See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/268688339

Montmorillonite-supported Ni nanoparticles for efficient

hydrogen production from ethanol steam reforming
Article in Fuel November 2015
Impact Factor: 3.52 DOI: 10.1016/j.fuel.2014.11.033

CITATIO
N

READS

75

5 authors, including:
Junfeng Zhang

Xianmei Xie

Chinese Academy of Sciences

Taiyuan University of Technology

18 PUBLICATIONS 111

29 PUBLICATIONS 245 CITATIONS

CITATIONS
SEE
PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate,

letting you access and read them immediately.

SEE PROFILE

Available from: Junfeng Zhang

Retrieved on: 18 April 2016

Fuel 143 (2015) 5562

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.c om/l ocate/fuel

Montmorillonite-supported Ni nanoparticles for efcient hydrogen production

from ethanol steam reforming
Tingting Li a, Junfeng Zhang b, Xianmei Xie a,, Xuemei Yin a, Xia An a
a
b

College of Chemistry and Chemical Engineering, Taiyuan University of Technology, Taiyuan 030024, China
State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, China

highl i ghts

graphical a bstrac t

Montmorillonite is an attractive catalyst

support for ethanol steam reforming
(ESR).
Ni0 as active sites are generated in the
montmorillonite clay.
The alkaline treatment created more
mesopores and neutralized acid sites.
Montmorillonite supported Ni0 show
excellent catalytic activity and
stability in ESR.

article

i nfo

Article history:
Received 4 September 2014
Received in revised form 28 October 2014
Accepted 11 November 2014
Available online 21 November 2014
Keywords:
Montmorillonite NiAl
catalysts Alkaline
treatment
Ethanol steam reforming
Hydrogen production

abstract
Alkaline-promoted of NiAlmontmorillonite (NiAlMt) catalysts were prepared by ion-exchange method, tested in
ethanol steam reforming (ESR) for hydrogen production, and characterized by a combination analysis of XRD,
TPR, SEM, HRTEM, EDS and N2 adsorptiondesorption. Characterization results indi- cated that the textural
properties of the as-prepared catalysts are strongly dependent on the alkaline treatment. More importantly, a clear
improvement on stability and catalytic activity, especially for the high selectivity to hydrogen, was observed in
ESR. For NiAlMt/NaOH catalyst, 74.7% selectivity of hydrogen was kept at 773 K during 10-h test, while that of
NiAlhydrotalcite (NiAl) catalyst decreased from 20.4% to 5.3% within 10 h, and the selectivity of hydrogen on NiAl
Mt was 55.1%, accompanying with byproducts of acetaldehyde and ethylene. The improved activity and stability were
attributed to the neutralization of acid sites by alkaline treatment, which accelerated the dehydrogenation and suppressed
the dehydration of ethanol.
2014 Elsevier Ltd. All rights reserved.

1. Introduction
With increasing concern about the shortage of fossil fuels and
environmental problems, certain renewable sources of energy are expected
to substitute for the conventional energy resources
Corresponding author. Tel./fax: +86 351 6018564.
E-mail addresses: lily8631@163.com (T. Li), xxmsxty@sohu.com (X. Xie).

http://dx.doi.org/10.1016/j.fuel.2014.11.033
0016-2361/ 2014 Elsevier Ltd. All rights reserved.

[1,2]. Given the increasing demand for new and clean energy, hydrogen
is considered to be an ideal energy carrier to support sustainable energy
development. It can be used as feed in fuel cells for power generation with
high efciency and it is extremely clean as the only by-product is water [3,4].
There are various resources for hydrogen production such as coal, natural
gas, propane, meth- anol, and ethanol. Among all these resources ethanol is
superiorly attractive because of its relatively high hydrogen
content,

56

T. Li et al. / Fuel 143 (2015) 5562

availability, non-toxicity, storage and handling safety, and more

importantly, it can be produced by fermentation of biomass [5]. The
target reaction of ethanol steam reforming (ESR) is represented as below:

C2H5OH 3H2O 2CO2 6H2

DH

H
298

347:4 kJ=mol

For the efcient production of hydrogen, the catalyst plays an

important role. In other words, how to design and develop efcient catalyst
with high activity and selectivity is very crucial. Up to now, many
kinds of metal catalysts have been employed in this reaction process,
including precious metal and non-noble metal catalysts [6]. Precious
metal catalysts, such as Ru, Pt, Rh and Ir exhibited higher activity and
selectivity, whereas extremely high cost associated with such catalytic
system and less availability limited their application [7,8]. Thus, many
studies tend to choose Ni, Cu, Co and non-noble metal as potential
alternatives [9,10]. Of all the non-noble metal catalysts, nickel catalysts
have been widely used in the ESR due to their high dehydrogenation activity
and low cost [11,12]. Unfortunately, Ni-based catalysts suffer from
signicant deactivation caused by sintering and carbon deposition. It has
been reported that alumina can improve the dispersibility of Ni and
efciently prohibit the carbon deposition and nickel sinter- ing during the
steam reforming reactions [13,14]. In the same way of the metallic phase, the
choice of the support is also very impor- tant in order to prevent these
drawbacks. The physical and chem- ical properties of the support as well
as the extent of interaction with the active metal would play a crucial
role in the complex chemistry of supported metal catalysts.
The montmorillonite (Mt) clay is considered as environmentally benign,
cheap, easily available and robust supporting/stabilizing material for
the synthesis of different metal nanoparticles [1518]. It has several
obviously advantages as support matrices: (1) Mt is abundant worldwide
and does not require laborious and time-consuming efforts for their
synthesis as well as surface functionalization, which may often poison the
catalytic active sites of the supports and the particles; (2) Mt has even
more open framework structure than other porous materials and does
not cause signicant mass-transfer limitation; (3) Mt has much higher
chemical and heat stability; and (4) the electrostatic force between Mt and
metal nanoclusters is stronger than the physical absorption existed in other
porous supports. To the best of our knowledge, few reports have studied on
the use of Mt in the ESR.
Herein, we want to report the utilization of the alkaline treat- ment in
Ni, Al intercalated Mt, which served as a novel catalyst in ESR
efcient for hydrogen production. The physicochemical properties of the
fabricated catalysts were investigated through a combination analysis of
XRD, XRF, TPR, SEM, HRTEM, N2BET, etc. The role of alkaline
treatment in the design of the catalysts for ESR and the structureactivity
relationship of the as-prepared catalyst were systematically studied.

