Effect of preparation methods on structure and catalytic activity of Ni loaded CexZr1?xO2 catalysts for hydrogen production via autothermal reforming of ethane

Article abstract of DOI:10.1007/s11164-016-2796-0

Abstract: Ni/Ce_0.75Zr_0.25O₂ catalysts were prepared using a variety of methods with water medium (Ni/CZ_W), urea-water medium (Ni/CZ_UW) and urea-water–alcohol medium (Ni/CZ_UWA). All of the catalysts were characterized by XRD, BET surface area, SEM, TEM, TPR and TGA analysis, respectively. The effect of the support composition as well as the metal loading on the autothermal reforming (ATR) of ethane was evaluated in a fixed-bed flow reactor at temperatures ranging from 300 to 700?°C. In the activity test, the Ni/CZ_UWA catalyst exhibited enhanced activity, which may be attributable to low levels of carbon deposition. The principal drawback—the decrease in the catalyst activity as the result of carbon deposition—was overcome by optimizing the catalyst preparation technique. From SEM analysis, Ni/CZ_UWA depicted nano-fiber morphology, whereas the other Ni/CZ_W and Ni/CZ_UW catalysts evidenced carbon clusters on its surface, which demonstrates clearly that the former is better than the latter for ethane reforming. The TPR results show that the Ni particles were more thoroughly dispersed in Ni/CZ_UWA catalyst than in the other catalysts. Depending on the composition and metal loading, a solid solution and/or a highly dispersed nickel oxide in ceria (or ceria–zirconia) can be obtained. The TGA data also supported the SEM observations, which illustrated approximately 5% less carbon deposition for the Ni/CZ_UWA catalyst. Graphical Abstract: [Figure not available: see fulltext.]

Full text of DOI:10.1007/s11164-016-2796-0

Res Chem Intermed
DOI 10.1007/s11164-016-2796-0
Effect of preparation methods on structure
and catalytic activity of Ni loaded Ce Zr12xO catalysts
x
2
for hydrogen production via autothermal reforming
of ethane
1
2
•
Dhanapalan Karthikeyan
Palaniappan Selvakumar
Mani Shanmugam
•
Raji Atchudan
Raji Sivakumar
1
1
•
•
3
Received: 7 July 2016 / Accepted: 19 October 2016
Ó Springer Science+Business Media Dordrecht 2016
Abstract Ni/Ce0.75Zr0.25O catalysts were prepared using a variety of methods with
2
water medium (Ni/CZ ), urea-water medium (Ni/CZUW) and urea-water–alcohol
W
medium (Ni/CZUWA). All of the catalysts were characterized by XRD, BET surface area,
SEM, TEM, TPR and TGA analysis, respectively. The effect of the support composition
as well as the metal loading on the autothermal reforming (ATR) of ethane was eval-
uated in a fixed-bed flow reactor at temperatures ranging from 300 to 700 °C. In the
activity test, the Ni/CZUWA catalyst exhibited enhanced activity, which may be
attributable to low levels of carbon deposition. The principal drawback—the decrease in
the catalyst activity as the result of carbon deposition—was overcome by optimizing the
catalyst preparation technique. From SEM analysis, Ni/CZUWA depicted nano-fiber
morphology, whereas the other Ni/CZW and Ni/CZUW catalysts evidenced carbon
clusters on its surface, which demonstrates clearly that the former is better than the latter
for ethane reforming. The TPR results show that the Ni particles were more thoroughly
dispersed in Ni/CZUWA catalyst than in the other catalysts. Depending on the compo-
sition and metal loading, a solid solution and/or a highly dispersed nickel oxide in ceria
(or ceria–zirconia) can be obtained. The TGA data also supported the SEM observations,
which illustrated approximately 5% less carbon deposition for the Ni/CZUWA catalyst.
&
&
1
Dhanapalan Karthikeyan
dkarthikeyan05@yahoo.co.in
Raji Atchudan
atchudanr@yu.ac.kr
Department of Chemistry, Arignar Anna College of Arts and Science, Krishnagiri,
Tamilnadu 635001, India
2
3
School of Chemical Engineering, Yeungnam University, Gyeongsan 38541, Republic of Korea
Department of Chemistry, Annamalai University, Annamalai Nagar, Chidambaram,
Tamilnadu 608002, India
123

D. Karthikeyan et al.
Graphical Abstract
Keywords Autothermal reforming Á Ethane Á Nickel nanoparticle Á Ceria–Zirconia Á
Hydrogen production
Introduction
Fossil fuels are non-renewable hydrocarbon energy sources, and the two best known
fossil fuels are coal and petroleum. Although many hydrogen-based power sources
are being sought, but remain hypothetical, massive quantities of hydrogen are
currently being generated from fossil materials such as natural gas, heavy oil
(
CH , C H , and C H . In order to utilize this hydrogen as a power source it must be
diesel), and coal. This generated hydrogen exists in compounds including H O,
2
4
2
6
3 8
liberated from these compounds, and a great many techniques have been developed
to do this. Reforming is the one of the most efficient known methods for isolating
hydrogen from its compound-bound form. Three major thermo-chemical reforming
techniques thus far developed for use in the production of hydrogen from methane
and other hydrocarbon fuels [1, 2] are steam reforming (SR), partial oxidation
reforming (POR), and autothermal reforming (ATR). Among these techniques,
steam reforming (SR) is the lowest-volume and lowest-pressure hydrogen produc-
tion technique, whereas partial oxidation reforming is an exothermic process, and
requires smaller reactor vessels [3, 4]. Autothermal reforming, therefore, has
attracted considerable interest in recent years due to its higher energy conversion
efficiency.
Autothermal reforming [5–10] combines the thermal effects of partial oxidation
reforming and steam reforming reactions by feeding the fuel, water, and air together
into a reactor. Autothermal reforming is an effective and energy-efficient method of
hydrogen or synthesis gas production. The thermal energy generated from partial
oxidation reforming is absorbed by steam reforming, and therefore, the overall
temperature is lower. The autothermal reactor is, thus, a more compact and practical
process than other processes, and this makes it potentially useful in mobile fuel
cells.
1
23

