Effects of ethylene/nitrogen mixtures on thermal chemical vapor deposition rates and microstructures of carbon films

Article abstract of DOI:10.1149/2.083206jes

When ethylenenitrogen (C₂H₄N₂) mixtures are used to deposit carbon films by thermal chemical vapor deposition (CVD), effects of C₂H₄(C₂H₄+N₂) ratios on the deposition rate and microstructures of carbon films are investigated. Experimental results reveal that the deposition rate of carbon films increases with the C₂H₄(C₂H ₄+N₂) ratio, and also, raises with the residence time, deposition temperature, and working pressure. The kinetics of this thermal CVD process is discussed. The deposition rate of carbon films is proportional to the C₂H₄(C₂H₄+N₂) ratio with a power of second order, which is resulted from the adsorption of remaining precursor gases C₂H₄ on the silica glass plate substrate. Few nitrogen and hydrogen atoms are incorporated into carbon films. As the partial pressure of C₂H₄ is smaller than a threshold pressure or the residence time is shorter than a threshold residence time, no film is formed. The activation energy (=448 kJmole) of carbon deposition is related to the activation energy of C₂H₄ dissociation. The degree of ordering and nano-crystallite size of carbon films decrease with increasing the C₂H₄(C₂H₄+N ₂) ratio, while the sp³ carbon atoms of carbon films increase. Finally, the results of thermal CVD carbon deposition using C ₂H₄ are compared with those using methane and acetylene.

Full text of DOI:10.1149/2.083206jes

Journal of The Electrochemical Society, 159 (6) D367-D374 (2012)
D367
0013-4651/2012/159(6)/D367/8/$28.00 © The Electrochemical Society
Effects of Ethylene/Nitrogen Mixtures on Thermal Chemical Vapor
Deposition Rates and Microstructures of Carbon Films
Liang-Hsun Lai,a Ke-Jie Huang,a Sham-Tsong Shiue,a,z Jing-Tang Chang,b and Ju-Liang Heb
aDepartment of Materials Science and Engineering, National Chung Hsing University, Taichung 402, Taiwan
bDepartment of Materials Science and Engineering, Feng Chia University, Taichung 407, Taiwan
When ethylene/nitrogen (C2H4/N2) mixtures are used to deposit carbon films by thermal chemical vapor deposition (CVD), effects of
C2H4/(C2H4 + N2) ratios on the deposition rate and microstructures of carbon films are investigated. Experimental results reveal that
the deposition rate of carbon films increases with the C2H4/(C2H4 + N2) ratio, and also, raises with the residence time, deposition
temperature, and working pressure. The kinetics of this thermal CVD process is discussed. The deposition rate of carbon films is
proportional to the C2H4/(C2H4 + N2) ratio with a power of second order, which is resulted from the adsorption of remaining
precursor gases C2H4 on the silica glass plate substrate. Few nitrogen and hydrogen atoms are incorporated into carbon films. As
the partial pressure of C2H4 is smaller than a threshold pressure or the residence time is shorter than a threshold residence time, no
film is formed. The activation energy (= 448 kJ/mole) of carbon deposition is related to the activation energy of C2H4 dissociation.
The degree of ordering and nano-crystallite size of carbon films decrease with increasing the C2H4/(C2H4 + N2) ratio, while the sp3
carbon atoms of carbon films increase. Finally, the results of thermal CVD carbon deposition using C2H4 are compared with those
using methane and acetylene.
© 2012 The Electrochemical Society. [DOI: 10.1149/2.083206jes] All rights reserved.
Manuscript submitted December 21, 2011; revised manuscript received March 8, 2012. Published April 10, 2012.
Because carbon possesses many allotropes with different bonding
types and structures, carbon films have many excellent properties in-
cluding wide bandgap, infrared transparency, high hardness, inertness
to chemical attack, and high water resistance.1,2 One kind of carbon
films called pyrolytic carbon films can be formed by decomposing
hydrocarbons in a heating reactor using thermal chemical vapor de-
position (CVD), and they were employed as hermetic optical fiber
coatings3 or graphite anodes of lithium ion secondary batteries.4,5
When carbon films are prepared by thermal CVD, their properties are
affected by many factors such as the precursor gas, deposition temper-
ature, working pressure, and mass flow rate of inlet gas.3–6 Among the
precursor gases, methane (CH4) remains a popular choice because it
is available in high purity, but its growth rate is lower.1 Alternatively,
acetylene (C2H2) is a very useful source gas for low pressure deposi-
tion, because its strong C≡C bond means it has a simple dissociation
= 12 mm, height = 1 mm) were cleaned in ultrasonic baths of ace-
tone and de-ionized water, in that order, to improve the adhesion of
carbon films onto these substrates. Then, the silica glass plates were
coated with carbon films by thermal CVD. The thermal CVD system
adopted a quartz tube as the reaction chamber, which had a length of
900 mm, an internal diameter of 25 mm, and a wall thickness of
1.5 mm. The deposition zone length of the reaction chamber was
60 mm, and the substrate was placed in the reaction chamber so that
the middle portion of the substrate’s length coincides with that of
the deposition zone. 99.9% C2H4 and 99.995% N2 were used as the
precursor gases. The mass flow rates of (C2H4 + N2) were kept at
40 sccm (standard cubic centimeter per minute, cm3/min), and five
carbon films were prepared with the C2H4/(C2H4 + N2) ratios of
60, 70, 80, 90, and 100%. The working pressure was maintained at
60 2 kPa by a mechanical pump. The temperature rose from room
temperature to deposition temperature at a rate of 15 K/min. The de-
position temperature and deposition time were set to 1033 1 K and
30 min, respectively. During the deposition process, a residual gas
analyzer (RGA, BCTECH-XT200M) was used to measure the partial
pressures of the residual gases. After the deposition process was fin-
ished, the temperature was quickly reduced to room temperature at a
rate of 250 K/min by cooling in air with a fan.
+
pattern, giving mainly C2Hn ions.7 Nevertheless, the degree of or-
dering of carbon films using C2H2 is lower, and the outlet of thermal
CVD system are covered with contaminants including asphalts.
