Synthetic Metals 219 (2016) 124134

Contents lists available at ScienceDirect

Synthetic Metals
journal homepage: www.elsevier.com/locate/synmet

Selective synthesis of carbon-nanotubes/graphite or carbon-nanotubes

/multi-graphene composites on 3-D nickel foam prepared with
different nickel catalyst and pre-treatment
Hsi-Chao Chena,b,c,* , Yin-Cheng Wanga , Kuo-Ting Huangb
a
b
c

Graduate School of Optoelectronics Engineering, National Yunlin University of Science and Technology, Yunlin 64002, Taiwan
Graduate School of Engineering Science and Technology, National Yunlin University of Science and Technology, Yunlin 64002, Taiwan
Department of Electronic Engineering, National Yunlin University of Science and Technology, Yunlin 64002, Taiwan

A R T I C L E I N F O

Article history:
Received 1 March 2016
Received in revised form 21 May 2016
Accepted 27 May 2016
Available online 17 June 2016
Keywords:
Chemical vapor deposition (CVD)
Nickel foam (NiF)
Graphene
Carbon nanotubes (CNTs)
Composite lm

A B S T R A C T

The research is to deposit carbon on porous 3-D nickel foam (NiF) as a gas diffusion electrode (GDE) by
chemical vapor deposition (CVD). Effects of three different preparations of the nickel (Ni) catalyst and
two pre-treatments of oxidation and reduction for carbon deposition were investigated. In the rst
method, the carbon was deposited on NiF with and without sputtering a Ni seed layer before the CVD. In
the second method, the NiF was oxidized, then reduced by hydrogen into Ni particle as a catalyst and the
carbon also was deposited by CVD. In the third method, a thin lm of SiO2 was deposited on the NiF as a
separation layer, then the Ni catalyst was deposited and nally, carbon was also deposited by CVD. The
results exhibited large grain sizes of Ni catalyst cannot grow carbon nanotube (CNTs) during CVD. Pretreatment of oxidation and the reduction reduce the grain size of the Ni catalyst, and an oxidation
temperature equal or higher than the reduction temperature always makes it easy to grow CNTs. A
smooth SiO2 thin lm was used as a separation layer to deposit the Ni catalyst to grow the SWCNTs and a
low area ratio of ID/IG for a non-crystallized sp3 graphite microstructure always creates a graphene
microstructure. A suitable hydrogen gas could help the amount of growth of CNTs. Anyway, the specied
Ni catalysis could promise strategy the selective synthesis of CNTs/graphite or CNTs/multi-graphene
composites on NiF.
2016 Elsevier B.V. All rights reserved.

1. Introduction
Carbon could be synthesized to a lot of structures such as
graphene, graphene oxide, graphite, oxided graphite and carbon
nanotubes (CNTs). In which graphene and CNTs have good
electricity and CNTs could become mature the 3-D structure.
Since graphene and CNTs combine a nanoscale size with unique
electrical, optical, mechanical, and electrochemical properties,
making them ideal candidate materials for high-impact applications in various elds, such as optoelectronics, energy, biomedicine, aerospace and so on [13]. With the raising of environmental
consciousness on the greenhouse effect and carbon dioxide
pollution in recent years, green-energy also catches the attention
as a project for our lives. Among these issues, hydrogen energy is
also a spotless energy; especially when applied as a fuel battery [4

* Corresponding author at: Graduate School of Optoelectronics Engineering,

National Yunlin University of Science and Technology, Yunlin 64002, Taiwan.
E-mail address: hcchen@yuntech.edu.tw (H.-C. Chen).
http://dx.doi.org/10.1016/j.synthmet.2016.05.029
0379-6779/ 2016 Elsevier B.V. All rights reserved.

6]. Its power density is always controlled by the transportation of

the proton; a nano-microstructure of the gas diffusion electrode
(GDE) plays an important role in enhancing the transportation of
the proton. The outstanding performance of graphene and CNTs
could be used for the GDE with a 3-D porosity substrate such as
metal foam [7,8], metal mesh [911], and so on. Metal foam
possesses a rich high strength, mechanical stability, high
conductivity and thermal dissipation. Simultaneously, the catalyst
of Fe, Co, and Ni was easily deposited on metal foam for the
deposition of CNTs [1214]. The response is very enthusiastic on
how to grow CNTs on metal foam. CNT deposition methods include
laser ablation [15,16], arc discharge [17], and chemical evaporation
deposition (CVD) [18,19]. Especially, the method of CVD can be
simple and easy to control with a high production of CNTs,
becoming the main fabrication process [20].
For this reason, many papers have researched the improvement
of gas diffusion on GDE with the base of CNTs [2123], showing
that CNT is better than other carbons, such as carbon black and
XC72 R [24]. Traditional GDE fabrication modulates the content of

H.-C. Chen et al. / Synthetic Metals 219 (2016) 124134

both carbon and polytetrauoroethene (PTFE) for gas diffusion

layers, but as people know PTFE is not a green material for our
environment. Since, CNTs maintain a high specic surface area,
perfect electricity (large mobility of 10,000 cm2/V/s), stable
mechanical properties and are an environmentally friendly
material they become a good succedaneum [2527]. The proposal
of this research is to use CNTs + nickel foam (NiF) to substitute for
carbon + PTFE on the GDE pole [28]. A total solution of CNTs
synthesized on nickel foam by CVD with different Ni catalysts was
investigated. These microstructures were checked by scanning
electron microscopy (SEM) imaging, X-ray diffraction (XRD) and
Raman spectra were used to check the crystal and bond. However,
the CNT deposition on 3-D NiF was not only dependent on the ow
rate, temperature and time of N2, H2, and C2H2 gases and the pretreatment of oxidation and reduction, but also dependent on the
microstructure and grain size of the nickel catalyst. However, these
results exhibited that the third method could fabricate better
single-walled carbon nanotubes (SWCNTs) layers than other
methods on NiF. The nanostructure of carbon-nanotubes/graphite
(CNTs/graphite) and carbon-nanotubes/multi-graphene (CNTs/Mgraphene) composites depended on both the deposition conditions
and Ni catalysis preparation.
2. Experimental
2.1. Deposition of a CNTs/NiF layer and pre-treatment
There are two pre-treatment processes on the 3-D NiF surface:
rst is the oxidation by oxygen and the second one is the reduction
by hydrogen. Then the Ni seed was got on the 3-D NiF surface, and
the reaction descriptions are below:
2xNi + O2 ! 2NixO

