Sunteți pe pagina 1din 43

Energy Conversion and Management

Volume 90, 15 January 2015, Pages 218229

Co-production of hydrogen and carbon nanofibers from


methane decomposition over zeolite Y supported Ni catalysts

Md. Nasir Uddina, ,


W.M.A. Wan Dauda, , ,
Hazzim F. Abbasb
Show more
doi:10.1016/j.enconman.2014.10.060
Get rights and content

Highlights

Methane cracking requires an optimum temperature range of 550600 C for


H2yield.

At 550 and 600 C, catalyst showed longer activity for the whole test.

At 600 C, a 614.25 gc/gNi of carbon was obtained using 30% Ni/Y zeolite
catalysts.

Produced filamentous carbon has the same diameter as the metallic nickel itself.

VHSV has reverse and non-linear relevancy to the weight of Ni/Y zeolite catalyst.

Abstract
The objective of this paper is to study the influences of different operating
conditions on the hydrogen formation and properties of accumulated carbon
from methane decomposition using zeolite Y supported 15% and 30% Ni,
respectively, at a temperature range between 500 and 650 C in a pilot scale
fixed bed reactor. The temperature ramp was showed a significant impact on
the thermo-catalytic decomposition (TCD) of methane. An optimum temperature
range of 550600 C were required to attain the maximum amount of methane
conversion and revealed that at 550 and 600 C, catalyst showed longer
activity for the whole studied of experimental runs. Additionally, at 550 C, the
methane decomposition is two times longer for 30% Ni/Y zeolite than that for
15% Ni/Y zeolite catalyst, whereas it is almost three times higher at 500 C. A
maximum carbon yield of 614.25 and 157.54 gc/gNi were reported after end of
the complete reaction at 600 C with 30% and 15% Ni/Y zeolite catalyst,
respectively. From BET, TPD, and XRD analysis, we had reported that how the
chemistry between the TCD of methane and metal content of the catalysts
could significantly affect the hydrogen production as well as carbon nano-fibers.
TEM analysis ensured that the produced carbon had fishbone type structures
with a hollow core and grew from crystallites of Ni anchored on the external
surface of the catalysts and irrespective of the metal loadings, the whisker
types of nano filaments were formed as confirmed from FESEM analysis.
Nevertheless, the effect of volume hourly space velocity (VHSV) on the
methane conversion was also investigated and reported that the methane
conversion increased as VHSV and nickel concentration in NiY catalysts
increased. Additionally, the initial methane decomposition rate increases with
VHSV and it has reverse and non-linear relevancy to the weight of Ni/Y zeolite
catalyst.

Keywords
Coproduction;
Hydrogen;
Carbon nanofiber;
Zeolite Y;
Supported Ni;

Methane decomposition

1. Introduction
The idea of co-production of hydrogen and carbon nanomaterials from thermocatalytic decomposition (TCD) of methane has been having a great mettlesome
attention among the Scientists and Engineers around the earth. This is because
it produces environmentally benign hydrogen and carbon nanomaterials that
seem to be a clean, halcyon and officious product in foreseeable
future [1], [2] and [3]. Both of the products are derived from the TCD of
methane, according to the following equation at low temperatures ranging from
400 to 700 C when the process associated with the schemes of metal
catalysts [4]:
equation(1)
Turn MathJax on

Because of its lower cost selectivity, and much higher active site/carbon
capacities, Ni has been broadly assessed in manufacturing of hydrogen from
methane decomposition as compared with the cobalt or iron [5] and [6].
Regards to catalytic materials, the use of zeolitic materials has of both
fundamental and practical significance due to their well-defined pore
distributions, high surface area and unique catalytic features as compared to
traditional catalyst supported on alumina, silica-alumina, CeO 2, peroviskite,
active carbon, La2O3, and MgO [7]. The Y-zeolite has comparatively larger pores
(Entry aperture of 0.74 nm, and a diameter of 1.3 nm) than the others such as
HZSM-5 (0.510.55 nm; 0.510.55 nm) and zeolite A (0.41 nm 1.1 nm).
Therefore, the introduction of nickel metals into Y zeolites by means of a
suitable method such as impregnation or ion-exchange, etc., would make it
suitable for TCD of methane. However, the research work on Y zeolite
supported Ni catalysts in hydrogen and carbon nanomaterials production from
TCD of methane is still very limited even though a great number of articles have
already been disclosed the application of the metallic nickel catalysts on the
activity of different support materials [2], [6] and [8].

The degree of hydrogen and carbon nanomaterials production from TCD of


methane depend as much on the reactor operating conditions such as feeding
rate, reaction temperature and volume hourly space velocity (VHSV), as on the
type of textural characteristics of support materials, catalysts composition, and
the additive concentration. Inaba et al. [9] have pointed out that the catalyst
optimization using silica and several types of zeolite supported Ni catalysts for
hydrogen yields from methane cracking at 650 C. They described that USYzeolite supported Ni catalysts showed higher catalytic longevitys (owing to
active for around 6 h) as compared to silica supported Ni catalyst. The highest
methane conversion was highlighted approximately 45%. They cited that the
quantity of solid acidity in the support was tending to be lower if nickel particles
have a smaller diameter. The accumulated carbon (AC) was estimated to be
higher on the surface of Ni-supported catalysts even after complete
deactivation when the supports have a larger external surface area. Very
similar results were reported by Ashok et al. [8] who found that the highest
activity subjected to Ni/HY catalyst (12 h) and higher activity of 955
molH2/molNi compared to other tested support on HY, USY, SiO 2and SBA-15 at
550 C. At a temperature range between 750 and 900 C and VHSV range
between 3.0 and 18.0 Lg1 h1 were set to investigate the methane
decomposition for hydrogen yield using activated alumina and carbon as
highlighted by Bai et al. [10]. At early of the decomposition reaction, the
methane conversion was reflected maximum value around 2535% for all types
of used catalysts and thereafter, it started to decline progressively over the
course of reaction stream (2 h). The catalytic activity took place primarily in the
micropores that is confirmed from the change of pore size of activated carbon
during the methane decomposition reaction. Apart from activated carbon
textural properties, the activated alumina catalysts showed mesoporous nature
that is subjected to the catalytic activity, resulting in a variation of carbon
deposits and textural properties. Rely on the different precursors of activated
carbon; two classes of AC can be appeared on the activated carbon either in
the form of carbon filaments or agglomerates.
The work reported here emphasized at clarifying the influences of different
operating conditions in a fixed bed reactor on hydrogen formation and
properties of AC using two different Ni loading on the catalytic activity of Y
zeolite catalyst. In addition, the structural and morphological features of the

produced carbon and the textural features of the samples used are also
examined for different operating parameters.

2. Experimental Methodology
2.1. Experimental view and operating conditions
Raw methane (99.9995%) purchased from Linde Compressed and nitrogen
(99.999%) served by Air products were utilized in the experiment. The zeolite Y
catalyst (surface area = 650 m2/g, Si/Al = 10) purchased from Tosoh
Corporation, Japan and used it as in the form of impregnation with nickel
metals. The catalysts preparation methods and experimental apparatus used
are described in details elsewhere [3]. In each experimental run, typically 3 g of
Ni/Y zeolite catalyst was loaded into the reactor. Then the reaction started with
the flow of 1 L/min of CH4 over the catalyst bed using 15% and 30% Ni/Y zeolite
catalysts at 500, 550, 600, and 650 C, respectively. Before reaction, the
catalyst is reduced by 20% hydrogen at 600 C (heating ramp was 10 C/min)
in nitrogen for 2 h in order to obtain active metallic state in the catalyst and this
is continued until system reached a set temperature. After that, a 1 L/min of
nitrogen gas was thoroughly flushed for another 1 h within the reactor before
feeding the set CH4 flow rate. The reaction continued until the spontaneous
hydrogen yield below 7%. The percentage of methane conversion calculated
from data obtained from outflow gas that was analyzed by an online gas
analyzer (GA). Using GA, other than hydrogen, no gaseous products detected.
A flow of nitrogen was purged through the reactor to cool down it to ambient
temperature after the reaction, and the total amount of solid carbon measured
by using the following equation:
equation(2)

Turn MathJax on

In Eq. (2), wtot and wcat represents the total and initial weight of the catalyst after
and before reaction, respectively. This yield was also expressed in the form of
gdep/gNi.
2.2. Characterization techniques

