- Email: [email protected]
alumina catalyst
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/alumina catalyst Jaekyeong Yoo, Seungwon Park, Ji Hwan Song, Sangbeom Yoo, In Kyu Song* School of Chemical and Biological Engineering, Institute of Chemical Processes, Seoul National University, Shinlimdong, Kwanak-ku, Seoul 151-744, South Korea
article info
abstract
Article history:
A series of butyric acid (BA)-assisted nickel/alumina catalysts (denoted as XBAN/A) with
Received 20 July 2017
different butyric acid/Ni molar ratio (X) were prepared by an impregnation method, and
Received in revised form
they were applied to the hydrogen production by steam reforming of natural gas. All the
5 September 2017
XBAN/A catalysts exhibited X-ray diffraction patterns indicative of highly stable nickel
Accepted 25 September 2017
aluminate phase. However, a XBAN/A catalyst with an appropriate amount of butyric acid
Available online xxx
showed the enhanced nickel dispersion because of steric hindrance of butyric acid shell surrounding nickel particle. Addition of butyric acid also increased methane adsorption
Keywords:
capacity of the catalysts, which was directly related to the catalytic performance. In the
Hydrogen production
steam reforming of natural gas, both natural gas conversion and hydrogen yield showed
Steam reforming of natural gas
volcano-shaped trends with respect to butyric acid/Ni molar ratio (X). Nickel dispersion of
Nickel/alumina catalyst
XBAN/A catalysts was well correlated with natural gas conversion and hydrogen yield over
Butyric acid
the catalysts. Natural gas conversion and hydrogen yield increased with increasing nickel dispersion. Among the catalysts tested, 0.25BAN/A catalyst with the highest nickel dispersion exhibited the best catalytic performance in the steam reforming of natural gas. © 2017 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
Introduction With increase in human population, global energy consumption increases every year [1]. The rapid population growth also causes imprudent industrial expansion, leading to serious environmental problems such as global warming and climate change [2]. Therefore, it is urgent to invest and develop sustainable alternative energy sources. Hydrogen is one of the most desirable energy carriers because of its cleanness and zero-emission property [3,4]. Stable supply chain and high energy density of hydrogen also facilitate the increase of annual hydrogen demand [5,6].
Most of hydrogen used in the world is produced by steam reforming processes, corresponding to the production of 96% of on-purpose hydrogen [7]. Natural gas, which is mainly composed of methane and ethane, is the most widely utilized hydrogen source in the steam reforming process because of its abundance. Steam reforming of natural gas can be described by the following equations.
CH4 þ H2O 4 CO þ 3H2 DHo298 K ¼ þ206.1 kJ/mol
(1)
C2H6 þ 2H2O 4 2CO þ 5H2 DHo298 K ¼ þ347.3 kJ/mol
(2)
* Corresponding author. E-mail address: [email protected] (I.K. Song). https://doi.org/10.1016/j.ijhydene.2017.09.148 0360-3199/© 2017 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
2
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
CO þ H2O 4 CO2 þ H2 DHo298
K
¼ 41.2 kJ/mol
(3)
Eqs. (1) and (2) represent steam reforming reactions to produce CO and H2 from methane and ethane, respectively. Water-gas shift reaction (Eq. (3)) accompanied by the reforming reactions allows an additional hydrogen production. Overall reactions in the presence of abundant water can be expressed as follows. CH4 þ 2H2O 4 CO2 þ 4H2 DHo298 K ¼ þ164.9 kJ/mol
(4)
C2H6 þ 4H2O 4 2CO2 þ 7H2 DHo298
(5)
K
¼ þ264.9 kJ/mol
According to the equations, steam reforming of natural gas is a highly endothermic reaction. Thus, typical steam reforming process is carried out at around 800 C to supply large amount of energy [8]. Such a high-temperature operation causes a hurdle for downsizing of reformers for mobile fuel cells. Therefore, it is needed to develop a new catalyst operated at low temperature with high catalytic activity and stability. Nickel-based catalysts have been commercially used in the steam reforming processes due to their high intrinsic activity for CeC bond cleavage and low cost [9,10]. However, nickelbased catalysts are known to show performance drop and catalyst deactivation due to carbon deposition at low reaction temperature [11,12]. According to the previous researches [13e15], these problems can be overcome by highly dispersing active metal species in the nickel-based catalysts. For example, Christensen et al. have reported that nickel catalysts supported on hydrotalcite materials showed the enhanced coke resistance with decreasing nickel particle size during the steam reforming of methane [13]. Zhu et al. have also reported that addition of Mg into nickel-silica catalyst induced high dispersion of active phase and inhibition of metal sintering, resulting in high catalytic activity and stability during the dry reforming of methane [14]. Furthermore, an ordered mesoporous nickel-alumina catalyst prepared with a structure modifier (trimethylbenzene) showed the enhanced mesoporous structure and formed small metallic nickel [15]. The modified catalyst exhibited a stronger resistance toward nickel sintering and a better catalytic performance than nickelalumina catalyst without modifier during the steam reforming of methane. In particular, it has been reported that addition of fatty acid into precursor solution increased nickel dispersion of nickel/silica catalysts prepared by an impregnation method, and the catalysts showed high catalytic activity and stability during the dry reforming of methane [16]. Therefore, modification of nickel dispersion supported on alumina with a suitable dispersing agent for enhancing catalytic performance in the steam reforming of natural gas would be worthwhile. In this work, a series of butyric acid-assisted nickel/ alumina catalysts were prepared using various compositions of butyric acid-nickel precursor solution. The catalysts were characterized by FT-IR, nitrogen adsorptionedesorption, XRD, TPR, H2-TPD, and CH4-TPD analyses. The effect of butyric acid addition on the characteristics and catalytic activities of nickel/alumina catalysts in the steam reforming of natural gas was investigated.
Experimental Preparation of catalysts Alumina support was preliminarily prepared via a sol-gel method. 25.7 g of aluminum (III) nitrate (Junsei, 98%) was dissolved in 160.5 ml of ethanol. 48.6 ml of propylene oxide (Samchun, 99%) was introduced to the aluminum precursor solution as a gelation agent. The mixed solution turned into a white gel composite after a few minutes. The gel was aged at room temperature for 2 days. The obtained gel was crushed into small pieces and washed with anhydrous ethanol. The gel was then placed in a convection oven at 80 C for 2 days for drying. The obtained matter was grinded into powder using a mortar and pestle. The powder was calcined at 700 C for 5 h. A series of butyric acid (BA)-assisted mesoporous nickel/ alumina catalysts were prepared via an impregnation method. 0.51 g of nickel (II) nitrate (Sigma-Aldrich, 97%) was dissolved in 10 ml of D.I. water, and different amount of butyric acid was dropped to the nickel precursor solution. The mixed solution was heated to 80 C, and 1.0 g of alumina support was added into the solution. Water was then evaporated at 80 C with vigorous stirring. The obtained grain was dried at 80 C overnight, and it was calcined at 700 C for 5 h. Nickel loading was kept at 10 wt% in all the catalysts, and butyric acid (BA)/Ni molar ratio was varied from 0 to 1. The prepared catalysts were denoted as XBAN/A (X ¼ 0, 0.1, 0.25, 0.5, and 1), where X represented the butyric acid (BA)/Ni molar ratio.