Table 1
The deposition conditions for the rst method.
Pre-treatment
(reduction)
Gas ow
(sccm)
N2

H2

0
80

0
20

Synthesis

Temp
( C)

Time (min)

0
800

0
5

Gas ow
(sccm)
N2

H2

C2H2

65
75

10

25

Temp
( C)

Time (min)

450
700

30

hydrogen (H2) was reduced from the carbon lm during the CVD.
The deposition temperatures in the furnace were 450  C and 700  C
for without and with reduction of H2, respectively.
2.1.2. The second method
Nickel foam substrate underwent pre-treatment of oxidization
with oxygen and then reduction by hydrogen, and then carbon was
deposited on the NiF by CVD. The clean NiF was oxidized and
reduced by H2 and carbon was deposited by CVD in the horizontal
furnace. The process involved oxidization, reduction and deposition of carbon by CVD; all were completed in the horizontal
furnace. The total experimental processes are shown in Fig. 2, and
relative fabrication parameters are listed in Table 2. The oxidation
temperature varied from 600  C, 650  C to 700  C, and the
reduction temperatures changed from 600  C, 650  C to 700  C.
The optimum oxidation and reduction temperatures were all the
same at 700  C. However, the synthesis temperatures of CVD varied
from 450  C, 500  C, 550  C, to 650  C and the synthesis times
changed from 5, 30, to 60 min at the same oxidation and reduction
temperatures of 700  C.

(1)

NixO + H2 ! xNi + H2O

(2)

CNTs/graphite composite and CNTs/M-graphene come from the

synthetized of the acetylene (C2H2) by the catalysis of Fe, Co, and Ni
to show below:
3H  C RC  H

125

Fe;Co;Ni

6C 3H2

Now, there are three methods to prepare the catalyst for

deposition of 3-D carbon/NiF by CVD. The deposition processes and
parameter conditions of the three methods are described below.
2.1.1. The rst method
Carbon was directly deposited on NiF with and without a Ni
catalyst during CVD. There were two different processes: one
deposited the carbon on the clear NiF by CVD without a Ni catalyst,
the other one sputtered the 5 nm Ni catalyst on NiF as a seed layer
and then deposited the carbon by CVD. The experimental processes
are shown in Fig. 1, and relative fabrication parameters are listed in
Table 1. The main working gas was C2H2 for carbon growth and

2.1.3. The third method

A 50 nm thickness of SiO2 lm was deposited on the NiF by
magnetron sputtering as a separation layer, and then 5 nm Ni was
deposited on it as a catalytic layer. The catalytic layer was smoother
and had a smaller grain size than previous methods. A pretreatment reduction of H2 was done and then, the CNTs were
synthesized on the substrate in the horizontal furnace. The
experimental process is shown in Fig. 3, and relative fabrication
parameters are listed in Table 3. The synthesis temperature varied
from 700  C to 800  C under a hydrogen reduction temperature of
700  C.The different synthesis times and hydrogen ows changed
from 5, 15, to 30 min and from 0, 5, to 40 sccm, respectively, under
the hydrogen reduction and deposition temperature of 700  C.
2.2. Measurements of microstructure, crystallization and bonds
The morphology of all lms was analysed with eld emission
scanning electron microscopy (FESEM, JSM-6701F). The maximum
multiple magnications could reach 100,000 to observe the
microstructure of the nickel foam, graphite, multi-graphene, and
CNTs. The grain size and surface morphology could be investigated

Fig. 1. The experimental process for the rst method.

126

H.-C. Chen et al. / Synthetic Metals 219 (2016) 124134

Fig. 2. The experimental process for the second method.

Table 2
The deposition conditions for the second method.
Pre-treatment
(oxidation)
Gas ow
(sccm)

Pre-treatment
(reduction)
Gas ow
(sccm)

Temp
( C)

Time (min)

N2

H2

60

80

60

Synthesis
Gas ow
(sccm)

Temp
( C)

Time (min)

N2

H2

C2H2

20

700

60

65

10

80

20

60

65

60

60

N2

O2

60

30

60

30

600
700
750
700

60

30

700

60

80

20

600
700
750
700

60

30

700

60

80

20

700

Temp
( C)

Time (min)

25

450

30

10

25

450

30

65

10

25

30

65

10

25

450
500
550
650
450

5
30
60

Fig. 3. The experimental process for the third method.

clearly by FESEM. The crystallization of thin lms was investigated

using X-ray diffraction (XRD, Bruker D8 Advance diffractometer).
Cobalt Ka radiation (lk = 17.8892 nm) from an X-ray tube with a
normal focus was employed. The XRD could observe the crystalline
of the nickel foam and carbon. The peak lines of (001) and graphite
(002) also were observed in detail. The characteristic bands of all
lms were measured by Raman spectroscopy (Jobin Yvon HORIBA,
RMS-iHR550). These Raman spectra were taken in a quasi-backscattering geometry using 100 mW at the 532.8 nm line of a
semiconductor laser focused on a square measuring 5  5 mm2, as

the excitation source. The Raman spectra could see the G and D
bands of carbon and the area ratio of ID/IG by Lorentzian tting
curvature.
3. Results and discussion
3.1. SEM images
The surface morphology of all lms was investigated by eld
emission SEM for carbon deposition with three different

