentally obtained CN
(B2
Σ+
-X2
Σ+
) spectra at 390 nm using simulation in the Massive OES
software [27]. Ethanol p.a. 99.8% from Penta was purchased from
Verkon (Czech Republic). Pure gas, argon 99.998%, was purchased
from Messer Technogas, s.r.o. (Czech Republic).
3. Results and discussion
3.1. Microwave plasma torch diagnostics and substrate temperature
measurement
Microwave plasma at atmospheric pressure represents high temperature
system reaching up to 5000 K during decomposition of ethanol
as shown by Tsyganov et al. [14] and Rincon et al. [28]. At such a high
temperature main products of plasma combustion of ethanol are C
atoms and C2, CO and H2 molecules. C2 molecules with small admixture
of C2Hx hydrocarbons are then the main precursors for the growth of
carbon nanostructures. In our experiment, the concentration of these
species was controlled by ethanol flow rate and absorbed microwave
power delivered to the plasma (PMW). The substrate temperature was
above all influenced by delivered microwave power and the distance
between the substrate and the discharge. It can be seen in Fig. 1 that the
delivered microwave power was the main deposition parameter influencing
discharge properties. The microwave power absorbed in the
plasma influenced not only its shape and gas flow dynamics but also
plasma temperature which increased with input power. Higher power
absorption led to decomposition of carbon molecules and the ratio of
C2/C plasma species was decreasing in case of lower flow rate of
ethanol precursor. On the other hand, higher flow rate of ethanol led to
growth of the ratio of C2/C even while delivered microwave power
increased. The main part of the discharge was excitation zone - plasma
plume, exhibiting typical green colour caused by emission of molecular
C2 Swan system between 420 and 620 nm with maximum at 517 nm.
This region's length was increasing as the plasma temperature increased
with absorbed microwave power.
Optical emission spectra of microwave plasma torch discharge in
argon with ethanol admixture can be seen in Fig. 2. We could observe
emission spectra of argon (4p-4s above 690 nm), hydrogen Hɑ
(656.3 nm) and C atomic line at 247.8 nm. Besides main molecular
bands of carbon radical C2 Swan system (d3
Πg-a3
Πu) at 420–620 nm we
observed molecular bands of hydroxyl radical OH (A2
Σ+
-X2
Π
303–315 nm), second positive system of N2 (C3
Πu-B3
Πg 315–380 nm),
NH (A3
Π+
-X3
Σ 336 nm) and violet system of CN (B2
Σ+
-X2
Σ+
,
344–460 nm) as well. The presence of OH and N related species in the
discharge was caused by small atmospheric impurities in the discharge
chamber. The large continuum background in high power regime was
caused by thermal radiation of carbon nanoparticles. The plasma temperature,
i.e. neutral gas temperature, determined by fitting of rotational
structure of CN (B2
Σ+
-X2
Σ+
) band at 390 nm depended linearly
on delivered microwave power and saturated with increasing ethanol
flow rate. The temperature increased from 3100 K for lowest Qs flow
rate and lowest power (Qs 35 sccm, PMW 105 W) to 4200 K for highest
Qs and PMW values. As the plasma temperature influenced the substrate
temperature, we carried out temperature measurement of the substrate
placed at variable distance from the plasma discharge. Because the
plasma discharge length changed, we used as a figure of merit the
distance of the substrate from the plasma nozzle. It can be seen in Fig. 3,
the substrate temperature was mainly influenced by plasma power i.e.
proximity of the discharge itself, and only weakly depended on gas and
precursor flow rates. The substrate position in the vicinity of the hot
plasma zone led to sharp increase of the substrate temperature. If the
substrate was placed directly into plasma hot zone it was irreversibly
damaged i.e. substrate temperature increased above its melting point.
In case of lower delivered microwave power, the discharge length reduced
and it was possible to place the substrate to lower position but
obviously, the maximum temperature of the substrate reachable
without damage remained the same.
3.2. Graphene layer deposition in dependence on precursor flow
The growth of carbon layer on dielectric substrate requires formation
of nucleation centers on the substrate surfaces from which the layer
continues to growth during the deposition phase. This can be achieved
by in situ process such as bias enhanced nucleation [29] in case of
nanocrystalline diamond or by mechanical exfoliation and transfer to
Fig. 2. Optical emission spectra of microwave plasma torch. Comparison of Qc
500 sccm, Qs 70 sccm, PMW 175 W and Qc 360 sccm, Qs 300 sccm, PMW 420 W.