Ethylene (C2H4) is an important product of petrochemical industry,
so it is also often chosen as the precursor gas to prepare carbon films
using different methods such as ion beam deposition,8 microwave
surface-wave plasma CVD,9,10 electron beam induced deposition,11
and dielectric barrier discharge plasma.12 Recently, we have adopted
CH4 and C2H2 as the precursor gas to study the properties of ther-
mal CVD carbon films.13,14 However, we have found no evidence of
previous works to investigate the effects of C2H4 on the properties
of carbon films using thermal CVD in detail. Hence, this study will
investigate the effect of C2H4/(C2H4 + N2) ratios on the thermal CVD
deposition rate and microstructures of carbon films on silica glass
plates. At a certain C2H4/(C2H4 + N2) ratio, the effects of the mass
flow rate of inlet gases, deposition temperature, and working pressure
on the deposition rate will be also considered. Furthermore, the ki-
netics of the thermal CVD process using C2H4/N2 mixtures will be
discussed, and the connection between the thermal CVD process and
microstructure of carbon films will be also considered. Finally, the
thermal CVD carbon deposition using C2H4 is compared with those
using CH4 and C2H2.
In thermal CVD process, the deposition rate depends on not only
the gas phase composition C2H4/(C2H4 + N2) but also the mass flow
rate (or residence time) of inlet gases, deposition temperature, and
working pressure. To understand the effect of the mass flow rate
of inlet gases, deposition temperature, and working pressure on the
deposition rate, other kinds of carbon films were prepared. In those
cases, the C2H4/(C2H4 + N2) ratio, mass flow rate of (C2H4 + N2),
deposition temperature, working pressure, and deposition time were
set to 80%, 40 sccm, 1033 K, 60 kPa, and 30 min, respectively, with
the exception of the specified parameters.
Characterization of carbon films.— The thicknesses and mor-
phologies of the carbon films were obtained by measuring the cross
sections of the silica glass plates located at the middle portion of
the length using a field emission scanning electron microscope (FE-
SEM, JEOL JSM-6700F). The operating voltage of the FESEM
was 3 kV. The microstructure of carbon films was investigated by
high-resolution X-ray diffraction meter (HRXRD, Brukers D8 Dis-
cover), Raman scattering spectrometer (RSS, Jobin Yvon Triax 550),
and X-ray photoelectron spectroscopy (XPS, ULVAC-PHI PHI 5000
VersaProbe). All the measurements of microstructures were made
on the carbon films located at the middle portion of the substrate.
Experimental
Preparation of carbon films.— The preparation of carbon films
proceeded as follows. The silica glass plate (length = 12 mm, width
zE-mail: stshiue@dragon.nchu.edu.tw
Downloaded on 2015-04-09 to IP 152.14.136.77 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D368
Journal of The Electrochemical Society, 159 (6) D367-D374 (2012)
Figure 1. FESEM images of cross sections of carbon films deposited at dif-
ferent C2H4/(C2H4 + N2) ratios.
HRXRD was performed using Cu Kα radiation (λ = 0.154 nm) with
an incident angle of 0.15 degrees. Diffraction peaks from the carbon
films were discerned by 2θ angles ranging between 10 and 50 degrees.
Data from the Joint Committee on Powder Diffraction File (JCPDF)
database (Number: 75-1621) were used to identify the microstructure
of the carbon films from the diffraction peaks. RSS was measured in
back-scattering geometry with the 633 nm line of a He-Ne laser at
room temperature in the spectral range of 1000–2000 cm−1. Alterna-
tively, XPS used Mg Kα radiation (Photo energy = 1253.6 eV) as an
excitation source was utilized to measure the binding energy spectra
of carbon films. All carbon core lines (C1s) spectra were acquired
when the X-ray incident angle was 54o to enhance the contribution of
carbon films on this core line shape.
Figure 2. (a) The deposition rate of carbon films as a function of the
C2H4/(C2H4 + N2) ratio, C2H4/(C2H4 + Ar) ratio, and residence time stayed
at the deposition zone. (b) The dependence of the deposition rate of carbon
films on the deposition temperature and working pressure.
C2H4/(C2H4 + N2) ratio can be expressed by a power function:
r f = k1[C2 H4/(C2 H4 + N2) − 0.063]n,
[1]
where constants k1 and n are obtained as 16.7 nm/min and 2.07,
respectively.
Deposition Rate
Effects of total mass flow rates.— To understand the effect of the
mass flow rate m on the deposition rate, seven carbon films are pre-
pared with the mass flow rates of (C2H4 + N2) being 20, 40, 60, 80,
100, 120, and 160 sccm. In this case, the other unspecified process
parameters are kept the same as those described in the experimental
section. Table I shows that the deposition rates are 48.2, 8.6, 2.8, 2.1,
1.1, 1.0, and 0.3 nm/min for the mass flow rates of (C2H4 + N2) being
20, 40, 60, 80, 100, 120, and 160 sccm, respectively. This indicates
that the deposition rates decrease with increasing the mass flow rates
of (C2H4 + N2). The mass flow rate of (C2H4 + N2) is related to
the residence time, τ, of the precursor gas stayed in the deposition
zone. Although the precursor gas is decomposed in the deposition
zone, the relation between the mass flow rate and residence time can
be approximately estimated as follows. Based on the conservation of
volume, the average flow velocity, vav, of the precursor gas in the
Effects of C2H4/(C2H4 + N2) ratios.— Figure 1 shows FESEM
images of cross sections of carbon films that are deposited on silica
glass plates with different C2H4/(C2H4 + N2) ratios. The thicknesses
tf of carbon films are obtained by measuring cross sections of the silica
glass plates located at the middle portion of the length. Experimental
results show that the thicknesses of carbon films are 133, 191, 259,
383, and 417 nm for the C2H4/(C2H4 + N2) ratio being 60, 70, 80,
90, and 100%, respectively. Notably, the carbon film was uniformly
deposited on the silica glass plate, and the error in the film thickness
was within 5%. After the thicknesses of the carbon films are obtained
using an FESEM, the deposition rate rf of the carbon films can be
calculated from the film thickness and deposition time. Figure 2a
shows that the deposition rate rf of carbon films on silica glass plates
are 4.4, 6.4, 8.6, 12.8, and 13.9 nm/min for the C2H4/(C2H4 + N2)
ratio being 60, 70, 80, 90, and 100%, respectively. This indicates
that the deposition rate of carbon films increases with increasing the
C2H4/(C2H4 + N2) ratio. Our experimental result also shows that as
the C2H4/(C2H4 + N2) is not larger than 6.3%, no films can be formed.
Hence, the relation between the deposition rate of carbon films and
•
•
deposition zone can be obtained as vav = V/(πd2/4 − t × w). Here
,
V
d (= 25 mm), t (= 1 mm), and w (= 12 mm) are the volume flow rate,
the internal diameter of the reaction chamber, the thickness of the silica
glass plate, and the width of t•he silica glass plate, respectively. More-
over, the volume flow rate
can be approximately obtained from
V
Downloaded on 2015-04-09 to IP 152.14.136.77 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 159 (6) D367-D374 (2012)
D369
Table I. The film thickness tf, deposition rate rf, average flow
velocity vav, and residence time τ at various mass flow rates of
(C2H4 + N2).