Characterizations In order to investigate surface chemical state of XBAN/A catalysts, infrared spectra of as-prepared catalysts were obtained using a Fourier transform infrared (FT-IR) spectrometer (ThermoScientific, Nicolet 6700) with a diffuse reflectance accessory. FT-IR spectra were recorded within the spectral range of 3500e650 cm1. Surface areas and porosity of XBAN/A catalysts were analyzed by nitrogen adsorptionedesorption method at 196 C (BELSORP-mini II, BEL Japan). For the measurement, each catalyst was preliminarily degassed at 150 C for 6 h. Crystallite phases of the catalysts were identified by X-ray diffraction (XRD) analyses with a D/ Max 2500V/PC X-ray diffractometer equipped with a Cu-Ka tube operated at 50 kV and 100 mA. The measurements were carried out at a rate of 10 /min within 2q range from 20 to 80 . For the XRD patterns of reduced XBAN/A catalysts, each calcined catalyst (50 mg) was preliminarily reduced under a hydrogen flow (3 ml/min) mixed with a nitrogen flow (30 ml/ min) at 700 C for 3 h. Temperature-programmed reduction (TPR) experiments were carried out in order to examine metal-support interaction of the catalysts. Each calcined catalyst sample (50 mg) was charged into a U-shaped quartz reactor, and it was reduced using a reducing gas mixture of hydrogen (2 ml/min) and nitrogen (20 ml/min) at a heating rate of 5 C/min from room temperature to 1000 C. TPR profiles of the catalysts were obtained using a thermal conductivity detector (TCD). Morphology and elemental
Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
3
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
Catalytic activity measurement Steam reforming of natural gas was conducted in a fixed-bed quartz reactor. Each catalyst (50 mg) was preliminarily reduced under a reducing gas of hydrogen (3 ml/min) and nitrogen (30 ml/min) at 700 C for 3 h. Steam reforming of natural gas was then carried out at 550 C for 1000 min. A feed stream composed of natural gas (5 ml/min), steam (11 ml/ min), and nitrogen (18 ml/min) was used during the reaction. Composition of natural gas was 92 vol% methane and 8 vol% ethane. The effluent gas was periodically sampled and analyzed using a gas chromatograph (ACME 6000, Younglin). The sampled gas was separated through Molecular Sieve 5 A (60/80 mesh, 2 ft 1/800 2 mm ID) and Porapak N (80/100 mesh, 20 ft 1/800 2 mm ID) columns. Each separated component was detected by a TCD. Conversion of natural gas and yield for hydrogen were defined as follows. Fi represents the molar flow rate of i species.
Natural gas conversionð%Þ ¼
FCH4 ;in þ FC2 H6 ;in FCН 4 ;out FC2 H6 ;out FCH4 ;in þ FC2 H6 ;in
Results and discussion Surface chemical state of as-prepared catalysts In order to investigate chemical state of butyric acid, asprepared XBAN/A (X ¼ 0, 0.1, 0.25, 0.5, and 1) catalysts were examined by FT-IR analyses (Fig. 1). XBAN/A catalysts with butyric acid addition (X 0.25) showed two small bands at 2975 cm1 and 2885 cm1, which were indicative of asymmetric CH2 stretch and symmetric CH2 stretch, respectively [17]. The intensity of CH2 stretch peaks increased with increasing butyric acid addition. Characteristic peaks indicative of asymmetric COO stretch (1565 cm1) and symmetric COO stretch (1465 cm1) were also detected in the catalysts with X 0.25 [18]. According to the previous research [17], bonding types of COO groups to catalyst surface can be determined by the wavenumber separation (△) between the asymmetric and the symmetric COO stretch. If △ is large (>200 cm1), monodentate interaction is expected. Small △ (<110 cm1) is for chelating bidentate mode. The medium range (140 cm1 < △ < 190 cm1) corresponds to bridging bidentate mode. In this work, △ (1565e1465 ¼ 100 cm1) was smaller than 110 cm1, indicating that butyric acid was bound to catalyst surface in a chelating bidentate mode. It should be noted that the peak at 1710 cm1, which was ascribed to C]O stretch of free butyric acid, appeared in the 1BAN/A catalyst [19]. This implies that butyric acid was presented as a separated form on the catalyst surface, when excess amount of butyric acid was introduced.
Textural properties of calcined catalysts Nitrogen adsorptionedesorption isotherms of calcined XBAN/ A catalysts are shown in Fig. 2. The obtained isotherms of all catalysts were type IV, indicating the formation of
CH2 stretch C=O stretch COO- stretch
Transmittance (A. U.)
distribution of the calcined catalysts were examined by scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS) mapping analyses using JSM-6701F (JEOL). High-resolution transmission electron microscopy (HR-TEM) experiments were carried out in order to examine the metal dispersion and lattice fringe of the reduced catalysts using a JEM-3010 (JEOL) apparatus operated at 300 kV. Hydrogen temperature-programmed desorption (H2-TPD) analyses were performed on reduced XBAN/A catalysts to measure nickel dispersion (BELCAT-B, BEL Japan). Each calcined catalyst (50 mg) was preliminarily reduced under a reducing gas of 5% hydrogen diluted with argon (50 ml/min) at 700 C for 3 h. Hydrogen adsorption was carried out under a 5% hydrogen flow diluted with argon (50 ml/min) at 50 C for 30 min. Elimination of physisorbed molecules was performed by purging with an argon gas (50 ml/min) at 100 C for 1 h. Desorption of chemisorbed molecules was monitored at 5 C/ min to 950 C under an argon flow (30 ml/min). A TCD was employed to detect the desorbed hydrogen. Methane temperature-programmed desorption (CH4-TPD) was performed to investigate methane adsorption capacity of reduced catalysts (BELCAT-B, BEL Japan). Each calcined catalyst (50 mg) was preliminarily reduced under a reducing gas of 5% hydrogen diluted with argon (50 ml/min) at 700 C for 3 h. Methane adsorption was carried out at 50 C under a mixed gas of 1% methane diluted with argon (50 ml/min) at 50 C for 30 min. Elimination of physisorbed methane was performed by purging with an argon gas (50 ml/min) at 250 C for 1 h. Desorption of chemisorbed methane was monitored at 5 C/ min within temperature range of 200e1000 C under an argon flow (30 ml/min).
0BAN/A 0.1BAN/A 0.25BAN/A 0.5BAN/A 1BAN/A
100 (6) Hydrogen yieldð%Þ ¼
FH2 ;out 100 4 FCH4 ;in þ 7 FC2 H6 ;in
(7)
3500
3000
2500
2000
1500
Wavenumber
(cm-1)
1000
Fig. 1 e FT-IR spectra of as-prepared XBAN/A catalysts.
Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
4
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
angle, indicating lattice expansion by the insertion of nickel species [24]. It is also noteworthy that broad and poorly defined peaks represent low crystallinity of XBAN/A catalysts, indicating fine dispersion of nickel species on g-Al2O3 support [25]. Fig. 3 (b) shows the X-ray diffractograms of reduced XBAN/A catalysts. Diffraction peaks which were attributed to nickel aluminate phase disappeared. Instead, all the catalysts showed diffraction peaks ascribed to metallic nickel (JCPDS 04-0850) phase. Moreover, the diffraction peak indicative of g-Al2O3 (440) shifted back to higher value, representing the exclusion of nickel cations from alumina lattice. Thus, it can be concluded that nickel species in the XBAN/A catalysts were sufficiently reduced from Ni2þ into Ni0 during the reduction step.