Fig. 3. Dependence of substrate temperature on delivered microwave power.
O. Jašek, et al. Diamond & Related Materials 105 (2020) 107798
3
substrate in the case of graphene [5]. In our experiment this process
was represented by in situ formation of carbon nanostructures in microwave
plasma in the vicinity of the substrate. At first, we investigated
influence of precursor flow rate on the formation of graphene nanosheets
and their deposition on the substrate. Because the aim of this
study was to form single layer of graphene material on the substrate we
chose low precursor liquid mass flow rates Qe from 2 to 9 mg/min of
ethanol i.e. low flow rate of argon carrier gas Qs (35–140 sccm) through
bubbler with liquid ethanol (Table 1). High flow rate of the ethanol Qs
(500–1400 sccm) led to formation of large amount of deposit as shown
in our previous work [26].
The substrate was placed 50 mm from the discharge nozzle. After
the deposition whole substrate was covered with graphene nanosheets
layer with small inhomogeneities around substrate edges due to the gas
flow turbulence. Detailed SEM analysis of the substrates showed gradual
increase of the density and size of graphene nanosheets on the
surface as can be seen on Fig. 4.
For lowest flow rate only very small carbon nanoparticles could be
found on the substrate. Median nanoparticle diameter of nanoparticles
was 38 nm and nanoparticles covered 5% of the substrate surface.
Further increase of ethanol flow rate led to graphene nanosheets coverage
of the substrate and the coverage increased for flow rate of 70
sccm to 105 sccm, respectively. The area covered by the graphene nanosheets
increased from 9 to 13% and median nanosheet size increased
as well, from 83 to 146 nm, for higher flow rate. For flow rate of 140
sccm the amount of sheets substantially increased and they started to
form 3D structure on the substrate. This result was in agreement with
our observation of increase of C2 molecules concentration with the increase
of ethanol flow rate as determined by OES. C2 molecule was
main plasma species responsible for graphene nanosheets growth i.e.
increase of their lateral size and amount. AFM analysis of substrate
surface showed that the height of the nanosheets covering the substrate
was between 1 and 2 nm. Raman analysis of samples showed presence
of D (1345 cm−1
), G (1585 cm−1
) and 2D (2694 cm−1
) band in the
Table 1
Deposition conditions and Raman analysis of prepared samples.
Sample No Qc (sccm) Qs (sccm) Qe (mg/min) Pmw (W) D (cm−1
) G (cm−1
) 2D (cm−1
) 2DFWHM (cm−1
) 2D/G D/G
GS35P175 500 35 2.2 175 NA NA NA NA NA NA
GS70P175 500 70 4.5 175 1346 1585 2694 49 0.76 1.48
GS105P175 500 105 6.7 175 1345 1578 2692 42 1.59 1.02
GS140P175 500 140 8.9 175 1348 1581 2695 42 1.55 1.15
GS70P105 500 70 4.5 115 NA NA NA NA NA NA
GS70P140 500 70 4.5 145 1347 1584 2685 56 2.05 1.29
GS70P140HT1 500 70 4.5 145 1345 1581 2689 63 0.94 3.48
GS70P140HT2 500 70 4.5 145 1344 1586 2695 79 0.39 2.59
GS70P140HT3 500 70 4.5 145 1348 1584 2696 60 1.01 1.07
GS300P420 360 300 19.2 420 1347 1579 2691 58 0.81 1.19
GS300P455 360 300 19.2 455 1346 1577 2691 59 1.29 1.1
Fig. 4. SEM micrograph of deposited layers on Si/SiO2 substrate for Qs a) 35 sccm b) 70 sccm c) 105 sccm and d) 140 sccm.
O. Jašek, et al. Diamond & Related Materials 105 (2020) 107798
4
Raman spectra (Fig. 5, left). For appropriate analysis of the Raman
spectra, fitting of all bands with Lorentz peak was performed with exception
of the D** band, where Gaussian profile is commonly used
[30,31]. The peak positions, their full width at half maximum (FWHM)
and intensity ratios can be found in Table 1. The reader should not
misinterpret the zero intensity red line as background, since the Lorentz
peaks were fitted without range restriction, thus it represent the gradual
(quadratic) decline of Lorentz peak function to zero. The background
was subtracted prior to peak fitting and was approximated by constant
value in the whole measured range.