Effect of preparation methods on structure and catalytic…
As ethane is one of the natural gas groups (methane, ethane, propane and butane),
there was less research done on hydrogen production from ethane via steam
reforming, partial oxidation or autothermal reforming that preformed ethane
individually. Some researchers have made a comparison between methane and
ethane, but more investigation has been about methane. Less research has been
about ethane (propane and butane as well) for hydrogen production via steam
reforming, partial oxidation and autothermal reforming, because researchers believe
that the same results will occur as well as the research that has been done on
methane using the three major techniques [11].
Some researches have been conducted on hydrogen production via catalytic
decomposition/cracking of ethane. Chin et al. [11] who has studied hydrogen
production via the catalytic cracking of ethane found that SiO -supported Ni
2
catalysts were the suitable catalytic decomposition of ethane. At temperatures above
5
00 °C and space velocity at 18,000 ml/g.h, almost complete initial conversion can
be achieved. In addition to the hydrogen and filamentous carbon, methane is also
formed as a by-product. The results indicate that although ethane decomposes faster
than methane, the problem becomes more complex in this case, due to the formation
of methane as a by-product. Consequently, a two-stage process may be necessary
for the production of hydrogen in this case.
Wang et al. [12] presented an experiment on the production of hydrogen from
ethane on a novel highly active catalyst system with a Rd-based membrane reactor.
The objective of the study is to make the dehydrogenation of ethane feasible at
relatively lower temperatures by using a novel highly active catalyst system and to
produce pure hydrogen and other more valuable hydrocarbons such as ethylene,
propylene and BTX (aromatics) by integration of the dehydrogenation of ethane and
hydrogen separation with a Pd-based membrane.
The coking problem is particularly relevant to nickel-based catalysts. Other
metals, such as Pt, Pd, Rh, and Ru, have been tested and shown to be more
resistant to carbon deposition than Ni [13–15]. However, these metals are all
quite expensive. On the other hand, nickel is fairly cheap and is an excellent
catalyst for the production of synthesis gas–unfortunately, it is also an excellent
catalyst for carbon formation. A considerable amount of research has been
aimed at improving the stability and coking resistance of Ni-based catalysts
[
16–19]. Ceria and metal oxide-doped ceria evidence high oxygen storage
capacity, redox, and thermal resistances, all of which are beneficial in oxidation
processes [1, 2, 5]. Ceria–zirconia mixed oxide is a catalytic promoter that is
very popular because of its marked oxygen storage capacity. The partial
4
?
4?
substitution of Ce with Zr in the CeO lattice results in the formation of a
2
solid solution [6–8]. Compared with conventional catalysts, the ceria–zirconia
support provides higher resistance against carbon formation. This study deals
with the ATR of ethane over a nickel catalyst immobilized on a ceria–zirconia
support under thermally neutral conditions, using different feedstock
compositions.
123

D. Karthikeyan et al.
Experimental
Catalyst preparation
Synthesis of Ni0.33Ce0.54Zr0.13O and Ni0.54Ce0.33Zr0.13O2
2
Ni Ce Zr (Ni0.33Ce0.54Zr0.13O and Ni0.54Ce0.33Zr0.13O ) catalysts were prepared
x
y
z
2
2
via co-precipitation. A predetermined amount of nickel nitrate (Ni (NO ) Á6H O,
3
2
2
E-Merck), cerium nitrate (Ce (NO ) Á6H O, Sisco Research Laboratory) and
3
3
2
zirconium oxychloride (ZrOCl .8H O, Central Drug House) were used to prepare an
2
2
aqueous solution ([Ce] ? [Zr] = 0.1 M, [Ce]/[Zr] = 3). An aqueous ammonia (20
wt%) was added dropwise to the above solution under stirring until the pH value of
the mixture adjusted to a pH of 9–10. Afterward, the mixture was stirred (100 rpm)
for 24 h at 60 °C. The precipitate prepared was washed with deionized water and
-
-
filtered until no Cl and NO
ions were detected in the filtrate. The sample was
3
then dried overnight at 110 °C for 1 h and then calcined at 800 °C for 6 h in an air
flow. As a reference, CeO was also prepared by the method according to the above
procedure.
2
Synthesis of Ni/CZW
Ce0.75Zr0.25O was prepared via co-precipitation. An aqueous solution of cerium
2
nitrate and zirconium nitrate ([Ce] ? [Zr] = 0.1 M, [Ce]/[Zr] = 3) was stirred for
1
h at 80 °C and then pH-adjusted to 9–10 with an aqueous ammonia (20 wt%)
solution and stirred for 24 h at 80 °C. The solution containing the precipitates was
then aged for 12 h at room temperature. The catalyst was dried overnight at 110 °C,
and calcined for 4 h in air at 800 °C. The 15 wt% Ni was loaded over Ce0.75Zr0.25
O
2
via the impregnation method. The catalyst was dried overnight at 110 °C, and then
calcined for 4 h in air at 500 °C. The catalyst obtained was designated as Ni/
Ce Zr O -water (Ni/CZ ).
0
.75 0.25
2
W
Synthesis of Ni/CZUW
Ce0.75Zr0.25O was prepared via co-precipitation. An aqueous solution of 0.4 M urea
2
was added to an aqueous solution containing cerium nitrate and zirconium nitrate
(
[Ce] ? [Zr] = 0.1 M, [Ce]/[Zr] = 3) was stirred for 1 h at 80 °C and then pH-
adjusted to 9–10 with an aqueous ammonia (20 wt%) solution and stirred for 24 h at
0 °C. The solution containing the precipitates was then aged for 12 h at room
8
temperature. The catalyst was dried overnight at 110 °C, and calcined for 4 h in air
at 800 °C. The 15 wt% Ni was loaded over Ce0.75Zr O via the impregnation
0.25
2
method. The catalyst was dried overnight at 110 °C, and then calcined for 4 h in air
at 500 °C. The catalyst obtained was designated as Ni/Ce Zr O -urea water
0
.75 0.25 2
(
Ni/CZ ).
UW
1
23