In order to investigate the textural features of the catalysts and nanomaterials,


the N2physisorption technique was done using nitrogen as adsorbent at 77 K
from the Micromeritics ASAP 2020 sorptometer. A Micromeritics TPD/TPR 2720
analyzer was used to characterize how NH3 molecules are strongly conjugated
to the acid sites qualitatively. Firstly, a typical 0.50 g catalyst samples was
heated in a flow system at a 30 mL/min helium flow rate with a temperature
ramp of 10 C/min increase to 25780 C, and at this temperature, the system
stays for 30 min. Next, a helium flow of 20 mL/min was performed to reduce the
catalyst until cool it to 210 C. Thereafter, ammonia was streamed on the
samples for 20 min. In order to reduce the physisorbed elements from the
samples, a purging procedure was introduced with helium for another one hour.
The chromatograms were obtained from the signal processing of thermal
conductivity detector using the temperature ramp of 10 C/min from 210 C to
750 C.
X-ray diffraction (XRD) patterns of the fresh and spent catalysts were
determined on a Rigaku miniflex using Cu K radiation with a generator voltage
and current of 45 kV and 40 mA, respectively. Using global Scherrer equation
(based on the half-width of diffraction lines assigned to planes (1 1 1), (2 0 0),
and (2 2 0) for NiO and Ni phase, respectively), the average crystallite size
(Davg) was obtained as follows [11]:
equation(3)

Turn MathJax on

where 0.9 is the Scherrer constant, is the full width at half maxima (in
radians) of a reflection located at 2.
The morphological structure and the diameter distribution of the catalysts and
carbon nanomaterials were performed on a Field Emission Scanning electron
microscopy, FESEM (QUANTA 450 FEG) and transmission electron
microscopy, TEM (Hitachi HT-7700). Totally, more than 50 carbon nanofibers
(CNFs) was measured on the TEM images, and the result reported as the
arithmetic mean value of the data.

3. Results and discussion


3.1. Thermal decomposition of Methane over zeolite Y supported Ni catalysts

3.1.1. Effect of temperature ramp on TCD of methane


In Table 1, we report the reaction yields in terms of the three main conjugating
reactions (initial H2%, initial CH4%, H2 formation rate, and average carbon
formation rate, ACFR) at the temperature range of 500650 C. In this
experiment, the hydrogen formation rate (HFR) was presented in terms of
average hydrogen formation rate (AHFR), and total hydrogen formation (THF)
at three representative reactions times for making a concise assessment of
TCD of methane over Ni/Y zeolite catalyst. Reported data in Table 1indicates
that the initial H2 and CH4 increase for both kinds of Ni-supported Y zeolite
catalyst as temperature increases. It can be observed from Table 1 that the
highest initial H2 and CH4 have to be found 41.02% and 25.76%, for 15% Ni/Y
zeolite catalyst whereas the value of 58.58% and 41.67% was noted for 30%
Ni/Y zeolite catalyst, respectively at 650 C. This is reasonable because TCD of
methane is an endothermic reaction, so equilibrium methane conversion is
increased with an increase in temperature [5]. However, the methane
decomposition showed a relatively shorter activity for more than 2 and 3 h for
15% and 30% Ni/Y zeolite catalyst, respectively (Fig. 1). At 600 C, the
decomposition rate is lesser than at 650 C, but shows higher catalytic activity
for more than 5 and 11 h before becomes negligible for 15% and 30% Ni/Y
zeolite catalyst, respectively. As can be noticed from Fig. 1, that the methane
decomposition at 550 C has maximum catalytic activity as compared with
600 C and it endures for more than 6 and 12 h for 15% and 30% Ni/Y zeolite
catalyst, respectively. On the other hand, at 500 C, the decomposition of
methane showed lower stability for more than 2 and 7 h for 15% and 30% Ni/Y
zeolite catalyst, respectively; before falling to almost zero. The above results
indicate that at 550 C, the methane decomposition is two times longer for 30%
Ni/Y zeolite than that for 15% Ni/Y zeolite catalyst, whereas it is almost three
times higher at 500 C. Interestingly, at 550 and 600 C, catalyst showed longer
activities for the whole studied of experiment runs. This is notifying that the
TCD of methane on Ni/Y zeolite catalyst required an optimum temperature and
in this case it is ranged from 550 to 600 C. As can be visualized from Fig. 1,
there was no induction period at which methane conversion pass through a
maximum level [12].
Table 1.
Experimentally obtained results from methane decomposition over Ni-supported Y zeolite under
selected reaction conditions.

Ru
n
no.

T(
C)

Initia
l
H2a(
%)

Initial
CH4b(
%)

Behavior of H2 production with respect to time

At time = 10 min

AHFR
THF
(mmol/
(mol/g (gcat.
min))
cat)
With 15% Ni-supported Y zeolite catalysts
1
500 19.0 10.52 36.14
4.88
9
2
550 30.2 17.73 59.35
8.56
7
3
600 36.3 22.11 70.21
10.46
2
4
650 41.0 25.76 80.62
12.40
2
With 30% Ni-supported Y zeolite catalysts
5
500 18.0 9.91
39.36
5.36
8
6
550 34.8 20.82 70.41
10.49
5
7
600 48.8 31.57 95.27
15.36
5
8
650 58.5 41.67 116.6
20.22
8
5

At time = 26 min

THF
(mol/g
cat)

AHFR
(mmol/
(gcat.
min))

81.08

4.67

133.3
3
159.3
4
182.9
1

8.19

84.47

4.89

159.2
3
216.3
1
265.8
0

10.10

10.11
11.97

14.82
19.57

ACFR,
(mmol/(gcat
min))

At time = 60 min

AHFR
(mmol/
(gcat.
min))

After end of
complete
reaction

161.5
4
270.7
0
331.5
5
372.3
4

4.17

1.35

7.41

2.06

9.39

2.87

10.82

3.61

172.4
0
324.5
8
449.1
5
557.1
9

4.47

1.79

9.17

2.45

13.66

4.32

18.16

6.01

THF
(mol/g
cat)

The initial hydrogen (%) was derived from the curve fitting of instantaneous hydrogen yield versus
reaction time using a proper nth order polynomial equation with regression coefficients greater
than 0.99 and then substituted of time equal to zero in the right hand side of the curve fitting
equation.
b

The initial methane (%) was calculated exactly as the same method as for initial hydrogen (%)
except, in this case, using the methane conversion against reaction time curve fitting.
Table options

Fig. 1.
Representation of methane conversion as a function of reaction time on stream for (a) 15% Ni/Y
zeolite catalyst, and (b) 30% Ni/Y zeolite catalyst.
Figure options

In accordance with the above results, Inaba et al. [9] has demonstrated that Nisupported on USY zeolite shows the longer activity for more than 6 h among
the others. Very similar results were reported by Ashok et al. [8] who found that
highest activity subjected to 30% Ni/HY catalyst (12 h) and higher activity of
955 molH2/molNi compared to other tested support on HY, USY, SiO2 and SBA-15
at 550 C. With 15% Ni/HY catalysts, they found hydrogen yield of 993
molH2/molNi, whereas 370 molH2/molNi of hydrogen production was noticed for
60% Ni/HY catalysts. Michalkiewicz and Majewska [13] reported about 55% of
hydrogen was produced using the Ni/ZSM-5 (3 0 0) catalysts from methane
decomposition at a temperature range of 650 C and 700 C. The highest
carbon yield was reported to be approximately 32 wt.% at 700 C by using Ni
particle size of about 3060 nm depending on the temperature. In comparison
to these results, ours results is a quite bit lower but our study examined the
temperature range which is much lower than do Michalkiewicz and Majewska
research.
The above discussion established that the carbon deposition is directly related
to the metal content of the catalyst. The highly distribution of nickel species into
the zeolite Y cages and likely more significantly, the synergistic influence
between the microporous surfaces of zeolite Y and metallic nickel were
considered to be the reasons for higher catalytic performance of 30% Ni/Y
zeolite [7]. A greater electron density on the surface of metallic nickel and
increment of the retention capacity of hydrogen in the Ni/Y zeolite catalysts
were caused from the synergistic effects due to it may cover the interactions
between the higher ionic microporous surface of zeolite Y and the nickel
particles. Eventually, a rapid catalytic deactivation of 30% Ni/Y zeolite was