Volume adsorbed (A. U.)
0BAN/A
0.1BAN/A
0.25BAN/A
0.5BAN/A
SEM-EDS mapping and TEM analyses
1BAN/A
0
0.2
0.4
0.6
0.8
1.0
Relative pressure (P/P0) Fig. 2 e Nitrogen adsorptionedesorption isotherms of calcined XBAN/A catalysts.
mesoporous structure [20]. In addition, H2-type hysteresis loops were observed for all catalysts, representing the existence of “ink-bottle” type pores [21]. Table 1 shows the detailed textural properties of XBAN/A catalysts. All the catalysts showed high surface area (>250 m2), large pore volume (>0.5 cm3/g), and large pore diameter (>8.0 nm). This result means that the mesoporous structure of alumina support used for the preparation of XBAN/A catalysts was successfully prepared via a sol-gel method in this work [22].
XRD result of calcined and reduced catalysts X-ray diffractograms of calcined XBAN/A catalysts are presented in Fig. 3 (a). The crystalline phases were identified by the JCPDS standards. All the catalysts showed the characteristic diffraction peaks of nickel aluminate (JCPDS 10-0339) phase. This means that nickel aluminate spinel phase was formed during the calcination step. The formation of nickel aluminate phase is known to take place by the insertion of nickel cations into the tetrahedral vacancies of g-Al2O3 lattice [23]. Accordingly, diffraction peak of g-Al2O3 (440) slightly shifted to lower
Table 1 e Textural properties of calcined XBAN/A catalysts. Catalyst 0BAN/A 0.1BAN/A 0.25BAN/A 0.5BAN/A 1BAN/A a b
Surface area (m2/g)a
Pore volume (cm3/g)b
Average pore diameter (nm)
257 276 265 262 274
0.55 0.67 0.55 0.64 0.69
8.5 9.7 8.3 9.7 10.1
Calculated by the BET equation at P/P0 ¼ 0.05e0.30. Total pore volume at P/P0 ~ 0.990.
Distributions of aluminum, nickel, and oxygen in the calcined 0.25BAN/A catalyst were examined by SEM-EDS mapping analyses as shown in Fig. 4. The EDS elemental mapping image shows that nickel species (green dots in Fig. 4) was homogeneously dispersed on alumina support. TEM images of reduced 0.25BAN/A catalyst are shown in Fig. 5. It was estimated that crystallite size of nickel particle was smaller than 10 nm (Fig. 5 (a)). A large agglomerate of nickel particle was not observed in this sample. From the SEM-EDS and TEM results, it can be said that nickel species in the 0.25BAN/A catalyst was finely dispersed on alumina support. Fig. 5 (b) shows a latticeresolved TEM image of the reduced 0.25BAN/A catalyst including inter-planar spacing of nickel crystallite. The periodic lattice fringe with a d-spacing of 0.20 ± 0.004 nm is attributed to the (111) plane of metallic nickel [26,27]. No lattice fringe indicative of nickel oxide phase was observed. Therefore, it is believed that the nickel species in the 0.25BAN/ A catalyst was fully reduced into the metallic nickel phase during the reduction step. This result is in good agreement with the XRD results in Section XRD result of calcined and reduced catalysts.
TPR result of calcined catalysts Fig. 6 shows the TPR profiles of calcined XBAN/A catalysts. All the XBAN/A catalysts showed a reduction peak at around 700 C due to highly dispersed nickel aluminate spinel [28]. It was observed that the peak temperature of XBAN/A (X > 0) catalysts with butyric acid addition showed peak shift to low temperature compared to 0BAN/A catalyst. It can be inferred that butyric acid weakened the interaction between metal and support by surrounding metal particles. In particular, 1BAN/A catalyst showed an additional reduction peak at 489 C, indicating the reduction of bulk nickel oxide to metallic nickel [29]. It is believed that the excessively added butyric acid surrounding nickel species lifts up the nickel cluster from alumina, inhibiting the interaction between nickel and alumina.
H2-TPD result of reduced catalysts H2-TPD experiments on reduced XBAN/A catalysts were performed and their corresponding profiles are shown in Fig. 7. Three different hydrogen desorption peaks appeared in all the
Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
5
Fig. 3 e XRD patterns of (a) calcined and (b) reduced XBAN/A catalysts.
Fig. 4 e SEM-EDS images of calcined 0.25BAN/A catalyst obtained by mapping on aluminum, nickel, and oxygen.
XBAN/A catalysts. According to the literature [30], hydrogen desorption peak centered at around 200 C is ascribed to weakly adsorbed hydrogen molecules on active metal sites, and hydrogen desorption peak centered at around 500 C is due to strongly adsorbed hydrogen on metal sites. However, desorption peak above 600 C is assigned to hydrogen spillover species. Therefore, total amount of hydrogen uptake on active nickel species can be obtained by integrating deconvoluted peaks below 600 C. Deconvolution of the TPD profiles was performed using the OriginLab software. Nickel dispersion was calculated using the amount of hydrogen uptake as
summarized in Table 2. It should be noted that the amount of total hydrogen uptake increased with increasing butyric acid/ Ni ratio up to 0.25, and then the amount of hydrogen uptake decreased with increasing butyric acid/Ni ratio over 0.25. The nickel dispersion calculated from the amount of hydrogen uptake showed the same trend. According to the previous research [16], an appropriate amount of fatty acid surrounding nickel particles acts as a dispersing agent, inhibiting sintering of nickel particles during the thermal treatment. In the presence of excess fatty acid, however, nickel particles were fully surrounded by a fatty acid shell and uplifted from support,
Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
6
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
Fig. 5 e TEM images of reduced 0.25BAN/A catalyst: (a) TEM image and (b) HR-TEM image with lattice fringes.
706
Hydrogen desorbed (A. U.)
Hydrogen consumption (A. U.)
0BAN/A 0BAN/A 691 0.1BAN/A
683
0.25BAN/A 691 0.5BAN/A
696 489
0.1BAN/A
0.25BAN/A
0.5BAN/A
1BAN/A
1BAN/A
0
200
400
600
800
1000
100 200 300 400 500 600 700 800 900
Temperature (oC) Fig. 6 e TPR profiles of calcined XBAN/A catalysts.
resulting in more opportunity to migrate and combine with other metal particles nearby. In a similar manner in this work, high dispersion of nickel species could be achieved when medium amount of butyric acid was introduced.
Temperature (oC) Fig. 7 e H2-TPD profiles of reduced XBAN/A catalysts.
Table 2 e H2-TPD results of reduced XBAN/A catalysts. Catalyst
CH4-TPD result of reduced catalysts CH4-TPD experiments on reduced XBAN/A catalysts were performed and their corresponding profiles are shown in Fig. 8. Broad desorption peaks were observed for all the catalysts. For precise quantification, deconvolution of the CH4TPD profiles was conducted. According to the literature [31], the peak appearing below 700 C is known to be related to the desorption of methane from the reduced nickel species. On the other hand, methane desorption peak at around 800 C is related to the desorption of methane from alumina support [32]. In this work, therefore, deconvoluted desorption peak
Nickel Amount of desorbed hydrogen dispersionb (%) (mmol-H2/g)a Weak site Strong site Total (<300 C) (300e600 C)
0BAN/A 0.1BAN/A 0.25BAN/A 0.5BAN/A 1BAN/A
18 26 26 28 33
38 33 75 64 35
56 59 101 92 68
7.2 7.6 13.0 11.9 8.8
Measurement errors for the amount of desorbed hydrogen and nickel dispersion are ± 4 mmol-H2/g and ±0.3%, respectively. a Calculated from deconvoluted peak area of H2-TPD profiles in Fig. 7. b Calculated by assuming H/Niatom ¼ 1.
Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
7
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
Methane desorbed (A. U.)
0BAN/A
0.1BAN/A 0.25BAN/A
0.5BAN/A
1BAN/A
200
400
600
800
1000
Temperature (oC)
affected by butyric acid/Ni molar ratio. The amount of desorbed methane decreased in the order of 0.25BAN/A > 0.5BAN/ A > 1BAN/A > 0.1BAN/A > 0BAN/A. This trend was well consistent with the trend of nickel dispersion calculated in Section H2-TPD result of reduced catalysts (Table 2). In other words, a XBAN/A catalyst with higher nickel dispersion showed higher methane adsorption capacity. Qian et al. have reported that partially or fully dehydrogenated methane is desorbed from nickel/alumina catalyst below 700 C [32]. It has also been reported that dehydrogenated methane species are major reaction intermediates in the methane reforming [33,34]. From the above researches, it can be inferred that a nickel/alumina catalyst with high methane adsorption capacity will dissociate much amount of methane effectively in the reforming reaction. The dissociative adsorption of methane into dehydrogenated carbon species is known as the rate-determining step of reforming reactions [35e37]. Therefore, it is expected that the XBAN/A catalyst with higher methane adsorption capacity will show a better catalytic performance in the steam reforming of natural gas.
Fig. 8 e CH4-TPD profiles of reduced XBAN/A catalysts.
Catalytic activity in steam reforming of natural gas
area below 700 C was only considered for quantification. The amount of desorbed methane was calculate by integrating the deconvoluted peaks below 700 C as summarized in Table 3. It was found that the amount of desorbed methane was greatly
Table 3 e CH4-TPD results of reduced XBAN/A catalysts. Catalyst
0BAN/A 0.1BAN/A 0.25BAN/A 0.5BAN/A 1BAN/A
Amount of desorbed methane (mmol-CH4/g)a Weak site (<400 C)
Strong site (400e700 C)
Total
6.5 6.9 8.9 8.2 7.6
12.0 12.0 15.4 13.8 13.2
18.5 18.8 24.2 22.0 20.8
Measurement error for the amount of desorbed methane is ±0.2 mmol-CH4/g. a Calculated from deconvoluted peak area of CH4-TPD profiles in Fig. 8.
Fig. 9 shows the catalytic activities of XBAN/A catalysts in the steam reforming of natural gas at 550 C. The reaction was performed for 1000 min. It was observed that all the catalysts maintained initial catalytic activity with no significant deactivation. In commercial steam reforming process, nickelbased catalysts undergo activity drop caused by the growth of large nickel particles and the consequent reduction of active surface area [38]. Carbon formation on the catalyst surface is another problem which results in the loss of active sites by encapsulating the catalyst surface or blocking pores of the catalysts [39]. When considering the stable catalytic performance of XBAN/A catalysts, all the catalysts are believed to retain a strong resistance toward sintering and carbon deposition because of finely dispersed nickel particles on alumina support with well-developed mesopores. Fig. 10 shows the catalytic activity of XBAN/A catalysts in the steam reforming of natural gas plotted as a function of butyric acid (BA)/Ni molar ratio. Volcano-shaped trends were observed with respect to butyric acid/Ni molar ratio.
70
(a)
70
(b) 65
65
60
60 55
55 0BAN/A 0.1BAN/A 0.25BAN/A 0.5BAN/A 1BAN/A
50
0BAN/A 0.1BAN/A 0.25BAN/A 0.5BAN/A 1BAN/A
50
45 0
200
400
600
800
1000 0
200
400
600
800
45 1000
Fig. 9 e Catalytic performances with time on stream: (a) Natural gas conversion and (b) hydrogen yield. Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
8
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
72 66 70 64 68
62
66
60
Fig. 11 shows the correlations between hydrogen yield and nickel dispersion, and between hydrogen yield and methane adsorption capacity of XBAN/A catalysts. A distinctive feature is that hydrogen yield increased with increasing nickel dispersion and with increasing methane adsorption capacity. On the basis of H2-TPD and CH4-TPD analyses, it is believed that an optimal butyric acid/Ni molar ratio was required for the highest nickel dispersion and methane adsorption capacity, and in turn, for the highest hydrogen yield in the steam reforming of natural gas.
64
58
62
56
60
54
Conclusions
52
A series of butyric acid (BA)-assisted nickel/alumina (XBAN/A, X ¼ 0, 0.1, 0.25, 0.5, and 1) catalysts were characterized and evaluated in order to find an optimal butyric acid/Ni molar ratio (X) in the steam reforming of natural gas. It was found that characteristics of XBAN/A catalysts were greatly affected by the butyric acid/Ni molar ratio. All the prepared catalysts showed a mesoporous structure. In addition, all the calcined catalysts retained a nickel aluminate phase. Nickel dispersion of XBAN/A catalysts was improved when an appropriate amount of butyric acid was introduced. This was due to the dispersing effect of butyric acid in thermal treatment step. A XBAN/A catalyst with high nickel dispersion also showed high methane adsorption capacity. Although all the XBAN/A catalysts showed a stable catalytic performance in the steam reforming of natural gas, high catalytic performance was obtained when butyric acid was introduced. Natural gas conversion and hydrogen yield increased with increasing nickel dispersion and with increasing methane adsorption capacity. Among the catalysts tested, 0.25BAN/A catalyst with the highest nickel dispersion and methane adsorption capacity showed the best catalytic performance in terms of natural gas conversion and hydrogen yield.
58 0
0.2
0.4
0.6
0.8
1.0
Butyric acid (BA)/Ni molar ratio Fig. 10 e Average catalytic performances during a 1000min reaction over XBAN/A catalysts, plotted as a function of butyric acid (BA)/Ni molar ratio.
26
20 0.25BAN/A
24
18 0.5BAN/A
16 14
22
1BAN/A 0.25BAN/A
0BAN/A 0.1BAN/A
12
18 0.5BAN/A
16
10
14
1BAN/A
8 6
0BAN/A
20
12
0.1BAN/A
10 59
60
61
62
63
64
65
66
Acknowledgements Fig. 11 e Correlations between hydrogen yield and nickel dispersion, and between hydrogen yield and methane adsorption capacity of XBAN/A catalysts.
0.25BAN/A catalyst treated with medium amount of butyric acid exhibited the best catalytic performance. This result was due to high methane adsorption capacity of 0.25BAN/A catalyst caused by high nickel dispersion. According to our previous studies [40,41], a nickel-based catalyst with high nickel dispersion retains a lot of active sites for steam reforming. Small nickel crystallite in finely dispersed nickelbased catalysts usually contains high proportion of step sites. The step sites are known to facilitate dehydrogenation of methane and ethane, which is the rate-determining step of steam reforming reaction. Thus, it is evident that the XBAN/A catalyst with high nickel dispersion and high methane adsorption capacity showed high catalytic performance in the steam reforming of natural gas.