Effect of preparation methods on structure and catalytic…
Synthesis of Ni/CZUWA
Ce0.75Zr0.25O was prepared via co-precipitation. An aqueous solution of 0.4 M urea
2
and ethanol was added to an aqueous solution containing of cerium nitrate and
zirconium nitrate ([Ce] ? [Zr] = 0.1 M, [Ce]/[Zr] = 3) was stirred for 1 h at 80 °C
and then pH-adjusted to 9–10 with an aqueous ammonia (20 wt%) solution and
stirred for 24 h at 80 °C. The solution containing the precipitates was then aged for
1
2 h at room temperature. The catalyst was dried overnight at 110 °C, and calcined
for 4 h in air at 800 °C. The 15 wt% Ni was loaded over Ce0.75Zr O via the
0.25
2
impregnation method. The catalyst was dried overnight at 110 °C, and then calcined
for 4 h in air at 500 °C. The catalyst obtained was designated as Ni/Ce Zr O -
0
.75 0.25 2
urea water alcohol (Ni/CZ
).
UWA
Characterization of the catalyst
X-ray diffraction measurements (XRD)
Powder X-ray diffraction patterns were recorded using an X’Pert X-ray diffrac-
tometer with Cu-Ka radiation (k = 0.154 nm). The diffraction data were recorded
in the scan range of 2h between from 10 to 90° with a 0.02° step size and 1 s step
time. The crystallite size of the samples was measured via X-ray broadening, using
the well-known Debye–Scherrer equation.
Surface area
Specific surface area and pore structure were evaluated with a Micromeritics
ASAP2010 apparatus using N adsorption and desorption at -196 °C (77 K). The
2
Brunauer–Emmett–Teller (BET) equation was used to calculate the specific surface
area. The Barrett–Joyner–Halenda (BJH) method applied over desorption isotherm
was used to elucidate the pore structure.
Scanning electron microscopy (SEM)
The surface morphologies of the supports were examined via scanning electron
microscopy (SEM) analysis (JEOL 640, Japan). Samples were gold-coated using an
Instrumental Scientific Instrument PS-2 coating unit. The SEM pictures were
developed on thin photographic paper.
Transmission electron microscopy (TEM)
Metal particle size was determined via transmission electron microscopy (TEM)
imaging (JEM-2000 FX-II-JEOL, Japan) at an accelerating voltage of 300 kV.
Specimens were enlarged using thin photographic paper. The size of all the metal
particles visible in each photograph was measured manually and averaged.
123

D. Karthikeyan et al.
Temperature programmed reduction (TPR)
Temperature-programmed reduction (TPR) was conducted on a Micrometrics
Autochem 2920 analyzer, and hydrogen consumption was measured with a TCD
detector: 20 mg of the sample was treated in the 5% H –95% Ar gas mixture (2
2
L/h). The temperature was increased to 800 °C at a heating rate of 10 °C/min. The
quantity of consumed H was determined using a thermal conductivity detector
(
2
TCD).
Thermogravimetric analysis (TGA)
The amounts of deposited carbon on deactivated by heating 0.5 g of catalyst
e
samples were determined by a Mettler Toledo TGA/SDTA851 thermogravimetric
analysis system under a 30 ml/min flow of 4% H in argon. The temperature was
2
increased from room temperature to 1000 °C with a rate of 10 °C/min.
Catalytic tests
The autothermal reactions were conducted in a conventional fixed-bed quartz
reactor under atmospheric pressure. About 500 mg of the catalyst was placed in a
Quartz glass tube with a 16 mm inner diameter. The temperature of the catalytic
reactor was measured using Cr–Al thermocouples. Steam was supplied by heating
distilled water in a pre-heater. Figure 1 shows the experimental setup for the
Fig. 1 Schematic diagram of the experimental set-up
1
23

Effect of preparation methods on structure and catalytic…
catalytic reaction. The catalyst was reduced under 40 ml/min 10% H /Ar flow at
2
8
C H = 8.75 and C H /O = 1.25. The reactions were conducted in a temperature
00 °C for 1.5 h prior to use. The volumetric ratio of the reactants was H O/
2
2
6
2
6
2
range from 300 to 700 °C. A gas chromatograph equipped with two columns, a
Molecular Sieve 5A and a Porapak Q column, was utilized for analysis of the
reactants and products. Ethane conversion, H selectivities, and yield were defined
2
as follows:
C H Conversion (%)
2
6
ꢀ
Þꢁ
1
00 ð%Ethane À %Ethane
in
out
XEthane ¼
ð1Þ
%
Ethane
in
H Selectivity (%)
2
ꢀ
ꢁ
1
00 ð%H2Þ
SH ¼
ð2Þ
ð3Þ
ð4Þ
ð5Þ
ð6Þ
2
7
ð%Ethanein À %EthaneoutÞ þ ð%H2Oin À %H2OoutÞ
ꢀ
ꢁ
ꢁ
ꢁ
1
00 ð%COÞ
SCO ¼
7
ð%Ethanein À %EthaneoutÞ
ꢀ
1
00 ð%CO2Þ
SCO ¼
2
7
ð%Ethanein À %EthaneoutÞ
ꢀ
1
00 ð%CH4Þ
SCH ¼
4
7
ð%Ethanein À %EthaneoutÞ
ꢀ
ꢁ
1
00 ð%C2H4Þ
SC H ¼
2
4
7
ð%Ethanein À %EthaneoutÞ
H Yield (%) = C H Conversion (%) 9 H Selectivity (%), % H : Hydrogen
2
2
2
6
2
concentration in the product gas, % Ethane : Ethane concentration in the feed gas,
in
%
tion in the feed gas, % H Oout: Water concentration in the product gas.
Ethaneout: Ethane concentration in the product gas, % H O : Water concentra-
2 in
2
Results and discussion
Catalyst characterization
XRD
Figure 2 reports the X-ray diffraction patterns of Ni/ZrO , Ni/CeO and Ni/
2
2
Ce0.75Zr0.25O (Ni 2-step impregnation) catalysts, respectively. In the XRD patterns
2
of the Ni impregnated catalysts, there were the diffraction peaks attributed to NiO,
and the intensity of the peaks decreased in the sequence of Ni/ZrO , Ni/CeO and
Ni/Ce0.75Zr0.25O (Ni 2-step impregnation) catalysts, respectively. Associated with
2
2
2
the BET surface areas of the supports, it is clear that larger surface area promotes
123