2.2. Catalyst preparation

Preparation of NiAlmontmorillonite (NiAlMt) catalysts. 5.0 g puried
NaMt was dispersed in 200 mL of an aqueous Ni(NO3)26H2O and Al(NO3)3 9H2O solution (the Ni/Al molar ratio was 5).
This slurry was magnetically stirred for 3 h to ensure ionicexchange adequately (5 wt% of Ni with respect to the clay weight).
The NiAlMt powder was collected by ltration and then thor- oughly
washed with deionized water to remove the excess salt. Then, the
sample was dried at 353 K, and calcined at 873 K under static air condition
for 3 h with a ramp of 5 K/min.
Preparation of NiAlhydrotalcite (NiAl) catalysts. The NiAl
derived catalyst was prepared by a coprecipitation technique. The Ni/Al
molar ratio was 5, and the Ni loading was xed at
5.0 wt%.
Preparation
of
NiAlmontmorillonite/NaOH
(NiAlMt/NaOH)
catalysts. The resulted NiAlMt was dispersed in deionized water and a
desired amount of aqueous NaOH (1 mol/L) was slowly added to the
solution with maintaining a constant pH of 910. The hydrothermal
treatment, drying process and calcination were identical with NiAlMt.
2.3. Characterization of catalysts
The X-ray diffraction (XRD) patterns were performed on Rigaku D/max2500 instrument (40 KV, 100 mA) using Cu Ka radiation at a scanning speed
of 8 (2h)/min, with the scanning range of 370 .
N2
adsorptiondesorption
isotherms
were
measured
with
Micromeritics ASAP 2020 instrument. Prior to each measurement, the
samples were degassed under vacuum at 573 K for 4 h. The spe- cic
surface area was calculated by the BrunauerEmmettTeller (BET)
method, and the pore size and pore volume were determined on the basis of
the BarrettJoynerHalenda (BJH) method using the data of adsorption
branches.
X-ray uorescence (XRF) spectra was recorded on S8 TIGER X- ray
uorescence spectrometer (Bruker, Germany).
H2temperature programmed reduction (TPR) was performed on a
Micromeritics AutoChem 2920 apparatus equipped with a TC detector.
Prior to each TPR, the samples, about 30 mg, were thermally treated
under air stream at 573 K to remove water and other contaminants. TPR
proles were obtained by heating the samples under a 10% H2/Ar ow (50
mL/min) from 298 to 1173 K at a linearly programmed rate of 10 K/min.
The surface morphology images were examined with the SU1500 (Hitachi, Tokyo, Japan) scanning electron microscope (SEM).
The microstructures of catalysts was investigated using trans- mission
electron microscope (TEM, JEM-1011) equipped with an energydispersive X-ray spectrometer (EDS).
2.4. Catalytic performance test

2. Experimental
2.1. Materials
Ni(NO3)2 6H2O (Aladdin, >98.0%), Al(NO3)3 9H2O (Aladdin,
>99.0%), NaOH (Kermel, >98.0%), ethanol (Kermel, >99.7%), Na
montmorillonite was purchased from Zhejiang Sanding Science and
Technology Co., Ltd. Namontmorillonite (NaMt) was used after
sedimentation to remove a small amount of impurity. The cation
exchange capacity (CEC) was around 110 meq per 100 g clay. The
chemical composition of Namontmorillonite wt%: SiO2
65.14, Al2O3 17.05, Fe2O3 3.12, TiO2 0.77, K2O 1.52, MgO 3.12,
CaO 0.35, Na2O 2.15, loss on ignition (LOI): 6.52.

Catalytic performance tests were performed in a tubular xed- bed steel

reactor at atmospheric pressure. 0.5 g of catalyst mixed with 2.5 g of quartz
particles (all catalysts were crushed into parti- cles of 2040 mesh in size).
Prior to each reaction, the system was out-gassed with N2 for 30 min, then
the catalysts were reduced in H2/Ar at 923 K for 2 h, and then switched to the
reaction tempera- ture with N2 ow protection. The liquid mixed
reactant with a water to ethanol molar ratio of 8:1 was fed into the reactor
through a electronic peristaltic pump at a rate of 0.1 ml/min before mixing
with N2 (as a carrier gas, 70 ml/min), and then vaporized at 423 K.
1
The total liquid hourly space velocity (LHSV) was 1.9 h . The efuent gaseous products were analyzed on-line by a haixin 950 gas
chromatograph equipped with a TCD and a FID detector and two packed
columns. TDX-01 packed column was used to separate H2