observed in the course of decomposition reaction due to a large number of


nickel species on the catalysts surface resulting in a carbon accumulation over
the active metal species. Accordingly, it is believed that much more carbon can
be generated only when the outer surface area of the catalysts covered by
metal particles and these sites are very much elevated when metal is deposited
by impregnation method [7].
According to stoichiometry calculation of methane decomposition, the AHFR
was computed assuming that only hydrogen and carbon were produced. The
THF was determined by summing on the total instantaneous hydrogen formed
per initial Ni-supported Y zeolite catalyst mass. It can be seen from Table 1 that
as time proceeds, THF is increased as temperature increased while AHFR is
almost straightly decreased for both type of catalyst. This is more likely that
TCD of methane with respect to time increase the deposition of carbon on the
Ni/Y zeolite catalyst surface. Additionally, the AHFR increased significantly with
increasing temperature but reduced the catalyst lifetime. This is reasonable
because at 650 C, the deactivation is rapid and results in a complete loss of
catalyst activity due to solid carbon formation rate was higher rather than the
carbon removal rate from the active surface, which led to the carbon
accumulation on the surface and thus, AHFR was reduced over the course of
decomposition reaction. On the other hand, the ACFR also increases as the
temperature increases from 500 to 650 C as shown in Table 1.
Fig. 2(a) and (b) shows the accumulated carbon (AC) against time plot for
temperatures of 500, 550, 600 and 650 C for 15% and 30% Ni/Y zeolite
catalyst, respectively. The AC was derived from the multiplication of carbon flux
and the integration of the curve fitting of methane conversion against reaction
time using a proper third order polynomial equations with regression
coefficients greater than 0.99. The initial AC over Ni/Y zeolite increases for
temperatures of 500650 C, while deactivation time decreases when the
reaction temperature increases from 550 to 600 C. The deactivation time at
500 and 650 C was almost identical and TAC was lower especially for 650 C
when compared with 550 and 600 C as shown in Table 2. Interestingly, the
highest TAC with 15% Ni/Y zeolite catalyst was almost equal to the lower TAC
value with 30% Ni/Y zeolite catalyst as shown in Table 2. In Fig. 2, we notice
two rather asymmetrical deactivation processes. In the first stage, at the
temperature range of 550600 C, a sluggish reduction of the decomposition
rate and at 500 and 650 C, the deactivation is faster and results in an absolute

decrease of the catalytic activity. We speculate that at 500 C, the carbon


solubility is lower within the metal and the presence of polyaromatic
hydrocarbon in the gas phase to start any type of carbon material growth is
limited [14]. Chen et al. [15] have demonstrated that the aggravated saturated
concentration of carbon materials was brought to the catalyst surface by low
driving force of carbon diffusion which in turns, show a low carbon formation
over the small size of Ni crystal sites which lead to a drastic deactivation of the
catalysts. Additionally, the degree of carbon formation in zeolites is restricted by
the limitations ascribed by the microporous structure; i.e., the loss of activity
seems to be a space restriction as the microporous network becomes confined
with carbon [7]. However, there is indication of that a decline in the rate of
carbon accumulation with expanded reaction times can be associated to the
poisoning of the active metal particle due to encapsulating carbon [7]. We can
speculate further that at 650 C, the free radical condensates or polyaromatic
hydrocarbons are the building blocks of carbon materials. The free radical
condensates are predicted to be a heap of carbon materials with differing
quantities of hydrogen atoms [16]. Reilly and Whitten [16]pointed out that the
free radical condensates are susceptible to quick rearrangement at 650 C, and
owing to the way in which hydrogen inattentive by radical recombination, a
diverse categories of carbon species can be generated. Structural and textural
rearrangements of the catalyst particle are happened with the increase of
carbon growth rate, which lead to form centers of growth in the carbon
fibers [17]. In this scenario, after reduction of metallic nickel particles are
saturated with carbon materials and their aggregate state may alter. The
synopsis from this phase is to the alteration of nickel metallic particles sizes
because of their schism and amalgamation.

Fig. 2.
Representation of accumulated carbon as a function of reaction time on stream for (a) 15% Ni/Y
zeolite catalyst, and (b) 30% Ni/Y zeolite catalyst.

Figure options

Table 2.
Experimentally observed the behavior of AC with total methane conversion for two types of Ni/Y
zeolite used catalyst.

Run
no.

T(C)

Total
methane
conversion
(mol/gcat)

Behavior of average AC (gdep/gcat)


yield with respect to time

t = 10 min t = 26 min
With 15% Ni-supported Y zeolite catalysts
1
500
197.08
0.151
0.378
2
550
526.40
0.258
0.652
3
600
586.21
0.323
0.82
4
650
405.47
0.378
0.958
With 30% Ni-supported Y zeolite catalysts
5
500
482.56
0.145
0.37
6
550
1047.30
0.476
0.92
7
600
1242.54
0.484
1.22
8
650
821.67
0.61
1.56

TACa(gdep/gcat)

Mc(%) Mc(gdep/gNi)
After end of
complete reaction

t = 60 min

After end of
complete
reaction

0.80
1.41
1.78
2.08

193.91
1248.97
1083.35
372.39

428
1013
921
684

66.56
157.54
143.23
106.37

0.836
1.80
2.70
3.44

1121.96
4097.12
6651.35
1075.19

923
1567
2629
1534

215.65
366.12
614.25
358.41

The TAC was determined by summing on the total instantaneous solid carbon accumulated per
initial Ni-supported Y zeolite catalyst mass.
Table options

In Table 2, the total methane conversion was determined by summing on the


total methane conversion per initial Ni-supported Y zeolite catalyst mass.
Inspection of yield data at times of 1060 min, the TAC was enhanced for each
run except 650 C. The total methane conversion was also mimicked the same
trends. For example, the total methane conversion was increased from 197.08
to 586.21 mol/gcat for temperature increased from 500 to 600 C for 15% Ni/Y
zeolite catalyst and afterwards, it started to decreased. The 30% Ni/Y zeolite
catalyst was followed the same behavior. The weight of carbon deposit (Mc)
was increased as temperature increased except 650 C as shown inTable 2.
For better assessment of the catalysts, Mc was calculated (using Eq. (2)) in
terms of the quantity of surface nickel due to the carbon formation rate was
highly subjected to the quantity of Ni active sites. Nevertheless, a maximum
carbon yield of 614.25 gc/gNi was observed after end of complete reaction at

600 C with 30% Ni/Y zeolite catalyst. On the other hand, the highest M c value
for 15% Ni/Y zeolite was 157.54 gc/gNi; which is even lower than from the lower
Mc obtained with 30% Ni/Y zeolite catalyst at 500 C. Evidently, the TCD of
methane over Ni/Y zeolite gave maximum amount of carbon yield in terms of
the quantity of surface nickel at 550 and 600 C. With increase of temperatures,
the carbon yield reached a maximum point and then starts to decrease till the
complete deactivation of the catalyst. Hernadi et al. [18] have reported the
carbon production up to 1.6 gC/gcat for the cracking of an array of organic
reactants over Co and Fe supported silica and zeolite. Park and
Keane [19] have remarked 70 gc/gNi carbon yield after 60 min reaction from the
decomposition of ethylene at 600 C using Ni supported NaY zeolite. Our
carbon yields compare favorably with those generated from methane over
supported Ni-alumina (244 gc/gNi) and from over Ni/Cu/Al (585 gc/gNi) [20].
3.1.2. Effect of VHSV on TCD of methane
Fig. 3 shows the effect of volume hourly space velocity (VHSV) where
percentage of methane conversion is typified as a function of reaction time for
various VHSV, 8.5760 Lgcat1 h1, at 600 C for different values of catalyst
weight (17 g) along with constant methane flow rate of 1 L/min. It can be
apparent from Fig. 3(a) that the deactivation rate of decomposition reaction is
almost half of the Fig. 3(b) for VHSV range between 8.57 and 60 Lgcat1 h1. For
example, at VHSV of 20 Lgcat1 h1, the catalytic activity endures a more than
10 h before falling almost zero with 30% Ni/Y zeolite catalyst, whereas, it was
only more than 5 h for 15% Ni/Y zeolite catalyst.