D. Karthikeyan et al.
2 2 2
Fig. 2 XRD patterns of various catalysts: a Ni/ZrO , b Ni/Ce0.75Zr0.25O and c Ni/CeO
the NiO dispersion. In addition, as shown in Table 1, the lattice parameters of
supports did not change markedly when nickel species were introduced to the
impregnated catalysts, indicating that nickel species were not incorporated into the
lattice of the solid solution. However, NiO was not detected in the co-precipitated
catalysts, indicating that the nickel species dispersed well. This is also related to
lower nickel contents in the co-precipitated catalysts. Moreover, the diffraction
peaks of the solid solution shifted to higher degrees due to the existence of nickel
species. Crystalline sizes of NiO and Ni calculated by the Scherrer equation are
listed in Table 1. The dimensions of CeO and ZrO crystallites were in the range
2
2
1
0.9–26.0 nm, whereas the NiO crystallites were in the range of 10.8–23.2 nm, and
the Ni crystallites were in the range of 12.8–24.7 nm Crystalline sizes of nickel
species were found to be somewhat reasonable. It is believed that the small particle
sizes of Ni species were attributed to the relatively large amount (15 wt%) of Ni
loading on the ceria zirconia supports with high surface area. Crystalline sizes of
nickel species in the Ni/CZ , Ni/CZ
W
and Ni/CZUWA catalysts decreased with
UW
addition of urea and water. Among the catalysts, Ni/CZUWA showed the smallest
crystalline sizes of NiO and Ni [17–19].
This result was also testified to by the following results of H -TPR. The ceria–
2
zirconia solid solution with the fluorite cubic structure existed in each sample. The
peaks appeared at a similar position regardless of the manufacturing method
utilized, but the peaks of the catalyst generated by co-precipitation evidenced a low
intensity, thus suggesting a low crystalline level.
Figure 3 shows the XRD patterns of the Ni/CZ , Ni/CZUW and Ni/CZUWA
W
catalysts, respectively. XRD revealed NiO, CeO (cubic), ZrO (tetragonal) and Ce–
2
2
ZrO2 on all of the catalysts. The catalyst structures did not change. Some
1
23

Effect of preparation methods on structure and catalytic…
123

D. Karthikeyan et al.
Fig. 3 X-ray diffraction patterns of Ni/Ce0.75Zr0.25
O
2
catalysts: a Ni/CZ
W
, b Ni/CZUW and c Ni/CZUWA
investigations also demonstrated that the preparation methods affected the lattice
parameters of the CeO –ZrO solid solutions [20], and the sample with the higher
2
2
surface area had a smaller lattice parameter [21]. According to the literature [20],
the presence of ZrO improves the thermal stability of CeO , thereby avoiding the
2
2
sintering process. However, the peaks of Ni/CZUWA exhibited high intensity,
suggesting that the catalysts are more crystalline than the other catalysts. Associated
with the BET surface areas of the supports, it is clear that a larger surface area
promotes NiO dispersion.
BET surface area
The surface areas of the Ni/ZrO , Ni/CeO , Ni/Ce0.75Zr0.25O (Ni 2-step impreg-
2
2
2
nation), Ni/CZ , Ni/CZ
W
and Ni/CZUWA catalysts were determined via the BET
UW
method and are presented in Table 1. The surface areas of the Ni/ZrO and Ni/CeO
2
2
catalysts are lower than other Ni loaded ceria zirconia catalysts and also decrease
the surface area for all the catalysts after the Ni loading. The surface area of the Ni-
impregnated catalysts evidenced a high surface area relative to the co-precipitated
catalysts. The order in which the catalyst surface area changes is: Ni/CZUWA [ Ni/
CZUW [ Ni/CZ . The observed reduction in surface area with increasing Ni
W
loading can be accounted for by the formation of large metal particles, as was shown
1
23

Effect of preparation methods on structure and catalytic…
by TEM. The large particles may block the pores of the support, hence, inducing a
reduction in the surface areas of the catalysts. Additionally, the total pore volume
and also the average pore diameter of all the catalysts evidenced insignificant
deviations upon the addition of Ni. The surface areas were comparable to those
reported by Laosiripojana and Assabumrungrat [22] for ceria–zirconia supported
nickel catalysts.
SEM analysis
Figure 4 shows SEM images of the Ni0.33Ce0.54Zr0.13O and Ni/Ce0.75Zr0.25O2
2
catalysts. In the Ni0.33Ce0.54Zr0.13O (Fig. 4a) catalyst, the particle size (50–60 nm)
2
is larger and occupies the active sites on the external surface. However, the particle
size is quite small (30–50 nm) for the Ni/Ce0.75Zr0.25O catalyst (Fig. 4b). Hence,
2
the particles might readily penetrate the internal structure. As a consequence, the Ni/
Ce0.75Zr0.25O catalyst evidenced a higher activity level (Table 2).
2
Figure 5 shows SEM images of the fresh and used Ni/CZ , Ni/CZUW and Ni/
W
CZUWA catalysts. The majority of the Ni/CZW and Ni/CZUW catalysts exhibited
irregular particle size. The particle size of the fresh Ni/CZW catalyst was
7
0–100 nm (Fig. 5a), and used catalyst was 200–300 nm (Fig. 5d). The particle
size of the fresh Ni/CZUW catalyst was 40–50 nm, and some of the particles
were [ 80 nm (Fig. 5b). Additionally, the particle size of the used catalyst differed
significantly. The particle size of the used Ni/CZUW catalyst was found to have
increased to 100–200 nm (Fig. 5e). On the other hand, the Ni/CZUWA catalyst
particles were generally 20–30 nm in size. The Ni/CZUWA catalyst particles were
smaller and more uniform than the Ni/CZUW catalyst particles (Fig. 5c). Therefore,
alcohol can prevent particle agglomeration [23]. Whereas the particle size of the
used Ni/CZUWA catalyst was 20–30 nm (Fig. 5f). No significant changes were
detected as compared with the fresh Ni/CZUWA catalyst prior to the reaction. Most
of the deactivation of the nickel catalyst was attributed to poisoning of the active
sites by carbon deposition [22]. The SEM images of all of the used catalysts
2 2 2
Fig. 4 SEM images of Ni0.33Ce0.54Zr0.13O and Ni/Ce0.75Zr0.25O catalysts: a Ni0.33Ce0.54Zr0.13O and
b Ni/Ce0.75Zr0.25
O
2
123