Mass flow rate of
(C2H4 + N2) (sccm)
20
40
60
80
100 120 160
tf (nm)
1446 258
48.2 8.6
84
2.8
63
2.1
33
1.1
30
1.0
10
0.3
rf (nm/min)
vav (mm/s)
τ (s)
4.43 8.85 13.3 17.7 22.1 26.6 35.4
6.8
3.4
2.3
1.7
1.4
1.1 0.85
•
•
n
the ideal gas equation as:
=
RT/p. In this study, the working
V
pressure p is 60 kPa; the gas constant R is 8.31 J/K · mole; the depo-
•
n
sition temperature T is 1033 K, and (mole/min) = m (sccm) × 4.46
× 10−5 (mole/cm3). Table I indicates that the average flow veloci-
ties, vav, are 4.43, 8.85, 13.3, 17.7, 22.1, 26.6, and 35.4 mm/s for
mass flow rates of (C2H4 + N2) being 20, 40, 60, 80, 100, 120, and
160 sccm, respectively. On the other hand, the length L of the deposi-
tion zone is 60 mm and the thickness of the carbon film was measured
at the place located at the middle of the deposition zone, and thus, the
residence time, τ, of the precursor gas stayed in the deposition zone
can be obtained as: τ = L /(2vav). Table I also show that the residence
times τ are 6.8, 3.4, 2.3, 1.7, 1.4, 1.1, and 0.85 s for the mass flow
rates of (C2H4 + N2) being 20, 40, 60, 80, 100, 120, and 160 sccm,
respectively. Figure 2a plots the relation between the deposition rate
and residence time, indicating that the deposition rate exponentially
increases with increasing the residence time. Notably, our experimen-
tal result also shows that if the mass flow rate of (C2H4 + N2) is larger
than 200 sccm (that is, the residence time is smaller than 0.68 s), no
film can be formed. Hence, the dependence of the deposition rate on
the residence time can be represented as
Figure 3. XRD patterns of carbon films deposited at different C2H4/(C2H4
+ N2) ratios. This figure reveals that XRD patterns include a (002) diffraction
peak of graphite and a (220) diffraction peak of silica glass.
working pressure changes from 12 to 67 kPa, the residence time varies
from 0 to 3.8 s. Therefore, the dependence of the deposition rate rf on
the working pressure p can be represented as
r f = k3[exp(qτ) − exp(qτc)][(p − pc)/(p0 − pc)]a,
[4]
where p0 (= 60 kPa) is the reference working pressure; pc (= 12 kPa)
is the threshold working pressure, and k3 and a are constants to be
determined. As stated in section 3.2, the residence time τ in Eq. 4 is
related to the working pressure p, and q (= 0.44/s) and τc (= 0.68 s)
are constants. When the data of the deposition rates at various working
pressures are substituted into Eq. 4, the constants k3 and a are found
to be 2.66 nm/min and 2.19, respectively, with an error well below
10%. The amount of C2H4 in the reaction gas not only is related to
the C2H4/(C2H4 + N2) ratio but also is proportional to the working
pressure, and thus, the constant a is close to n in Eq. 1. Eqs. 1 and 4
show that the deposition rate is proportional to the partial pressure (or
pressure) of C2H4 with a power of second order, and thus, the thermal
CVD process of this work is a second-order reaction.
r f = k2[exp(qτ) − exp(qτc)],
where τc (= 0.68 s) is the threshold residence time, and constants k2
and q are obtained as 2.69 nm/min and 0.44/s, respectively, with an
error within 10%.
[2]
Effects of deposition temperatures.— To understand the effect of
the deposition temperature T on the deposition rate, five carbon films
were prepared with the deposition temperatures of 1013, 1023, 1033,
1043, and 1053 K. In this case, the other unspecified process param-
eters are kept the same as those described in the experimental details.
Figure 2b shows that if the deposition temperatures are 1013, 1023,
1033, 1043, and 1053 K, the deposition rates are 2.2, 4.2, 8.6, 11.6, and
21.0 nm/min, respectively. This indicates that the dependence of the
deposition rate on the deposition temperature follows the Arrhenius
law. Hence, the constant k1 in Eq. 1 can be replaced by
Structural Analyzes
Figure 3 shows HRXRD patterns of carbon films that are prepared
with different C2H4/(C2H4 + N2) ratios. This figure displays a (220)
diffraction peak around 20.8 degree for the silica glass substrate and
a (002) diffraction peak around 25.5 degree for the graphite, and thus,
the (220) peak of the silica glass substrate should be extracted from the
HRXRD pattern to obtain the (002) peak of the graphite. Moreover,
to eliminate the intrinsic instrumental broadening, the B parameter
k1 = k0 exp(−E/RT ),
[3]
where E (= 448 kJ/mole) is the activation energy with a derivation
within 10%; R (= 8.31 J/K · mole) is the gas constant, and k0 = 6.13
× 1023 nm/min. Notably, as the temperature changes from 1013 to
1053 K, the residence time varies from 3.47 to 3.35 s. Because the
variation of the residence time is small, we assume that the residence
time is not changed here.
2
2
was calculated using the equation, B = (Bc −BSi
)
1/2, where Bc is the
full-width-at-half-maximum (FWHM) of the (002) peak of the car-
bon films, and BSi is the experimentally obtained FWHM of the (220)
peak of a standard silicon sample.15 Because the thickness of the car-
bon film with the C2H4/(C2H4 + N2) ratio of 60% is relatively small
(= 133 nm), the (220) diffraction peak of the silica glass substrate
is larger than the (002) diffraction peak of graphite. Hence, the fit-
ting result of (002) diffraction peak of graphite with the C2H4/(C2H4
+ N2) ratio of 60% has less accuracy. Experimental results reveal
that the diffraction angle of the (002) peak of the graphite decreases
from 25.4 to 25.2 degrees as the C2H4/(C2H4 + N2) ratio increases
from 70 to 100%, while B parameter of carbon films increases from
4.5 to 5.7 degrees. These results indicate that the degree of order-
ing of carbon films decreases with increasing the C2H4/(C2H4 + N2)
ratio.5 The mean crystallite size, Lc, of carbon films can be estimated
Effects of working pressures.— To further understand the effect of
the working pressure on the deposition rate, seven carbon films were
prepared with the working pressures of 23, 33, 40, 47, 53, 60, and
67 kPa, where the derivation of working pressures is within 2 kPa.