and N2, whereas CO, CH4, CO2, C2H4 were separated using a GDX401 column. Methane was used as an external standard to
calculate the carbon and hydrogen balance. The liquid products
collected in the gasliquid separator were analyzed off-line on another
gas chromatograph equipped with a Porapack Q column by manual
injection. As a result of the products variation, the catalyst performance
was characterized by two parameters. Si rep- resents
the
product
distribution of i, and ethanol conversion denoted as Cethanol. They
were calculated according to Eqs. (2) and (3):

nethanolin nethanolout
Cethanol

Si Pn

nethanol

in

npi
100%
i1 n pi

be intercalated into the interlayer region of Mt and oxide pillars are

formed after calcination. Thus, the slightly expanded interlayer space of the
NiAlMt/NaOH might be originated from the oxide pillars in the
interlayer region [20]. The X-ray diffraction patterns of the reduced
catalysts and the used catalysts are displayed in Fig. S1. In reduced
0
catalysts (Fig. S1A), the diffraction peaks of Ni 2h = 44.5 , 51.8 and
76.4 (JCPDS78-0643) related to the crystalline planes of Ni0 (111),
2+
0
(200) and (220), indicating the reduction of Ni to Ni ; no diffraction
peaks indicative of NiO were observed. This means that NiO in the NiAl
composite structure was completely reduced into metallic nickel during the
reduction process employed in this work. For NiAlMt and NiAlMt/NaOH, there
is no peaks assigned to be Ni0 detected by XRD, which suggested
0
the Ni species should be well dispersed that veried by TEM. Sub- stantial
0
evidences for the existence of Ni in Mt are still being
investigated by HRTEM.
3.2. N2-adsorption and desorption

3. Results and discussion

3.1. X-ray diffraction patterns
The X-ray diffraction patterns of the samples after drying and
calcination are given in Fig. 1. For the dried samples (Fig. 1A), the NiAl
exhibits a pure well-crystallized and the diffraction peaks at
2h = 9.9 , 20.0 , 35.0 . The (001) diffraction of NaMt is at 2h of
about 7.5 , corresponding to a d001 spacing of 1.20 nm. However, the
d001 spacing of the NiAlMt increase slightly. For the dried samples,
water entrapped in the Mt interlayer cannot be removed at a temperature of
333 K. Moreover, the Ni2+, Al3+ cations are located in the Mt
interlayer, and are present in a hydrated form. The interlayer distance
of NiAlMt/NaOH is more increased in comparison to that of NaMt, that
2+
3+
is because the Ni and Al is converted into their hydroxides, so it is
exerted pillared effect in the interlayer of Mt. After calcination at 773
K (Fig. 1B), the NiAl was converted to its mixed oxide showing only
the diffraction peaks of the NiAlO solid solution. The NaMt and
NiAlMt exhibits the same d001 spacing of 0.96 nm, which can be reasonably
ascribed to the loss of the water molecules hydrated to the
interlayer cations [19]. This observation strongly supports our
explanation on the d001 spacings of the dried samples. Thus, after ionexchanging and calcination, the Mt laminar structure was not destroyed.
The d001 spacing of NiAlMt/NaOH (1.00 nm) is observed slightly
higher and the intensity of the diffractions is slightly decreased in
comparison with NaMMT. That indicates that an insignicantly
decreased crystallinity and layered Mt struc- ture may be partially destroyed.
A part of silica and alumina in the Mt layer is dissolved in alkaline
solutions. Moreover, the layered structure of Mt is still preserved,
although a partial destruction of the Mt layer occurred under NaOH
treatment. The silica and alumina species from the desilication and
dealumination could

Fig. 2 depicts typical nitrogen adsorptiondesorption isotherms for Na

Mt, NiAl, NiAlMt, NiAlMt/NaOH, and the calculated tex- tural properties
are summarized in Table 1. The shape of the N2 adsorptiondesorption
isotherm and the average pore size indi- cated that the raw NaMt was
practically a nonporous material. Moreover, its small pore volume (0.08
cm3 g 1) was possible due to the agglomeration of the lamellar Mt
akes. The NiAl showed H3 hysteresis loops, which demonstrated
that the plate-like particles with slit-shaped pores were successfully
prepared. On the contrary, a narrow slit-shaped porous structure was
observed for the NiAlMt and NiAlMt/NaOH, which was clearly
indicated the mesoporous structure and as suggested from the type IV isotherm together with an H4-type hysteresis loop [21]. The volume of the
adsorbed N2 increased steeply with the relative P/P0 in the range of P/P0 <
0.1, indicating the presence of rich micropores. Moreover, the slightly
gradual increase of the amount of N2 adsorbed in the medium P/P0
and the near parallel part of the hysteresis loop above P/P0 of 0.8
indicated the presence of open mesopores [22,23]. Noticeably, both
micropores and open mesop- ores were created in NiAlMt/NaOH catalyst,
which showed a clear triangular shape, the desorption branch falling sharply
close to the lower limit of the hysteresis loop. This observation was
well matched with BJH pore size distribution (Fig. S2).
As shown in Table 1, the surface area of the raw NaMt its surface
area was 10.7 m2/g and surface area of the NiAlcatalyst was 22.3 m2/g.
2+
On the contrary, by simply ion exchanging the NaMMT with Ni and
3+
2
Al , its BET surface area was increased to 94.4 m /g. This indicated
that many pores accessible to N2 were produced after exchanging Na
2+

3+

MMT with Ni
and Al . If the pH value of the ion-exchange solution
was examined (Table 1), this may be because the removal of mineral
impurities present in
the void between the Mt platelets in an acidic environment, which

Intensity

Intensity

d
d
d=1.50 nm

d=1.28 nm

d=1.20 nm
10

20

d=1.00 nm
c
b
a

d=0.96 nm

a
30

40

2 /

50

60

70

10

20

30

40

50

60

70

2 /

Fig. 1. XRD patterns of samples (a) NaMt, (b) NiAlMt, (c) NiAlMt/NaOH, and (d) NiAl. The dried (A) and calcined (B) samples.