D. Karthikeyan et al.
Table 2 Ethane conversion and hydrogen Yield over the Ni/Ce Zr
x
y
O
2
and Ni
yield (%)
500
x
Ce
y
Zr
z
O
2
catalysts
C
2
H
6
conversion (%)
H
2
Temperature (°C)
400
0
500
100
100
100
600
700
100
100
100
400
0
600
700
Ni(15wt.%)/Ce(0.75)Zr(0.25)
O
2
100
100
100
28.5
10.3
12.8
38.8
21.7
29.1
45.5
30.2
31.8
Ni(0.33)Ce(0.54)Zr(0.13)
O
O
2
2
0
0
Ni(0.54)Ce(0.33)Zr(0.13)
0
0
Reaction conditions: Feed of H
alyst weight: 0.5 g Time: 1 h, Pressure: 1 atm
2
2 6 2 6 2
O/C H = 8.75 and C H /O = 1.25 at a GHSV of 9500 ml/gcath, cat-
W
Fig. 5 SEM images of the catalyst prepared by: a fresh Ni/CZ , b fresh Ni/CZUW, c fresh Ni/CZUWA,
W
d used Ni/CZ , e used Ni/CZUW and f used Ni/CZUWA
evidenced carbon deposition. However, there were some differences. The carbon
appeared as clusters on both Ni/CZW and Ni/CZUW catalysts, but as nano-fibers on
the Ni/CZUWA catalyst (Fig. 5d–f). The nickel on the Ni/CZW and Ni/CZUW
catalysts surface can gradually become coated with carbon, resulting in the
encapsulation of the NiO active site and an interruption of the catalytic cycle [22].
However, carbon deposited on the Ni/CZUWA catalyst in the form of nano-fibers
exhibited higher activation levels. Overall, the nano-fiber form of carbon causes
relatively less deactivation and Ni evidences stable catalytic activity as a result of
nano-fiber carbon formation during ethane decomposition on the supported Ni
catalysts [22].
TEM analysis
Figure 6 shows TEM images of the fresh and used Ni/CZUW and Ni/CZUWA
catalysts. The particle size of the fresh Ni/CZUWA catalysts was much smaller than
1
23

Effect of preparation methods on structure and catalytic…
Fig. 6 TEM images of the catalyst prepared by: a fresh Ni/CZUW, b used Ni/CZUW, c fresh Ni/CZUWA
and d used Ni/CZUWA
that of the fresh Ni/CZUW catalyst (Fig. 6a,c). After the reaction, the particle size of
the Ni/CZUW used catalyst was markedly increased. Additionally, the degree of
dispersion of the Ni/CZUWA particles was higher than the Ni/CZUW particles prior to
and after the reaction. When alcohol was added during the preparation of the
catalyst, the used Ni/CZUWA catalyst exhibited remarkably low levels of carbon
deposition (Fig. 6b, d). This result was consistent with the TGA profiles.
TPR
The results of TPR of the nickel-loaded ceria zirconia catalysts are shown in Fig. 7.
For the Ni/CZ , Ni/CZ
and Ni/CZUWA catalysts, a sharp peak at 350 °C and
W
UW
another low broad peak at 760 °C were obtained, respectively. The first peak is
0
associated with the reduction of NiO to Ni , and the second peak at 760 °C is related
4
?
3?
to the partial reduction of Ce
corresponds to a high degree of uniformity in the NiO crystallite size. For the co-
to Ce
[24]. The peak obtained at 350 °C
123

D. Karthikeyan et al.
W
Fig. 7 TPR profiles of Ni supported ceria zirconia catalysts. a Ni/CZ , b Ni/CZUW, c Ni/CZUWA,
d Ni0.33Ce0.54Zr0.13O and e Ni0.54Ce0.33Zr0.13O
2 2
precipitated (Ni0.33Ce0.54Zr0.13O2 and Ni0.54Ce0.33Zr0.13O ) catalysts, one broad
2
peak was detected at 430 °C for the reduction of incorporated Ni cations into the
fluorite structure, thus indicating strong interaction with the host oxide [25], and
4?
another low broad peak was detected at 780 °C for the partial reduction of Ce to
3
?
Ce , respectively. The width of the peaks is reflective of a broad particle size
distribution, and NiO exhibited a stronger interaction with the support than the NiO
present in the Ni Ce Zr catalysts. The impregnated nickel catalysts were reduced at
x
y
z
a lower starting temperature relative to the co-precipitated supports. This is
2
?
associated with the existence of Ni
ions. Because the starting reduction
temperature of bulk NiO was higher than that of the impregnated catalysts, it
may be that NiO particles facilitate the reduction of the solid solution via hydrogen
spillover [26]. The TPR results indicated the existence of a distribution of NiO
crystallite sizes interacting almost identically with the support; however, these
support interactions of the Ni Ce Zr catalysts differed profoundly from those
x
y
z
observed in Ni/CZ , Ni/CZ
W
and Ni/CZUWA, respectively. Among the Ni/CZW,
UW
Ni/CZUW and Ni/CZUWA catalysts, the Ni/CZUWA catalysts evidenced well-
dispersed NiO particles as compared to the other catalysts.
Catalytic activity
The activity test results of the Ni Ce Zr O catalysts (co-precipitation) are shown in
z 2
x
y
Table 2. Table 2 shows that the activities observed on the Ni/Ce Zr O catalysts
x 1-x 2
were superior to the Ni Ce Zr O catalysts. All of the catalysts show a 100% ethane
x
y
z 2
conversion at temperatures range from 500 to 700 °C. However, the hydrogen yield
1
23