In this case, the other unspecified process parameters are kept the
same as those described in the experimental details. Figure 2b reveals
that if the working pressures are 23, 33, 40, 47, 53, 60, and 67 kPa,
the deposition rates are 0.3, 0.6, 1.1, 1.9, 4.5, 8.6, and 14.2 nm/min,
respectively. Our experimental result also shows that if the working
pressure p is not over 12 kPa, no film is formed. Notably, when the
Downloaded on 2015-04-09 to IP 152.14.136.77 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D370
Journal of The Electrochemical Society, 159 (6) D367-D374 (2012)
composed into five absorption peaks: D band at about 1350 cm−1
,
G band at about 1580 cm−1, Dꢀ band at about 1620 cm−1, D1 band
at about 1200 cm−1, and D2 band at about 1500 cm−1, where the D,
G, Dꢀ, and D1 bands were fitted with Lorentzian function, and the
D2 band was fitted with Gaussian function.17 The D band is associ-
ated with the breathing mode of sixfold sp2 rings and identified as
the disorder-activated band, this mode is forbidden in perfect graphite
and only becomes active in the presence of disorder;18 the G band is
associated with the bond stretching vibrations of all sp2 sites;18 the Dꢀ
band is related to disordered graphitic lattice;17 the D1 band is also re-
lated to the disorder graphitic lattice17 but not uniquely defined,19 and
the D2 band is attributed to amorphous carbon.17 Figure 4b shows the
peak shift of the D band (ωD) and G band (ωG), the full-width-at-half-
maximum of D band (FWHMD) and G band (FWHMG), and integrated
intensity ratio of the D band to G band (ID/IG) for carbon films pre-
pared with different C2H4/(C2H4 + N2) ratios. This figure shows that
as the C2H4/(C2H4 + N2) ratio increases, ωD and ωG have no apparent
trend. Alternatively, this figure also reveals that as the C2H4/(C2H4
+ N2) ratio increases from 60 to 100%, FWHMD increases from 154 to
162 cm−1 and ID/IG values increase from 2.90 to 3.28, while FWHMG
values decrease from 54 to 50 cm−1. This indicates that the degree of
ordering of carbon films decreases with increasing the C2H4/(C2H4
+ N2) ratio.18 The ratio of ID and IG is commonly used to approx-
imately estimate the crystallite size of nanographite. An expression
that gives the crystallite size (La) from the integrated intensity ratio
ID/IG is given by:15 La(nm) = (2.4 × 10−10)λo4(ID/IG)−1, where λo
(= 633 nm) is the laser line wavelength used in the Raman experiment.
This expression indicates that the crystallite size of nanographite in-
creases with decreasing the ID/IG value. Hence, Fig. 4b shows that
ID/IG increases from 2.90 to 3.28 as the C2H4/(C2H4 + N2) ratio in-
creases from 60 to 100%, while La decreases from 13.3 to 11.7 nm.
The decrease of the crystallite size (La) of carbon films also indicates
that the degree of ordering of carbon films decreases. This result is
consistent with that of XRD.
On the other hand, XPS is adopted to investigate the chemical
bonding of carbon films. Figure 5a shows XPS patterns of carbon
films that are prepared with different C2H4/(C2H4 + N2) ratios. This
figure indicates that the C1s core around 285 eV and O1s core around
531 eV are revealed, while the N1s core around 400 eV is not found.
This implies that little nitrogen is incorporated into the carbon films.
The O1s core shows that oxygen existed in the carbon films. Remark-
ably, the presence of oxygen is probably related to the incorporation of
oxygen due to prolonged exposure of the sample to the external atmo-
sphere before XPS measurements and accidental incorporation during
deposition.20 Figure 5b shows C1s patterns of carbon films that are
prepared with different C2H4/(C2H4 + N2) ratios. The C1s spectra of
carbon films can be decomposed into five components: 284.2, 285.0,
286.5, 287.9, and 289.6 eV attributed to sp2 carbon atoms, sp3 carbon
atoms, C-O bonds, C=O bonds, and COOR, respectively.21–23 The
C–O bonds, C=O bonds, and COOR are assigned to atomic carbon
bonded to surface adsorbents from the ambient, hereafter referred to
as satellite peaks.23 Thus, the sum of the satellite peak areas indicates
the fraction of carbon bonding with surface adsorbents. Table II shows
the peak position and peak area fraction for all kinds of carbon atoms.
This result indicates that the peak positions of sp2 and sp3 carbon
atoms are slightly changed. Alternatively, if the C2H4/(C2H4 + N2)
ratio changes from 60 to 100%, the sp2 peak area fraction decreases
from 35.9 to 19.1%, but the sp3 peak area fraction increases from 35.0
to 51.4%. This implies that if the C2H4/(C2H4 + N2) ratio increases,
the structure of the carbon films shifts to diamond-like. Table II reveals
that the area fraction of satellite peaks is about 30%, which is changed
at different C2H4/(C2H4 + N2) ratios. This is because the surface of
carbon films is exposed to the atmosphere, and the amount of surface
adsorbents from the ambient for various carbon films is different.
Figure 4. (a) Raman spectra of carbon films deposited at different
C2H4/(C2H4 + N2) or C2H4/(C2H4 + Ar) ratios. The Raman spectra con-
tain the D band at about 1350 cm−1, G band at about 1580 cm−1, Dꢀ band
at about 1620 cm−1, D1 band at about 1200 cm−1, and D2 band at about
1500 cm−1. (b) Fitting results of Raman spectra for carbon films deposited at
different C2H4/(C2H4 + N2) or C2H4/(C2H4 + Ar) ratios.
using Scherrer’s formula, Lc = 0.9λ/(BcosθB). Here λ (= 0.154 nm)
is the wavelength of the copper Kα X-ray line; B is the FWHM of an
HRXRD peak, and θB is the diffraction angle of the HRXRD peak.16
The calculated result shows that the mean crystallite size (Lc) of car-
bon films decreases from 2.0 to 1.5 nm as the C2H4/(C2H4 + N2) ratio
increases from 70 to 100%. The decrease of the mean crystallite size
(Lc) of carbon films indicates that the degree of ordering of carbon
films decreases. This also implies that the increase of the C2H4/(C2H4
+ N2) ratio raises the deposition rate of carbon films, but decreases
the degree of ordering of carbon films.
