Journal of the Energy Institute xxx (2016) 1e9

Contents lists available at ScienceDirect

Journal of the Energy Institute

journal homepage: http://www.journals.elsevier.com/journal-of-the-energy-
institute

Investigation of Ni/SiO2 catalysts prepared at different conditions for

hydrogen production from ethanol steam reforming
Chunfei Wu a, b, **, Valerie Dupont a, Mohamad Anas Nahil a, Binlin Dou c,
Haisheng Chen d, Paul T. Williams a, *
a
School of Chemical & Process Engineering, University of Leeds, Leeds, LS2 9JT, UK
b
School of Chemical Engineering, University of Hull, Hull, HU6 7RX, UK
c
School of Energy and Power Engineering, Dalian University of Technology, 116023, Dalian, China
d
Institute of Engineering Thermophysics, Chinese Academy of Sciences, Beijing, China

a r t i c l e i n f o a b s t r a c t

Article history: Ni/SiO2 catalysts prepared by a solegel method have been investigated for hydrogen production via
Received 13 February 2015 steam reforming of ethanol using a continuous ow, xed bed reactor system. Chemical equilibrium
Received in revised form calculations were also performed to determine the effects of temperature and molar steam to carbon
9 December 2015
ratio on hydrogen production. The acidity of the preparation solution (modied by nitric acid and
Accepted 4 January 2016
ammonia) and calcination atmosphere (air and N2) were investigated in the preparation of the catalysts.
Available online xxx
BET surface area and porosity, temperature-programmed oxidation (TPO), X-ray diffraction (XRD),
transmission electron microscopy (TEM) and scanning electron microscopy (SEM) were used to char-
Keywords:
Catalyst acterise the prepared catalysts. The BET surface area was reduced when the solution acidity was lowered
Nickel during the solegel preparation process. A pH value less than 2.0 was necessary to achieve high metal
Hydrogen dispersion in the catalyst. Smaller NiO particles were obtained when the catalyst was calcined in N2.
Ethanol Material balances on ethanol steam reforming at 600
C using the prepared Ni/SiO2 catalysts were
Reforming determined, and higher hydrogen production with lower coke deposition on the reacted catalysts were
also obtained from the catalysts calcined in N2 atmosphere.
2016 Energy Institute. Published by Elsevier Ltd. All rights reserved.

1. Introduction

Due to the environmental problems derived from utilization of fossil fuels and the consideration of energy security, more attention has
been paid to the development of new energy fuels [1e3]. Hydrogen is regarded as one of the promising energy carriers due to its high energy
mass density, waste-free combustion and the vast variety of raw materials that can be used for its production [4]. Among the oxygenated
hydrocarbons used for hydrogen production, ethanol is highly advantageous as it is readily available, safe to transport and handle [5,6], and
offers a high maximum theoretical yield of H2 via steam reforming (26.3 wt% of the ethanol feed) compared to other renewable feedstocks
(e.g. glycerol, acetic acid). For these reasons it has been extensively investigated in the steam reforming process [7e9].
Catalysts play an important role in hydrogen production from ethanol steam reforming by improving the production of hydrogen and the
efciency of the energy balance [10,11]. However, catalyst development remains an interesting topic for research due to the deactivation of
the catalyst during the ethanol steam reforming process. Although some studies have revealed that noble metal-based catalysts performed
well for ethanol steam reforming [12e14], nickel based catalysts are one of the most attractive catalysts that have been investigated because
of their effective catalytic activity and relatively low cost [15e17].
The solegel catalyst preparation method has been shown to confer high surface area and pore volume as well as high Ni dispersion for
nickel-based catalysts [18e20]. In addition, the physical and chemical properties of the produced catalyst have been signicantly inuenced
by the preparation methods using, for example, different Ni contents [21], solution acidity [22], solution reagent (citric acid etc.) [23], and

* Corresponding author. Tel.: 44 113 3432504.

** Corresponding author. Tel.: 44 1482466464.
E-mail addresses: [email protected] (C. Wu), [email protected] (P.T. Williams).

http://dx.doi.org/10.1016/j.joei.2016.01.002
1743-9671/ 2016 Energy Institute. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: C. Wu, et al., Investigation of Ni/SiO2 catalysts prepared at different conditions for hydrogen production from
ethanol steam reforming, Journal of the Energy Institute (2016), http://dx.doi.org/10.1016/j.joei.2016.01.002
2 C. Wu et al. / Journal of the Energy Institute xxx (2016) 1e9

catalyst calcination atmosphere [24]. Ni/SiO2 catalysts have been widely used in hydrogen production [25]. However, studies of the inuence
of different preparation conditions using solegel methods for the preparation of Ni/SiO2 catalysts and the inuence on ethanol steam
reforming are very limited.
In this paper, Ni/SiO2 catalysts have been prepared under different acidities of the solegel solution, and the catalyst precursor was
calcined under different atmospheres (air and N2). The physical and chemical properties of the catalysts were analysed and related to their
performance in relation to hydrogen production from the steam reforming of ethanol. The objective of the paper was to provide information
concerning the inuence of preparation conditions of Ni/SiO2 catalysts on hydrogen production from the catalytic steam reforming of
ethanol.