Effect of preparation methods on structure and catalytic…
varied over the range of tested catalysts. The maximum hydrogen yield, 45.5%, was
observed over the Ni-impregnated catalyst at 700 °C. These results demonstrate that
the active nickel particles are present on the internal surfaces; this might be induced
for the catalytic activity required for ethane reforming. XRD and TPR showed that
NiO is the active phase of the Ni/Ce Zr O catalysts (Figs. 3, 7). In comparison,
x
1-x 2
the Ni Ce Zr O catalyst contained urea also showed better activity. The activity
x
y
z 2
and conversion were similar. However, the hydrogen yield was approximately 15%
higher. According to these results, one can surmise that metallic nickel is necessary
for ethane reforming over the studied catalytic system, which is consistent with the
findings of previous studies [27, 28]. Additionally, the profound interaction between
Ni and Ce–ZrO solid solution over the Ni/Ce–ZrO catalysts with low Ni loading
2
2
results in a retardation of the reduction of NiO, and a broadening of the range of
reduction as compared with pure NiO. Thus, it is expected that relatively small and
finely dispersed Ni metal particles would be obtained during reduction treatment. In
this case, the ensemble size on the metal surface becomes smaller, resulting in good
coke resistance, as the ensemble size required for carbon formation is larger than the
size required for ethane formation [29].
The ethane conversion and the hydrogen yield over Ni/Ce Zr1-xO catalysts (Ni
x
2
2
-step impregnation) are shown in Figs. 8, 9, respectively. Figure 8 shows that Ni/
CZUWA catalyst achieved 100% ethane conversion at 500 °C, but Ni/CZW catalyst
and Ni/CZUW catalyst reached 100% ethane conversion at 600 and 650 °C,
respectively. However, the hydrogen yield varied over all the catalysts (Fig. 9).
Among these catalysts, the Ni/CZUWA catalyst showed a higher hydrogen yield of
Fig. 8 C
and C
2
H
6
conversion over Ni/CZ
= 1.25 at a GHSV of 9500 ml/gcath, catalyst weight: 0.5 g Time: 1 h, Pressure: 1 atm)
W 2 2 6
, Ni/CZUW and Ni/CZUWA catalysts. (Feed of H O/C H = 8.75
2
H
6
/O
2
123

D. Karthikeyan et al.
Fig. 9 H
H
2 W 2 2 6
yield of the Ni/CZ , Ni/CZUW and Ni/CZUWA catalysts. (Feed of H O/C H = 8.75 and
2
/O = 1.25 at a GHSV of 9500 ml/gcath, catalyst weight: 0.5 g Time: 1 h, Pressure: 1 atm)
C
2
6
6
conversion and the hydrogen yield was: Ni/CZ
0.5% at 700 °C compared to the other tested catalysts. The order of ethane
ꢀ Ni/CZ [ Ni/CZ . At
UWA
UW
W
temperatures ranging from 300 to 700 °C, increases in temperature were associated
with increases in both ethane conversion and hydrogen yield (Figs. 8, 9). On the Ni/
CZUWA catalyst, hydrogen was manufactured at low temperatures. As can be seen in
Fig. 9, the hydrogen yield is increased by 20% when a mixture of water and alcohol
was used to prepare the catalyst, because the alcohol prevented particle agglom-
eration. Additionally, the catalyst was more resistant to deactivation from carbon
deposition. The hydrogen yield was better on the Ni/CZUWA catalyst. In particular,
the Ni/CZUWA catalyst evidenced a 60.5% yield at 700 °C, which is a higher yield
than was detected with the other catalysts. These results were supported further by
the BET surface area, TEM images, and TGA data (Table 1; Figs. 7, 11).
The reaction was also carried out at 700 °C in the presence of Ni/CZUWA
catalyst, product distribution of ethane reforming and selectivities for H , CO, CO ,
CH , and C H were measured as compiled in Table 3. In the presence of catalyst,
2
2
4
2 4
the various gas species observed at the outlet were C H , CH , CO, CO , and H .
2
2
4
4
2
Selectivities for major carbon products (CO, CO , CH , and C H ) were calculated
2
4
2 4
according to Eqs. (3–6). Although the conversion of ethane was 100% for all the Ni
loaded ceria zirconia catalyst, only the selectivities of products were completely
different. The production of H was extremely higher for the ATR of ethane over
2
Ni/CZUWA catalyst than other Ni/CZW and Ni/CZUW catalysts. This means the
reforming reaction proceeded mostly on the Ni surface. Therefore, it is considered
1
23