H.-C. Chen et al. / Synthetic Metals 219 (2016) 124134

Table 3
The deposition conditions for the third method.
Pre-treatment
(reduction)
Gas ow
(sccm)

Synthesis

Temp
( C)

Time (min)

Gas ow
(sccm)

Temp
( C)

Time (min)

30

N2

H2

N2

H2

C2H2

80

20

700

75

25

80

20

700

75

25

700
800
700

80

20

700

75

0
5
40

25

700

5
15
30
5

deposition methods. The grain size, diameter size, and microstructures were all studied with different deposition conditions.
3.1.1. The rst method
The SEM images of Figs. 4(a)-(d) were of the clean NiF, 5 nm Ni
sputtering on NiF, and carbon deposited on NiF without and with a
5 nm Ni sputtering catalyst by CVD, respectively. In Figs. 4(a) and
(b) the grain size of the clean NiF was about 15 mm in the righttop drawing with the yardstick of 2 mm, so the grain size was too
big to synthesize the growth of CNTs but only grew graphite sheet.
Although 5 nm Ni catalyst on NiF could change grain structure to
515 nm Ni sheet it was also not easy to grow CNTs; only grew
multi-graphene and sometimes a very small amount of CNT
appeared, as shown in Figs. 4(c) and (d). People know the grain size
of 5 nm Ni sputtering seed also depends on both grain size and 3-D
structure of the NiF substrate, so the structure was too complex to
synthesize the CNTs. However, the 5 nm Ni sputtering seed allowed
smooth multi-graphene deposition on the clear NiF rather than no
sputtering seed, as shown in Figs. 4(b) and (d).

127

3.1.2. The second method

In order to understand the effect of the pretreatment of
oxidation temperatures, clear NiF was oxidized in a thermal
furnace with different temperatures. Fig. 5(a)(c) show the surface
morphology of carbon lm with different oxidation temperatures
of 600  C, 700  C and 750  C after CVD, respectively. The grain size
of a reduced Ni catalyst by hydrogen was increased with the
decrease in oxidation temperature, so the oxidation temperature of
600  C was too low to grow the CNTs resulting from the large grain
size of a Ni catalyst, as shown in Fig. 5(a). The phenomenon shows
that the high oxidation temperature will increase the thermal
stress to reduce the grain size of a reduced Ni catalyst. The
oxidation temperatures of 700  C and 750  C could successfully
grow CNTs, as shown in Fig. 5(b) and (c). At oxidation temperatures
of 700  C the grain size was a little larger and rougher than at
750  C, so the CNTs have slightly bigger diameter, and it is not easy
to grow the graphite sheet. However, the pretreatment of an
oxidation temperature of 700  C has a low stress on the substrate to
easily grow the CNTs.
Then, in order to improve the performance of a Ni catalyst, the
NiF was reduced by hydrogen with different temperatures.
Fig. 5(d)(f) show the surface morphology of carbon lms with
different reduction temperatures of 600  C, 700  C and 750  C,
respectively, at the oxidization temperature of 700  C. When the
reduction temperature was less than the oxidation temperature it
did not grow CNTs because of the large grain size. The phenomenon
also states that the low reduction temperature easily induced low
stress and a large grain size, so it only grew graphite. When the
reduction temperature was equal to the oxidization temperature
the CNTs could grow and accompany some graphite sheet, but
when the reduction temperature was higher than the oxidation
temperature the graphite sheet grew more than the CNTs. This
phenomenon resulted from the high reduction temperature
increasing roughness for the deposition of graphite sheet. The
optimal reduction temperature was chosen as 700  C under the
oxidation temperature of 700  C.

Fig. 4. The SEM images of surface morphology of (a) clean NiF, (b) carbon deposited on NiF, (c) NiF with 5 nm Ni sputtering catalyst, and (d) carbon deposited on Ni/NiF.

128

H.-C. Chen et al. / Synthetic Metals 219 (2016) 124134

Fig. 5. The SEM images of carbon lm with different oxidation temperatures from (a) 600  C, (b) 700  C to (c) 750  C and reduction temperatures from (d) 600  C, (e) 700  C to
(f) 750  C.

In order to understand the effect of the synthesis temperature

during CVD, the synthesis temperatures were changed from 450  C,
500  C, 550  C, to 650  C under pre-treatment oxidation and
reduction temperatures of 700  C. Fig. 6(a)(d) show the surface
morphology of CNTs with different synthesis temperatures. In
Fig. 6(a) and (b), the low synthesis temperature resulted from the
low degradation speed of a hydrocarbon (C2H2) and low deposition
speed of carbon, so the catalytic pellet of the nickel surface could
stack CNT microstructures. Simultaneously, the CNTs/graphite
composite both could be observed clearly, but in Fig. 6(c) and (d)
the increase in synthesis temperature led to the deposition speed
of carbon being faster than that of the CNTs growth. Therefore, the
Ni particles were covered by carbon losing catalytic activity. This

explains why we could only observe the multi-graphene layer and

some holes but little CNTs. Once the synthesis temperature was
increased over 650  C, the SP2 multi-graphene increased from the
degradation of C2H2 [29]. The carbon layer on the surface of the NiF
turned into a CNTs/M-graphene composite structure. This phenomenon would be proven by XRD and Raman spectra.
Figs. 7(a)(c) show synthesis times of 5, 30 and 60 min during
CVD at the synthesis temperature of 450  C. In Fig. 7(a), there are
only graphite sheet on the surface and no CNTs; the phenomenon
suggests that the synthesis time was too short to stack the CNTs. In
Fig. 7(b), there was CNTs/graphite on the surface. The situation
suggests that the synthesis time was enough to begin degradation
and diffuse the carbon particles, so the NiF surface appeared as a

Fig. 6. The SEM images of surface morphology of carbon lm with different synthesis temperatures of (a) 450  C, (b) 500  C, (c) 550  C, and (d) 650  C during CVD.