2. Experimental

2.1. Preparation of Ni/SiO2 catalyst

Ni/SiO2 catalysts with a Ni content of 20 wt.% were prepared by a simple solegel method adapted from the literature [26]. Ni(NO3)26H2O
(SigmaeAldrich), anhydrous citric acid (Alfa Aesar), deionized water, absolute ethanol (SigmaeAldrich) and tetraethyl silicate (TEOS)
(SigmaeAldrich) were used as raw materials. 0.01 mol of Ni(NO3)26H2O and 0.02 mol citric acid were rst dissolved into 200 ml of absolute
ethanol and stirred at 60
C for 3 h. Then, 7 ml of deionized water and a given amount of nitric acid (HNO3/TEOS molar ratio of 0.04, 0.12 or
0.20) or given amount of ammonia solution (NH4OH/TEOS molar ratio of 0.04 or 0.20) were added into the solution. The pH value of the
solution was recorded. The molar ratio of the TEOS/HNO3 or TEOS/NH4OH was selected according to reported work [27,28], and the iso-
electric point of silica (pH value of 2.0). 8.7 ml of TEOS were then added to the solution. After drying at 80
C in a water bath, the precursor
was calcined at 450
C in an air or N2 atmosphere for 3 h. The prepared catalysts were assigned as Ni/SiO2-1 to Ni/SiO2-8, respectively, with
the preparation conditions reported in Table 1 (A-agent; B-agent/TEOS molar ratio; C-calcination atmosphere). All the catalysts used in this
paper were crushed and sieved to granules with a size between 0.08 and 0.20 mm.

2.2. Characterization of catalysts

The BET (Brunauer, Emmett and Teller) surface area and the porosity of the prepared catalysts were determined using a NONA 2200e
Surface Area and Pore Size Analyzer. Samples were initially degassed under vacuum for 3 h before surface analysis. The system operates by
measuring the quantity of nitrogen adsorbed onto or desorbed from a solid sample at different equilibrium vapour pressures.
The temperature-programmed oxidation (TPO) of the reacted catalysts was carried out using a Stanton-Redcroft thermogravimetric
analyser (TGA and DTG) to determine the properties of the coked carbons deposited on the reacted catalysts. About 20 mg of the reacted
catalyst was heated in air at 15
C min1 to a nal temperature of 800
C, with a dwell time of 10 min with air ow around 50 ml min1.
Ni/SiO2-1, Ni/SiO2-4, Ni/SiO2-5 and Ni/SiO2-8 catalysts, i.e. representatives of the low/high pH and the Air/N2 calcination atmosphere
used in the preparation method, were analysed by X-ray diffraction (XRD). The analysis was carried out with a Philips PW 1050 Goniometer
using a PW 1730 with a CuKa radiation X-ray tube. The sample was ground to less than 75 mm size and loaded into the 20 mm aperture of an
aluminium sample holder.
A high resolution scanning electron microscope (SEM, LEO 1530) was used to characterise and examine the fresh catalysts and the
characteristics of the carbon deposited on the coked catalysts. Transmission electron microscopy (TEM) (FEI Tecnai TF20 FEG) coupled with
an energy dispersive X-ray spectroscope (EDXS) was used to determine the fresh and reacted catalysts. For the TEM analysis, the samples
were ground, dispersed with methanol, and then deposited on a Cu grid covered with a perforated carbon membrane.

2.3. Ethanol steam reforming with the Ni/SiO2 catalysts

The prepared Ni/SiO2 catalysts were investigated for hydrogen production via steam reforming of ethanol using a bench scale xed bed
reaction system. A schematic diagram of the reaction system and experimental process have been presented in our previous work [29].
During the experiment, a mixture of ethanol and deionized water was introduced into a pre-heated reactor section (190
C) with a total
ow rate of 3.44 g h1, and was catalytically reformed in a second reactor section, where 0.8 g of the prepared Ni/SiO2 catalyst (not pre-
reduced) was placed. N2 was used as a carrier gas during the experiments with a ow rate of 80 ml min1. It should be noted that the
catalyst would be expected to be initially auto-reduced during the reforming process, and as shown later, that the consumption of ethanol
for catalyst reduction was negligible in this work. Chemical equilibrium of the ethanol/water/N2 system was predicted at different molar
steam to carbon ratios (S/C) up to 4, and for a range of temperatures up to 700
C, using the code Chemical Equilibrium and Applications

Table 1
Surface properties of investigated catalysts prepared at different conditions.