Effect of preparation methods on structure and catalytic…
Table 3 Product distributions of ethane autothermal reforming over Ni-loaded ceria zirconia catalysts at
700 °C
Catalyst
C
2
H
6
H
2
yield
CO
selectivity
(%)
CO
selectivity
(%)
2
CH
selectivity
(%)
4
2 4
C H
conversion
%)
(%)
selectivity
(%)
(
Ni/CZ
W
100
100
100
41.5
52.7
60.5
30.7
26.4
20.8
18.4
14.8
12.7
5.2
3.7
3.1
4.2
3.4
2.9
Ni/CZUW
Ni/CZUWA
2 2 6 2 6 2
Reaction conditions: Feed of H O/C H = 8.75 and C H /O = 1.25 at a GHSV of 9500 ml/gcath, cat-
alyst weight: 0.5 g Time: 1 h, Pressure: 1 atm
that the conversion of ethane is initiated by the decomposition into carbon nuclear
and lower hydrocarbons at the inlet zone of the catalyst bed, then the autothermal
reforming of lower hydrocarbons proceeded in the rear zone of the catalyst bed.
Selectivity for H over Ni-loaded ceria zirconia catalysts increased with addition of
2
water and alcohol. On the other hand, selectivities for CO, CO , CH , and C H over
2
4
2 4
Ni/CZ catalysts decreased with addition of water and alcohol. It should be noted that
composition of C compound such as C H over Ni/CZ catalyst was much higher
than that over Ni/CZUW and Ni/CZUWA catalysts. The above results indicate that the
2
2
4
W
C–C bond cleavage reaction (C H ? 2H O = CO ? CH ? 2H ), the water–gas
2
4
2
2
4
2
shift reaction (CO ? H O = CO ? H ), and the methane steam reforming
2
2
2
reaction (CH ? H O = CO ? 3H ) were accelerated with the developing tetrag-
4
2
2
onal phase of ceria zirconia in the Ni/CZ catalysts. This is well consistent with the
fact that the tetragonal phase of zirconia was more efficient than the monoclinic
phase of zirconia for improving the catalytic performance of supported Ni catalysts
[
30].
Effects of the Ce/Zr ratio on ethane conversion and hydrogen yield
The effects of the Ce/Zr ratio on ethane conversion and hydrogen yield at 550 °C
are illustrated in Fig. 10. As a consequence, the Ni/Ce Zr
O
catalyst
evidenced higher C H conversion and hydrogen yield than was seen with the
0
.75 0.25
2
2
6
other catalysts. This indicates that the Ce/Zr ratio also performs a crucial role in
4
ethane reforming. Additionally, the incorporation of Zr into the CeO lattice
?
2
resulted in structural distortions, which promoted the reduction of both surface and
bulk oxygen species; as a result, the activity levels of the Ni-loaded ceria zirconia
solid solutions were higher than those of Ni/CeO . The H -TPR results demonstrate
2
2
that the interaction between nickel species and supports was stronger in Ni/
Ce0.75Zr0.25O than with any of the other catalysts. A stronger interaction was
2
shown to be more conducive to close contact between Ni and the support. This
promotes the dispersion of nickel species and probably enhances oxygen transfer
from the ceria zirconia supports to Ni. Moreover, Ni/Ce0.75Zr0.25O evidenced
2
higher activity levels than other catalysts, principally as the result of its larger
surface area and pore volume. In general, the metal-doped ceria zirconia catalysts
had higher oxygen storage capacity (OSC) and were more resistant to thermal and
123

D. Karthikeyan et al.
1
00
8
0
0
0
0
6
4
2
Ethane Conversion
Hydrogen Yield
0
0
0.2
0.4
0.6
0.8
1
Ce/Ze molar ratio
Fig. 10 C
0
g
3
H
8
Conversion and H
.5, 0.75 and 1.0 wt%) catalysts. (Feed of H
cath, catalyst weight: 0.5 g Time: 1 h, Pressure: 1 atm)
2
Yield at 550 °C of Ni/Ce
x
Zr1-x
O
H
2
(where x is the respective 0, 0.25,
/O = 1.25 at a GHSV of 9500 ml/
2
O/C = 8.75 and C
2
H
6
2
6
2
redox aging. The origin of the enhanced reducibility and improved OSC may be
attributable to several factors, including the presence of structural defects, which
induce the high mobility of bulk oxygen ions, even at moderate temperatures [31].
The autothermal reforming of ethane over Ni/Ce Zr O catalysts evidenced better
x
1-x 2
catalytic activity, higher conversion, and better hydrogen selectivity at an optimal
ratio of x = 3:1 (Ni/Ce0.75Zr0.25O2).
TGA analysis
Figure 11 shows the TGA results of the fresh and used Ni/CZ , Ni/CZ
W
and Ni/
UW
CZUWA catalysts. TGA demonstrated the formation of some carbon compounds on
the fresh and used Ni/CZ , Ni/CZ and Ni/CZUWA catalysts. The weight losses of
W
UW
the used Ni/CZ , Ni/CZ
W
and Ni/CZUWA catalysts were approximately 20, 14,
UW
and 5%, respectively. There was less weight loss of used Ni/CZUWA catalyst
compared with used Ni/CZW and Ni/CZUW catalyst. The level of coking on Ni/
CZUWA was relatively low. The activation tests indicated that the carbon deposits
had deactivated the surface.
Time on stream
Figure 12 shows the performance of Ni/CZUWA catalyst with time-on-stream at
7
observed during the time-on-stream. In the XRD pattern of the spent Ni/CZ
00 °C, and significant reductions in the ethane conversion and hydrogen yield were
UWA
1
23

Effect of preparation methods on structure and catalytic…
Fig. 11 TGA profiles of the Ni/CZ
W
, Ni/CZUW and Ni/CZUWA catalysts before and after the reaction
1
00
9
0
0
0
0
0
0
0
0
0
8
7
6
5
4
3
2
1
Ethane Conversion
Hydrogen Yield
0
0
5
10
15
20
25
Time (h)
Fig. 12 C
reforming of ethane over Ni/CZUWA catalyst at 700 °C. (Feed of H
at a GHSV of 9500 ml/gcath, catalyst weight: 0.5 g Time: 1 h, Pressure: 1 atm)
2
H
6
Conversion and H
2
Yield as a function of time-on-stream [TOS (h)] for autothermal
2
O/C = 8.75 and C /O = 1.25
2
H
6
2
H
6
2
123