2D/G band intensity ratio increased from 0.8 to 1.6 with increase of
Qs flow rate from 70 to 140 sccm, respectively. Only very weak signal
with high noise was obtained on sample GS35P175 and the spectra
could not be evaluated. The full width at half maximum (FWHM) of 2D
band varied from 42 to 49 cm−1
corresponding to few layer graphene
nanosheets [32] and was consistent with results of AFM, 0.34 nm per
layer. Other bands observed in Raman spectra were D' at 1620 cm−1
corresponding to defects in the structure, and second order bands at
2450 cm−1
and 2940 cm−1
(D + G). As a consequence of very thin
layer deposited on the substrate, silicon peak at 521 cm−1
and the
second order modes between 900 and 1000 cm−1
were observed as
well. XPS analysis and deconvolution of C1s peak region for sample
with 140 sccm flow rate (Fig. 5, right) revealed presence of sp.
(283.5 ± 0.1 eV), sp2
(284.4 ± 0.1 eV), sp3
(285.1 ± 0.1 eV) and
oxygen related groups CeO (286.0 ± 0.2 eV), C]O (286.9 ± 0.2 eV)
and O-C=O (288.1 ± 0.2 eV). Similarly, as in the case of well graphitised
graphene nanosheets [26], the dominant carbon phase was sp2
hybridisation. However, the sp2
peak was also complemented with 8.4
at.% of sp3
phase and also, a peak at lower binding energies, was observed
which is mostly assigned to sp. hybridized carbon [33] or vacancies
in the nanosheet structure [34,35]. TEM analysis of synthesized
nanosheets (Fig. 6a) showed few-layer graphene structure with interlayer
distance of 0.34 nm (image inset). The size and thickness of the
nanosheets was consistent with results of SEM and curved and overlapping
edges of the nanosheets could be observed as well. This can be
seen in detail in the inset of Fig. 6a, were gradual increase of nanosheet
thickness by two layers can be seen on top left forming few-layer graphene
structure (FLG). FFT analysis of the in-plane structure in the
bottom right section showed 6-fold symmetry of overlapping twisted
bilayer (TBL) by 29° consistent with observations of Limbu et al. [36]
and Wong et al. [37]. The formation of curved edges is known phenomena
to minimize energy of nanosheet structure due to the presence
of high energy dangling bonds on its edges [38]. The presence of the
corrugations, edges and topological defects caused the relatively high
intensity of D band in our samples in comparison with CVD growth
graphene. The degree of disorder in our graphene nanosheets can be
seen in Fig. 6b, were single layer structure (1 L), as proved by 6 single
points in its FFT image analysis (Fig. 6b inset), is overlaid with incomplete
second (2 L) and third layer (3 L). These observations are in
agreement with TEM analysis of graphene nanosheets by Dato et al.
[15]. We did not observe any carbon nanotubes in SEM or TEM mi-
crographs.
3.3. Influence of delivered microwave power on deposition of graphene
layer
We further investigated influence of delivered microwave power on
carbon nanostructures deposited on the substrate. Delivered microwave
power was limited by discharge stability. In regime with low precursor
flow rate, below Qs of 100 sccm, the discharge became unstable for
microwave power of 200 W and more. In this regime stable discharge
transitioned into pulsing mode with high erosion of the electrode and
contamination of the deposited layer. Low power mode was investigated
for PMW of 105 W and 145 W for Qs flow rate of 70 sccm. The
samples were placed 50 mm from the carbon nozzle and the substrate
temperature was 900 K. The result of deposition with low delivered
microwave power can be seen Fig. 7.
Similarly to the deposition conditions of GS35P175 sample with
PMW of 175 W, we obtained small carbon nanostructures with median
nanoparticle diameter of 48 nm but their density on the substrate was
lower under selected conditions, area covered by the nanoparticles was
2%. For the intermediate setting of 145 W, we could observe sparse
carbon nanosheets on the substrate. The area covered by the nanosheets
was around 1.5% and their size was around 350 nm. Microwave power,
105 vs 175 W, had thus much higher influence on the deposition of
carbon nanostructures and horizontally aligned graphene nanosheets
than precursor flow rate (70 vs 35 sccm). This can be explained by
different concentration of C2 molecules in the discharge, due to lower
decomposition rate of ethanol. The decomposition rate was influenced
by plasma dynamics and its temperature that was 3100 and 3800 K for
105 and 145 W, respectively. Raman spectroscopy of sample GS70P140
(Fig. 8) showed high intensity of 2D band at 2685 cm−1
with FWHM of
56 cm−1
corresponding to bi- or trilayer graphene nanosheets.