Catalyst Conditionsa A-B-C pHb BET (m2 g1) BJH pore volume (cm3 g1) BJH pore diameter (nm)
Ni/SiO2-1 HNO3-0.20-Air 1.00 829 0.665 3.8
Ni/SiO2-2 HNO3-0.04-Air 1.10 737 0.698 3.8
Ni/SiO2-3 NH4OH-0.04-Air 1.50 707 0.640 3.8
Ni/SiO2-4 NH4OH-0.20-Air 2.08 605 0.575 3.2e5.0
Ni/SiO2-5 HNO3-0.20-N2 1.00 810 0.581 3.8
Ni/SiO2-6 HNO3-0.04-N2 1.10 807 0.590 3.8
Ni/SiO2-7 NH4OH-0.04-N2 1.50 797 0.619 3.8
Ni/SiO2-8 NH4OH-0.20-N2 2.08 545 0.309 3.8e5.0
a
A-agent; B-agent/TEOS molar ratio; C-calcination atmosphere.
b
pH value was determined after addition of agent (HNO3 or NH4OH).

Please cite this article in press as: C. Wu, et al., Investigation of Ni/SiO2 catalysts prepared at different conditions for hydrogen production from
ethanol steam reforming, Journal of the Energy Institute (2016), http://dx.doi.org/10.1016/j.joei.2016.01.002
 C. Wu et al. / Journal of the Energy Institute xxx (2016) 1e9 3

(CEA) [30] which relies on minimisation of Gibbs energy. As shown in Fig. 1, and in agreement with Le Chatelier's principle, for a given
temperature, hydrogen yield in wt.% of the ethanol feed increased signicantly with the rise in S/C ratio from 0 to 3; however, only a slight
incremental increase was obtained at a S/C of 4. We have shown in a previous study that as S/C approaches 6, the energy balance of ethanol
steam reforming at atmospheric pressure equals that of thermal water splitting [31], negating the benets of using ethanol as a source of
hydrogen. This is caused by the energy burden of raising excess steam. It then becomes inefcient to carry out ethanol steam reforming at a
S/C larger than 4 at atmospheric pressure. Fig. 1 also shows that the equilibrium hydrogen yield was maximized at the reforming tem-
perature of 600
C with the S/C of 4. For temperatures above 600
C, reverse water gas shift was favoured, thus reducing the overall H2 yield.
Based on these chemical equilibrium calculations, the S/C ratio of 4 and reforming temperature of 600
C were used to evaluate experi-
mentally the performance of the catalysts in this work.
Condensable products were collected by using an air cooled condenser and a dry ice cooled condenser. The non-condensed gases,
derived from around 1.5 h reaction time, were collected with a Tedlar gas sample bag and analyzed for their components concentration by
gas chromatography (GC). The conditions of GC and the gas analysis process have been described in our previous report [29]. Gas yield,
calculated from the mass of gaseous products divided by the mass of the injected sample, and liquid yield, determined from the mass of
condensed products divided by the mass of the injected sample, are presented in this work. Experiments were repeated to ensure the
reliability of the results.

3. Results and discussion

3.1. Characterization of the Ni/SiO2 catalysts

3.1.1. XRD analysis

Crystal phases in the prepared Ni/SiO2 catalysts were identied using XRD analysis. As shown in Fig. 2, an amorphous peak at around 23
,
assigned to SiO2, was obtained for all the catalysts. Only a crystalline NiO phase was observed for the catalysts calcined in air, while a
crystalline Ni phase was observed in the catalyst calcined in N2 (Fig. 2). Similar XRD patterns for Ni/SiO2 catalyst calcined in air have been
reported in other work [32,33]. The Ni crystal phase was also found in the catalyst calcined in N2 [24].

3.1.2. Surface analysis

Surface properties of the prepared Ni/SiO2 catalysts are shown in Table 1. BET surface area was reduced from 829 to 605 m2 g1 for the
catalyst calcined in air when the solution acidity was reduced, corresponding to an increase of pH value from 1 to 2; however, there was no
obvious relationship between the solution acidity and the BJH pore volume. Small changes were observed for the BET surface area for the
catalysts calcined in N2; however, the lowest BET surface area (545 m2 g1) was obtained for the Ni/SiO2-8 catalyst which was prepared
under the most basic solegel solution (Table 1). Results of BJH pore diameter indicate that about 3.8 nm pore diameter was obtained for all
the prepared catalysts except those produced under a NH4OH/TEOS molar ratio of 0.2.
The pore size distribution (Fig. 3 and 4) shows that the Ni/SiO2-4 and Ni/SiO2-8 catalysts prepared with the lowest acidity had a smaller
fraction of pores at ~3.8 nm, compared with catalysts prepared under other acidities. The N2 adsorption/desorption isotherms of the selected
catalysts are shown in Fig. 3, which suggest that the catalysts show a combination of type I and IV (IUPAC classication) indicating a meso-
porous structure in the catalysts [34]. Compared with catalysts calcined in N2, the Ni/SiO2-1 and Ni/SiO2-4 prepared in air showed higher N2
adsorption when the P/P0 ratio was higher than 0.5. It is suggested that the catalysts prepared in air had more meso-pores compared with
the catalysts prepared in N2. In terms of the pH of the preparation solution, a lower pH value seemed to produce a catalyst with more meso-

25
S/C = 0
S/C = 0.5
20
S/C = 1.5
H2 yield (wt% etOH)

S/C = 3
15
S/C = 4

10

0
200 300 400 500 600 700
T (C)
Fig. 1. H2 yield weight percent of ethanol feed from the EtOH/H2O/N2 equilibrium system at various reforming temperatures and steam/carbon ratios (equilibrium study). Other
experimental conditions: preheating temperature-190
C, carrier gas (N2) ow rate-80 ml min1; total raw material (ethanol and water) ow: 3.44 g h1.