Quantity Adsorbed cm3/g STP

100

NiAlMt, it is noteworthy that an additionally small but clear reduction

peak at about 1090 K was observed for NiAlMt/NaOH. There are c-type
NiO species, ascribed to the stable nickel aluminate spine [28]. This phenomenon can be explained on the basis of the
N2-adsorption and desorption and XRD analysis. After alkaline
treatment, the mesopores created because of the desilication and dealumination of Mt and the formation of oxide pillars.
wherein, the Ni2+, Al3+ can be easily dispersed into the mesopores
on the framework of the Mt layers, leading to the retarded reduction
a
of the nickel species because of the strong Ni-support interac- tions. These results seem to
indicate that the alkaline treatment
increases the interaction between nickel sites and the support [13].

80

d
60

40

20
0
0.0

0.5

1.0

3.4. The scanning electron microscopy

Relative Pressure p/p0

Fig. 2. N2 adsorptiondesorption isotherms of (a) NaMt, (b) NiAl, (c) NiAlMt, (d) NiAl
Mt/NaOH.

has been reported for the acid activated Mt [24]. Moreover, the sur- face
2
areas of NiAlMt/NaOH (168.2 m /g) was much higher than those of
other samples. This may be explained based on the follow- ing factors: (1)
under the conditions of alkaline activation, more mesopores were
created over the activated Mt because of the destruction of the Mt
layered structure, and (2) a part of silica and alumina species from the
desilication and dealumination can be intercalated into the interlayer
region of Mt and oxide pillars were formed after calcination, which
increased the surface area. These results were consistent with the XRD
and XRF analysis. Consequently, NiAlMt/NaOH had huge surface area
and can pro- vide a large number of active for ESR. When the pore
properties are concerned, some differences can be found. In comparison with
the raw NaMt, the total pore volume increased while the average pore size
decreased for other samples. In addition to the above rea- sons, the pore
volume and the average pore size were dependent on the type of catalyst and
preparation method.

To reveal the effect of the NaOH treatment on the morphologic changes

of Mt, the NaMt, NiAlMt and NiAlMt/NaOH were sub- jected to SEM
observations, and the results are given in Fig. 4. Gen- erally, SEM images
of the catalysts whether or not processed alkaline activation are greatly
different. The typical SEM image of the raw NaMt were composed of plates
and represents irregular akes with small patches of platelets. The NiAlMt
exhibited sim- ilar morphologies, and there were on signicant changes on
SEM images. Most of the platelets of NiAlMt were unaffected by the
ion-exchanged. However, for NiAlMt/NaOH, clumps of agglomer- ated
particles attached on Mt akes are clearly observable and some cracks
are found, reecting the partial collapse of the Mt lam- inar structure.
Damage of the alkaline activation on the Mt laminar structure can be
distinctly determined from the SEM observations, which is agreeable with
the results of XRD and XRF.
3.5. Catalytic ESR performance
3.5.1. Effect of reaction temperature
The catalytic performance of samples in the ESR under different reaction
temperature (TR) described in Fig. 5. As previous reports observed [29],
Ni could act as the active center and Al was often

3.3. Temperature programmed reduction

726 K

Based on the peak positions of the H2TPR curves (Fig. 3), the reduced
TCD Signal (a.u.)

NiO species are usually classied into four types of inter- action forces: a,
b1, b2, c. The a-type is assigned to the free nickel oxide species possessing a
weak interaction with support and can be reduced in the temperature range of
573673 K. The medium- temperature peaks (6731023 K) represent btype (b1, b2) NiO species which have a stronger interactions with support
than the a-type NiO. The c-type (>1023 K) belongs to the much less reducible NiO in nickel aluminate phase with the spinel structure. The NiAl
showed only one reduction peak at 663 K, which was attrib- uted to a-type
NiO species with weak interaction with alumina
[25]. As for NiAlMt, two main reduction zones can be distinguished, one reduction peaks appeared at below 726 K associated with btype (b1) NiO species and another at higher temperatures (965 K, b2 NiO
species) [26], that is because the metal oxide parti- cles have strong
interaction with Mt [27]. Compared to NiAl and

965 K

663 K
1090 K

c
b

a
273

473

673

873

1073

1273

Temperature (K)
Fig. 3. H2TPR proles for the catalysts (a) NiAl, (b) NiAlMt/NaOH, and (c) NiAl Mt.

Table 1
Summary of the textural properties of different samples.

Sample

pH of solution

BET surface area (m g )

Pore volume (cm g )

Average pore size (nm)

SiO2 (wt%)

Al2O3 (wt%)

NaMMT
NiAl
NiAlMMT
NiAlMMT/NaOH

7.83
9.21
5.53
9.45

10.7
22.3
94.4
168.2

0.08
0.10
0.16
0.39

33.0
10.8
6.4
14.3

65.1

64.8
55.8

17.1

18.8
13.6

The content of SiO2 and Al2O3 over the MMTs was determined by XRF.

Fig. 4. SEM images of (a) NaMt, (b) NiAlMt, and (c) NiAlMt/NaOH.