Fig. 3.
Representation of methane conversion versus reaction time on the stream at different volume
hourly space velocity for (a) 15% Ni/Y zeolite catalyst, and (b) 30% Ni/Y zeolite catalyst.
Figure options

It can be ensured from Fig. 3(a) and (b) that the methane conversion increased
as VHSV and nickel concentration in NiY catalysts increased. The more
insightful observation is reintroduced that, with increasing VHSV, the time
needed for catalyst deactivation gradually increases [20] and [21]. This is
because a lower contact time results from a higher VHSV, the catalyst
deactivation increased as the VHSV increased. The initial hydrogen yield was
65.90% at a low spatial velocity of 8.57 Lgcat1 h1 while its representative stability
time was more than 5 h with 30% nickel concentration in Y zeolite catalyst. As
compared with this result, 15% Ni/Y zeolite catalyst gave initial hydrogen of
50.35%, having stability time more than 2 h at the same spatial velocity of
8.57 Lgcat1 h1. Experimental observation marks that the studied range of VHSV
still figured a high catalytic stability in originating hydrogen production. Using
Ni/Y zeolite catalyst and different VHSV at 600 C, the following Table 3 is
represented an inclusion of experimentally obtained results. From Fig. 3(a) and
(b), it can be visualized that the initial methane conversion was decreased
gradually with increasing VHSV, indicating a continuous diminishing in the
hydrogen yield over the course of the reaction. As a matter fact that the
methane conversion was noticeably increased from 23.10 to 50.40 vol% as with
decreased space velocity from 60 to 8.57 Lgcat1 h1 for 30% Ni/Y zeolite catalyst.
In case of 15% Ni/Y zeolite catalyst, these values increased from 16.65 to
33.92 vol% as shown in Table 3. This carried on to a lower constant hydrogen
yield (below 7%) whilst the complete deactivation of the catalyst.
Table 3.
Summarized experimental results with different VHSV at 600 C using Ni/Y zeolite catalyst.
Initial
TCFR (mmol/(gcat min))
Ru VHSV
H2yie
1
CH (
n
(Lgcat h ld
Ro(mmol/gcat
no. 1)
(%)
%)
min)
At
At
At
time = 10
time = 26
time = 60
min
min
min
With 15% Ni-supported Y zeolite catalysts
1
8.57
50.3
33.92
3.62
22.89
51.52
104.84
5
2
12
46.6
30.31
4.33
25.75
58.27
120.70
9
3
20
36.3
22.11
5.29
31.90
70.78
139.94
2
4
60
28.5
16.65
12.08
73.66
162.70
317.85
3
4

ACFR
(mmol/
(gcatmin))
After end of
complete
reaction
2.59
2.22
2.17
5.66

Ru
n
no.

VHSV
(Lgcat1 h
1
)

Initial
H2yie
ld
(%)

TCFR (mmol/(gcat min))

CH (
4

%)

At
time = 10
min

At
time = 26
min

At
time = 60
min

ACFR
(mmol/
(gcatmin))
After end of
complete
reaction

32.56

72.38

144.24

1.97

29.30

66.56

139.13

2.28

35.15

80.22

169.46

2.86

71.23

161.03

334.81

6.91

Ro(mmol/gcat
min)

With 30% Ni-supported Y zeolite catalysts


5
8.57
65.9
50.40
5.10
0
6
12
49.7
33.43
4.70
7
7
20
46.9
28.15
7.47
9
8
60
34.8
23.10
16.33
5

Table options

It can be clearly obvious from Table 3 that ACFR and initial methane conversion
move in reverse directions. Before catalysts completely deactivated, an
optimum ACFR of 2.86 and 2.17 mmol/(gcat min) can be obtained at space
velocity of 20 Lgcat1 h1 due to a high stability is more pronounced for this
experimental run for 30% and 15% Ni/Y zeolite catalysts, respectively.
Moreover, a higher VHSV, increases the TCFR in the reported results for metal
catalyst is not far away from our observation as reported in Table 3.
Fig. 4 reveals the methane decomposition rate (RD) in (mmol/(gcat min)) as a
function of reaction time for both type of catalyst. The decomposition rate can
be described as the ratio between product of molar flow rates of methane
(mmol/min) and methane fractional decomposition to weight of the Ni/Y zeolite
catalyst [5]. From Fig. 4, it may be inferred that a high initial RD followed by a
dramatic reduction. The initial methane reaction rate (Ro) in (mmol/(gcat min))
was derived from the curve fitting of Fig. 4 using a proper nth order polynomial
equations with regression coefficients greater than 0.99 and then equalizing the
time to zero in the right hand side of the curve fitting equation. From Table
3 and Fig. 5, it can be seen that Ro increases with VHSV and it has a reverse
and non-linear relevancy to the weight of Ni/Y zeolite catalysts. The higher
concentration of nickel in Ni/Y zeolite causes a higher reduction for Ro. On the
other hand, when higher weight of Ni/Y zeolite was used, the yield of hydrogen
was reduced because of the declines of contact time between the reactant
molecule and catalyst, resulting from the increased VHSV. At the same time,
the Ro enhanced due to decreasing the influence of methane pressure drop that
causes from the greater methane decomposition along with dilution by the

formed hydrogen [22]. However, in Fig. 5, the upper curve indicates that 30%
Ni/Y zeolite catalysts was produced maximum amount of hydrogen as well
carbon nanofiber.

Fig. 4.
Representation of methane decomposition rate as a function of time for (a) 15% Ni/Y zeolite
catalyst, and (b) 30% Ni/Y zeolite catalyst.
Figure options

Fig. 5.
Representation of initial methane decomposition versus (a) weight of Ni/Y zeolite, and (b) VHSV
for both 15% and 30% Ni/Y zeolite catalysts at constant methane flow rate of 1 L/min.
Figure options

3.2. Characterization of the catalysts


3.2.1. BET and TPD analysis
In Table 4, we report that the BET surface and micropore area of the support
materials were marginally reduced due to the proposition of nickel species onto
the Y zeolite. This may be interpreted the effect of higher pore blockage of Y
zeolite pore channel resulted from the existence of highly concentrated small
Ni2+ ions within the pore layout of the Y zeolite. Conversely, the catalyst surface
area was changed to a small extent because of the intensity of metal loadings.
This can be anticipated from the influence of partial blocking of Y zeolite pore

channel caused the existence of metallic oxides on the surface of support.


Fairly, the degree of interaction within the support and metal precursor is a
function of maximum reduction temperature [15].
Table 4.
Textural properties of the fresh and used catalyst.
Single
point
BET
Micropor
surfac surfac e
Sampl e area e area areab(m2/
Mesopore + exter
e
(m2/g)a (m2/g) g)
nal areac (m2/g)
Y
644.4
633.6
549.91
83.69
zeolite 0
1
15%
451.7
443.4
360.18
83.24
Ni/Y
6
2
zeolite
30%
335.5
329.6
258.97
70.72
Ni/Y
1
9
zeolite
With 15% Ni/Y zeolite catalysts
500
175.1
174.7
86.00
C
9
2
550
155.2
154.2
85.11
C
7
0
600
137.5
136.8
73.39
C
3
2
650
173.9
172.7
94.27
C
0
8
With 30% Ni/Y zeolite catalysts
500
158.3
157.2
91.90
C
8
5
550
100.3
100.2
44.39
C
0
3
600
96.75
97.15
35.16
C
650
104.3
103.6
51.69
C
1
9

Total pore
volumee(cm3/
g)
0.388

0.1756

0.272

0.1266

0.237

2.88

9.90

88.71

0.0415

0.274

6.28

17.16

69.09

0.0413

0.209

5.42

19.45

63.43

0.0356

0.201

5.87

21.92

78.50

0.0457

0.242

5.62

17.36

65.35

0.044

0.211

5.39

19.07

55.83

0.0213

0.152

6.06

29.93

61.99

0.0167

0.138

7.49

30.87

52.00

0.0250

0.221

8.55

28.93

Represents the values calculated at a relative pressure (P/Po) of N2 equal to 0.251.


bd

Represents the values calculated from t-plot method.