Please cite this article in press as: C. Wu, et al., Investigation of Ni/SiO2 catalysts prepared at different conditions for hydrogen production from
ethanol steam reforming, Journal of the Energy Institute (2016), http://dx.doi.org/10.1016/j.joei.2016.01.002
4 C. Wu et al. / Journal of the Energy Institute xxx (2016) 1e9

Fig. 2. XRD analysis for selected prepared catalysts; (a) Ni/SiO2-4, (b) Ni/SiO2-8, (c) Ni/SiO2-1, (d) Ni/SiO2-5.

Fig. 3. N2 adsorption/desorption isotherms of the selected fresh catalysts.

and larger pores. For example, the Ni/SiO2-1 catalyst (pH 1.0) exhibited much higher N2 adsorption when the P/P0 ratio was higher than
0.5, compared with the Ni/SiO2-4 catalyst (pH 2.08) (Fig. 3). Therefore, from Table 1 and Fig. 3, it is suggested that total surface area and
pore volume are slightly changed, when the pH value of the catalyst preparation solution was lower than 2.0. However, both pore size and
surface area are signicantly reduced resulting from the reduction of pore size, when the pH value of catalyst preparation solution was above
2.0.

3.1.3. TEM analysis

Transmission electron microscope analysis was carried out on the fresh calcined catalysts, and the results are shown in Fig. 4. Smaller
particle sizes were obtained for the catalysts prepared in N2 compared to those prepared in air. In addition, similar particle dispersion was
obtained for the catalysts calcined under air and nitrogen atmosphere, respectively, except those prepared at a NH4OH/TEOS molar ratio of
0.20. Unexpectedly large particles were observed for the Ni/SiO2-4 and Ni/SiO2-8 catalysts (Fig. 4). It should be noted that precipitation
occurred when the ammonia solution (NH4OH/TEOS molar ratio of 0.20) was added to the solution during the solegel preparation process;
and this was observed before adding the TEOS to the solution. The large particles in the Ni/SiO2-4 and Ni/SiO2-8 catalysts were determined
to be SiO2 through EDXS analysis. The observation of large SiO2 particles with the Ni/SiO2-4 and Ni/SiO2-8 catalysts is also consistent with
the surface analysis, where these two catalysts exhibited obvious lower porosity compared to the other catalysts. It can be suggested that
condensation was dominant (Reaction (2) and (3)) during the preparation of Ni/SiO2-4 and Ni/SiO2-8 catalysts where the co-precipitated
solution was more basic [24], resulting in the production of large SiO2 particles. Therefore, in this work, a pH value of the preparation
solution of less than 2.0 is suggested in order to obtain a high NiO dispersion.

Si(OC2H5)4 4H2O / Si(OH)4 4C2H5OH (1)

Si(OH)4 Si(OH)4 / SiOSi 4H2O (2)

Si(OH)4 Si(OC2H5)4 / SiOSi 4C2H5OH (3)

Please cite this article in press as: C. Wu, et al., Investigation of Ni/SiO2 catalysts prepared at different conditions for hydrogen production from
ethanol steam reforming, Journal of the Energy Institute (2016), http://dx.doi.org/10.1016/j.joei.2016.01.002
 C. Wu et al. / Journal of the Energy Institute xxx (2016) 1e9 5

Fig. 4. TEM analysis of the prepared fresh catalysts.

Please cite this article in press as: C. Wu, et al., Investigation of Ni/SiO2 catalysts prepared at different conditions for hydrogen production from
ethanol steam reforming, Journal of the Energy Institute (2016), http://dx.doi.org/10.1016/j.joei.2016.01.002
6 C. Wu et al. / Journal of the Energy Institute xxx (2016) 1e9

3.2. Ethanol steam reforming using the Ni/SiO2 catalyst

3.2.1. Mass balance and hydrogen production

As seen previously in Fig. 1, hydrogen yield, expressed as weight percent of ethanol feed is expected, from equilibrium calculations, to
increase with the increase of temperature to 600
C as well as with S/C, and then declines slightly above 600
C due to reverse water gas
shift.
The experimental hydrogen production and mass balance during ethanol steam reforming with the eight prepared Ni/SiO2 catalysts at
600
C and S/C of 4 are shown in Table 2. Results in Table 2 reveal notable differences between the catalysts calcined in air and in N2.
Measured gas yields were higher for the catalysts calcined in N2, with a yield around 64 wt.% of the total feed (ethanol water), compared to
a gas yield around 57 wt.% for the catalysts calcined in air. Measured liquid products show a slight difference. Measured carbon deposits on
the catalyst calculated from the weight loss proles under TPO were about 0.15 wt.% of feed. Overall the products balance closure was
signicantly better for the N2 than for the air-calcined catalysts (2.5 wt.% vs. 6.4 wt.%). It is estimated that the balance closure value (sum of
the measured product yields as a difference to 100 wt.%) was caused by carbon rich deposits (e.g tars) forming on the reactor which could
not subsequently be recovered, as well as by experimental errors. It is likely that the larger this value, the more signicant were the un-
measured carbon rich deposits.
Finally, the measured H2 yield expressed in percent of the equilibrium value of 23.2 wt.% of EtOH was higher for the N2 than for the air
calcined catalysts, averaging 71 wt.% vs. 67 wt.%. These values also indicated the reactor conditions were some distance from equilibrium,
which allowed for better comparison between catalyst's activities. A calculation was performed to assess whether initial auto-reduction of
the catalysts calcined in air as per reaction (4), which would have used the ethanol feed without net H2 production, could have accounted for
the measured difference in H2 yields between the air-calcined and the N2-calcined catalysts.