This research was supported by C1 Gas Refinery Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (2016M3D3A1A01913252). This research was also supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (MSIP) (NRF-2016R1A5A1009592).
references
[1] Dincer I, Acar C. Innovation in hydrogen production. Int J Hydrogen Energy 2017;42:14843e64. [2] Momirlana M, Veziroglu TN. The properties of hydrogen as fuel tomorrow in sustainable energy system for a cleaner planet. Int J Hydrogen Energy 2005;30:795e802. [3] El-Melih AM, Iovine L, Shoaibi AA, Gupta AK. Production of hydrogen from hydrogen sulfide in presence of methane. Int J Hydrogen Energy 2017;42:4764e73.
Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
[4] Palma V, Martino M, Meloni E, Ricca A. Novel structured catalysts configuration for intensification of steam reforming of methane. Int J Hydrogen Energy 2017;42(3):1629e38. [5] Haryanto A, Fernando S, Murali N, Adhikari S. Current status of hydrogen production by steam reforming of ethanol: a review. Energy Fuels 2005;19:2098e106. [6] Geissler K, Newson E, Vogel F, Truong T-B, Hottinger P, Wokaun A. Autothermal methanol reforming for hydrogen production in fuel cell applications. Phys Chem Chem Phys 2001;3:289e93. [7] Gillan C, Fowles M, French S, Jackson SD. Ethane steam reforming over a platinum/alumina catalyst: effect of sulfur poisoning. Ind Eng Chem Res 2013;52:13350e6. [8] Matsumura Y, Nakamori T. Steam reforming of methane over nickel catalysts at low reaction temperature. Appl Catal A Gen 2004;258:107e14. [9] Hoshi N, Nakamura M, Kida K. Structural effects on the oxidation of formic acid on the high index planes of palladium. Electrochem Commun 2007;9(2):279e82. [10] Ali S, Al-Marri MJ, Abdelmoneim AG, Jumar A, Khader MM. Catalytic evaluation of nickel nanoparticles in methane steam reforming. Int J Hydrogen Energy 2016;41(48):22876e85. [11] Evans SE, Staniforth JZ, Darton RJ, Ormerod RM. A nickel doped perovskite catalyst for reforming methane rich biogas with minimal carbon deposition. Green Chem 2014;16:4587e94. [12] Shin G, Yun J, Yu S. Thermal design of methane steam reformer with low-temperature non-reactive heat source for high efficiency engine-hybrid stationary fuel cell system. Int J Hydrogen Energy 2017;42(21):14697e707. [13] Christensen KO, Chen D, Lødeng R, Holmen A. Effect of supports and Ni crystal size on carbon formation and sintering during steam methane reforming. Appl Catal A Gen 2006;314:9e22. [14] Zhu J, Peng X, Yao L, Tong D, Hu C. CO2 reforming of methane over Mg-promoted Ni/SiO2 catalysts: the influence of Mg precursors and impregnation sequences. Catal Sci Technol 2012;2:529e37. [15] Bang Y, Han SJ, Yoo J, Choi JH, Kang KH, Song JH, et al. Hydrogen production by steam reforming of liquefied natural gas (LNG) over trimethylbenzene-assisted ordered mesoporous nickel-alumina catalyst. Int J Hydrogen Energy 2013;38:8751e8. [16] Mo L, Saw ET, Du Y, Borgna A, Ang ML, Kathiraser Y, et al. Highly dispersed supported metal catalysts prepared via insitu self-assembled core-shell precursor route. Int J Hydrogen Energy 2015;40:13388e98. [17] Mourdikoudis S, Liz-Marzan LM. Oleylamine in nanoparticle synthesis. Chem Mater 2013;25:1465e76. [18] Pei Z-F, Ponec V. On the intermediates of the acetic acid reactions on oxides: an IR Study. Appl Surf Sci 1996;103:171e82. [19] Wu N, Fu L, Su M, Aslam M, Wong KC, Dravid VP. Interaction of fatty acid monolayers with cobalt nanoparticles. Nano Lett 2004;4(2):383e6. [20] Morris SM, Fulvio PF, Jaroniec M. Ordered mesoporous alumina-supported metal oxides. J Am Chem Soc 2008;130:15210e6. [21] Gobara HM, Gomaa II MM. Electrical properties of Ni/silica gel and Pt/alumina catalysts in relation to catalytic activity. Pet Sci Technol 2009;27:1572e91. [22] Bang Y, Seo JG, Youn MH, Song IK. Hydrogen production by steam reforming of liquefied natural gas (LNG) over mesoporous Ni-Al2O3 aerogel catalyst prepared by a singlestep epoxide-driven sol-gel method. Int J Hydrogen Energy 2012;37(2):1436e43. [23] Nazemi MK, Sheibani S, Rashchi F, Gonzalez-DelaCruz VM, Caballero A. Preparation of nanostructured nickel aluminate
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33] [34]
[35] [36]
[37]
[38]
[39]
[40]
[41]
9
spinel powder from spent NiO/Al2O3 catalyst by mechanochemical synthesis. Adv Powder Technol 2012;23:833e8. Dimotakis ED, Pinnavaia TJ. New route to layered double hydroxides intercalated by organic anions: precursors to polyoxometalate-pillared derivatives. Inorg Chem 1990;29:2393e4. Zangouei M, Moghaddam AZ, Arasteg M. The influence of nickel loading on reducibility of NiO/Al2O3 catalysts synthesized by sol-gel method. Chem Eng Res Bull 2010;14:97e102. Zhang L, Zhang Y. An active and stable Ni/Al2O3 nanosheet catalyst for dry reforming of CH4. RSC Adv 2015;5:62173e8. Goula MA, Charisiou ND, Papageridis KN, Delimitis A, Pachatouridou E, Iliopoulou EF. Nickel on alumina catalysts for the production of hydrogen rich mixtures via the biogas dry reforming reaction: influence of the synthesis method. Int J Hydrogen Energy 2015;40(30):9183e200. Zhu X, Huo P, Zhang Y, Cheng D, Liu C. Structure and reactivity of plasma treated Ni/Al2O3 catalyst for CO2 reforming of methane. Appl Catal B Environ 2008;81:132e40. Youn MH, Seo JG, Cho KM, Jung JC, Kim H, La KW, et al. Effect of support on hydrogen production by auto-thermal reforming of ethanol over supported nickel catalysts. Korean J Chem Eng 2008;25:236e8. Velu S, Gangwal SK. Synthesis of alumina supported nickel nanoparticle catalysts and evaluation of nickel metal dispersions by temperature programmed desorption. Solid State Ionics 2006;177:803e11. Park S, Yoo J, Han SJ, Song JH, Lee EJ, Song IK. Steam reforming of liquefied natural gas (LNG) for hydrogen production over nickel-boron-alumina xerogel catalyst. Int J Hydrogen Energy 2017;42:15096e106. Qian L, Yan Z. Studies on adsorption and dissociation of methane and carbon dioxide on nickel catalyst. J Nat Gas Chem 2002;11:151e8. Hou K, Hughes R. The kinetics of methane steam reforming over a Ni/a-Al2O catalyst. Chem Eng J 2001;82:311e28. Wang C, Sun N, Wei W, Zhao Y. Carbon intermediates during CO2 reforming of methane over Ni-CaO-ZrO2 catalysts: a temperature-programmed surface reaction study. Int J Hydrogen Energy 2016;41(42):19014e24. Sehested J. Four challenges for nickel steam-reforming catalysts. Catal Today 2006;111:103e10. Bernardo CA, Alstrup I, Rostrup-Nielsen JR. Carbon deposition and methane steam reforming on silicasupported Ni-Cu catalysts. J Catal 1985;96:517e34. Munster P, Grabke HJ. Kinetics of the steam reforming of methane with iron, nickel, and iron-nickel alloys as catalysts. J Catal 1981;72:279e87. Sehested J, Gelten JAP, Remediakis IN, Bengaard H, Nørskov JK. Sintering of nickel steam-reforming catalysts: effects of temperature and steam and hydrogen pressures. J Catal 2004;223:432e43. Annesini MC, Piemonte V, Turchetti L. Carbon formation in the steam reforming process: a thermodynamic analysis based on the elemental composition. Chem Eng Trans 2007;11:21e6. Seo JG, Youn MH, Bang Y, Song IK. Hydrogen production by steam reforming of simulated liquefied natural gas (LNG) over mesoporous nickel-alumina (M¼Ni, Ce, La, Y, Cs, Fe, Co, and Mg) aerogel catalysts. Int J Hydrogen Energy 2011;36:3505e14. Bang Y, Park S, Han SJ, Yoo J, Song JH, Choi JH, et al. Hydrogen production by steam reforming of liquefied natural gas (LNG) over mesoporous Ni/Al2O3 catalyst prepared by an EDTA-assisted impregnation method. Appl Catal B Environ 2016;180:179e88.
Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/alumina catalyst Jaekyeong Yoo, Seungwon Park, Ji Hwan Song, Sangbeom Yoo, In Kyu Song* School of Chemical and Biological Engineering, Institute of Chemical Processes, Seoul National University, Shinlimdong, Kwanak-ku, Seoul 151-744, South Korea
article info
abstract
Article history:
A series of butyric acid (BA)-assisted nickel/alumina catalysts (denoted as XBAN/A) with
Received 20 July 2017
different butyric acid/Ni molar ratio (X) were prepared by an impregnation method, and
Received in revised form
they were applied to the hydrogen production by steam reforming of natural gas. All the
5 September 2017
XBAN/A catalysts exhibited X-ray diffraction patterns indicative of highly stable nickel
Accepted 25 September 2017
aluminate phase. However, a XBAN/A catalyst with an appropriate amount of butyric acid
Available online xxx
showed the enhanced nickel dispersion because of steric hindrance of butyric acid shell surrounding nickel particle. Addition of butyric acid also increased methane adsorption
Keywords:
capacity of the catalysts, which was directly related to the catalytic performance. In the
Hydrogen production
steam reforming of natural gas, both natural gas conversion and hydrogen yield showed
Steam reforming of natural gas
volcano-shaped trends with respect to butyric acid/Ni molar ratio (X). Nickel dispersion of
Nickel/alumina catalyst
XBAN/A catalysts was well correlated with natural gas conversion and hydrogen yield over
Butyric acid
the catalysts. Natural gas conversion and hydrogen yield increased with increasing nickel dispersion. Among the catalysts tested, 0.25BAN/A catalyst with the highest nickel dispersion exhibited the best catalytic performance in the steam reforming of natural gas. © 2017 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
Introduction With increase in human population, global energy consumption increases every year [1]. The rapid population growth also causes imprudent industrial expansion, leading to serious environmental problems such as global warming and climate change [2]. Therefore, it is urgent to invest and develop sustainable alternative energy sources. Hydrogen is one of the most desirable energy carriers because of its cleanness and zero-emission property [3,4]. Stable supply chain and high energy density of hydrogen also facilitate the increase of annual hydrogen demand [5,6].
Most of hydrogen used in the world is produced by steam reforming processes, corresponding to the production of 96% of on-purpose hydrogen [7]. Natural gas, which is mainly composed of methane and ethane, is the most widely utilized hydrogen source in the steam reforming process because of its abundance. Steam reforming of natural gas can be described by the following equations.
CH4 þ H2O 4 CO þ 3H2 DHo298 K ¼ þ206.1 kJ/mol
(1)
C2H6 þ 2H2O 4 2CO þ 5H2 DHo298 K ¼ þ347.3 kJ/mol
(2)
* Corresponding author. E-mail address: [email protected] (I.K. Song). https://doi.org/10.1016/j.ijhydene.2017.09.148 0360-3199/© 2017 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
2
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
CO þ H2O 4 CO2 þ H2 DHo298
K
¼ 41.2 kJ/mol
(3)
Eqs. (1) and (2) represent steam reforming reactions to produce CO and H2 from methane and ethane, respectively. Water-gas shift reaction (Eq. (3)) accompanied by the reforming reactions allows an additional hydrogen production. Overall reactions in the presence of abundant water can be expressed as follows. CH4 þ 2H2O 4 CO2 þ 4H2 DHo298 K ¼ þ164.9 kJ/mol
(4)
C2H6 þ 4H2O 4 2CO2 þ 7H2 DHo298
(5)
K
¼ þ264.9 kJ/mol
According to the equations, steam reforming of natural gas is a highly endothermic reaction. Thus, typical steam reforming process is carried out at around 800 C to supply large amount of energy [8]. Such a high-temperature operation causes a hurdle for downsizing of reformers for mobile fuel cells. Therefore, it is needed to develop a new catalyst operated at low temperature with high catalytic activity and stability. Nickel-based catalysts have been commercially used in the steam reforming processes due to their high intrinsic activity for CeC bond cleavage and low cost [9,10]. However, nickelbased catalysts are known to show performance drop and catalyst deactivation due to carbon deposition at low reaction temperature [11,12]. According to the previous researches [13e15], these problems can be overcome by highly dispersing active metal species in the nickel-based catalysts. For example, Christensen et al. have reported that nickel catalysts supported on hydrotalcite materials showed the enhanced coke resistance with decreasing nickel particle size during the steam reforming of methane [13]. Zhu et al. have also reported that addition of Mg into nickel-silica catalyst induced high dispersion of active phase and inhibition of metal sintering, resulting in high catalytic activity and stability during the dry reforming of methane [14]. Furthermore, an ordered mesoporous nickel-alumina catalyst prepared with a structure modifier (trimethylbenzene) showed the enhanced mesoporous structure and formed small metallic nickel [15]. The modified catalyst exhibited a stronger resistance toward nickel sintering and a better catalytic performance than nickelalumina catalyst without modifier during the steam reforming of methane. In particular, it has been reported that addition of fatty acid into precursor solution increased nickel dispersion of nickel/silica catalysts prepared by an impregnation method, and the catalysts showed high catalytic activity and stability during the dry reforming of methane [16]. Therefore, modification of nickel dispersion supported on alumina with a suitable dispersing agent for enhancing catalytic performance in the steam reforming of natural gas would be worthwhile. In this work, a series of butyric acid-assisted nickel/ alumina catalysts were prepared using various compositions of butyric acid-nickel precursor solution. The catalysts were characterized by FT-IR, nitrogen adsorptionedesorption, XRD, TPR, H2-TPD, and CH4-TPD analyses. The effect of butyric acid addition on the characteristics and catalytic activities of nickel/alumina catalysts in the steam reforming of natural gas was investigated.