e

Averag
e
particle
size
(nm)
4.73

Micropo
re
volume
(cm3/g)d
0.2684

Pore
size
(nm)
2.45
2
2.45
6

6.76

Represents the total pore volume evaluated from nitrogen uptake at a relative pressure (P/Po) of
N2equal to 0.99.
Table options

It has reported that a fewer quantities of weak acid sites were enwrapped to the
raw Y zeolite [19]. The Na+ ions of the Y zeolite were substituted by Ni2+ ions
mimicked the hydrogen reduction to create Bronsted acid sites (protons)
through the preparation method, resulting in acidity to a greatest extent. It has
suggested that the reduction of nickel species in the impregnated catalyst was
brought to a fewer surface acidity, which in turns fit in reposing the smaller
particles [7]. Actually, the counter diffusion between the cations and protons of
the metal is providing higher numbers of metal ions into the pore mouths;
resulting in form a zerovalent external metal phase followed a reduction. This
type of counter diffusion is significantly accelerated when there is a high
protonic concentration was present in the Y zeolite matrix. The assemblage
process which is known to as responsible to form the particle growth may be
disrupted because of the presence of high volume of cation density in the
diffusion path of Ni content. Thereat, the carbon growths is far more prevailing if
the outer surface areas of Y zeolite are covered by the metal particles and
these sites are exclusively forcible if impregnation processes is employed to
introduce the metal precursor [7]. This phenomenon may be anticipated by
considering the Y zeolite structure. Structurally, zeolite Y has 8 supercages, 8
sodalite cages and 16 hexagonal prism cages per unit cell [23]. It has pore
mouth size of 0.74 nm, and the supercage size has 1.3 nm, suggesting that the
void volume of Y zeolite (0.85 nm3) is big enough to accommodate the
dispersed metallic nickel in the supercages. Nevertheless, the production of
carbon nanomaterials with a diameter below 1.3 is possible if Ni species lodged
in the Y zeolite supercages. Indeed, such type of production is not promising
energetically at all [7].
Table 4 shows that the surface areas subjected to the formation of CNFs are
higher than that of the surface area subjected to the formation of carbon
nanotube, but having a position between the amorphous carbon (surface
area = 670 m2/g) and the graphite (surface area = 7 m2/g) [24]. With the raise of
temperatures ramp from 500 to 600 C, the BET surface area decreased and
afterwards, initiated to increase. The total pore volume and micropore volume
has followed the same trends as BET surface area do until 600 C and then
onset to increase even though a smaller micropore volume was observed in all

the samples. As a consequence, at a temperature greater than 600 C, the


surface of carbon could be manifested mainly as mesoporous with an average
pore diameter of 6 and 7.50 nm for 15% and 30% Ni/Y zeolite catalysts
respectively.
To characterize the NH3 adsorbed on the various catalysts, the chromatograms
are typified in Fig. 6. The peak location differs with a great extent of different
metal loadings. It has been suggested that the characterization of the acid sites
through NH3-TPD method does not have a close similarity to the quantitative
measurement of the actual acidity strength due to the onset of diffusional
limitations [25]. However, this type of measurement gives the qualitative
indication of how NH3 molecules are strongly conjugated to the acid sites. At
temperature ranges of 172220, 340450 and greater than 450 C are
assigned for the peak maxima of the weak, medium, and strong acid sites,
respectively. It has been reported that at lower temperature (weak acid sites),
the quantity of adsorbed NH3 was always be higher than at higher temperature
(strong acid sites) [26]. Additionally, one should note that when the metallic
function is loaded into the Y zeolite, the acidity of Y zeolite decreases. This is
reasonable because, through a catalyst preparation method, the small fraction
of OH groups from the Y zeolite framework was eliminated by the metal salt
precursors. This phenomenon is especially serious about the case of Ni/Y
zeolite catalyst with the largest metal content, whose chromatograms has
shifted by 30% from the original support and this metal content can be taken
into consideration by means of few acid sites occupation by covered nickel
species. With enhancing Ni content into Y zeolite, a slow decline in surface
area was also visualized from Table 4. The increase of nickel species showed a
slight impact on the maximum temperature corresponding to the desorption
profile of NH3 as shown in Fig. 6. This would assuredly reflect that the Ni
species marginally influenced the strength of the acid sites.

Fig. 6.
Representation of NH3-TPD profile for selected sample catalyst.
Figure options

3.2.2. XRD analysis


The degree of structural order, longevity of catalyst and catalyst activity in fresh
and deactivated samples is usually related with the apparent size of the
crystallites determined by X-ray diffraction (XRD). The d-spacing and mean
crystallite size along c axis is useful indicators in evaluating of catalyst
performance during TCD of methane.Fig. 7 shows the XRD patterns for
calcined and reduced catalysts for 15% and 30% Ni-supported Y zeolite
catalyst. The diffraction peaks located at 2 = 10.098, 11.84, and 15.59
corresponds to the d-spacings of 8.752, 7.464 and 5.679 , respectively for
calcined fresh catalysts as shown in Fig. 7(A) and (B). The positions of the
diffraction peaks in the sample are in good agreement with those given in ICDS
NO: 74-1192 for Y zeolite phase. It may be observed from Fig. 7(C) and (D) that
the reduction of catalyst with 20% hydrogen in N 2 at 600 C was not sufficient to
reduce NiO phase from the fresh catalyst. It has suggested that the
decomposition of nickel nitrate from the preparation and calcination of catalysts
could form the NiO phase in the fresh catalysts. In addition, this NiO phase
interacting with supports on the surface of the catalyst, which creates a much
more complex catalysis system. As a result, the active site for the adsorption
and de-adsorption of gases may alter in the conjugated process, which likely
shifts to a different reaction course during the TCD of methane.

Fig. 7.

Powder XRD patterns of fresh catalyst after calcination and reduction: (A and C) with 15% Ni/Y
zeolite catalyst and (B and D) with 30% Ni/Y zeolite catalyst, respectively. Quadrilateral () and
square () represents Y zeolite phase and NiO phase, respectively.
Figure options

Li et al. [27] have demonstrated that at the low temperature (550 C), the
reduction of catalyst is far from complete. However, after reduction of catalyst,
the intensity of Y and NiO phase is slightly reduced in 30% Ni/Y zeolite (Fig.
7(D)) but significantly reduced in 15% Ni/Y zeolite (Fig. 7(C)). The diffraction
peaks at 2 = 37.44, 43.47, and 63.20 are corresponding to crystalline NiO
phases (ICDS NO: 01-1239) and their interlayer distance are pointed out as
2.40, 2.08, and 1.47 respectively. However, the mean diameter of NiO
crystallites in 30% Ni/Y zeolite was determined of 39.54 nm, while it was 41.69
for 15% Ni/Y zeolite attributing to the larger NiO particles lodged on the zeolite
surface in before reduction. On the other hand, after reduction, the mean
diameter of Ni crystallites was 34.84 nm for 30% Ni/Y zeolite, while it was
36.57 nm for 15% Ni/Y zeolite attributing to the larger Ni particles anchored on
the zeolite surface in after reduction.
At the temperature range of 500650 C, the X-ray diffraction profiles of
deactivated catalysts in presence of 30% and 15% Ni/Y zeolite catalysts are
shown in Fig. 8 and Fig. 9, respectively. As can be seen from Fig. 8 and Fig.
9 that, the broad diffraction peak at 2 = 26.34, 44.83, 54.23, and 77.54 are
characteristic to the graphitic carbon (ICDS NO: 01-0640) and their d values
are defined as 3.38, 2.02, 1.69 and 1.23 , respectively. The presence of
metallic Ni phase (ICDS NO: 04-0850) in the deactivated catalyst gave the
diffraction peak at 2 = 44.5, 51.84, and 76.37 and their corresponding d
values of 2.03, 1.76, and 1.24 , respectively.

Fig. 8.