C2H5OH 6 NiO 4 6 Ni 2CO2 3H2O (4)

For 0.8 g of catalyst with 20 wt.% Ni loading, 2.726 mmol of NiO would have required reduction, representing an ethanol consumption of
1.7 wt.% of the ethanol feed over 1.5 h of experiment. Correcting the H2 yield in wt.% of ethanol feed when the latter was diminished by
1.7 wt.% only increased the H2 yield by 1 wt.%, whereas the average gap between H2 yields by air-calcined (uncorrected for autoreduction)
and N2-calcined catalysts was of 4 wt.%. This indicated that the benets in H2 yield of performing catalyst calcination in a nitrogen at-
mosphere rather than air on the H2 yield were genuine. Clearly signicant additional benets were shown in the improved balance closure,
lower coking on the catalyst, and higher gas yield to the detriment of the liquid yield, which includes unconverted ethanol and water.
When examining more closely which of the N2-calcined catalysts performed the most effectively, catalyst Ni/SiO2-7 provided the highest
H2 yield and purity due to a combination of lower carbon-products selectivity to methane and to CO (Table 3). Methane and CO by-product
represent a penalty in H2 yield and purity due to lack of conversion by steam methane reforming and by water gas shift respectively,
therefore low selectivity in both CH4 and CO is desirable. Table 3 shows the selectivity to the carbon and hydrogen containing products for all
the catalysts, compared to the calculated equilibrium values. These show that selectivity to CH4 in all the experiments was signicantly
higher and selectivity to CO2 lower than the equilibrium value for both calcination atmospheres (~11% for air-calcined catalysts, ~12% for N2-
calcined catalysts, compared to ~2% at equilibrium), consistent with a kinetically favoured exothermic methanation to the detriment of the
endothermic steam methane reforming.
The more effective performance of the catalysts for hydrogen production prepared under N2 calcination is ascribed to the higher Ni metal
dispersion in the catalyst. From XRD analysis (Fig. 2), Ni diffraction peaks are barely observed for the catalysts calcined under N2, indicating
ne Ni particles are presented. In contrast, sharp diffraction could be clearly observed for the catalysts calcined under air atmosphere,
indicating the presence of large NiO particles. Additionally, a particle size of around 20 nm can be clearly observed from TEM analysis for the
Ni/SiO2 (1e3) catalysts calcined under air, while much smaller particles of around 4 nm could be found for the Ni/SiO2 (5e7) catalysts
calcined in N2.
Although calcination atmosphere has shown signicant inuence on hydrogen production from ethanol steam reforming, little changes
in gas yield could be observed for the catalysts prepared under different acidities (Table 2). This phenomenon might be due to the simi-
larities of surface properties and chemical properties of the catalysts (similar level of surface area and metal particle size) prepared at

Table 2
Mass balance and gas compositions for ethanol steam reforming with the prepared catalysts, carrier gas is included in the gas composition. Equilibrium H2 yield was 23.2 wt%
of etOH.

Reducing atmosphere Air N2

Catalyst Ni/SiO2-x. x 18 1 2 3 4 5 6 7 8

Mass balance (wt.% of feed)
Measured gas yield 57.5 56.7 58.9 58.2 65.3 66.4 62.2 63.0
Measured liquid yield 35.9 35.5 36.1 34.8 32.3 31.3 35.4 33.6
Calculated balance closure 6.5 7.6 4.8 6.8 2.3 2.2 2.2 3.3
Coke formation (wt.%) 1.8 2.9 3.8 4.0 1.5 1.6 2.8 2.3
Hydrogen yield (% Eq. value) 66.5 66.6 68.4 66.7 70.7 71.6 72.0 69.2
Gas composition (Vol.%)
CO 4.2 4.3 4.8 4.5 5.3 5.4 4.5 5.4
H2 24.3 24.6 26.3 25.6 25.1 25.0 23.9 26.3
N2 63.8 63.4 60.8 61.7 61.3 61.4 64.2 59.7
CO2 6.3 6.2 6.6 6.7 6.5 6.5 6.0 6.8
CH4 1.4 1.5 1.6 1.5 1.8 1.9 1.4 1.8
H2 purity in gas (N2 free, Vol.%) 67.1 67.2 66.9 66.8 64.9 64.4 66.8 65.3

Please cite this article in press as: C. Wu, et al., Investigation of Ni/SiO2 catalysts prepared at different conditions for hydrogen production from
ethanol steam reforming, Journal of the Energy Institute (2016), http://dx.doi.org/10.1016/j.joei.2016.01.002
 C. Wu et al. / Journal of the Energy Institute xxx (2016) 1e9 7

Table 3
Experimental selectivity (%) to gas products and calculated equilibrium selectivity.