High power mode was represented by delivered microwave power
of 420 and 455 W. In this case the size of the plasma increased substantially
and the substrate had to be placed 10 cm from the carbon
Fig. 5. Measured Raman spectra (black) with fitted peaks (red) of deposited layer for Qs of 140 sccm (left) and its XPS analysis (right). (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of this article.)
O. Jašek, et al. Diamond & Related Materials 105 (2020) 107798
5
nozzle. To obtain the stable plasma the central channel flow rate was
lowered to 360 sccm and Qs flow rate was increased to 300 sccm. The
deposition time was 300 s and the substrate temperature was 870 and
940 K for 420 and 455 W, respectively. SEM analysis of the sample
showed mixture of carbon nanoparticles with embedded graphene nanosheets
as can be seen in Fig. 9.
Raman spectroscopy of the sample showed broad structure of the D
peak at 1347 cm−1
which was fitted with shoulder peaks D* and D** at
1250 and 1500 cm−1
(Fig. 10, left). The highly defective structure was
caused by high content of carbon atoms and C2 molecules in the plasma
which resulted in growth of amorphous phase in the gas phase and
which was deposited on the substrate and continued to grow there. This
was in agreement with results of Dato et al. [15] where high content of
free carbon atoms led to formation of amorphous phase. This result was
further confirmed by XPS analysis of C1s spectra of the sample (Fig. 10,
right). The content of sp3
phase at 285.1 eV was 20.4 at.% and 44.5 at.
% of sp2
phase without a signal at lower binding energies signifying
presence of sp. carbon phase as in previous, lower ethanol flow
rate and microwave power case. Such high content of sp3
phase together
with an intensive D band region in Raman spectroscopy (especially
the broad D** band) corresponded to the amorphous carbon
nanoparticles structure. The content of carbon‑oxygen bonds also increased
to 29.1 at.%.
Fig. 6. TEM analysis a) of graphene nanosheets synthesized in microwave plasma torch (deposition time 5 min, substrate distance 300 mm) with detail of few-layer
graphene layer and b) detail of single layer graphene nanosheet with overlaid second and third layer, inset- single layer FFT analysis.
Fig. 7. SEM micrograph of graphene layers deposited at low power mode of a) 105 W and b) 145 W, substrate temperature 900 K.
Fig. 8. Measured Raman spectrum (black) with peaks fitted (red) of graphene
layer deposited at Qs 70 sccm and PMW of 145 W. (For interpretation of the
references to colour in this figure legend, the reader is referred to the web
version of this article.)
O. Jašek, et al. Diamond & Related Materials 105 (2020) 107798
6
3.4. Influence of close proximity of the plasma active zone on deposition of
graphene layer
As discussed above, flow of growth species towards substrate surface
played important role in the deposition process. Therefore we investigated
the influence of close vicinity of the plasma region on the
deposition process. In the experiment we used precursor flow rate of 70
sccm and applied microwave power of 140 W. The distance between the
substrate and plasma nozzle was decreased to 40 mm. We also carried
out this experiment also for PMW of 175 W but at distance of 40 mm the
silicon substrate was damaged at the point of contact with the plasma.
Due to the very small distance between active plasma region, below
10 mm, there was an inhomogeneous flow of the active plasma species
towards substrate surface and radial temperature profile was formed as
well i.e. temperature in the center of the substrate was higher than at its
edges. According to our measurement the substrate temperature was
above 1000 K and as previously mentioned, at higher power, the silicon
substrate could reach on the surface closest to the plasma melting
temperature of silicon (1687 K). The results of the deposition can be
seen in Fig. 11.