H.-C. Chen et al. / Synthetic Metals 219 (2016) 124134

129

Fig. 7. The SEM images of surface morphology of carbon lm with different synthesis times of (a) 5, (b) 30, and (c) 60 min after CVD.

compound of thick-short CNTs and graphite sheet. In Fig. 7(c), the

synthesis time of 60 min was too long to allow the NiF to be
carbonized.

the diameter of CNTs to become large which is not good for the
growth of CNTs, as seen in Fig. 9(f).
3.2. Measurement of X-ray diffraction (XRD)

3.1.3. The third method

The third method was rst to sputter a smooth layer of 50 nm
SiO2 on the surface of the NiF as a separation layer, and then,
sputter 5 nm Ni as a catalytic seed on the smooth SiO2 surface.
Fig. 8(a) and (b) show the SEM images of the synthesis
temperatures of 700  C and 800  C after CVD at a synthesis time of
30 min. The growth distribution of the CNTs was clearly better than
the rst and second methods. Simultaneously, CNT growth at
deposition temperatures of 700  C was better than that at 800  C
with small diameters of 15 nm. This phenomenon suggests that
the high synthesis temperature resulted from the degradation of
C2H2 being too quick, so the carbon particle covered the Ni catalyst
to reduce the catalytic activity and then grew thick-short CNTs, as
shown in Fig. 8(b).
Fig. 9(a)(c) show the surface morphology of CNTs with
different synthesis times of 5, 15 and 30 min, respectively, at a
synthesis temperature of 700  C. In Fig. 9(a) some nucleation Ni
metal catalytic also were found on the top of the CNTs, but at a
synthesis time of 15 min the surface will be covered with luxuriant
CNTs, as shown in Fig. 9(b). When the synthesis time increased to
30 min the diameter and size of the CNTs became large and some
small carbon clusters appeared. Therefore, a synthesis time of
15 min was the most suitable synthesis time. In order to nd the
effect of the reduction of hydrogen, hydrogen gas was introduced
into the furnace during the CVD process at the synthesis
temperature and time of 700  C and 5 min, respectively. Figs. 9(d)(f) show the SEM images of the surface morphology of CNTs with
different ows of 0, 5 and 40 sccm of hydrogen gas. A suitable
hydrogen ow could reduce the Ni catalyst on the surface to allow
the CNTs to become smooth. A hydrogen ow of 5 sccm allows
CNTs growth to become more luxuriant with a slim diameter, as
seen in Fig. 9(e), but the greater hydrogen ow of 40 sccm allows

The deposition of the second and third methods has more

carbon microstructures, so XRD was used to detect these relative
peak lines of carbon structures. There were two main diffraction
peaks of carbon that appeared in the XRD spectra: one (001) was
about 2u = 10.03 which was the oxided graphite peak and the
other one (002) was about 2u = 26.65 which was the graphite or
CNT peak [30,31].
3.2.1. XRD spectra for the second method
Fig. 10(a)(c) shows the XRD spectra of carbon deposition with
different oxidization temperatures of 600  C, 700  C and 750  C,
respectively, under reduction temperatures of 700  C. When
oxidization temperatures increased, the peak (001) of oxided
graphite decayed, such as in Figs. 10(b),(c). When oxidization
temperatures at 600  C, which was lower than the reduction
temperature, there was more oxygen bonding between carbons
resulting in the peak of oxided graphite shifting from 10.03 to
11.36 , such as in Fig. 10(a). These phenomena also can be observed
in Fig. 10(d)(f) for reduction. Fig. 10(d)(f) exhibits the XRD
spectra of carbon deposition with different reduction temperatures of 600  C, 700  C and 750  C, respectively, under an
oxidization temperature of 700  C. When the reduction temperature increased the oxygen bonding in the graphite would increase,
so the peak of oxided graphite would shift from 10.03 to 11.36 ,
such as in Fig. 10(f). When the reduction temperatures of 700  C
and 750  C were the same as or higher than the oxidation
temperature of 700  C the oxygen bonding would be increased, so
the peak (001) of oxided graphite could be observed. Figs. 10(a)(f)
show the (002) peaks of graphite at all different oxidation and
reduction temperatures; the situation is similar to the observations
of SEM images of graphite or graphite sheet in the previous section.

Fig. 8. The SEM images of surface morphology of CNTs with different deposition temperatures of (a) 700  C and (b) 800  C after CVD.

130

H.-C. Chen et al. / Synthetic Metals 219 (2016) 124134

Fig. 9. The SEM images of CNTs with different synthesis times of (a) 5, (b) 15 and (c) 30 min and hydrogen ows of (d) 0, (e) 5, and (f) 40 sccm.