Experimental Preparation of catalysts Alumina support was preliminarily prepared via a sol-gel method. 25.7 g of aluminum (III) nitrate (Junsei, 98%) was dissolved in 160.5 ml of ethanol. 48.6 ml of propylene oxide (Samchun, 99%) was introduced to the aluminum precursor solution as a gelation agent. The mixed solution turned into a white gel composite after a few minutes. The gel was aged at room temperature for 2 days. The obtained gel was crushed into small pieces and washed with anhydrous ethanol. The gel was then placed in a convection oven at 80 C for 2 days for drying. The obtained matter was grinded into powder using a mortar and pestle. The powder was calcined at 700 C for 5 h. A series of butyric acid (BA)-assisted mesoporous nickel/ alumina catalysts were prepared via an impregnation method. 0.51 g of nickel (II) nitrate (Sigma-Aldrich, 97%) was dissolved in 10 ml of D.I. water, and different amount of butyric acid was dropped to the nickel precursor solution. The mixed solution was heated to 80 C, and 1.0 g of alumina support was added into the solution. Water was then evaporated at 80 C with vigorous stirring. The obtained grain was dried at 80 C overnight, and it was calcined at 700 C for 5 h. Nickel loading was kept at 10 wt% in all the catalysts, and butyric acid (BA)/Ni molar ratio was varied from 0 to 1. The prepared catalysts were denoted as XBAN/A (X ¼ 0, 0.1, 0.25, 0.5, and 1), where X represented the butyric acid (BA)/Ni molar ratio.
Characterizations In order to investigate surface chemical state of XBAN/A catalysts, infrared spectra of as-prepared catalysts were obtained using a Fourier transform infrared (FT-IR) spectrometer (ThermoScientific, Nicolet 6700) with a diffuse reflectance accessory. FT-IR spectra were recorded within the spectral range of 3500e650 cm1. Surface areas and porosity of XBAN/A catalysts were analyzed by nitrogen adsorptionedesorption method at 196 C (BELSORP-mini II, BEL Japan). For the measurement, each catalyst was preliminarily degassed at 150 C for 6 h. Crystallite phases of the catalysts were identified by X-ray diffraction (XRD) analyses with a D/ Max 2500V/PC X-ray diffractometer equipped with a Cu-Ka tube operated at 50 kV and 100 mA. The measurements were carried out at a rate of 10 /min within 2q range from 20 to 80 . For the XRD patterns of reduced XBAN/A catalysts, each calcined catalyst (50 mg) was preliminarily reduced under a hydrogen flow (3 ml/min) mixed with a nitrogen flow (30 ml/ min) at 700 C for 3 h. Temperature-programmed reduction (TPR) experiments were carried out in order to examine metal-support interaction of the catalysts. Each calcined catalyst sample (50 mg) was charged into a U-shaped quartz reactor, and it was reduced using a reducing gas mixture of hydrogen (2 ml/min) and nitrogen (20 ml/min) at a heating rate of 5 C/min from room temperature to 1000 C. TPR profiles of the catalysts were obtained using a thermal conductivity detector (TCD). Morphology and elemental
Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
3
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
Catalytic activity measurement Steam reforming of natural gas was conducted in a fixed-bed quartz reactor. Each catalyst (50 mg) was preliminarily reduced under a reducing gas of hydrogen (3 ml/min) and nitrogen (30 ml/min) at 700 C for 3 h. Steam reforming of natural gas was then carried out at 550 C for 1000 min. A feed stream composed of natural gas (5 ml/min), steam (11 ml/ min), and nitrogen (18 ml/min) was used during the reaction. Composition of natural gas was 92 vol% methane and 8 vol% ethane. The effluent gas was periodically sampled and analyzed using a gas chromatograph (ACME 6000, Younglin). The sampled gas was separated through Molecular Sieve 5 A (60/80 mesh, 2 ft 1/800 2 mm ID) and Porapak N (80/100 mesh, 20 ft 1/800 2 mm ID) columns. Each separated component was detected by a TCD. Conversion of natural gas and yield for hydrogen were defined as follows. Fi represents the molar flow rate of i species.
Natural gas conversionð%Þ ¼
FCH4 ;in þ FC2 H6 ;in FCН 4 ;out FC2 H6 ;out FCH4 ;in þ FC2 H6 ;in
Results and discussion Surface chemical state of as-prepared catalysts In order to investigate chemical state of butyric acid, asprepared XBAN/A (X ¼ 0, 0.1, 0.25, 0.5, and 1) catalysts were examined by FT-IR analyses (Fig. 1). XBAN/A catalysts with butyric acid addition (X 0.25) showed two small bands at 2975 cm1 and 2885 cm1, which were indicative of asymmetric CH2 stretch and symmetric CH2 stretch, respectively [17]. The intensity of CH2 stretch peaks increased with increasing butyric acid addition. Characteristic peaks indicative of asymmetric COO stretch (1565 cm1) and symmetric COO stretch (1465 cm1) were also detected in the catalysts with X 0.25 [18]. According to the previous research [17], bonding types of COO groups to catalyst surface can be determined by the wavenumber separation (△) between the asymmetric and the symmetric COO stretch. If △ is large (>200 cm1), monodentate interaction is expected. Small △ (<110 cm1) is for chelating bidentate mode. The medium range (140 cm1 < △ < 190 cm1) corresponds to bridging bidentate mode. In this work, △ (1565e1465 ¼ 100 cm1) was smaller than 110 cm1, indicating that butyric acid was bound to catalyst surface in a chelating bidentate mode. It should be noted that the peak at 1710 cm1, which was ascribed to C]O stretch of free butyric acid, appeared in the 1BAN/A catalyst [19]. This implies that butyric acid was presented as a separated form on the catalyst surface, when excess amount of butyric acid was introduced.
Textural properties of calcined catalysts Nitrogen adsorptionedesorption isotherms of calcined XBAN/ A catalysts are shown in Fig. 2. The obtained isotherms of all catalysts were type IV, indicating the formation of
CH2 stretch C=O stretch COO- stretch
Transmittance (A. U.)
distribution of the calcined catalysts were examined by scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS) mapping analyses using JSM-6701F (JEOL). High-resolution transmission electron microscopy (HR-TEM) experiments were carried out in order to examine the metal dispersion and lattice fringe of the reduced catalysts using a JEM-3010 (JEOL) apparatus operated at 300 kV. Hydrogen temperature-programmed desorption (H2-TPD) analyses were performed on reduced XBAN/A catalysts to measure nickel dispersion (BELCAT-B, BEL Japan). Each calcined catalyst (50 mg) was preliminarily reduced under a reducing gas of 5% hydrogen diluted with argon (50 ml/min) at 700 C for 3 h. Hydrogen adsorption was carried out under a 5% hydrogen flow diluted with argon (50 ml/min) at 50 C for 30 min. Elimination of physisorbed molecules was performed by purging with an argon gas (50 ml/min) at 100 C for 1 h. Desorption of chemisorbed molecules was monitored at 5 C/ min to 950 C under an argon flow (30 ml/min). A TCD was employed to detect the desorbed hydrogen. Methane temperature-programmed desorption (CH4-TPD) was performed to investigate methane adsorption capacity of reduced catalysts (BELCAT-B, BEL Japan). Each calcined catalyst (50 mg) was preliminarily reduced under a reducing gas of 5% hydrogen diluted with argon (50 ml/min) at 700 C for 3 h. Methane adsorption was carried out at 50 C under a mixed gas of 1% methane diluted with argon (50 ml/min) at 50 C for 30 min. Elimination of physisorbed methane was performed by purging with an argon gas (50 ml/min) at 250 C for 1 h. Desorption of chemisorbed methane was monitored at 5 C/ min within temperature range of 200e1000 C under an argon flow (30 ml/min).