100
80
60
40
20
0
600 650 700 750 800 850 900
950

Temperature /K

Conversion or Selectivity /%

C
100

Conversion or Selectivity /%

Conversion or Selectivity /%

100
80
60
40
20
0
600 650 700 750 800 850 900 950

Temperature /K

CH3CHO

80

H2
CO
CO2

60
40

CH4

20
0
600 650 700 750 800 850 900 950

CH
2

Temperature /K
Fig. 5. The conversion of ethanol and the selectivity to the different products vs temperature over (A) NiAl, (B) NiAlMt, and (C) NiAlMt/NaOH. Reaction conditions: H2O/ EtOH molar ratio = 8:1,
1

LHSV = 1.9 h , and P = 1 bar.

used to enhance the mechanical stability and chemical resistance of the

resulting nanocomposite [30,31], the acid-rich surface was resulted by Al
atoms [13], so undesired ethanol dehydration reac- tion was more prone to
occur on NiAlcatalyst, resulting in the low hydrogen selectivity (Fig. 5A).
At 823 K, the amount of ethylene was already signicant being the
largely predominant product. This was due to the fact that large amount of
acid sites promoted the dehydration of ethanol to ethylene (Eq. (4)) [32,33]:

CH3CH2OH C3H4 H2O

Small differences in the amount of ethylene and H2 formed (the

selectivity of H2 was slightly higher than that of when TR above
873 K), indicating further decomposition of ethylene to carbon or undergo
steam reforming to H2 and C1 (Eqs. (5)(7)). At 923 K, a slight amount
of ethane occurred because ethylene was probably hydrogenated to ethane
(Eq. (8)):

C2H4 polymerize coke H2

1=2C2H4 H2O CO 2H2

1=2C2H4 2H2O CO2 3H2

C2H4 H2 C2H6

For NiAlMt catalyst (Fig. 5B), the ethanol conversation and H2

selectivity increased with TR. The selectivity of H2 approached to
58.7%. Acetaldehyde formation showed a signicant decline if TR

increased, and disappeared at 923 K. The CO2 started to rise from

773 K and rapidly reached 20%. Some other undesirable products, such as
CH4 and CO were also obtained. CH4 presented a maximum of 12% at 823
K. CO underwent monotonically increased with TR and reached a
maximum of 20% at 923 K. The catalytic perfor- mance of NiAlMt
catalyst was better than NiAlcatalyst, that was mainly due to the
promoted effect from the Mt support. The layers of MMT can generate
multiple mixture of metal oxide (e.g. SiO2, CaO, MgO, Fe2O3, etc.) after
calcination, so it is not only comprehensive support but also twodimensional reaction vessel. In the reaction, Mt could stabilized the
metal ions, increased the surface area and enhanced the interaction between
Ni and Mt.
The NiAlMt/NaOH catalyst exhibited high activity for ESR as
shown in Fig. 5C. The ESR over NiAlMt/NaOH generally divided into
three stages: the rst stage was ethanol dehydrogenation (Eq. (9)) over
the support of Mt (multiple mixture of metal oxide) to yield acetaldehyde and
hydrogen [34,35]. Previous reports sug- gest that the ethanol dissociates into
ethoxy species, which further decomposed into acetaldehyde and hydrogen
in rst stage [36]. The selectivity to acetaldehyde dramatically
decreased with the increase of TR, and no acetaldehyde was detected. At the
same time, the fed of ethanol completely converted at 773 K and
products mainly contained H2, CH4, CO and CO2. This indicated that
the second stage: acetaldehyde steam reforming (Eq. (10)) and acetal- dehyde
0
decomposition (Eq. (11)) were preceded. The Ni had high ability in
CC bond cleavage and causes the bond breaking of ethanol on pure
Ni (111) surface in the following order:

OH, CH2, CC, and CH3 [37], the acetaldehyde-steam mixture was
transformed over the Ni0 into mixture of hydrogen, carbon monoxide,
carbon dioxide and methane in the second stage. In
third stage, it can be noticed a growth in CO2 selectivity, whereas CH4 and
CO ones decrease. This may be as consequence of methane steam reforming
(MSR Eq. (12)) and water gas shift reaction (WGSR, Eq. (13)) are
being taking place. That can also account for the rise in hydrogen
selectivity. An interesting point is the mini- mum in CO concentration,
at about 773 K, with an increase in hydrogen selectivity and a slightly
decrease for methane selectiv- ity. This probably occurs because MSR
begins to reach a larger extension, while WGSR equilibrium is displaced
towards products formation. When TR is above 823 K, the activity of MSR
and meth- ane dry reforming reaction (Eq. (14)) are greatly enhanced,
leading to the conversion of CO2 and CH4CO and H2. This was in
agree- ment with the thermodynamic analysis via response reactions conducted by Fishtik et al. [38]. However WGSR reaches equilibrium, so that
CO cannot be consumed, which leads to a growth in CO selectivity. It
is well known that CO is a strong poison for fuel cell [39], so excessively
high temperature is unfavorable for the appli- cation of ESR in fuel cell.

CH3CH2OH CH3CHO H2

CH3CHO 2H2O 2CO 3H2

10

CH3CHO CO CH4

11

CH4 H2O CO 3H2

12

CO H2O CO2 H2

13

CH4 CO2 2CO 2H2

14

Accordingly, we can obtain a preliminary reaction pathway of ESR

on nickel based catalysts (as shown in Fig. 6). The NiAl and NiAl
Mt/NaOH showed the catalytic activity in an opposite way: ethanol
dehydration to ethylene is favoured on acidic sites of NiAl. Coke deposition
results in deactivation of the NiAl when large amounts of ethylene are
produced. On the contrary, Na+ neutral- izes the surface acidicbasic
properties of the support oxide, so dehydrogenation of ethanol is preferred
on NiAlMt/NaOH, leading to acetaldehyde and hydrogen (stage 1).
Acetaldehyde is sequen- tially decomposed and converted into syngas
through steam reforming (stage 2). Hereafter, MSR and WGSR are
performed in stage 3 to reduce the CH4 and CO content. Although
acidity of the catalyst is not the sole factor determining the catalytic performance in the steam reforming of ethanol, it greatly affected the
reaction path [40].