Fig. 11 shows the XRD spectra of carbon deposition with

different synthesis temperatures of 450  C, 500  C, 550  C and
650  C during CVD. When the synthesis temperatures changed
from 450  C to 500  C a oxided graphite peak (001) was observed,
but when the synthesis temperature was higher than 550  C or
from 550  C to 650  C the oxided graphite peak disappeared and a
peak of graphite (001) appeared [32,33]. These phenomena were
also found by SEM images, as seen in Fig. 6. The low synthesis
temperatures of 450  C and 500  C made it easy to deposit oxided
graphite so the morphology was rough, as seen in Figs. 6(a) and (b),
but the high synthesis temperature of 550  C and 650  C could
cause the multi-graphene to stack so that the morphology was
smooth, as seen in Figs. 6(c) and (d). Finally, Fig. 12 shows the XRD
spectra of carbon deposition with different synthesis times of 5, 30,
and 60 min during CVD. The oxided graphite peak (001) was only
found at a synthesis time of 5 min, but the graphite peak (002) was
found at synthesis times of 30 and 60 min, and it was especially
protruded at 60 min for the carbonization. These phenomena
responded to SEM images as seen in the previous section.

3.2.2. XRD spectra for the third method

Figs. 13 and 14 show the XRD spectra for the third method. The
2u peaks ((111), (200), (220)) of 44.5 , 51.9 , and 76.4 are the peak
lines of the metal Ni for the 5 nm sputtering Ni particle [34,35], but
22.06 is the peak line of graphite or CNTs. Fig. 13 exhibits the XRD
spectra of SiO2/Ni/CNT lm with different synthesis temperatures
of 700  C and 800  C at a reduction temperature of 800  C. There are
three clear peak lines for the Ni metal, but the peak of 22.06 (002)
only appears at a synthesis temperature of 700  C; it does not
appear at 800  C. This phenomenon shows that when the synthesis
temperature increased to 800  C the microstructure began to
appear multi-graphene structure and, like the CNTs, also turned
from single-walled carbon nanotubes (SWCNTs) to multi-walled
carbon nanotubes (MWCNTs). Fig. 14(a)(c) shows the XRD spectra
of the SiO2/Ni/CNT lm with different synthesis times of 5, 15 and
30 min at reduction and synthesis temperatures of 800  C and
700  C, respectively. A 22.06 peak can be found at a synthesis time
of 15 min showing a good SWCNT microstructure. Fig. 14(d)(f)
also shows the XRD spectra of the SiO2/Ni/CNTs lm with different
H2 ows of 0, 5 and 40 sccm at a reduction and synthesis
temperature of 800  C and 700  C, respectively. The 22.06 peak
can be found for no hydrogen; once the hydrogen gas was
introduced to the CVD the SWCNTs turn to an MWCNT
microstructure. These phenomena can be double checked by
Raman spectra as in the next section.
3.3. Measurement of Raman spectra
The second and third deposition methods show CNT microstructures, so Raman spectra were used to investigate the band line
of G & D and the area ratio of ID/IG by Lorentzian tting to show the
performance of the CNTs. In the Raman spectra, the radial
breathing mode (RBM) can be used to study the nanotube
diameter (dt) through its frequency of appearance at 248/dt
cm1 for SWCNTs [36,37] and Table 4 shows these nanotube
diameters with different frequency.
WRBM = 248/dt

(4)
1

Fig. 10. XRD spectra of carbon deposition by the second method with different
oxidization temperatures of (a) 600  C, (b) 700  C, and (c) 750  C, and reduction
temperatures of (d) 600  C, (e) 700  C, and (f) 750  C.

C
The G-band usually appears at around 1600 cm for the C
tangential stretching (TM) mode from the sp2 graphite-like
microstructure; the area of the G-band (IG) shows the level of
graphite. The D-band appears at around 1300 cm1 for the defect

H.-C. Chen et al. / Synthetic Metals 219 (2016) 124134

Fig. 11. XRD spectra of carbon deposition by the second method with different
synthesis temperatures of (a)450  C, (b)500  C, (c)550  C and (d)650  C during CVD.

131

Fig. 14. XRD spectra of SiO2/Ni/CNT lm with different synthesis times of (a) 5, (b)
15 (c) 30 min and H2 ows of (d) 0, (e) 5 and (f) 40 sccm at a reduction and synthesis
temperature of 800  C and 700 .

Table 4
Nanotube diameter with different frequency for SWCNTs.
WRBM (cm1)
dt (nm)

125.79
1.97

130.01
1.90

162.34
1.52

183.84
1.35

196.74
1.26

Table 5
The ID/IG area ratio for the second method with different conditions.
Condition/Item

Fig. 12. XRD spectra of carbon deposition by the second method with different
synthesis times of (a)5, (b)30 and (c)60 min during CVD.

Oxidation temp. ( C)
Reduction:700  C
Synthesis:450  C
Time:30 min
Reduction temp. ( C)
Oxidation:700  C
Synthesis:450  C
Time:30 min
Synthesis temp. ( C)
Oxidation:700  C
Reduction:700  C
Time:30 min

Sample No.

D-band (cm1)

G-band (cm1)

ID/IG

(a)
(b)
(c)

600
700
750

1331.67
1329.67
1326.55

1579.85
1575.28
1577.12

3.37
2.00
1.99

(a)
(b)
(c)

600
700
750

1328.22
1329.67
1326.17

1576.68
1575.28
1575.51

1.82
2.00
2.22

(a)
(b)
(c)
(d)

450
500
550
650

1329.67
1325.29
1337.71
1337.01

1575.28
1574.16
1566.05
1564.47

2.00
1.91
0.06
0.45

5
30
60

1328.46
1329.67
1325.40

1577.12
1575.28
1571.98

2.14
2.00
1.89

Synthesis time (min)

Oxidation:700  C
(a)
Reduction:700  C (b)
Synthesis:450  C
(c)

Table 6
The ID/IG area ratio for the third method with different conditions.
Condition/Item

Sample No.