The analysis showed that the growth transitioned from horizontal to
vertical with increasing flux of active plasma species and the substrate
temperature. This observation was in agreement with work of
Malesevic et al. [8] who suggested that the initial horizontal growth
converts into vertical growth by partial cracking and delamination of
nucleation layer. This conclusion was also supported by our early
observation of layers deposited with variable flow rate, where part of
the nanosheets edges was delaminated from the surface and formed
natural nucleation sites for vertical growth. High temperature and high
flux of C2 were also suggested as necessary condition for the preferential
growth of carbon nanowalls by Cheng et al. [39] during the
synthesis of nanodiamond layers. Observed growth at the substrate
temperature of 1000 K was also in agreement with the values between
970 and 1100 K reported by Bo et al. [10] and Meško et al. [12] for
atmospheric pressure DC PECVD growth of carbon nanowalls. Raman
spectra (Fig. 11, top right) of the graphene nanosheets in Fig. 11a exhibited
high intensity D peak at 1345 cm−1
and D′ peak at 1608 cm−1
with low intensity of second order peaks. This observation was consistent
with features found in Raman spectra of vertically aligned gra-
phene.
4. Conclusions
Horizontally aligned layer of graphene nanosheets was successfully
deposited on Si/SiO2 substrate by decomposition of ethanol in dualchannel
microwave plasma torch at atmospheric pressure. Median size
and density of the graphene nanosheets forming the layer increased,
from 83 to 146 nm, with increasing ethanol flow rate and delivered
microwave power which determined amount of C atoms and C2 molecules
in the plasma. Simultaneously, the 2D/G band ratio increased
from 0.76 to 1.55 and 2D peak FWHM decreased from 49 to 42 cm−1
.
The composition of the carbon layer on the substrate could be changed
Fig. 9. Carbon layer deposited in high power mode Qc 360 sccm, Qs 300 sccm a) 420 W and b) 455 W.
Fig. 10. Measured Raman spectra (black) with fitted peaks (red) of deposited layer for PMW of 455 W (left) and its XPS analysis (right). (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of this article.)
O. Jašek, et al. Diamond & Related Materials 105 (2020) 107798
7
in controlled manner from horizontally aligned graphene nanosheets to
vertically aligned few layer graphene by increasing substrate temperature
above 1000 K or to amorphous carbon nanoparticles by increasing
ethanol flow rate and delivered microwave power. For both
types of layers, the intensity of D*, D and D** bands in Raman spectra
increased substantially, with D/G band ratio between 1.2 and 3.5, while
2D/G band ratio decreased below 1. The amount of disorder in the
amorphous nanoparticles structure was also accompanied with high
content of 20.4 at.%. of sp3
phase determined by deconvolution of C1s
peak region in XPS spectra. The ability to controllably grow wide range
of carbon based layers directly on dielectric substrate using simple
precursor at atmospheric pressure plasma system represents unique
opportunity for future applications such as transparent conductive
layers, sensors and carbon functional coatings.
CRediT authorship contribution statement
Ondřej Jašek:Conceptualization, Investigation, Writing - original
draft, Writing - review & editing.Jozef Toman:Investigation, Formal
analysis, Writing - original draft.Jana Jurmanová:
Investigation.Miroslav Šnírer:Investigation, Writing - original
draft.Vít Kudrle:Writing - review & editing.Vilma
Buršíková:Investigation.
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgements
This work was supported by the Czech Science Foundation under
project 18-08520S and in part by project LM2018097 funded by The
Ministry of Education, Youth and Sports of the Czech Republic. We
acknowledge CEITEC Nano Research Infrastructure supported by MEYS
CR (LM2018110). We would like to thank Jiri Bursik for TEM analysis.
References
[1] A.C. Ferrari, F. Bonaccorso, V. Fal'ko, K.S. Novoselov, S. Roche, P. Bøggild,
S. Borini, F.H.L. Koppens, V. Palermo, N. Pugno, J.A. Garrido, R. Sordan, A. Bianco,
L. Ballerini, M. Prato, E. Lidorikis, J. Kivioja, C. Marinelli, T. Ryhänen, A. Morpurgo,
J.N. Coleman, V. Nicolosi, L. Colombo, A. Fert, M. Garcia-Hernandez, A. Bachtold,
G.F. Schneider, F. Guinea, C. Dekker, M. Barbone, Z. Sun, C. Galiotis,
A.N. Grigorenko, G. Konstantatos, A. Kis, M. Katsnelson, L. Vandersypen,
A. Loiseau, V. Morandi, D. Neumaier, E. Treossi, V. Pellegrini, M. Polini,
A. Tredicucci, G.M. Williams, B.H. Hong, J.-H. Ahn, J.M. Kim, H. Zirath, B.J. van
Wees, H. van der Zant, L. Occhipinti, A. Di Matteo, I.A. Kinloch, T. Seyller,
E. Quesnel, X. Feng, K. Teo, N. Rupesinghe, P. Hakonen, S.R.T. Neil, Q. Tannock,
T. Löfwander, J. Kinaret, Science and technology roadmap for graphene, related
two-dimensional crystals, and hybrid systems, Nanoscale 7 (2015) 4598–4810.