Powder XRD patterns of used (a) 500 C, (b) 550 C, (c) 600 and (d) 650 C with 30% Ni/Y
zeolite catalysts. Circle (o) and square () represents graphitic carbon phase and metallic Ni
phase, respectively.
Figure options

Fig. 9.
Powder XRD patterns of used (a) 500 C, (b) 550 C, (c) 600 and (d) 650 C with 15% Ni/Y
zeolite catalysts. Circle (o) and square () represents graphitic carbon phase and metallic Ni
phase, respectively.
Figure options

From Fig. 8, it is apparent that, with increasing temperature from 500 to 650 C,
the graphitization intensity enhanced and altered to a higher 2 angle which
indicates a progressive graphitization of the CNFs. For example, the degree of
graphitization in deactivated catalysts increased from 23.25% to 81.40% as
temperature increased from 500 to 600 C and afterwards, slightly reduced
(69.76%) at 650 C for 30% Ni/Y zeolite. One should note that at 600 C, the
degree of graphitization was almost identical to that of natural graphite
(0.335 nm). Of beneficiary feedback from such graphitization is that it can
positively be good for use in composites applications [7]. Table 5 shows the
crystallite sizes of Ni metallic phases for both catalysts from XRD analysis.
Table 5.
Crystallite sizes of deactivated catalysts after TCD of methane from XRD analysis at different
temperatures.
Samples
Ni (1 1 1) (nm)
With 15% Ni-supported Y zeolite
500 C
67.11
550 C
20.97
600 C
83.91
650 C
18.64

Ni (2 0 0) (nm)

Ni (2 2 0) (nm)

57.54
28.77
19.18
28.77

14.10
16.46
16.46
14.10

Samples
Ni (1 1 1) (nm)
With 30% Ni-supported Y zeolite
500 C
83.91
550 C
33.55
600 C
55.92
650 C
27.96

Ni (2 0 0) (nm)

Ni (2 2 0) (nm)

34.53
49.32
28.77
57.54

14.10
49.37
32.92
32.92
Table options

Reported data in Table 5 reflects that at 500 and 550 C, the mean diameter of
metallic Ni crystallites after reaction were 44.18 and 44.08 nm, while it was
39.20 and 39.47 nm, respectively at 600 and 650 C for 30% Ni/Y zeolite.
Conversely, the mean diameter of metallic Ni crystallites in deactivated samples
were 46.25 nm, and 22.06 nm at 500 and 550 C, while it was 39.85 nm, and
20.50 nm at 600 and 650 C, respectively for 15% Ni/Y zeolite catalyst. It can
be observed that the most of the Ni particles of bigger sizes are located on the
zeolite surface especially for 30% Ni/Y zeolite catalyst. Very clearly, as can be
observed that at 600 C, the crystallite size of metallic Ni was almost same for
both used catalyst and a smaller crystallite size of Ni was also appeared
(20.50 nm) among the reported data. Nevertheless, it has suggested that the
beginning of carbon filament growth was utilized diameters of 550 nm of the
Ni metal particles.
3.2.3. FESEM analysis
Fig. 10(a) shows that the particle size of parent Y zeolites has appearance in
two forms, either in single crystallites or in the form of stacked-together small
crystallites form. These small crystallites have an irregular polyhedra structure.
After Ni-impregnation onto the Y zeolite channel, a significant number of
agglomerates, like the one in Fig. 10(b), presented in the zeolite samples, with
mean dimensions of nearly 4.5610.17 m. As can be seen, after nickel
dispersion, the catalyst particles are more uniform and distributed in
homogeneously (Fig. 10b), indicating that the used catalyst preparation
methods in the foregoing experiments are well enough to aggregate the Ni
atoms in the Y zeolite channels or cages. This is indicative that the ligands of
surface hydroxyl groups (SiOH) of nearby crystallites could be broken-down,
resulting in the generation of agglomeration. This agglomeration function is
dependent on the catalyst preparation and subsequent calcination processes.
We can speculate that this type of behavior are only possible when the aperture
diameter of supercage of Y zeolite has much enough larger than the diameter

of metallic nickel ion, which in turns, forms catalyst particles with the same
diameter as that of the cage in zeolite.

Fig. 10.
Representation of SEM morphology of fresh Y zeolite (a), and Ni-supported Y zeolite (b) catalyst,
respectively.
Figure options

Fig. 11 and Fig. 12 are typified with FESEM images along with different
magnification after the complete deactivation of methane at temperatures of
500, 550, 600 and 650 C, respectively using 15% and 30% Ni supported Y
zeolite. As can be seen from Fig. 11 and Fig. 12, that a whisker type of nano
filaments were generated onto the catalyst surface, in regardless of the metal
loadings. Additionally, the lengths and diameter of the formed nano fiber are, of
course, different. The particle size of Ni and the support has significant effect on
the generation of carbon nano fiber (CNF) and the nature of type carbon
produced [8]. FESEM images (Fig. 11) show that the produced CNF had
uniform diameters of about 52.85, 28.41, 45.22, and 30.50 nm, respectively at
temperatures of 500650 C. Moreover, they have no impurities on the
surfaces and curved together. On the other hand, Fig. 12 shows the CNF
formation with a higher metal loading, which yields also a huge quantity of
CNFs, suggesting that a good co-relationship between selectivity and
productivity.

Fig. 11.
Representative FESEM image of TCD methane over 15% Ni/Y zeolite at (a): 500, (b): 550, (c):
600, and (d): 650 C, respectively.
Figure options

Fig. 12.
Representative FESEM image of TCD methane over 30% Ni/Y zeolite at (a): 500, (b): 550, (c):
600, and (d): 650 C, respectively.
Figure options

Generally, CNFs has also uniform diameters of about 51.80, 48.45, 43.71, and
41.84 nm, respectively for all studied temperatures. This result is in line with our
XRD analysis of the spent catalyst that the TCD of methane over metallic nickel
particles has tendency to produce graphitic CNF. Nevertheless, the higher
magnification FESEM images (Fig. 12) reflects that the quality and morphology
of the CNFs produced at higher catalyst density are almost identical to those
produced at the lower one, indicating that without a significant defects in the
quality of CNFs, the productivity shows an incremental pattern. It has reported
that the methane decomposition over Ni supported oxidized diamond produced
a whisker type of CNF, suggesting that the catalyst could be stable for a long
time over the course of reaction stream, resulting in a higher catalytic
activity [28]. On the other hand, a filamentous type of CNF formation is also
possible from the methane decomposition using Ni/SiO 2 and Ni/HY
catalysts [12]. Apart from higher catalytic activities, the accumulated carbon on

the catalysts active surface causes a significant reduction in catalytic activities.


The FESEM images give information that the formation of CNF can be of a
bottom or tip mechanism according to the linkage between catalysts and
supports. The plethora of bright spots in the tips of almost every CNFs are
pertaining to the Ni surface, suggesting a tip mechanism for CNF formation.
The diameters of CNFs are about the same as that of Ni particles which is in
well consistent with the reported results [7] and [29]. Nevertheless, a broad
range of variation in structural and textural features of CNF could observe.
3.2.4. TEM analysis
Representative TEM images at different levels of magnification of the carbon
nanomaterials grown from the 15% nickel loaded impregnated catalyst are
given in Fig. 13. However, we can notice that a mixture of carbon nanofibers
and carbon nanotubes with thick wall in Fig. 13(a) and (d) were formed on the
surface of Ni/Y zeolite catalyst. In this experiment, we used the definitions only
for the filamentous carbons conferred by the carbon nanofibers (CNFs) for the
sake of simplicity.

Fig. 13.
Representative TEM image of TCD methane over 15% Ni/Y zeolite at 550 C of (a) and (b) and at
600 C of (c), and (d), respectively.
Figure options