D-band (cm1)

G-band (cm1)

ID/IG

Synthesis temp. ( C)
Reduction:800  C
Time:30 min
Synthesis time (min)
Reduction:800  C
Synthesis:700  C
Fig. 13. XRD spectra of the SiO2/Ni/CNT lm with different synthesis temperatures
of (a) 700  C and (b) 800  C at a reduction temperature of 800  C.

induced band and dispersive from the sp3 microstructure; the area
of the D-band (ID) shows the defect of the graphite. The 2D band is
its second-order harmonic band and is highly dispersive always
appearing at 2700 cm1 [38,39]. Hence, Tables 5 and 6 exhibit the

H2 ow (sccm)
Reduction:800  C
Synthesis:700  C
Time: 5 min

(a)
(b)

700
800

1326.99
1335.45

1571.31
1562.63

1.64
0.09

(a)
(b)
(c)

5
15
30

1332.71
1326.87
1326.99

1573.92
1571.19
1571.31

1.22
1.49
1.64

(a)
(b)
(c)

0
5
40

1332.71
1329.94
1326.47

1573.92
1572.54
1570.62

1.22
1.28
1.76

132

H.-C. Chen et al. / Synthetic Metals 219 (2016) 124134

Fig. 15. Raman spectra of the second method with different oxidation temperatures from(a) 600  C, (b) 700  C to (c)750  C and reduction temperatures from (d) 600  C, (e)
700  C to (f)750  C.

values of both band lines of G & D and the area ratio of IG/ID for the
second and third methods.
3.3.1. Raman spectra for the second method
Figs. 15 show the Raman spectra of carbon deposition with
different oxidization and reduction temperatures, respectively. In
Fig. 15(a)(c), the carbon structure was graphite at an oxidization
temperature of 600  C, and the structure changed from high
graphite to low graphite with the increase in oxidation temperature to 750  C by the appearance of 2D peaks, as seen in Fig. 15(c)
[40,41]. Simultaneously, the area ratio of ID/IG decayed from 3.37,
2.00, to 1.99 for the different oxidation temperatures of 600  C,
700  C and 750  C, and they were all graphite or CNTs/graphite
composite structure. In Fig. 15(d),(e), the carbon structure was
changed from low graphite to high graphite by the increase in the
reduction temperature from 600  C, 700  C, to 750  C with the
disappearance of 2D peaks at 750  C, as seen in Fig. 15(d),(e).
Simultaneously, the area ratio of IG/ID increased from 1.82, 2.00
2.22 at reduction temperatures from 600  C, 700  C, to 750  C. The
phenomenon shows the growth condition for CNTs at an oxidation
temperature equal to the reduction temperature of 700  C. These
phenomena also could be observed by SEM images and XRD
spectra. Fig. 16 shows the Raman spectra of carbon deposition with
different synthesis temperatures during CVD. It was found that the
microstructure changed from CNTs/graphite to CNTs/M-graphene
composite
at
synthesis
temperatures
above
550  C.

Simultaneously, the area ratio of ID/IG decayed from 2.00, 1.91,

0.06, to 0.45 for synthesis temperatures from 450  C, 500  C, 550  C,
to 650  C in Table 5. Especially, the ID/IG ratio was reduced to 0.06
and 0.45 leading the graphite sheet change to multi-graphene

Fig. 16. Raman spectra of the second method with different synthesis temperatures
of (a) 450  C, (b) 500  C, (c) 550  C and (d) 650  C during CVD.

Fig. 18. Raman spectra of the third method with different synthesis temperatures of
(a) 700  C and (b) 800  C during CVD.

Fig. 17. Raman spectra of the second method with different synthesis times of (a)5,
(b)30, and (c)60 min during CVD.

H.-C. Chen et al. / Synthetic Metals 219 (2016) 124134

133

Fig. 19. Raman spectra of the third process with different synthesis times of (a) 5, (b) 15 and (c) 30 min and H2 ows of (d) 0, (e) 5 and (f) 40 sccm during CVD.

microstructure with a 2D band in the Raman spectra, since surfaces

become very smooth as seen in the SEM image and the (001) peak
appeared in the XRD spectra. Fig. 17 shows the Raman spectra of
carbon with different synthesis times during CVD. The ID/IG was
reduced from 2.14, 2.001.85 with the increase in time from 5, 30,
to 60 min, as shown in Table 5. This said, these microstructures all
were
CNTs/graphite composite, but in the synthesis time of 60 min the
nickel foam showed carbonization all over.
3.3.2. Raman spectra for the third method
Figs. 18 and 19 show the Raman spectra of the third method
with a SiO2 thin lm as a separation layer. Fig. 18 shows the Raman
spectra of the third method with different synthesis temperatures
of 700  C and 800  C during CVD. At a deposition temperature of
700  C, an RBM peak line was found in the Raman spectra, so the
microstructure of the carbon was SWCNTs. But at a synthesis
temperature of 800  C the carbon almost changed to an sp2
structure when the ID/IG ratio reduced from 1.64 to 0.09, as seen in
Table 6, since the microstructure changed from single-wall carbon
nanotubes (SWCNTs) to multi-wall carbon nanotubes (MWCNTs)
as the synthesis temperature changed from 700  C to 800  C, as
seen in the Raman spectra [42]. Fig. 19(a)-(c) exhibits the Raman
spectra of the third method with different synthesis times of 5, 15
and 30 min during CVD. The peak size of the RBM decays with the
increase in synthesis time; this said the SWCNTs would change to
MWCNTs with the increase in the ID/IG ratio, as seen in Table 6.
Fig. 19(d)(f) shows the Raman spectra with different ows of
hydrogen during CVD. When the hydrogen ow was zero, the RBM
mode could be clear obverse to show the microstructure of
SWCNTs such as in Fig. 19(d) and we could nd these responded
lines of 125.79, 130.01, 162.34, 183.84, and 196.74 cm1 as showed
in Table 4. When the hydrogen was introduced to the deposition,
the RBM peaks disappeared to support the microstructural change
from SWCNTs to MWCNTs. So, we found thick-short CNTs after the
introduction of hydrogen, as seen in the SEM images. The SEM
images and XRD spectra always respond to the Raman spectra.
4. Conclusions
The research investigated carbon deposited on porous nickel
foam by CVD with different preparations of Ni catalyst. The
analysis of the surface morphology tells us that large grain sizes of