Discussion
Figure 4a shows Raman spectra (RS) of carbon films that are
prepared with different C2H4/(C2H4 + N2) ratios. The RS can be de-
The kinetics of this thermal CVD process is discussed, and the
details are described in the Appendix. In brief, the precursor gases
Downloaded on 2015-04-09 to IP 152.14.136.77 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 159 (6) D367-D374 (2012)
D371
resulted from the adsorption of the remaining precursor gases C2H4
on the silica glass plate substrate. The carbon species adsorbed on
the substrate surface will further diffuse and conjugate to form the
laminar graphite structure.
The dissociation energy of N2 is very high (= 942 kJ/mole),24
and XPS results of this study reveal that little nitrogen is incor-
porated into the carbon films. Hence, most of N2 is exhausted to
the atmosphere. In other words, the N2 gas is only used to reduce
the partial pressure of C2H4. In the literature, N2 was often added
in the precursor gas to form various kinds of nitrides.25–29 Neverthe-
less, this study reveals that adding N2 in C2H4 is not able to produce ni-
trides in thermal CVD process. To verify the above statement, an addi-
tional experiment was executed as follows. The deposition parameter
of this additional experiment is the same as that of various C2H4/(C2H4
+ N2) ratios with the exception that N2 is replaced by Ar (99.999%
in purity). Notably, Ar is an inert gas, so it is only used to reduce the
partial pressure of C2H4. The deposition rate and RS results of carbon
films deposited at various C2H4/(C2H4 + Ar) ratios are also shown in
Figs. 2a and 4b, respectively. Figure 2a reveals that the deposition rate
using C2H4/Ar mixtures is quite close to that with C2H4/N2 mixtures
at the same deposition condition. Alternatively, Fig. 4b also shows that
the RS result of carbon films with C2H4/Ar mixtures is very similar to
that with C2H4/N2 mixtures. Hence, this is evident that the N2 gas is
only used to reduce the partial pressure of C2H4, as the Ar gas does.
The deposition of carbon films on the substrate needs adequate
amount of carbon species in the gas phase (or sufficient residence
time for carbon species to stay in the deposition zone). Hence, if
the partial pressure (or working pressure) of C2H4 is smaller than the
threshold pressure (or if the residence time is shorter than the thresh-
old residence time), no film can be formed. Moreover, as the residence
time of the precursor gas stayed in the deposition zone increases, the
amount of decomposed radicals also raises. As RGA results show that
C2H4 is decomposed to form C1 and C2 radicals, and thus, the total
amount of C1 and C2 species is increasing with the residence time.
Consequently, the carbon fragment converted from these decomposed
precursor gases to form carbon films on the substrate also exponen-
tially increases.30 The dissociation energy of C2H4 was reported as
411 kJ/mole,31,32 and thus, the activation energy of 448 kJ/mole in
this study is mainly correlated to the dissociation energy of C2H4
dissociation.
Figure 1 also illustrates that carbon films exhibit a laminar struc-
ture, and there are many particles appeared on the carbon film surface.
The pyrolytic carbon films prepared by thermal CVD usually ex-
hibits a laminar structure.33 The growth of these particles had been
studied previously, which proceeded as follows.34,35 Initially, a small
(nanoscale) particle is nucleated on the carbon film. The laminar car-
bon film is then deposited around the small particle, extending the
shape of the particle to that of a headstand cone with a spherical base.
The spherical base of the cone can be evident from the surface of the
carbon film. The surface area of this spherical base increases with the
carbon film thickness. Additionally, the number of nucleated particles
increases with the deposition rate, so the size and number of parti-
cles increase as the thickness of the carbon film increases.34 Notably,
the degree of ordering of carbon films decreases with increasing the
size and number of particles, and thus, XRD and RS results of this
study show that the degree of ordering of carbon films decreases with
increasing the deposition rate (and also the C2H4/(C2H4 + N2) ratio).
Sobol-Antosiak and Ptak36 reported that the degree of surface
hydrogen coverage tends to increase in nucleation processes, leading
to the formation of carbon films, from gas phase by the CVD methods.
The presence of hydrogen is considered to enhance the process of
forming films with high sp3 contribution. In this work, the RGA result
shows that the gas phase hydrogen increases as the C2H4/(C2H4 +
N2) increases. Consequently, the effect of gas phase hydrogen on
the formation of carbon films may be taken to explain the reason
why the sp3 fraction in the carbon films increases with increasing the
C2H4/(C2H4 + N2) ratio.
Figure 5. (a) XPS of carbon films deposited at different C2H4/(C2H4
+ N2) ratios. (b) Decomposed curves of C1s core for carbon films deposited
at different C2H4/(C2H4 + N2) ratios.
(C2H4 and N2) are flowed through the reactor and heated up, and thus,
some of precursor gases are decomposed. The RGA result (mentioned
in the Appendix) reveals that the residual gases in the gas phase
mainly contain the remaining precursor gases C2H4 (and/or N2) and
the product gases C2H2, C2H3, and H2. The deposition rate of carbon
films is proportional to the partial pressure of the inlet gas C2H4 with
a power of second order, and the pyrolysis of C2H4/N2 mixtures is
Table II. The peak positions and peak area fractions of carbon
atoms analyzed by XPS at different C2H4/(C2H4 + N2) ratios.
C2H4/(C2H4 + N2) ratio (%)
60
70
80
90
100
sp2 peak position (eV)
sp3 peak position (eV)
sp2 fraction (%)
284.5 284.6 284.4 284.5 284.4
285.1 285.2 285.2 285.1 285.0
35.9
35.0
29.1
34.1
38.7
27.2
26.4
40.5
33.1
24.1
45.0
30.9
19.1
51.4
29.3
sp3 fraction (%)
In our previous works, we adopted CH413 and C2H214 as the precur-
sor gases to study the effect of deposition parameters on the properties
Satellite fraction (%)
Downloaded on 2015-04-09 to IP 152.14.136.77 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D372
Journal of The Electrochemical Society, 159 (6) D367-D374 (2012)
Table III. A comparison between the thermal CVD carbon deposition of this work and that of our previous works13,14 at some specified process
parameters.