However, it is apparent from TEM computation that the predominant CNFs


diameters formed by impregnation method were closer to the Ni crystallite size
in the deactivated catalyst. The average outer diameter of CNFs from 15% Ni/Y
zeolite was 30.76 nm (Fig. 13(a) and (b)) at 550 C, while it was 43.58 nm (Fig.
13(c) and (d)) at 600 C. On the other hand, the average outer diameter of
CNFs at 550 C was 50.45 nm, while it was 46.2 nm at 600 C with higher
loaded Ni/Y zeolite (Fig. is not shown here due to space limitation). This result
is in well consistent with our XRD results that showed that the Ni crystallite size
at 550 and 600 C were 44.08 nm and 39.20 nm, respectively with 30% Ni/Y
zeolite catalyst, while it was 22.06 and 39.85 nm at 550 and 600 C,
respectively from 15% Ni/Y zeolite catalyst. It can be observed that with
increasing nickel loading in Y zeolite channel, the diameters of formed CNFs

also increased. Interestingly, one should note that with temperature increases,
the diameter of CNFs increases for 15% Ni/Y zeolite from 30.76 to 43.58 nm,
but this tendency is completely reverse in 30% Ni/Y zeolite catalyst. The
diameter of the carbon nanomaterials at higher temperatures was smaller, and
after decomposition of methane, the predominant size of nickel particles on the
surface of zeolite Y-supported Ni catalysts was lower. These results are
attributed to the melting temperature of the nickel nanoparticles, which was
linearly reduced and inversely enhanced the diameter. This phenomenon can
be explained by the Lindemann effect [13]: the melting point of a solid material
may be reduced when a nanometer particle size range could be reduced from
the original particle size. Nickel has a melting point of about 1455 C. Upon
heating the nickel nanoparticles at a temperature of 500 C, the softened nickel
nanoparticles may be formed because of the applied temperature, which is
almost one-third of the melting temperature of nickel [13]. This response
indicates that at the highest temperature, the average diameter of carbon
nanomaterials could be lowered. A liquefied nickel form may exist at higher
temperatures and thus cause the deformation of nickel particles. In addition,
melted nickel interacts with carbon. A fraction of liquefied nickel was passed
within the carbon nanomaterials, and the fraction of liquefied nickel particle was
restricted to the external side. Moreover, the reduction temperature does not
affect the diameter of the nickel particle. In the course of methane
decomposition, a miscellaneous trend was observed. Methane decomposition
caused the carbon to accumulate on the surface of zeolite Y-supported nickel
catalysts and to interact with the liquefied nickel. Nickel particle deformation
was noticed as a result of carbonnickel interaction [13]. Nevertheless, Fig.
13(a) and (c) clearly shows that a highly homogeneous crystalline structure with
the graphitic sheets heaped almost collateral to the fiber axis and a hollow
channel obviously observe along the axis of filament, appearing like bamboo at
550600 C [30]. The interaction within the metal-support and metal reactants
are controlled the surface propensity of metal, which is directly related with the
stacking of the graphite layers [31].
It is worthwhile to mention that the tip growths of carbon nano-filament were
observed due to most of the active Ni particles were anchored on top of carbon
nano-filaments. This may be explained from the metal oxides dispersion in the
zeolite channel. On the zeolite Y, the NiO particles have propensity to be more
globular and so, there is a probability to show a weak interaction with the

support due to NiO particles on Y zeolite surface reflects a large contact


angle [32]. As a result, they could simply be withdrawn from it forming the tip
grown carbon nano-filaments in the carbon decomposition. Most of the carbon
nano-filaments grown on Ni/Y zeolite exhibited that a waved structure of the
graphitic sheets were formed on a short range and so, most of them carried the
defects in their structures. Conversely, a continuous growth of carbon filaments
with reaction time was resulting from the higher carbon atoms liberation in the
metal-support interface, leads to Ni particle on the tip of the carbon
filaments [33]. However, this phenomenon is subjected to the preparation
methods of catalyst, types of metal, nature of support material, distribution of
metal particle on support matrix and the electrostatic attraction between
them [34].

4. Conclusion
The thermo-catalytic decomposition (TCD) of methane on zeolite Y supported
Ni catalysts were investigated for hydrogen and carbon nanofibers (CNFs)
production at 500, 550, 600, and 650 C using a pilot scale fixed bed reactor.
Experimental investigation had described that the highest initial H 2 and
CH4 have to be found 41.02% and 25.76%, for 15% Ni/Y zeolite catalyst
whereas the value of 58.58% and 41.67% was noted for 30% Ni/Y zeolite
catalyst, respectively at 650 C. The average hydrogen formation rate
increased significantly with increasing temperature but reduced the catalyst
lifetime and the average carbon formation rate also increased as the
temperature rose from 500 to 650 C. Nevertheless, the TCD of methane in
presence of Ni/Y zeolite catalyst required an optimum temperature and in this
case it is ranged from 550 to 600 C. A maximum carbon yield of 614.25 and
157.54 gc/gNi was obtained after end of complete reaction at 600 C with 30 and
15% Ni/Y zeolite catalyst, respectively. The initial methane decomposition rate
increases with VHSV and it has a reverse and non-linear relevancy to the
weight of Ni/Y zeolite catalyst. BET, TPD, and XRD results expressed that how
metallic nickel species were lodged onto the exterior surface of Y zeolite to the
greatest extent using the impregnation system and how it influenced the
catalyst activity of the support materials. The FESEM images gave information
that the produced CNFs are in the form of whisker types, suggesting the
longevity of the catalyst, in regardless of the metal loadings. TEM analysis has
confirmed that the produced CNFs had a fishbone type structures with a hollow

core and the CNFs has identical or very similar as pear-shape of Ni particle,
suggesting the tip growths mechanism of CNFs formation.

Acknowledgment
The authors acknowledge the financial support from the postgraduate
Research Fund (UM.C/HIR/MOHE/ENG/11) University of Malaya, Malaysia.

References
1.
o

[1]

N.Z. Muradov

CO2-free production of hydrogen by catalytic pyrolysis of hydrocarbon fuel

Energy Fuels, 12 (1998), pp. 4148

View Record in Scopus


|
Full Text via CrossRef
|
Citing articles (164)

2.
o

[2]

A.E. Awadallah, A.A. Aboul-Enein, A.K. Aboul-Gheit

Effect of progressive Co loading on commercial CoMo/Al 2O3 catalyst for natural


gas decomposition to COx-free hydrogen production and carbon nanotubes

Energy Convers Manage, 77 (2014), pp. 143151

Article
|
PDF (2403 K)
|
View Record in Scopus
|
Citing articles (16)

3.
o

[3]

Md. Nasir Uddin, W. Wan Daud, H.F. Abbas

Kinetics and deactivation mechanisms of the thermal decomposition of methane in


hydrogen and carbon nanofiber co-production over Ni-supported Y zeolite-based
catalysts

Energy Convers Manage, 87 (2014), pp. 796809

Article
|
PDF (2003 K)
|
View Record in Scopus
|
Citing articles (3)

4.
o

[4]

F. Gutirrez, F. Mndez

Entropy generation minimization for the thermal decomposition of methane gas in


hydrogen using genetic algorithms

Energy Convers Manage, 55 (2012), pp. 113

Article
|
PDF (650 K)
|
View Record in Scopus
|
Citing articles (10)

5.
o

[5]

H.F. Abbas, W. Daud

Hydrogen production by thermocatalytic decomposition of methane using a fixed


bed activated carbon in a pilot scale unit: apparent kinetic, deactivation and
diffusional limitation studies

Int J Hydrogen Energy, 35 (2010), pp. 1226812276

Article
|
PDF (638 K)

|
View Record in Scopus
|
Citing articles (18)
6.
o

[6]

K. Svoboda, A. Siewiorek, D. Baxter, J. Rogut, M. Pohoel

Thermodynamic possibilities and constraints for pure hydrogen production by a


nickel and cobalt-based chemical looping process at lower temperatures

Energy Convers Manage, 49 (2008), pp. 221231

Article
|
PDF (227 K)
|
View Record in Scopus
|
Citing articles (38)

7.
o

[7]

A. de Lucas, A. Garrido, P. Snchez, A. Romero, J.L. Valverde

Growth of carbon nanofibers from Ni/Y zeolite based catalysts: effects of Ni


introduction method, reaction temperature, and reaction gas composition

Ind. Eng Chem Res, 44 (2005), pp. 82258236

View Record in Scopus


|
Full Text via CrossRef
|
Citing articles (31)

8.
o

[8]

J. Ashok, S.N. Kumar, M. Subrahmanyam, A. Venugopal

Production of COx free hydrogen by catalytic decomposition of methane over Ni/HY


catalysts

Catal Lett, 118 (2007), pp. 139145

View Record in Scopus


|
Full Text via CrossRef
|
Citing articles (16)

9.
o

[9]

M. Inaba, K. Murata, M. Saito, I. Takahara, N. Mimura

Hydrogen production by conversion of methane over nickel-supported USY-type


zeolite catalysts

React Kinetics Catal Lett, 77 (2002), pp. 109115

View Record in Scopus


|
Full Text via CrossRef
|
Citing articles (16)

10.
o

[10]