Ni catalyst cannot grow CNTs during CVD. Pre-treatment of

oxidation and the reduction of hydrogen reduce the grain size of
the Ni catalyst, and an oxidation temperature equal or higher than
the reduction temperature always makes it easy to grow CNTs. A
suitable synthesis temperature and time will allow the growth of
CNTs on nickel foam; however, a smooth SiO2 thin lm was used as
a separation layer to deposit the Ni catalyst to grow the SWCNTs.
The deposition temperature and time could affect the size and
length of the SWCNTs changing them to MWCNTs. A low area ratio
of ID/IG for a non-crystallized sp3 graphite microstructure always
creates a graphene microstructure. A suitable hydrogen gas could
help the amount of growth of CNTs, but hydrogen gas also allows
the CNT microstructure to change from SWCNTs to MWCNTs.
However, the specied Ni catalysis could promise strategy the
selective synthesis of CNTs/graphite or CNTs/M-graphene composites on 3-D nickel foam.
Acknowledgement
The authors would like to thank the Ministry of Science and
Technology of Taiwan for nancially supporting this research
under contract no. MOST 104-2622-E-224-008-CC3 and MOST
104-2220-E-224-001 .
References
[1] M.M. Shulaker, G. Hills, N. Patil, H. Wei, H.Y. Chen, H.S.P. Wong, et al., Carbon
nanotube computer, Nature 501 (7468) (2013) 526530.
[2] M.W. Rowell, M.A. Topinka, M.D. McGehee, H.J. Prall, G. Dennler, N.S. Sariciftci,
et al., Organic solar cells with carbon nanotube network electrodes, Appl. Phys.
Lett. 88 (23) (2006) 233506.
[3] H.C. Chen, S.F. Lin, K.T. Huang, Thorn-like ZnO/CNT composites via the
hydrothermal method with different seed layer, Appl. Opt. 53 (4) (2014)
242247.
[4] B.C.H. Steele, A. Heinzel, Materials for fuel-cell technologies, Nature 14 (6861)
(2001) 345352.
[5] M. Han, J.H. Xu, S.H. Chan, S.P. Jiang, Characterization of gas diffusion layers for
PEMFC, Electrochim. Acta 53 (16) (2008) 53615367.
[6] M.P. Manahan, M.C. Hatzell, E.C. Kumbur, M.M. Mench, Laser perforated fuel
cell diffusion media. Part I: related changes in performance and water content,
J. Power Sources 196 (13) (2011) 55735582.
[7] S. Park, J.W. Lee, B.N. Popov, Effect of carbon loading in microporous layer on
PEM fuel cell performance, J. Power Sources 163 (2006) 357363.
[8] J.H. Chun, K.T. Park, D.H. Jo, J.Y. Lee, S.G. Kim, S.H. Park, et al., Development of a
novel hydrophobic/hydrophilic double micro porous layer for use in a cathode
gas diffusion layer in PEMFC, Int. J. Hydrogen Energy 36 (2011) 84228428.
[9] A.L. Ong, A. Bottino, G. Capannelli, A. Comite, Effect of preparative parameters
on the characteristics of poly(vinylidene uoride)-based microporous layer for
proton exchange membrane fuel cells, J. Power Sources 183 (2008) 6268.

134

H.-C. Chen et al. / Synthetic Metals 219 (2016) 124134

[10] J. Chen, H. Xu, H. Zhang, B. Yi, Facilitating mass transport in gas diffusion layer
of PEMFC by fabricating micro-porous layer with dry layer preparation, J.
Power Sources 182 (2008) 531539.
[11] T.F. Hung, J. Huang, H.J. Chuang, S.H. Bai, Y.J. Lai, Y.W. Chen, Highly efcient
single-layer gas diffusion layers for the proton exchange membrane fuel cell, J.
Power Sources 184 (1) (2008) 165171.
[12] Y. Xing, Synthesis and electrochemical characterization of uniformlydispersed high loading Pt nanoparticles on sonochemically-treated carbon
nanotubes, J. Phys. Chem. B 108 (50) (2004) 1925519259.
[13] S. Iijima, Helical microtubules of graphitic carbon, Nature 354 (6348) (1991)
5658.
[14] S. Iijima, T. Ichihashi, Single-shell carbon nanotubes of 1-nm diameter, Nature
363 (6430) (1993) 603605.
[15] T. Guo, P. Nikolaev, A. Thess, D.T. Colbert, R.E. Smalley, Catalytic growth of
single-walled nanotubes by laser vaporization, Chem. Phys. Lett. 243 (1) (1995)
4954.
[16] H. Katauraa, Y. Kumazawa, Y. Maniwa, Y. Ohtsuk, R. Sen, S. Suzuki, et al.,
Diameter control of single-walled carbon nanotubes, Carbon 38 (12) (2000)
16911697.
[17] E.T. Thostenson, Z.F. Ren, T.W. Chou, Advances in the science and technology of
carbon nanotubes and their composites: a review, Compos. Sci. Technol. 61
(13) (2001) 18991912.
[18] C.J. Lee, J.H. Park, J.A. Yu, Catalyst effect on carbon nanotubes synthesized by
thermal chemical vapor deposition, Chem. Phys. Lett. 360 (3) (2002) 250255.
[19] P.N. Nirmalraj, P.E. Lyons, S. De, J.N. Colem, J.J. Boland, Electrical connectivity in
single-walled carbon nanotube networks, Nano Lett. 9 (11) (2009) 38903895.
[20] P. Ramesh, T. Okazaki, T. Sugai, J. Kimura, N. Kishi, K. Sato, et al., Purication and
characterization of double-wall carbon nanotubes synthesized by catalytic
chemical vapor deposition on mesoporous silica, Chem. Phys. Lett. 418 (4)
(2006) 408412.
[21] A.M. Kannan, V. Veedu, L. Munukutla, M.N. Ghasemi-Nejhad, Nano structured
gas diffusion and catalyst layers for PEMFCs, Solid State Lett. 10 (3) (2007)
B4750.
[22] X. Wang, H. Zhang, J. Zhang, H. Xu, X. Zhu, J. Chen, et al., A bi-functional microporous layer with composite carbon black for PEM fuel cells, J. Power Source
162 (2006) 474479.
[23] E.T. Thostenson, T.W. Chou, Processing structure-multi-functional property
relationship in carbon nanotube/epoxy composites, Carbon 44 (14) (2006)
28693148.
[24] R. Kannan, B.A. Kakade, V.K. Pillai, Polymer electrolyte fuel cells using Naonbased composite membanes with functionalized carbon nanotubes, Angew.
Chem. Int. Ed. 47 (14) (2008) 26532656.
[25] A. Jarrah Nabeel, J.G. van Ommen, L. Lefferts, Mechanistic aspects of the
formation of carbon-nanobers on the surface of Ni foam: a new
microstructured catalyst support, J. Catal. 239 (2) (2006) 460469.