This work
Reference 13
Reference 13
Reference 14
Precursor gas
C2H4 (32 sccm)
N2 (8 sccm)
Silica glass plate
1033
CH4 (32 sccm)
N2 (8 sccm)
Silica glass fiber
1248
CH4 (32 sccm)
N2 (8 sccm)
Silica glass plate
C2H2 (32 sccm)
N2 (8 sccm)
Silica glass fiber
Substrate
Deposition temperature (K)
Working pressure (kPa)
Deposition rate (nm/min)
C/H ratio
1248
77
39.2
1/4
1003
13.3
5.1
60
8.6
1/2
77
56.7
1/4
1/1
Activation energy (kJ/mole)
Power order of controlled process
Outlet Contaminants
448
456
Null
1.17
Small
484
4.83
Large
2.07 ∼ 2.19
1.15 ∼ 1.23
Moderate
Small
of thermal CVD carbon films. A comparison between the results of
this work and those of the previous works13,14 at a specified process
parameter is listed in Table III. Notably, the specified deposition pa-
rameters listed in Table III for various precursor gases are obtained
by the authors to form a carbon film with smooth surface and good
adhesion on the substrate surface, so they are different. The working
pressure in previous work14 was 13.3 kPa, but it was misprinted as
133 kPa. Table III shows that the deposition temperature of C2H4 is
smaller than that of CH4, but it is slightly larger than that of C2H2.
Alternatively, the working pressure of C2H4 is slightly smaller than
that of CH4, but it is larger than that of C2H2. C2H4 is a sp2 hybridized
hydrocarbon, while CH4 and C2H2 are sp3 and sp1 hybridized hy-
drocarbons, respectively. Hence, the hydrogen/carbon (H/C) ratio of
C2H4 is 1/2, while the H/C ratios of CH4 and C2H2 are 1/4 and 1,
respectively. The activation energy of C2H4 (= 448 kJ/mole) is lower
that those of C2H2 (= 484 kJ/mole) and CH4 (= 456 kJ/mole). The
pyrolysis of C2H4 with added N2 is controlled by the second-order
process to create C2 species; the pyrolysis of CH4 with added N2 is
mainly controlled by the first-order process to create C1 species, and
the pyrolysis of C2H2 with added N2 is controlled by the fifth-order
process to create C10 species. As the process parameters of thermal
CVD are kept the same, the deposition rate of CH4 is much smaller
than those of C2H4 and C2H2. Alternatively, if the deposition param-
eters are selected as: deposition temperature T = 1003 K, working
pressure p = 13.3 kPa, and mass flow rate of inlet gas m = 40 sccm,
our experimental results show that the deposition rates of carbon films
on silica glass plate with C2H4/(C2H4 + N2) = 80% and C2H2/(C2H2
+ N2) = 80% are 0 and 2.6 nm/min, respectively. This means that
the deposition rate of thermal CVD carbon deposition using C2H4 is
smaller than that using C2H2 at the same deposition condition. More-
over, it was found that the amount of contaminants in the outlet of
thermal CVD system using CH4 is small; the amount of contaminants
in the outlet of thermal CVD system using C2H4 is moderate, and the
amount of contaminants in the outlet of thermal CVD system using
C2H2 is large.
or the residence time is shorter than the threshold residence time, no
carbon film is formed. The activation energy (= 448 kJ/mole) of car-
bon deposition is shown to correlate to the activation energy of C2H4
dissociation. XRD and RS results show that the degree of ordering
and nano-crystallite size of carbon films decrease with increasing the
C2H4/(C2H4 + N2) ratio, this is because the size and number of parti-
cles in the carbon films increase. Alternatively, when the C2H4/(C2H4
+ N2) ratio increases, the hydrogen in the gas phase also increases,
and thus, XPS results reveal that sp3 carbon atoms in carbon films
raise. The results of this work are compared to those using CH4/N2
and C2H2/N2 mixtures at a specified process parameter.
Acknowledgment
This work was supported by the National Science Council,
Taiwan, under grant No. NSC 99-2221-E-005-054-MY2, and was
also supported in part by the Ministry of Education, Taiwan, under
the ATU plan.
Appendix
The kinetics of this thermal CVD process is described here. In this work, substrates
are heated in a deposition reactor and well-mixed precursor gases C2H4 and/or N2 are
flowed through the reactor and heated up. After heated up in the deposition reactor, the
precursor gas undergoes reactions in the gas phase. Simultaneously, complex hetero-
geneous reactions occur at the surface of the silica glass substrate. Deposition of the
solid-phase carbon materials thus is a consequence of complex homogeneous gas-phase
and heterogeneous surface reactions.31 The partial pressure of the residual gases in the
gas phase was analyzed by residual gas analyzer (RGA). The RGA result reveals that the
residual gases contain the remaining precursor gases C2H4 and/or N2; the main product
gases C2H2, C2H3, and H2; and also, some other secondary product gases CH, CH2, CH3,
CH4, C2, C2H, and C2H5. Notably, the partial pressures of the secondary product gases
are much small than those of the excess precursor gases and main product gases, so only
the latter gases are considered here. Figure A1 shows the partial pressures, piout , of the
remaining precursor gases and main product gases as a function of the partial pressure,
pCin , of the inlet gas C2H4. Notably, pCin
is the multiplication of the C2H4/(C2H4
H
4
+
H
2
4
N2) ratio and 60 kPa. Figure A1 reveals t2hat the partial pressure of the outlet gas C2H4
(pCout ) is proportional to (pCin 4 )2. Alternatively, the RGA result also shows that the
H
H
2
2
4
partial pressure of the outlet gas C2H2 (pCout ) almost equals to that of C2H3 (pCout 3 ),
which is proportional to (pCin
also proportional to (pCin
H
H
2
2
2
)
0.5. The partial pressure of the outlet gas H2 (poHu2t ) is
H
4
Conclusions
2
)
0.5. When the conversion of C2H4 into the product gases is
H
4
When C2H4/N2 mixtures are used to form carbon films by ther-
mal CVD, effects of deposition parameters on the deposition rate and
microstructures of carbon films are investigated. Experimental results
show that as the C2H4/(C2H4 + N2) ratio increases from 60 to 100%,
the deposition rate increases from 4.4 to 13.9 nm/min. Alternatively,
if the residence time, deposition temperature, and working pressure
raise, the deposition rate of carbon films also increases. The depo-
sition rate of carbon films is proportional to the partial pressure of
the inlet gas C2H4 with a power of second order, and the pyrolysis
of C2H4/N2 mixtures is resulted from the adsorption of the remaining
precursor gases C2H4 on the silica glass plate substrate. Few N and
H atoms are incorporated into the carbon films. If the partial pressure
(or working pressure) of C2H4 is smaller than the threshold pressure
2
a nth-order reaction, the rate equation can be expressed as: d(pCin )/dt = −kn(pCin 4 )n,
H
H
2
2
4
and thus, one obtains
1−n
1−n
(pCout
)
− (pCin
)
H
4
= kn(n − 1)tr ,
[A1]
H
2
4
2
where kn is a constant and tr is the residence time of the precursor gases stayed in the
deposition zone. By substituting the data of pCout and pCin 4 into Eq. A1, one obtains
H
H
2
4
that n = 0.5. This means that the conversion of C22H4 to C2H2 (also C2H3) is a half-order
reaction. Therefore, the partial pressures of the main product gases, pCout , pCout 3 , and
H
H
2
2
2
poHut , are proportional to (pCin
) .