Fig. 6. Reaction scheme of ESR on NiAl and NiAlMt/NaOH.

3.5.2. Stability
To further investigate the catalytic behavior of the samples, the
conversion of ethanol and selectivities to different products as function
of time on stream are illustrated in Fig. 7. For NiAl (Fig. 7A),
hydrogen selectivity reached 20.4% at the beginning, but decreased
gradually to below 10%, and remained around
5.7%. Meanwhile, similar phenomena were observed on selectivi- ties to
methane and carbon monoxide, which emerged at high value of 17.8%
and 25.5% then decreased with time, and stabilized at 6% and 9%,
respectively. In contrast, a bias toward ethylene was observed: the selectivity
to ethylene appears at 22.3% at beginning, then increased and remained at a
higher level of 60.7%. The carbon dioxide showed slight variation between
15% and 20%. No acetal- dehyde was detected throughout the test. NiAlMt
showed no sig- nicant changes of selectivity for all products during 7 h (Fig.
7B). Ethylene started to be increased just after 7 h on stream and still in
relatively small amounts. Furthermore, the hydrogen selectivity was
close to 55%, and the carbon dioxide selectivity was nearly
15%, remaining unchanged for the whole experiment. After alka- line
treatment, an important increase in the hydrogen selectivity was
obtained over NiAlMt/NaOH, reaching the 74.7%, as well as an
improvement in stability (Fig. 7C). The selectivity of CH4, CO and CO2
almost kept at about 4.8%, 8.0%, and 16.4%, respectively. Ethylene was not
produced, and the catalytic behavior was com- pletely stable at least up to
10 h on stream.
The NiAl has very small surface area, which is unfavorable for
steam reaction, and its acidic property would generated large
amounts of ethylene, resulting in coke deposition on the surface of
NiAlcatalyst. In contrast, on NiAlMt/NaOH, dehydra- tion route of
ethanol is suppressed, and the dehydrogenation route is enhanced, thus
resulting in higher selectivity to non- ethylene originated C1 products
and H2, and exhibited better stability. Necessary to say, the pore size,
BET surface area, and acidbase properties of catalysts play a
synergistic role for improving H2 production and stability. Although
the porous properties of the catalysts were not key factors in
determining the ESR activity, the relatively rich mesopores and the
partial destruction of the MMT layer of NiAlMt/NaOH, may facilitate
the hydrogen radicals spillover from Ni, leading to the dominant
dehydrogenation [41]. The inhibition of ethylene formation was
attributed to the high-strength basic sites on NiAlMt/NaOH by alkaline
treatment, which neutralized the acid sites, improving its selectivity to
hydrogen and stabilizing the catalyst again deactivation [42].
3.6. High resolution transmission electron microscopy
0

To gain some insight into the location of the nickel particles (Ni )
on the support of Mt, the catalysts after reduction and after 10 h of
ESR reaction were conducted on the high resolu- tion transmission
electron microscopy (HRTEM) equipped with an energy-dispersive X-ray
spectrometer (EDS) analysis. Fig. 8a shows the TEM images of the
reduced catalyst, it is seen that the Ni0-nanoparticles are anchored on
the heterogeneous meso- porous surface of the Mt matrix with sizes of
10 nm. The EDS results are performed to conrm the presence of
elemental Ni in the alumino-silicate constituent of the Mt matrix
(Fig. S3A). HRTEM image of reduced catalyst (Fig. 8b) shows the
reticular lattice planes inside the nanoparticles. The lattice planes
contin- uously extended throughout the whole particles without stacking faults or twins, indicating the single crystalline nature. The
measured inter-planar lattice fringe spacing is about 0.203 nm, which
0
corresponds to the (111) plane of fcc Ni crystals. The corresponding
Fast Fourier Transformation (FFT) was performed to examine closely
the ow-induced orientational patterns as shown in the inset of Fig.
8b. The formation of hexagonal

B 60

A 60

50

Selectivity /%

Selectivity /%

50
40
30
20
10

40
30
20
10

0
0 1 2 3 4 5 6 7 8 9 10 11

0 1 2 3 4 5 6 7 8 9 10 11

Time-on-stream /h

Time-on-stream /h

CH3CHO

C 70
Selectivity /%

60

H2

50

CO
CO2

40
30
20

CH4

10

C2H4

0
0 1 2 3 4 5 6 7 8 9 10 11

Time-on-stream /h
1

Fig. 7. The selectivity to the different products vs time over (A) NiAl, (B) NiAlMt, and (C) NiAlMt/NaOH. Reaction conditions: H2O/EtOH molar ratio = 8:1, LHSV = 1.9 h ,
T = 773 K and P = 1 bar.