Reducing atmosphere Air N2 Calc.

Catalyst Ni/SiO2-x. x 18 1 2 3 4 5 6 7 8 Equil.

C-productsa
Sel CO 35.3 35.8 36.9 35.4 39.0 39.1 37.8 38.6 25.6
Sel CO2 52.9 51.7 50.8 52.8 47.8 47.1 50.4 48.6 72.1
SelC CH4 11.8 12.5 12.3 11.8 13.2 13.8 11.8 12.9 2.3
H-productsb
SelH CH4 10.3 10.9 10.8 10.5 12.5 13.2 10.5 12.0 1.7
Sel H2 89.7 89.1 89.2 89.5 87.5 86.8 89.5 88.0 98.2
a
e.g. Sel CO 100  Vol.% CO/Vol.% (CO CO2CH4)).
b
Sel H2 100  Vol.% H2/Vol.% (H22CH4); SelH CH4 100  2  vol% CH4/vol% (H22CH4)).

different acidity conditions calcined under the same atmosphere (Table 1). In addition, although some large particles could be observed for
the Ni/SiO2-4 and Ni/SiO2-8 catalysts, the majority of small metal particles, similar to other catalysts, could also be observed from the TEM
analysis (Fig. 4), for the catalysts calcined under N2 and air, respectively.

3.2.2. Coke formation on the reacted catalysts

Coke deposition is one of the major challenges for catalyst development during ethanol steam reforming. In this section, the reacted
catalysts were characterised by TPO experiments and SEM and TEM analysis for discussion of coke formation. TPO-TGA and TPO-DTG results
are shown in Fig. 5. The increasing mass peak in the TPO-TGA thermogram is assigned to oxidation of Ni particles which were reduced from
the NiO phases by the reducing gases such as ethanol, its reaction intermediates, H2 and CO during ethanol steam reforming. Previous work
[31] showed how ethanol alone and an ethanol/bio-oil mixture could successfully autoreduce NiO catalysts at 600
C and a S/C of 3.3. From
Fig. 5, one oxidation peak in the TPO-DTG analysis was obtained due to combustion of deposited carbons, which were assigned to la-
mentous carbons. In addition, the lamentous carbons were conrmed from the SEM and TEM results (Fig. 6).
Coke deposition after ethanol steam reforming for each reacted catalyst is presented in Table 2. It is noted that the possible overlapping
between Ni and coke oxidation was neglected during the TPO experiment. The amount of coke deposition was obtained from the TPO
analysis; calculated as the mass difference between the sample weight (after water evaporation) and the mass of residue divided by the
sample weight (after water evaporation). Low coke deposition was found on the Ni/SiO2 catalysts, however, the reaction time used in this
work of 1.5 h was short in relation to industrial scale processes. Longer reaction times, catalyst deactivation studies and catalyst recycling
studies are recommended for future work to determine the effectiveness of the catalysts in relation to coking characteristics. From Fig. 5 and

Fig. 5. TGA-TPO and DTG-TPO results of the selected reacted catalyst.

Please cite this article in press as: C. Wu, et al., Investigation of Ni/SiO2 catalysts prepared at different conditions for hydrogen production from
ethanol steam reforming, Journal of the Energy Institute (2016), http://dx.doi.org/10.1016/j.joei.2016.01.002
8 C. Wu et al. / Journal of the Energy Institute xxx (2016) 1e9

Fig. 6. SEM and TEM results of reacted Ni/SiO2 catalyst; (a) (b) typical SEM results, (c) (d) typical TEM results.

Table 2, more carbon deposition was observed on the reacted catalyst calcined in air. It is suggested that smaller Ni particle size benets the
prohibition of coke formation on the surface of the catalyst. Kong et al. [35] investigated toluene reforming by various nickel catalysts, and
reported that a larger amount of coke deposition was obtained for the catalyst (Ni/ZrO2 and Ni/SiO2) with large Ni particle size (23.6, and
26.8 nm, respectively) compared with the Ni/Al2O3 and Ni/MgO catalysts with Ni particle sizes less than 10 nm. Small Ni particle size was
also reported to prohibit coke formation on a Ni-based catalyst during steam methane reforming [36].