0.5
H
2 Norinaga and Deutschm2an4n37 (also with their coworkers38) had investigated the de-
tailed kinetic modeling of gas-phase reactions in the chemical vapor deposition of carbon
from light hydrocarbons. Nevertheless, the RGA result of this study shows that the resid-
ual gases mainly include C2H2, C2H3, and C2H4 species, so only these species will
Downloaded on 2015-04-09 to IP 152.14.136.77 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 159 (6) D367-D374 (2012)
D373
rd1 = kd1C (H) − k−d1C ( )pout
,
[A3d]
∞
∞
H
2
rd2 = kd2C (H) − k−d2C ( )pout
,
.
[A3e]
[A3f]
∞
∞
H
2
rd3 = kd3C (H) − k−d3C ( )pout
∞
∞
H
2
The total concentration of active site, C, is the summation of that of free active site and
active site with adsorbed hydrogen, that is
C = C ( ) + C (H).
[A4]
∞
∞
At steady-state conditions, the total adsorption rate is equal to the total desorption rate,
and thus, we obtain
ra1 + ra2 + ra3 = rd1 + rd2 + rd3 = rs ,
[A5]
where rs = pyrocarbon deposition rate, which is proportional to the deposition rate of the
carbon film, rf. Substituting Eqs. A3a–A3f and A4 into Eq. A5, one obtains
rs = r f /k4
out
2
out
2
out
2
C(k
p
+k
p
+k
p
)
4
a1
a2
a3
C
H
C
H
C
H
2
3
=
,
1+[(k−d1+k−d2+k−d3)pout +(k
p
+k
p
+k
p
C H
2
4 )]/(kd1+kd2+kd3
)
out
out
out
a1
a2
a3
H
C
H
C
H
2
2
2
2
3
[A6]
Figure A1. The partial pressures of the residual gases as a function of the
partial pressure of the inlet gas C2H4.
where k4 is a constant. We had adopted Fourier transfer infrared (FTIR) spectroscope
to measure the carbon films, and found that little hydrogen was within the carbon film.
Hence, we believe that the hydrogen radicals H• in the carbon film is desorbed. The
desorbed hydrogen radicals H• may be conjugated together to produce hydrogen gases
H2, and then is exhausted to the atmosphere. Hence, we can predict that the rate constants
of desorption, kd1, kd2, and kd3, are much larger than that of adsorption. Accordingly, the
second term on the denominator of Eq. A6 can be neglected. Therefore, Eq. A6 is reduced
to
be supposed to be the main species able to adsorb at the deposition surface. Based on
Langmuir-Hinshelwood kinetics,39,40 we assume that the chemisorption of hydrocarbon
species from the gas phase is irreversible, and the surface reactions by ring formation
from carbon-hydrocarbon surfaces complexes are fast in comparison to hydrogen desorp-
tion; thus, only hydrogen desorption by back-formation of free sites is considered. The
adsorption equilibria of chemisorbed hydrocarbon species and surface reactions of C2H2,
C2H3, and C2H4 species can be expressed as
out
2
out
2
out
2
r f = k4C(k
p
+ k
p
+ k
p
a3
4 ).
H
[A7]
a1
a2
C
H
C
H
C
2
3
The RGA results (as shown in Fig. A1) reveal that pCout or pCout is proportional to
H
H
3
k
a
2
2
2
1
C
C
C
( ) + C2 H2 −−−−−−−→ C (C2 H2),
[A2a]
[A2b]
[A2c]
[A2d]
[A2e]
[A2f]
[A2g]
[A2h]
[A2i]
[A2j]
[A2k]
[A2l]
∞
∞
(pCin
)
0.5, but pCout is proportional to (pCin 4 )2. Hence, based on Eqs. 1 and 4, we
H H H
4 2
beli2eve that ka3 is mu4ch larger than ka1 and ka22, and thus, Eq. A7 can be further reduced
k
ꢁk
a1
to
+1
C
(C2 H2) −−−−−−−→ C (C2 H2),
∞
∞
r f = k4k5Cka3(pCin 4 )2,
[A8]
H
2
k
a
2
where k5 is a constant. As stated in Eqs. 1 and 4, the deposition rate of carbon films is
proportional to the partial pressure of the inlet gas C2H4 with a power of second order, and
thus, the pyrolysis of C2H4 with added N2 is resulted from the adsorption of the remaining
precursor gases C2H4 on the silica glass plate substrate. Remarkably, the partial pressure
( ) + C2 H3 −−−−−−−→ C (C2 H3),
∞
∞
k
ꢁk
a2
+2
C
(C2 H3) −−−−−−−→ C (H3),
∞
∞
of C2H4, pCin , in Eq. A8 is related to the [C2H4/(C2H4 + N2)−0.063] ratio in Eq. 1
H
and the work2ing pressure (p-pc)/(p0-pc) in Eq. 4; the rate constant ka3 in Eq. A8 is related
to the residence time [exp(qτ)−exp(qτc)] in Eq. 2 and the temperature exp(-E/RT) in
Eq. 3, and the total concentration of active sites C in Eq. A8 is dependent on the specific
surface area of the substrate. In this study, the substrate is not changed, so C is a constant.
Furthermore, the carbon species adsorbed on the substrate surface will further diffuse and
conjugate to form the laminar graphite structure.
4
k
a
3
( ) + C2 H4 −−−−−−−→ C (C2 H4),
∞
∞
k
ꢁk
a3
+3
C
(C2 H4) −−−−−−−→ C (H4),
∞
∞
k
d1
C
H2 −−−−−−−→ C ( ) + H2,
∞
∞
References
k
−d1
C
( ) + H2 −−−−−−−→ C (H2),
∞
∞
1. A. Grill, Diam. Relat. Mater., 8, 428 (1999).
2. J. Robertson, Mater. Sci. Eng. R, 37, 129 (2002).
k
d2
3
3. C. A. Taylor and W. K. S. Chiu, Surf. Coat. Technol., 168, 1 (2003).
4. Y. S. Han and J. Y. Lee, Electrochim. Acta, 48, 1073 (2003).
5. Y. S. Ding, W. N. Li, S. Iaconetti, X. F. Shen, J. DiCarlo, F. S. Galasso, and S. L.
Suib, Surf. Coat. Technol., 200, 3041 (2006).