Z. Bai, H. Chen, B. Li, W. Li

Catalytic decomposition of methane over activated carbon

J Anal Appl Pyrolysis, 73 (2005), pp. 335341

Article
|
PDF (276 K)
|
View Record in Scopus
|
Citing articles (46)

11.
o

[11]

B.M. Manohara, H. Nagabhushana, D.V. Sunitha, K. Thyagarajan, B. Daruka Prasad,


S.C. Sharma

Synthesis and luminescent properties of Tb3+ activated cadmium silicate


nanophosphor

J Alloys Compd, 592 (2014), pp. 319327

Article
|
PDF (2220 K)
|
View Record in Scopus
|
Citing articles (5)

12.
o

[12]

T. Choudhary, C. Sivadinarayana, C.C. Chusuei, A. Klinghoffer, D. Goodman

Hydrogen production via catalytic decomposition of methane

J Catal, 199 (2001), pp. 918

Article
|
PDF (322 K)
|
View Record in Scopus
|
Citing articles (172)

13.
o

[13]

B. Michalkiewicz, J. Majewska

Diameter-controlled carbon nanotubes and hydrogen production

Int J Hydrogen Energy, 39 (2014), pp. 46914697

Article
|
PDF (1103 K)
|
View Record in Scopus
|
Citing articles (2)

14.
o

[14]

A. Shaikjee, N.J. Coville

The role of the hydrocarbon source on the growth of carbon materials

Carbon, 50 (2012), pp. 33763398

Article
|
PDF (3190 K)
|
View Record in Scopus
|
Citing articles (22)

15.
o

[15]

D. Chen, K.O. Christensen, E. Ochoa-Fernndez, Z. Yu, B. Ttdal, N. Latorre

Synthesis of carbon nanofibers: effects of Ni crystal size during methane


decomposition

J Catal, 229 (2005), pp. 8296

Article
|
PDF (920 K)
|
View Record in Scopus
|
Citing articles (190)

16.
o

[16]

P.T. Reilly, W.B. Whitten

The role of free radical condensates in the production of carbon nanotubes during
the hydrocarbon CVD process

Carbon, 44 (2006), pp. 16531660

Article
|
PDF (380 K)
|
View Record in Scopus

|
Citing articles (45)
17.
o

[17]

S.G. Zavarukhin, G.G. Kuvshinov

The kinetic model of formation of nanofibrous carbon from CH 4H2 mixture over a
high-loaded nickel catalyst with consideration for the catalyst deactivation

Appl Catal A: Gen, 272 (2004), pp. 219227

Article
|
PDF (190 K)
|
View Record in Scopus
|
Citing articles (38)

18.
o

[18]

K. Hernadi, A. Fonseca, J.B. Nagy, A. Siska, I. Kiricsi

Production of nanotubes by the catalytic decomposition of different carboncontaining compounds

Appl Catal A: Gen, 199 (2000), pp. 245255

Article
|
PDF (2358 K)
|
View Record in Scopus
|
Citing articles (153)

19.
o

[19]

C. Park, M.A. Keane

Controlled growth of highly ordered carbon nanofibers from Y zeolite supported


nickel catalysts

Langmuir, 17 (2001), pp. 83868396

View Record in Scopus


|
Full Text via CrossRef
|
Citing articles (41)

20.
o

[20]

Y. Li, J. Chen, Y. Qin, L. Chang

Simultaneous production of hydrogen and nanocarbon from decomposition of


methane on a nickel-based catalyst

Energy Fuels, 14 (2000), pp. 11881194

View Record in Scopus


|
Full Text via CrossRef
|
Citing articles (135)

1.
o

[21]

A. Cunha, J. Orfao, J. Figueiredo

Catalytic decomposition of methane on Raney-type catalysts

Appl Catal A: Gen, 348 (2008), pp. 103112

Article
|
PDF (1896 K)
|
View Record in Scopus
|
Citing articles (55)

2.
o

[22]

H.F. Abbas, W. Daud

Influence of reactor material and activated carbon on the thermocatalytic


decomposition of methane for hydrogen production

Appl Catal A: Gen, 388 (2010), pp. 232239

Article
|
PDF (849 K)
|
View Record in Scopus
|
Citing articles (8)

3.
o

[23]

P. Payra, P.K. Dutta

Zeolites: a primer

S.M. Auerbach, K.A. Carrado, P.K. Dutta (Eds.), Handbook of zeolite science and
technology, Dekker, New York (2003), pp. 119

View Record in Scopus


|
Citing articles (1)

4.
o

[24]

L.D. Cherukuri, G. Yuan, M.A. Keane

Catalytic growth of structured carbon via the decomposition of chlorobenzene


over Ni/SiO2

Topics Catal, 29 (2004), pp. 119128

Full Text via CrossRef

[25]

A. Romero, A. Garrido, A. Nieto-Mrquez, P. Snchez, A.D. Lucas, J.L. Valverde

Synthesis and structural characteristics of highly graphitized carbon nanofibers

5.

produced from the catalytic decomposition of ethylene: influence of the active


metal (Co, Ni, Fe) and the zeolite type support
o

Micropor Mesopor Mater, 110 (2008), pp. 318329

Article
|
PDF (1437 K)
|

View Record in Scopus


|
Citing articles (20)
6.
o

[26]

E. Selli, L. Forni

Comparison between the surface acidity of solid catalysts determined by TPD and
FTIR analysis of pre-adsorbed pyridine

Micropor Mesopor Mater, 31 (1999), pp. 129140

Article
|
PDF (268 K)
|
View Record in Scopus
|
Citing articles (59)

7.
o

[27]

Y. Li, J. Chen, L. Chang, Y. Qin

The doping effect of copper on the catalytic growth of carbon fibers from methane
over a Ni/Al2O3 catalyst prepared from feitknecht compound precursor

J Catal, 178 (1998), pp. 7683

Article
|
PDF (344 K)
|
View Record in Scopus
|
Citing articles (87)

8.
o

[28]

K. Nakagawa, M. Nishitani-Gamo, T. Ando

Hydrogen production from methane for fuel cell using oxidized diamond-supported
catalysts

Int J Hydrogen Energy, 30 (2005), pp. 201207

Article
|
PDF (475 K)
|
View Record in Scopus
|
Citing articles (19)

9.
o

[29]

V.R. Choudhary, S. Banerjee, A.M. Rajput

Continuous production of h2 at low temperature from methane decomposition over


Ni-containing catalyst followed by gasification by steam of the carbon on the
catalyst in two parallel reactors operated in cyclic manner

J Catal, 198 (2001), pp. 136141

Article
|
PDF (82 K)
|
View Record in Scopus
|
Citing articles (58)

10.
o

[30]

T. Reshetenko, L. Avdeeva, Z. Ismagilov, V. Pushkarev, S. Cherepanova, A. Chuvilin, et


al.

Catalytic filamentous carbon: structural and textural properties

Carbon, 41 (2003), pp. 16051615

Article
|
PDF (1259 K)
|
View Record in Scopus
|

Citing articles (115)


11.
o

[31]

Z. Yu, D. Chen, B. Ttdal, A. Holmen

Effect of support and reactant on the yield and structure of carbon growth by
chemical vapor deposition

J Phys Chem B, 109 (2005), pp. 60966102

View Record in Scopus


|
Full Text via CrossRef
|
Citing articles (46)

12.
o

[32]

S.-Y. Lee, M. Yamada, M. Miyake

Synthesis of carbon nanotubes over gold nanoparticle supported catalysts

Carbon, 43 (2005), pp. 26542663

Article
|
PDF (764 K)
|
View Record in Scopus
|
Citing articles (69)

13.
o

[33]

R. Baker, M. Barber, P. Harris, F. Feates, R. Waite

Nucleation and growth of carbon deposits from the nickel catalyzed decomposition
of acetylene

J Catal, 26 (1972), pp. 5162

Article
|
PDF (1470 K)
|

View Record in Scopus


|
Citing articles (803)
14.
o

[34]

H. Wang, R.T.K. Baker

Decomposition of methane over a NiCuMgO catalyst to produce hydrogen and


carbon nanofibers

J Phys Chem B, 108 (2004), pp. 2027320277

View Record in Scopus


|
Full Text via CrossRef
|
Citing articles (60)

Corresponding author. Tel.: +60 1111200713; fax: +60 379675319.


Copyright 2014 Elsevier Ltd. All rights reserved.