D. Karthikeyan et al.
catalyst (not shown in the Figure), no metal nickel phase was detected, thereby
indicating the good sintering resistance of nickel species. In the TGA curves
(
Fig. 10) of the spent Ni/CZUWA catalyst, a marked loss in mass (5%) was detected
during the reaction. High nickel dispersion could favor the extent of the boundary
between the nickel and the support. This would result in the ready migration of the
carbon species formed on the nickel surface to the nickel-support interface, thereby
promoting carbon elimination via the oxygen species from the ceria–zirconia
support. According to Mattos et al. [32], the partial oxidation of the ethane reaction
is related directly to metallic dispersion–that is, for higher metal dispersions, the
conditions for ethane dissociation and the formation of H and CO are better. Xu
2
and Wang [33], in their study of Ni catalysts supported by CeO , ZrO , and
2
2
Ce Zr O , verified that at high temperatures the catalytic activity was dependent
x
1-x 2
on the BET surface area and Ni dispersion on the catalyst surface. Owing to the
minimal differences in dispersions and BET areas, the catalysts evidenced similar
performance during the partial oxidation of ethane reaction.
Conclusions
Ni/Ce0.75Zr0.25O was prepared using water (Ni/CZ ), urea water (Ni/CZUW), and
2
W
urea water alcohol (Ni/CZUWA). The Ni/Ce Zr O catalysts (Ni 2-step impreg-
x
1-x 2
nation) evidenced a relatively higher level of activity, because nickel particles might
exist within the internal structure. The Ni/CZUWA catalyst evidenced higher ethane
conversion and hydrogen yield than the other catalysts. The observed activities were
ordered as follows: Ni/CZUWA ꢀ Ni/CZ [ Ni/CZ . BET revealed that the
UW
W
surface area of the Ni/CZUWA catalyst was greater than that of the other catalysts.
SEM and TEM showed carbon clusters on the used Ni/CZW and Ni/CZUW catalysts,
whereas carbon nano-fibers were noted on the used Ni/CZUWA catalyst. The TPR
results showed that the Ni particles were well dispersed in the impregnated catalysts
as compared to the co-precipitated catalysts, particularly in the Ni/CZUWA catalyst.
TGA showed approximately 5% less carbon deposition on the Ni/CZUWA catalyst
than on other Ni/CZW and Ni/CZUWA catalysts. Overall, the activity and hydrogen
yield were higher over the Ni/CZUWA catalyst than the other catalysts. The Ni/
CZUWA catalyst also evidenced improved stability as the result of the higher degree
of dispersion and lower carbon deposition level.
References
1. S. Ahmed, M. Krumpelt, Int. J. Hydrogen Energy 26, 291 (2001)
2. S.H. Chan, O.L. Ding, D.L. Hoang, Int. J. Green Energy 1, 265 (2004)
3. M.H. Halabi, M.H.J.M. de Croon, J. van der Schaaf, P.D. Cobden, J.C. Schouten, Chem. Eng. J. 137,
568 (2008)
4. P.K. Cheekatamarla, C.M. Finnerty, J. Power Sources 160, 490 (2006)
5. D.L. Hoang, S.H. Chan, Appl. Catal. A 268, 207 (2004)
6. T. Takeguchi, S.N. Furukawa, M. Inoue, K. Eguchi, Appl. Catal. A 240, 223 (2003)
1
23

Effect of preparation methods on structure and catalytic…
7. J.A.C. Dias, J.M. Assaf, J. Power Sources 130, 106 (2004)
8. M.M.V.M. Souza, M. Schmal, Appl. Catal. A 281, 19 (2005)
9. K. Nagaoka, A. Jentys, J.A. Lercher, J. Catal. 229, 185 (2005)
10. S.H.D. Lee, D.V. Applegate, S. Ahmed, S.G. Calderone, T.L. Harvey, Int. J. Hydrogen Energy 30,
829 (2005)
1
1
1
1
1
1. S.Y. Chin, Y.H. Chin, M.D. Amiridis, Appl. Catal. A 300, 8 (2005)
2. L. Wang, K. Murata, M. Inaba, Catal. Today 82, 99 (2005)
3. S. Tang, J. Lin, K.L. Tan, Catal. Lett. 59, 129 (1999)
4. T. Wu, Q. Yan, F. Mao, Z. Niu, Q. Zhang, Z. Li, H. Wan, Catal. Today 93, 121 (2004)
5. S. Albertazzi, P. Arpentinier, F. Basile, P. Del Gallo, G. Fornasari, D. Gary, A. Vaccari, Appl. Catal.
A 247, 1 (2004)
16. A.C.W. Koh, L. Chen, W.K. Leong, B.F.G. Johnson, T. Khimyak, J. Lin, Int. J. Hydrogen Energy 32,
725 (2007)
1
1
1
2
2
2
2
7. J. Chen, Q. Wu, J. Zhang, J. Zhang, Fuel 87, 2901 (2008)
8. S.C. Dantas, J.C. Escritori, R.R. Soares, C.E. Hori, Chem. Eng. J. 156, 380 (2010)
9. P. Biswas, D. Kunzru, Int. J. Hydrogen Energy 32, 969 (2007)
0. J. Ka sˇ par, P. Fornasiero, M. Graziani, Catal. Today 50, 285 (1999)
1. P. Fornasiero, E. Fonda, R.D. Monte, G. Vlaic, J. Ka sˇ par, M. Graziani, J. Catal. 187, 177 (1999)
2. N. Laosiripojana, S. Assabumrungrat, Appl. Catal. A 290, 200 (2005)
3. N. Laosiripojana, Reaction engineering of indirect internal steam reforming of methane for appli-
cation in solid oxide fuel cells, PhD Thesis, University of London UK (2003)
4. J.A. Montoya, E.R. Pascual, C. Gimon, P.D. Angel, A. Monzon, Catal. Today 63, 71 (2000)
5. F. Ocampo, B. Louis, A.C. Roger, Appl. Catal. A 369, 90 (2009)
6. T. Takeguchi, S.N. Furukawa, M. Inoue, J. Catal. 202, 14 (2001)
7. W.S. Dong, H.S. Roh, K.W. Jun, S.E. Park, Y.S. Oh, Appl. Catal. A 226, 63 (2002)
8. M.H. Youn, J.G. Seo, K.M. Cho, S. Park, D.R. Park, J.C. Jung, I.K. Song, Int. J. Hydrogen Energy
2
2
2
2
2
33, 5052 (2008)
2
3
3
3
3
9. J.R. Rostrup-Nielsen, Promotion by poisoning. Stud. Surf. Sci. Catal. 68, 85 (1991)
0. M. Benito, R. Padilla, L. Rodrıguez, J.L. Sanz, L. Daza, J. Power Sources 169, 167 (2007)
1. J. Kaspar, P. Fornasiero, J. Solid State Chem. 171, 19 (2003)
2. L.V. Mattos, E.R. Oliveira, P.D. Resende, F.B. Noronha, F.B. Passos, Catal. Today 77, 245 (2002)
3. S. Xu, X. Wang, Fuel 84, 563 (2005)
123