C
C
(H3) −−−−−−−→ C ( ) + H2,
∞
∞
∞
2
k
−d2
3
( ) + H2 −−−−−−−→ C (H3),
∞
6. W. N. Li, Y. S. Ding, S. L. Suib, J. F. DiCarlo, and F. S. Galasso, Surf. Coat. Technol.,
2
190, 366 (2005).
k
d3
7. R. Kleber, M. Weiler, A. Kruger, S. Sattel, G. Kunz, K. Jung, and H. Ehrhardt, Diam.
Relat. Mater., 2, 246 (1993).
8. S. Pirzada, J. J. Liu, F. Li, B. Demczyk, and D. Spaulding, J. Appl. Phys., 91, 7562
(2002).
9. M. Umeno and S. Adhikary, Diam. Relat. Mater., 14, 1973 (2005).
10. D. C. Ghimire, S. Adhikari, H. R. Aryal, S. M. Mominuzzamn, T. Soga, T. Jimbo,
H. Uchida, and M. Umeno, Diam. Relat. Mater., 17, 1724 (2008).
11. O. Guise, H. Marbach, J. Levy, J. Ahner, and J. T. YatesJr, Surf. Sci., 571, 128
(2004).
C
H4 −−−−−−−→ C ( ) + 2H2,
∞
∞
k
−d3
C
( ) + 2H2 −−−−−−−→ C (H4).
∞
∞
Where C ( ) = free active site; C (C2H2), C (C2H3), C (C2H4) = active sites with
∞
∞
∞
∞
adsorbed hydrocarbon; C (H2), C (H3) C (H4) = active sites with adsorbed hydrogen,
∞
∞
∞
and ka1, k+1, ka2, k+2, ka3, k+3, kd1, k-d1, kd2, k-d2, kd3, k-d3 = rate constants.
The rate equations of above adsorption and desorption reactions can be expressed as:
12. J. Niu, L. Zhanga, Z. Zhang, D. Liu, Y. Liu, and Z. Feng, Appl. Surf. Sci., 256, 6887
(2010).
13. S. T. Shiue, P. Y. Chen, R. H. Lee, T. S. Chen, and H. Y. Lin, Surf. Coat. Technol.,
ra1 = ka1C ( )pout
,
[A3a]
[A3b]
[A3c]
∞
C
2
H
2
205, 780 (2010).
14. R. H. Lee, L. H. Lai, and S. T. Shiue, Thin Solid Films, 518, 7267 (2010).
15. L. G. Cancado, K. Takai, T. Enoki, M. Endo, Y. A. Kim, H. Mizusaki, A. Jorio,
L. N. Coelho, R. Magalha˜es-Paniago, and M. A. Pimenta, Appl. Phys. Lett., 88,
163106 (2006).
ra2 = ka2C pout
,
∞
C
H
2
3
ra3 = ka3C ( )pout
,
∞
C
2
H
4
Downloaded on 2015-04-09 to IP 152.14.136.77 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D374
Journal of The Electrochemical Society, 159 (6) D367-D374 (2012)
16. B. D. Cullity and S. R. Stock, Elements of X-ray Diffraction, Third ed., Prentice Hall,
New Jersey (2001).
28. M. Jurzecka, S. Kluska, S. Jonas, H. Czternastek, and K. Zakrzewska, Vacuum, 82,
1128 (2008).
17. A. Sadezky, H. Muckenhuber, H. Grothe, R. Niessner, and U. Poschl, Carbon, 43,
1731 (2005).
29. M. H. Kang, D. S. Kim, A. Ebong, B. Rounsaville, A. Rohatgi, G. Okoniewska, and
J. Hong, J. Electrochem. Soc., 156, H495 (2009).
18. A. C. Ferrari and J. Robertson, Phys. Rev., B, 61, 14095 (2000).
19. A. C. Ferrari and J. Robertson, Phys. Rev., B, 63, 121405 (2001).
20. M. Rusop, S. M. Mominuzzaman, X. M. Tian, T. Soga, T. Jimbo, and M. Umeno,
Appl. Surf. Sci., 197, 542 (2002).
21. P. Me´rel, M. Tabbal, M. Chaker, S. Moisa, and J. Margot, Appl. Surf. Sci., 136, 105
(1998).
30. A. Becher and K. J. Huttinger, Carbon, 36, 213 (1998).
31. A. Becker and K. J. Huttinger, Carbon, 36, 177 (1998).
32. J. H. Kiefer and W. A. von Drasek, Int. J. Chem. Kin., 22, 747 (1990).
33. X. Bourrat, J. Lavenac, F. Langlais, and R. Naslain, Carbon, 39, 2376 (2001).
34. S. S. Chen, S. T. Shiue, K. J. Cheng, P. Y. Chen, and H. Y. Lin, Opt. Eng., 47, 045005
(2008).
22. G. L. Duˆ, N. Celini, F. Bergaya, and F. Poncin-Epaillard, Surf. Coat. Technol., 201,
5815 (2007).
23. H. S. Zhang and K. Komvopoulos, J. Appl. Phys., 106, 093504 (2009).
24. R. H. Christian, R. E. Duff, and F. L. Yarger, J. Chem. Phys., 23, 2045
(1955).
35. P. Y. Chen, S. T. Shiue, and H. Y. Lin, Thin Solid films, 518, 2883 (2010).
36. J. Sobol-Antosiak and W. S. Ptak, Mater. Lett., 56, 842 (2002).
37. K. Norinaga and O. Deutschmann, Ind. Eng. Chem. Res., 46, 3547 (2007).
38. K. Norinaga, O. Deutschmann, N. Saegusa, and J. I. Hayashi, J. Analytical Appl.
Pyrolysis, 86, 148 (2009).
25. T. Otani and M. Hirata, Thin Solid Films, 442, 44 (2003).
26. R. P. Parikh and R. A. Adomaitis, J. Cryst. Growth, 286, 259 (2006).
27. V. Woods and N. Dietz, Mater. Sci. Eng., B, 127, 239 (2006).
39. G. F. Froment and K. B. Bischoff, Chemical Reactor Analysis and Design, John
Wiley & Sons, New York (1979).
40. K. Norinaga and K. J. Huttinger, Carbon, 41, 1509 (2003).
Downloaded on 2015-04-09 to IP 152.14.136.77 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).