4. Conclusions

The results show that the catalyst prepared using the solegel method has high surface area (>700 m2 g1) and narrow pore size dis-
tribution (pore diameter is around 3.8 nm), except for the catalyst prepared at a TEOS/NH4OH ratio of 1:0.20. When the catalysts (Ni/SiO2-4
and Ni/SiO2-8 catalysts) were prepared with the lowest solution acidity, large SiO2 particles and the lowest porosity were obtained.
However, the solution acidity showed little inuence on the gas yield and hydrogen concentration.
Ni and NiO crystal phases were identied by XRD analysis in the catalyst calcined in N2, however, only one NiO phase was identied in
the catalyst calcined in air. Catalysts prepared in N2 showed a higher Ni dispersion and resulted in higher gas yield with higher hydrogen
production during ethanol steam reforming, compared with those calcined in air. The coke formation on the catalyst increased with the
increase of basicity of the preparation solution for the Ni/SiO2 catalyst.

Acknowledgements

This work was supported by the International Exchange Scheme from the Royal Society (IE121244).

References

[1] Y. Himri, A.S. Malik, A. Boudghene Stambouli, S. Himri, B. Draoui, Review and use of the Algerian renewable energy for sustainable development, Renew. Sust. Energy
Rev. 13 (2009) 1584e1591.
[2] S.H. Jensen, P.H. Larsen, M. Mogensen, Hydrogen and synthetic fuel production from renewable energy sources, Int. J. Hydrogen Energy 32 (2007) 3253e3257.
[3] F. Marias, S. Letellier, P. Cezac, J.P. Serin, Energetic analysis of gasication of aqueous biomass in supercritical water, Biomass Bioenergy 35 (2011) 59e73.

Please cite this article in press as: C. Wu, et al., Investigation of Ni/SiO2 catalysts prepared at different conditions for hydrogen production from
ethanol steam reforming, Journal of the Energy Institute (2016), http://dx.doi.org/10.1016/j.joei.2016.01.002
 C. Wu et al. / Journal of the Energy Institute xxx (2016) 1e9 9