[2] F. Bonaccorso, A. Lombardo, T. Hasan, Z. Sun, L. Colombo, A.C. Ferrari, Production
and processing of graphene and 2d crystals, Mater. Today 15 (2012) 564–589,
https://doi.org/10.1016/s1369-7021(13)70014-2.
[3] A. Dato, Graphene synthesized in atmospheric plasmas—a review, J. Mater. Res. 34
(2019) 214–230, https://doi.org/10.1557/jmr.2018.470.
[4] Q.-Q. Zhuo, Q. Wang, Y.-P. Zhang, D. Zhang, Q.-L. Li, C.-H. Gao, Y.-Q. Sun, L. Ding,
Q.-J. Sun, S.-D. Wang, J. Zhong, X.-H. Sun, S.-T. Lee, Transfer-free synthesis of
doped and patterned graphene films, ACS Nano 9 (2015) 594–601.
[5] D. Wei, Y. Lu, C. Han, T. Niu, W. Chen, A.T.S. Wee, Critical crystal growth of
graphene on dielectric substrates at low temperature for electronic devices, Angew.
Chem. Int. Ed. Eng. 52 (2013) 14121–14126.
[6] D. Wei, L. Peng, M. Li, H. Mao, T. Niu, C. Han, W. Chen, A.T.S. Wee, Low
Fig. 11. Deposition of vertically oriented graphene with substrate in the vicinity of the discharge a) 3 mm from the edge of the substrate and its Raman spectra measured
spectra (black) and fitted peaks (red), b) 4 mm from the edge of the substrate and c) center of the substrate. (For interpretation of the references to colour in
this figure legend, the reader is referred to the web version of this article.)
O. Jašek, et al. Diamond & Related Materials 105 (2020) 107798
8
temperature critical growth of high quality nitrogen doped graphene on dielectrics
by plasma-enhanced chemical vapor deposition, ACS Nano 9 (2015) 164–171.
[7] M. Meyyappan, J.-S. Lee, Graphene growth by plasma-enhanced chemical vapor
deposition (PECVD), Plasma Processing of Nanomaterials (2017) 231–243, https://
doi.org/10.1201/b11473-9.
[8] A. Malesevic, R. Vitchev, K. Schouteden, A. Volodin, L. Zhang, G. Van Tendeloo,
A. Vanhulsel, C. Van Haesendonck, Synthesis of few-layer graphene via microwave
plasma-enhanced chemical vapour deposition, Nanotechnology 19 (2008) 305604.
[9] Z. Bo, Y. Yang, J. Chen, K. Yu, J. Yan, K. Cen, Plasma-enhanced chemical vapor
deposition synthesis of vertically oriented graphene nanosheets, Nanoscale 5 (2013)
5180–5204.
[10] Z. Bo, K. Yu, G. Lu, P. Wang, S. Mao, J. Chen, Understanding growth of carbon
nanowalls at atmospheric pressure using normal glow discharge plasma-enhanced
chemical vapor deposition, Carbon 49 (2011) 1849–1858, https://doi.org/10.
1016/j.carbon.2011.01.007.
[11] K. Yu, Z. Bo, G. Lu, S. Mao, S. Cui, Y. Zhu, X. Chen, R.S. Ruoff, J. Chen, Growth of
carbon nanowalls at atmospheric pressure for one-step gas sensor fabrication,
Nanoscale Res. Lett. 6 (2011) 202.
[12] M. Meško, V. Vretenár, P. Kotrusz, M. Hulman, J. Šoltýs, V. Skákalová, Carbon
nanowalls synthesis by means of atmospheric dcPECVD method, Phys. Status Solidi
B 249 (2012) 2625–2628, https://doi.org/10.1002/pssb.201200144.
[13] A. Vesel, R. Zaplotnik, G. Primc, M. Mozetič, Synthesis of vertically oriented graphene
sheets or carbon nanowalls-review and challenges, Materials. 12 (2019).
doi:https://doi.org/10.3390/ma12182968.
[14] N.M. Santhosh, G. Filipič, E. Tatarova, O. Baranov, H. Kondo, M. Sekine, M. Hori, K.