Fig. 8. TEM image of reduced NiAlMt/NaOH (a); HRTEM image of reduced Ni -nanoparticle (b); TEM image of used NiAlMt/NaOH (c); HRTEM image of used Ni - nanoparticle.

symmetrical diffraction spot patterns indicates that crystalline nature of

0
0
the Ni nanoparticles. After 10 h ESR experiment (Fig. 8c), the Ni
particles are still homogeneously distributed throughout the Mt matrix
without any signicant agglomera- tion. It is noteworthy that some
graphitic type carbon is observed on the catalyst surface (Fig. 8d),
which is quite consis- tent with the EDS analysis (Fig. S3B). However,
heavy carbon deposition did not occurred and the active site Ni0
was not completely enwrapped in the layers of graphitic carbon, which
is attributed to the neutralization of acid sites by alkaline treat- ment,
so NiAlMt/NaOH shows excellent catalytic activity and stability in
ESR.

4. Conclusions
Natural NaMt was initially ion exchanged with Ni2+ and Al3+ by
controlling the pH value about 910, the obtained NiAlMt cat- alysts was
modied by NaOH treatment, and applied to H2 produc- tion in ESR.
Through the systematical characterizations and study of the relationship
between structure and activity, the following conclusions can be reached:
(1) the structural and textural proper- ties of Mt are strongly dependent upon
the alkaline treatment. A larger surface area and more mesopores are
generated. Even after the desilication and dealumination, the layered
structure of Mt is still partially maintained. (2) The novel NiAl
Mt/NaOH catalyst

exhibited improved catalytic activity and high hydrogen selectivity as well as

the durability in ESR, which was superior to NiAlMt and NiAlcatalysts.
The alkaline treatment strategy has been success- fully developed in this
work, which is promising for the general use in the steam reforming
reaction.
Acknowledgments
This work was supported by the National Natural Science Foun- dation
of China (Grant No. 50872086) and the Science-Technology key projects in
Shanxi Province (Grant No. 20130321033-02).
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in the
online version, at http://dx.doi.org/10.1016/j.fuel.2014.11.033.
References
[1] Schrope M. Which way to energy utopia? Nature 2001;414:6824.
[2] Llorca J, Piscina PR, Sales J, Homs N. Direct production of hydrogen from ethanolic
aqueous solutions over oxide catalysts. Chem Commun
2001;7:6412.
[3] Lee G, Kim D, Kwak BS, Kang M. Hydrogen rich production by ethanol steam
reforming reaction over Mn/Co10Si90MCM-48 catalysts. Catal Today
2014;232:13950.
[4] Yan K, Wu GS, Laeur T, Jarvis C. Production, properties and catalytic
hydrogenation of furfural to fuel additives and value-added chemicals. Renew
Sust Energy Rev 2014;38:66376.
[5] Haryanto A, Fernando S, Murali N, Adhikari S. Current status of hydrogen
production techniques by steam reforming of ethanol: a review. Energy Fuels
2005;19:2098106.
[6] Ni M, Leung DYC, Leung MKH. A review on reforming bio-ethanol for hydrogen
production. Int J Hydrogen Energy 2007;32:323847.
[7] Divins NJ, Lpez E, Rodrguez , Vega D, Llorca J. Bio-ethanol steam reforming and
autothermal reforming in 3-lm channels coated with RhPd/CeO2 for hydrogen
generation. Chem Eng Process 2013;64:317.
[8] Wu Y, He YM, Chen T, Weng WZ, Wan HL. Low temperature catalytic
performance of nanosized TiNiO for oxidative dehydrogenation of propane to propene.
Appl Surf Sci 2006;252:52206.
[9] Abell S, Bolshak E, Montan D. NiFe catalysts derived from hydrotalcite-like
precursors for hydrogen production by ethanol steam reforming. Appl Catal A
2013;450:26174.
[10] He L, Chen D. Hydrogen production from glucose and sorbitol by sorptionenhanced steam reforming: challenges and promises. ChemSusChem
2012;5:58795.
[11] Szijjrt GP, Pszti Z, Saj I, Erdhelyi A, Radnczi G, Tompos A. Nature of the active
sites in Ni/MgAl2O4-based catalysts designed for steam reforming of ethanol. J Catal
2013;305:290306.
[12] Garbarino G, Campodonico S, Perez AR, Carnasciali MM, Riani P, Finocchio E, et
al. Spectroscopic characterization of Ni/Al2O3 catalytic materials for the steam
reforming of renewables. Appl Catal A 2013;452:16373.
[13] Lindo M, Vizcano AJ, Calles JA, Carrero A. Ethanol steam reforming on Ni/Al- SBA-15
catalysts: effect of the aluminium content. Int J Hydrogen Energy
2010;35:5895901.
[14] Devianto H, Li ZL, Yoon SP, Han J, Nam SW, Lim TH, et al. The effect of Al
addition on the prevention of Ni sintering in bio-ethanol steam reforming for molten
carbonate fuel cells. Int J Hydrogen Energy 2010;35:25916.
[15] Huang ZJ, Wu PX, Li HL, Li W, Zhu YJ, Zhu NW. Synthesis and catalytic
properties of La or Ce doped hydroxyFeAl intercalated montmorillonite used as
heterogeneous photo Fenton catalysts under sunlight irradiation. RSC Adv
2014;4:65007.
[16] Yamauchi Y, Itagaki T, Yokoshima T, Kuroda K. Preparation of Ni nanoparticles between
montmorillonite layers utilizing dimethylaminoborane as reducing agent. Dalton Trans
2012;41:12105.
[17] Lee SD, Park MS, Kim WD, Kim I, Park WD. Catalytic performance of ionexchanged
montmorillonite
with
quaternary
ammonium
salts
for
the
glycerolysis of urea. Catal Today 2014;232:12733.