[4] C. Wu, P.T. Williams, Hydrogen production by steam gasication of polypropylene with various nickel catalysts, Appl. Catal. B Environ. 87 (2009) 152e161.
[5] M. Murdoch, G.I.N. Waterhouse, M.A. Nadeem, J.B. Metson, M.A. Keane, R.F. Howe, J. Llorca, H. Idriss, The effect of gold loading and particle size on photocatalytic
hydrogen production from ethanol over Au/TiO2 nanoparticles, Nat. Chem. 3 (2011) 489e492.
[6] R. Espinal, A. Anzola, E. Adrover, M. Roig, R. Chimentao, F. Medina, E. Lo
pez, D. Borio, J. Llorca, Durable ethanol steam reforming in a catalytic membrane reactor at
moderate temperature over cobalt hydrotalcite, Int. J. Hydrogen Energy 39 (2014) 10902e10910.
[7] W. Cai, F. Wang, A. van Veen, C. Descorme, Y. Schuurman, W. Shen, C. Mirodatos, Hydrogen production from ethanol steam reforming in a micro-channel reactor, Int. J.
Hydrogen Energy 35 (2010) 1152e1159.
[8] A. Carrero, J.A. Calles, A.J. Vizcano, Effect of Mg and Ca addition on coke deposition over CueNi/SiO2 catalysts for ethanol steam reforming, Chem. Eng. J. 163 (2010)
395e402.
[9] S.I. Ito, K. Tomishige, Steam reforming of ethanol over metal-oxide-promoted Pt/SiO2 catalysts: effects of strong metal-oxide interaction (SMOI), Catal. Commun. 12
(2010) 157e160.
[10] A.L. Alberton, M.M.V.M. Souza, M. Schmal, Carbon formation and its inuence on ethanol steam reforming over Ni/Al2O3 catalysts, Catal. Today 123 (2007) 257e264.
[11] J. Comas, F. Marin ~ o, M. Laborde, N. Amadeo, Bio-ethanol steam reforming on Ni/Al2O3 catalyst, Chem. Eng. J. 98 (2004) 61e68.
[12] G.A. Deluga, J.R. Salge, L.D. Schmidt, X.E. Verykios, Renewable hydrogen from ethanol by autothermal reforming, Science 303 (2004) 993e997.
[13] F. Frusteri, S. Freni, Bio-ethanol, a suitable fuel to produce hydrogen for a molten carbonate fuel cell, J. Power Sources 173 (2007) 200e209.
[14] H. Idriss, M. Scott, J. Llorca, S.C. Chan, W. Chiu, P.Y. Sheng, A. Yee, M.A. Blackford, S. Pas, A.J. Hill, F.M. Alamgir, R. Rettew, C. Petersburg, S.D. Senanayake, M.A. Barteau, A
phenomenological study of the metaleoxide interface: the role of catalysis in hydrogen production from renewable resources, Chem. Sus. Chem. 1 (2008) 905e910.
[15] A.E. Galetti, M.F. Gomez, L.A. Arrua, A.J. Marchi, M.C. Abello, Study of CuCoZnAl oxide as catalyst for the hydrogen production from ethanol reforming, Catal. Commun. 9
(2008) 1201e1208.
[16] J.Y. Liu, C.C. Lee, C.H. Wang, C.T. Yeh, C.B. Wang, Application of nickelelanthanum composite oxide on the steam reforming of ethanol to produce hydrogen, Int. J.
Hydrogen Energy 35 (2010) 4069e4075.
[17] H. Muroyama, R. Nakase, T. Matsui, K. Eguchi, Ethanol steam reforming over Ni-based spinel oxide, Int. J. Hydrogen Energy 35 (2010) 1575e1581.
[18] M. Numata, R. Takahashi, I. Yamada, K. Nakanishi, S. Sato, Solegel preparation of Ni/TiO2 catalysts with bimodal pore structures, Appl. Catal. A Gen. 383 (2010) 66e72.
[19] R. Takahashi, S. Sato, S. Tomiyama, T. Ohashi, N. Nakamura, Pore structure control in Ni/SiO2 catalysts with both macropores and mesopores, Micropor. Mesopor. Mater.
98 (2007) 107e114.
[20] L. Zhao, K. Fang, D. Jiang, D. Li, Y. Sun, Solegel derived NieMo bimetallic carbide catalysts and their performance for CO hydrogenation, Catal. Today 158 (2010) 490e495.
[21] K. Peng, L. Zhou, A. Hu, Y. Tang, D. Li, Synthesis and magnetic properties of NieSiO2 nanocomposites, Mater. Chem. Phys. 111 (2008) 34e37.
[22] K.H. Wu, W.C. Huang, Effect of varying the acid to metal ion ratio R on the structural and magnetic properties of SiO2-doped NieZn ferrite, J. Solid State Chem. 177 (2004)
3052e3057.
[23] J.B. Pang, K.Y. Qiu, Y. Wei, Preparation of mesoporous silica materials with non-surfactant hydroxy-carboxylic acid compounds as templates via solegel process, J. Non-
Cryst. Solids 283 (2011) 101e108.
[24] R. Takahashi, S. Sato, T. Sodesawa, M. Kato, S. Takenaka, S. Yoshida, Structural and catalytic properties of Ni/SiO2 prepared by solution exchange of wet silica gel, J. Catal.
204 (2001) 259e271.
[25] S.K. Saraswat, K.K. Pant, Synthesis of hydrogen and carbon nanotubes over copper promoted Ni/SiO2 catalyst by thermocatalytic decomposition of methane, J. Nat. Gas.
Sci. Eng. 13 (2013) 52e59.
[26] N. Tang, L. L, W. Zhong, C. Au, Y. Du, Synthesis and magnetic properties of carbon-coated Ni/SiO2 core/shell nanocomposites, Sci. China Ser. G Phys. Mech. Astro. 52
(2009) 31e34.
[27] D.W. Lee, S.K. Ihm, K.H. Lee, Mesoporous silica framed by sphere-shaped silica nanoparticles, Micropor. Mesopor. Mater. 83 (2005) 262e268.
[28] R. Takahashi, S. Sato, T. Sodesawa, M. Kawakita, K. Ogura, High surface-area silica with controlled pore size prepared from nanocomposite of silica and citric Acid, J. Phys.
Chem. B 104 (2000) 12184e12191.
[29] C. Wu, P.T. Williams, Hydrogen production from steam reforming of ethanol with nano-Ni/SiO2 catalysts prepared at different Ni to citric acid ratios using a solegel
method, Appl. Catal. B Environ. 102 (2011) 251e259.
[30] B.J. McBride, S. Gordon, Computer Program for Calculation of Complex Chemical Equilibrium Compositions and Applications II, Users manual and program description,
NASA Lewis Research Centre, 1996.
[31] R. Md Zin, A.B. Ross, M. Jones, V. Dupont, Hydrogen from ethanol reforming with aqueous fraction of pine pyrolysis oil with and without chemical looping, Bioresour.
Technol. 176 (2015) 257e266.
[32] R. Takahashi, S. Sato, T. Sodesawa, M. Suzuki, N. Ichikuni, Ni/SiO2 prepared by solegel process using citric acid, Micropor. Mesopor. Mater. 66 (2003) 197e208.
[33] R. Takahashi, S. Sato, T. Sodesawa, M. Yoshida, S. Tomiyama, Addition of zirconia in Ni/SiO2 catalyst for improvement of steam resistance, App. Catal. A Gen. 273 (2004)
211e215.
[34] K.S.W. Sing, Reporting physisorption data for gas/solid systems with special reference to the determination of surface area and porosity (Recommendations 1984), Pure
Appl. Chem. (1985).
[35] M. Kong, J.M. Fei, S. Wang, W. Lu, X. Zheng, Inuence of supports on catalytic behavior of nickel catalysts in carbon dioxide reforming of toluene as a model compound of
tar from biomass gasication, Bioresour. Technol. 102 (2011) 2004e2008.
[36] K.O. Christensen, D. Chen, R. Ldeng, A. Holmen, Effect of supports and Ni crystal size on carbon formation and sintering during steam methane reforming, Appl. Catal. A
Gen. 314 (2006) 9e22.

Please cite this article in press as: C. Wu, et al., Investigation of Ni/SiO2 catalysts prepared at different conditions for hydrogen production from
ethanol steam reforming, Journal of the Energy Institute (2016), http://dx.doi.org/10.1016/j.joei.2016.01.002