K. Ostrikov, U. Cvelbar, Oriented carbon nanostructures by plasma processing: recent
advances and future challenges, Micromachines (Basel). 9 (2018). doi:https://
doi.org/10.3390/mi9110565.
[15] A. Dato, M. Frenklach, Substrate-free microwave synthesis of graphene: experimental
conditions and hydrocarbon precursors, New J. Phys. 12 (2010) 125013,
https://doi.org/10.1088/1367-2630/12/12/125013.
[16] E. Tatarova, A. Dias, J. Henriques, A.M.B. do Rego, A.M. Ferraria, M.V. Abrashev, C.
C. Luhrs, J. Phillips, F.M. Dias, C.M. Ferreira, Microwave plasmas applied for the
synthesis of free standing graphene sheets, Journal of Physics D: Applied Physics. 47
(2014) 385501. doi:https://doi.org/10.1088/0022-3727/47/38/385501.
[17] D. Tsyganov, N. Bundaleska, E. Tatarova, A. Dias, J. Henriques, A. Rego, A. Ferraria,
M.V. Abrashev, F.M. Dias, C.C. Luhrs, J. Phillips, On the plasma-based growth of
“flowing” graphene sheets at atmospheric pressure conditions, Plasma Sources Sci.
Technol. 25 (2016) 015013, , https://doi.org/10.1088/0963-0252/25/1/015013.
[18] C. Melero, R. Rincón, J. Muñoz, G. Zhang, S. Sun, A. Perez, O. Royuela, C. GonzálezGago,
M.D. Calzada, Scalable graphene production from ethanol decomposition by
microwave argon plasma torch, Plasma Phys. Controlled Fusion 60 (2018) 014009,
, https://doi.org/10.1088/1361-6587/aa8480.
[19] R. Rincón, C. Melero, M. Jiménez, M.D. Calzada, Synthesis of multi-layer graphene
and multi-wall carbon nanotubes from direct decomposition of ethanol by microwave
plasma without using metal catalysts, Plasma Sources Sci. Technol. 24 (2015)
032005, , https://doi.org/10.1088/0963-0252/24/3/032005.
[20] R. Rincón, M. Jiménez, J. Muñoz, M. Sáez, M.D. Calzada, Hydrogen production
from ethanol decomposition by two microwave atmospheric pressure plasma
sources: surfatron and TIAGO torch, Plasma Chem. Plasma Process. 34 (2014)
145–157, https://doi.org/10.1007/s11090-013-9502-4.
[21] N. Bundaleska, N. Bundaleski, A. Dias, F.M. Dias, M. Abrashev, G. Filipič,
U. Cvelbar, Z. Rakočević, Z. Kissovski, J. Henriques, E. Tatarova, Microwave N2-Ar
plasmas applied for N-graphene post synthesis, Mater. Res. Express 5 (2018)
095605, , https://doi.org/10.1088/2053-1591/aad7e9.
[22] N. Bundaleska, J. Henriques, M. Abrashev, A.M. Botelho do Rego, A.M. Ferraria, A.
Almeida, F.M. Dias, E. Valcheva, B. Arnaudov, K.K. Upadhyay, M.F. Montemor, E.
Tatarova, Large-scale synthesis of free-standing N-doped graphene using microwave
plasma, Sci. Rep. 8 (2018) 12595.
[23] O. Jašek, M. Eliáš, L. Zajíčková, Z. Kučerová, J. Matějková, A. Rek, J. Buršík,
Discussion of important factors in deposition of carbon nanotubes by atmospheric
pressure microwave plasma torch, J. Phys. Chem. Solids 68 (2007) 738–743,
https://doi.org/10.1016/j.jpcs.2007.01.039.
[24] A.G. Bannov, O. Jasek, A. Manakhov, M. Marik, D. Necas, L. Zajickova, High-performance
ammonia gas sensors based on plasma treated carbon nanostructures,
IEEE Sensors J. 17 (2017) 1964–1970, https://doi.org/10.1109/jsen.2017.
2656122.
[25] A.G. Bannov, J. Prášek, O. Jašek, L. Zajíčková, Investigation of pristine graphite
oxide as room-temperature chemiresistive ammonia gas sensing material, Sensors.
17 (2017). doi:https://doi.org/10.3390/s17020320.