[18] Khanikar N, Bhattacharyya KG. Cu(II)kaolinite and Cu(II)montmorillonite as catalysts

for wet oxidative degradation of 2-chlorophenol, 4-chlorophenol and
2,4-dichlorophenol. Chem Eng J 2013;233:8897.
[19] Hao QQ, Wang GW, Zhao YH, Liu ZT, Liu ZW. FischerTropsch synthesis over
cobalt/montmorillonite promoted with different interlayer cations. Fuel
2013;109:3342.
[20] Zhao YH, Hao QQ, Song YH, Fan WB, Liu ZT, Liu ZW. Cobalt supported on
alkaline-activated montmorillonite as an efcient catalyst for FischerTropsch synthesis.
Energy Fuels 2013;27:636271.
[21] Sing KSW, Everett DH, Haul RAW, Moscou L, Pierotti RA, Rouquerol J, et al.
Reporting physisorption data for gas/solid systems with special reference to the
determination of surface area and porosity. Pure Appl Chem
1985;57:60319.
[22] de Jong KP, Zecevic J, Friedrich H, de Jongh PE, Bulut M, van Donk S, et al.
Zeolite Y crystals with trimodal porosity as ideal hydrocracking catalysts. Angew
Chem Int Ed 2010;49:100748.
[23] Yan K, Wu GS, Jarvis C, Wen JL, Chen AC. Facile synthesis of porous
microspheres composed of TiO2 nanorods with high photocatalytic activity for
hydrogen production. Appl Catal B 2014;148:2817.
[24] Kumar P, Jasra RV, Bhat TSG. Evolution of porosity and surface acidity in
montmorillonite clay on acid activation. Ind Eng Chem Res 1995;34:14408.
[25] Mas V, Baronetti G, Amadeo N, Laborde M. Ethanol steam reforming using
Ni(II)Al(III) layered double hydroxide as catalyst precursor: kinetic study. Chem Eng
J 2008;138:6027.
[26] Hu DC, Gao JJ, Ping Y, Jia LH, Gunawan P, Zhong ZY. Enhanced investigation of
CO methanation over Ni/Al2O3 catalysts for synthetic natural gas production. Ind Eng
Chem Res 2012;51:487586.
[27] Tanksale A, Beltramini JN, Dumesic JA, Lu GQ. Effect of Pt and Pd promoter on
Ni supported catalysts-A TPR/TPO/TPD and microcalorimetry study. J Catal
2008;258:36677.
[28] Zhang JF, Bai YX, Zhang QD, Wang XX, Zhang T, Tan YS, et al. Low-temperature
methanation of syngas in slurry phase over Zr-doped Ni/c-Al2O3 catalysts prepared
using different methods. Fuel 2014;132:2118.
[29] He L, Berntsen H, Ochoa-Fernndez E. CoNi catalysts derived from
hydrotalcite-like
materials
for
hydrogen
production
by
ethanol
steam
reforming. Top Catal 2009;52:20617.
[30] Snchez-Snchez MC, Navarro RM, Fierro JLG. Ethanol steam reforming over
Ni/MxOyAl2O3 (M = Ce, La, Zr and Mg) catalysts: inuence of support on the
hydrogen production. Int J Hydrogen Energy 2007;32:146271.
[31] Hung CC, Chen SL, Liao YK, Chen CH, Wang JH. Oxidative steam reforming of ethanol
for hydrogen production on M/Al2O3. Int J Hydrogen Energy
2012;37:495566.
[32] Luy JC, Parera JM. Acidity control in alcohol dehydration. Appl Catal
1986;26:295304.
[33] Di Cosimo JI, Dez VK, Xu M, Iglesia E, Apestegu CR. Structure and surface and catalytic
properties of MgAl basic oxides. J Catal 1998;178:499510.
[34] Garca V, Lpez E, Serra M, Llorca J. Dynamic modelling of a three-stage low- temperature
ethanol reformer for fuel cell application. J Power Sources
2009;192:20815.
[35] Garca V, Lpez E, Serra M, Llorca J, Riera J. Dynamic modeling and
controllability analysis of an ethanol reformer for fuel cell application. Int J
Hydrogen Energy 2010;35:976875.
[36] Wang JH, Lee CS, Lin MC. Mechanism of ethanol reforming: theoretical
foundations. J Phys Chem C 2009;113:66818.
[37] Xu J, Zhang X, Zenobi R, Yoshinobu J, Xu Z, Yates Jr JT. Ethanol decomposition on Ni (1 1
1): observation of ethoxy formation by IRAS and other methods. Surf Sci 1991;256:288
300.
[38] Fishtik I, Alexander A, Datta R. A thermodynamic analysis of hydrogen
production bysteam reforming of ethanol via response reactions. Int J Hydrogen
Energy 2000;25:3145.
[39] Garca V, Serra M, Llorca J, Riera J. Design of linear controllers applied to an
ethanol steam reformer for PEM fuel cell applications. Int J Hydrogen Energy
2013;38:76406.
[40] Mattos LV, Jacobs G, Davis BH, Noronha FB. Production of hydrogen from ethanol:
review of reaction mechanism and catalyst deactivation. Chem Rev
2012;112:4094123.
[41] Pirez C, Capron M, Jobic H, Dumeignil F, Jalowiecki-Duhamel L. Highly efcient and
stable CeNiHZOY nano-oxyhydride catalyst for H2 production from ethanol at room
temperature. Angew Chem Int Ed 2011;50:101937.
[42] Choong CKS, Huanga L, Zhong ZY, Lin JY, Hong L, Chen LW. Effect of calcium
addition on catalytic ethanol steam reforming of Ni/Al2O3: II. Acidity/basicity, water
adsorption and catalytic activity. Appl Catal A 2011;407:15562.