[26] N. Jeong, J.H. Lee, Growth of lamentous carbon by decomposition of ethanol

on nickel foam: inuence of synthesis conditions and catalytic nanoparticles
on growth yield and mechanism, J. Catal. 260 (2) (2008) 217226.
[27] D. Zhao, Z. Yang, L. Zhang, X.L. Feng, Y.F. Zhang, Electrodeposited manganese
oxide on nickel foamsupported carbon nanotubes for electrode of
supercapacitors, Electrochem. Solid-State Lett. 14 (2011) A9396.
[28] J. Xiong, X.F. Dong, Y.C. Dong, X.F. Hao, S. Hampshire, Dual-production of nickel
foam supported carbon nanotubes and hydrogen by methane catalytic
decomposition, Int. J. Hydrogen Energy 37 (17) (2012) 1230712316.
[29] H.T. Fang, C.G. Liu, C. Liu, F. Li, M. Liu, H.M. Cheng, Purication of single-wall
carbon nanotubes by electrochemical oxidation, Chem. Mater. 16 (26) (2004)
57445750.
[30] M. Fang, K.G. Wang, H.B. Lu, Y.L. Yang, S. Nutt, Covalent polymer
functionalization of graphene nanosheets and mechanical properties of
composites, J. Mater. Chem. 19 (38) (2009) 70987105.
[31] M. Khenfouch, U. Buttner, M. Batoul, M. Maaza, Synthesis and characterization
of mass produced high quality few layered graphene sheets via a chemical
method, Graphene 3 (2014) 713.
[32] K. Zhang, Y. Zhang, S. Wang, Enhancing thermoelectric properties of organic
composites through hierarchical nanostructures, Sci. Rep. 3 (2013) 17.
[33] M.H. Choi, Y.J. Min, G.H. Gwak, S.M. Paek, J.M. Oh, A nanostructured Ni/
graphene hybrid for enhanced electrochemical hydrogen storage, J. Alloys
Compd. 610 (2014) 231235.
[34] Syukri, T. Ban, Y. Ohya, Y. Takahashi, A simple synthesis of metallic Ni and Ni
Co alloy ne powders from a mixed-metal acetate precursor, Mater. Chem.
Phys. 78 (2003) 645649.
[35] L.J. Wang, H.B. Liu, Electroless nickel plating on chitosan-modied wood
veneer, BioResources 6 (2) (2011) 20352044.
[36] H. Kataura, Y. Kumazawa, Y. Maniwa, I. Umezu, S. Suzuki, Y. Ohtsuka, et al.,
Optical properties of single-wall carbon nanotubes, Synth. Met. 103 (13)
(1999) 25552558.
[37] M.J. O'Connell, S.M. Bachilo, C.B. Huffman, V.C. Moore, M.S. Strano, E.H. Haroz,
et al., Band gap uorescence from individual single-walled carbon nanotubes,
Science 297 (5581) (2002) 593596.
[38] M.S. Dresselhaus, G. Dresselhaus, A. Jorio, A.G. Souza Filho, R. Saito, Raman
spectroscopy on isolated single wall carbon nanotubes, Carbon 40 (12) (2002)
20432061.
[39] H. Kawamoto, T. Uchida, K. Kojima, M. Tachibana, The feature of the BreitWigner-Fano Raman line in DNA-wrapped single-wall carbon nanotubes,
Appl. Phys. Lett. 99 (9) (2006) 094309.
[40] R. Saito, G. Dresselhaus, M.S. Dresselhaus, Physical Properties of Carbon
Nanotubes, College Press, London Imperial, 1998.
[41] M.S. Dresselhaus, P.C. Eklund, Phonons in carbon nanotubes, Adv. Phys. 49 (6)
(2000) 705814.
[42] M.S. Dresselhausa, G. Dresselhausb, R. Saito, A. Joriod, Raman spectroscopy of
carbon nanotubes, Phys. Rep. 409 (2005) 4799.