[26] J. Toman, O. Jasek, M. Snirer, V. Kudrle, J. Jurmanova, On the interplay between
plasma discharge instability and formation of free-standing graphene nanosheets in
a dual-channel microwave plasma torch at atmospheric pressure, J. Phys. D. Appl.
Phys. 52 (2019) 265205, https://doi.org/10.1088/1361-6463/ab0f69.
[27] J. Voráč, P. Synek, L. Potočňáková, J. Hnilica, V. Kudrle, Batch processing of
overlapping molecular spectra as a tool for spatio-temporal diagnostics of power
modulated microwave plasma jet, Plasma Sources Sci. Technol. 26 (2017) 025010, ,
https://doi.org/10.1088/1361-6595/aa51f0.
[28] R. Rincón, J. Muñoz, M. Sáez, M.D. Calzada, Spectroscopic characterization of atmospheric
pressure argon plasmas sustained with the Torche à Injection Axiale sur
Guide d'Ondes, Spectrochim. Acta B At. Spectrosc. 81 (2013) 26–35, https://doi.
org/10.1016/j.sab.2012.12.006.
[29] W. Kulisch, L. Ackermann, B. Sobisch, On the mechanisms of bias enhanced nucleation
of diamond, Phys. Status Solidi A 154 (1996) 155–174, https://doi.org/10.
1002/pssa.2211540113.
[30] T. Jawhari, A. Roid, J. Casado, Raman spectroscopic characterization of some
commercially available carbon black materials, Carbon 33 (1995) 1561–1565,
https://doi.org/10.1016/0008-6223(95)00117-v.
[31] A. Sadezky, H. Muckenhuber, H. Grothe, R. Niessner, U. Pöschl, Raman microspectroscopy
of soot and related carbonaceous materials: spectral analysis and
structural information, Carbon. 43 (2005) 1731–1742, https://doi.org/10.1016/j.
carbon.2005.02.018.
[32] A.C. Ferrari, J.C. Meyer, V. Scardaci, C. Casiraghi, M. Lazzeri, F. Mauri, S. Piscanec,
D. Jiang, K.S. Novoselov, S. Roth, A.K. Geim, Raman spectrum of graphene and
graphene layers, Phys. Rev. Lett. 97 (2006) 187401.
[33] M. Rybachuk, J.M. Bell, Electronic states of trans-polyacetylene, poly(p-phenylene
vinylene) and sp-hybridised carbon species in amorphous hydrogenated carbon
probed by resonant Raman scattering, Carbon 47 (2009) 2481–2490, https://doi.
org/10.1016/j.carbon.2009.04.049.
[34] A. Barinov, O. Bariş Malcioǧlu, S. Fabris, T. Sun, L. Gregoratti, M. Dalmiglio,
M. Kiskinova, Initial stages of oxidation on graphitic surfaces: photoemission study
and density functional theory calculations, J. Phys. Chem. C 113 (2009)
9009–9013, https://doi.org/10.1021/jp902051d.
[35] K. Ganesan, S. Ghosh, N.G. Krishna, S. Ilango, M. Kamruddin, A.K. Tyagi, A comparative
study on defect estimation using XPS and Raman spectroscopy in few layer
nanographitic structures, Phys. Chem. Chem. Phys. 18 (2016) 22160–22167,
https://doi.org/10.1039/c6cp02033j.
[36] T.B. Limbu, J.C. Hernández, F. Mendoza, R.K. Katiyar, J.J. Razink, V.I. Makarov,
B.R. Weiner, G. Morell, A novel approach to the layer-number-controlled and grainsize-controlled
growth of high quality graphene for nanoelectronics, ACS Appl.
Nano Mater. 1 (2018) 1502–1512, https://doi.org/10.1021/acsanm.7b00410.
[37] H.S. Wong, C. Durkan, Unraveling the rotational disorder of graphene layers in
graphite, Physical Review B. 81 (2010). doi:https://doi.org/10.1103/physrevb.81.
045403.
[38] G. Yang, L. Li, W.B. Lee, M.C. Ng, Structure of graphene and its disorders: a review,
Sci. Technol. Adv. Mater. 19 (2018) 613–648.
[39] C.Y. Cheng, K. Teii, Control of the growth regimes of nanodiamond and nanographite
in microwave plasmas, IEEE Trans. Plasma Sci. 40 (2012) 1783–1788,
https://doi.org/10.1109/tps.2012.2198487.
O. Jašek, et al. Diamond & Related Materials 105 (2020) 107798
9