0BAN/A 0.1BAN/A 0.25BAN/A 0.5BAN/A 1BAN/A
100 (6) Hydrogen yieldð%Þ ¼
FH2 ;out 100 4 FCH4 ;in þ 7 FC2 H6 ;in
(7)
3500
3000
2500
2000
1500
Wavenumber
(cm-1)
1000
Fig. 1 e FT-IR spectra of as-prepared XBAN/A catalysts.
Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
4
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
angle, indicating lattice expansion by the insertion of nickel species [24]. It is also noteworthy that broad and poorly defined peaks represent low crystallinity of XBAN/A catalysts, indicating fine dispersion of nickel species on g-Al2O3 support [25]. Fig. 3 (b) shows the X-ray diffractograms of reduced XBAN/A catalysts. Diffraction peaks which were attributed to nickel aluminate phase disappeared. Instead, all the catalysts showed diffraction peaks ascribed to metallic nickel (JCPDS 04-0850) phase. Moreover, the diffraction peak indicative of g-Al2O3 (440) shifted back to higher value, representing the exclusion of nickel cations from alumina lattice. Thus, it can be concluded that nickel species in the XBAN/A catalysts were sufficiently reduced from Ni2þ into Ni0 during the reduction step.
Volume adsorbed (A. U.)
0BAN/A
0.1BAN/A
0.25BAN/A
0.5BAN/A
SEM-EDS mapping and TEM analyses
1BAN/A
0
0.2
0.4
0.6
0.8
1.0
Relative pressure (P/P0) Fig. 2 e Nitrogen adsorptionedesorption isotherms of calcined XBAN/A catalysts.
mesoporous structure [20]. In addition, H2-type hysteresis loops were observed for all catalysts, representing the existence of “ink-bottle” type pores [21]. Table 1 shows the detailed textural properties of XBAN/A catalysts. All the catalysts showed high surface area (>250 m2), large pore volume (>0.5 cm3/g), and large pore diameter (>8.0 nm). This result means that the mesoporous structure of alumina support used for the preparation of XBAN/A catalysts was successfully prepared via a sol-gel method in this work [22].
XRD result of calcined and reduced catalysts X-ray diffractograms of calcined XBAN/A catalysts are presented in Fig. 3 (a). The crystalline phases were identified by the JCPDS standards. All the catalysts showed the characteristic diffraction peaks of nickel aluminate (JCPDS 10-0339) phase. This means that nickel aluminate spinel phase was formed during the calcination step. The formation of nickel aluminate phase is known to take place by the insertion of nickel cations into the tetrahedral vacancies of g-Al2O3 lattice [23]. Accordingly, diffraction peak of g-Al2O3 (440) slightly shifted to lower
Table 1 e Textural properties of calcined XBAN/A catalysts. Catalyst 0BAN/A 0.1BAN/A 0.25BAN/A 0.5BAN/A 1BAN/A a b
Surface area (m2/g)a
Pore volume (cm3/g)b
Average pore diameter (nm)
257 276 265 262 274
0.55 0.67 0.55 0.64 0.69
8.5 9.7 8.3 9.7 10.1
Calculated by the BET equation at P/P0 ¼ 0.05e0.30. Total pore volume at P/P0 ~ 0.990.
Distributions of aluminum, nickel, and oxygen in the calcined 0.25BAN/A catalyst were examined by SEM-EDS mapping analyses as shown in Fig. 4. The EDS elemental mapping image shows that nickel species (green dots in Fig. 4) was homogeneously dispersed on alumina support. TEM images of reduced 0.25BAN/A catalyst are shown in Fig. 5. It was estimated that crystallite size of nickel particle was smaller than 10 nm (Fig. 5 (a)). A large agglomerate of nickel particle was not observed in this sample. From the SEM-EDS and TEM results, it can be said that nickel species in the 0.25BAN/A catalyst was finely dispersed on alumina support. Fig. 5 (b) shows a latticeresolved TEM image of the reduced 0.25BAN/A catalyst including inter-planar spacing of nickel crystallite. The periodic lattice fringe with a d-spacing of 0.20 ± 0.004 nm is attributed to the (111) plane of metallic nickel [26,27]. No lattice fringe indicative of nickel oxide phase was observed. Therefore, it is believed that the nickel species in the 0.25BAN/ A catalyst was fully reduced into the metallic nickel phase during the reduction step. This result is in good agreement with the XRD results in Section XRD result of calcined and reduced catalysts.
TPR result of calcined catalysts Fig. 6 shows the TPR profiles of calcined XBAN/A catalysts. All the XBAN/A catalysts showed a reduction peak at around 700 C due to highly dispersed nickel aluminate spinel [28]. It was observed that the peak temperature of XBAN/A (X > 0) catalysts with butyric acid addition showed peak shift to low temperature compared to 0BAN/A catalyst. It can be inferred that butyric acid weakened the interaction between metal and support by surrounding metal particles. In particular, 1BAN/A catalyst showed an additional reduction peak at 489 C, indicating the reduction of bulk nickel oxide to metallic nickel [29]. It is believed that the excessively added butyric acid surrounding nickel species lifts up the nickel cluster from alumina, inhibiting the interaction between nickel and alumina.
H2-TPD result of reduced catalysts H2-TPD experiments on reduced XBAN/A catalysts were performed and their corresponding profiles are shown in Fig. 7. Three different hydrogen desorption peaks appeared in all the
Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
5
Fig. 3 e XRD patterns of (a) calcined and (b) reduced XBAN/A catalysts.
Fig. 4 e SEM-EDS images of calcined 0.25BAN/A catalyst obtained by mapping on aluminum, nickel, and oxygen.
XBAN/A catalysts. According to the literature [30], hydrogen desorption peak centered at around 200 C is ascribed to weakly adsorbed hydrogen molecules on active metal sites, and hydrogen desorption peak centered at around 500 C is due to strongly adsorbed hydrogen on metal sites. However, desorption peak above 600 C is assigned to hydrogen spillover species. Therefore, total amount of hydrogen uptake on active nickel species can be obtained by integrating deconvoluted peaks below 600 C. Deconvolution of the TPD profiles was performed using the OriginLab software. Nickel dispersion was calculated using the amount of hydrogen uptake as
summarized in Table 2. It should be noted that the amount of total hydrogen uptake increased with increasing butyric acid/ Ni ratio up to 0.25, and then the amount of hydrogen uptake decreased with increasing butyric acid/Ni ratio over 0.25. The nickel dispersion calculated from the amount of hydrogen uptake showed the same trend. According to the previous research [16], an appropriate amount of fatty acid surrounding nickel particles acts as a dispersing agent, inhibiting sintering of nickel particles during the thermal treatment. In the presence of excess fatty acid, however, nickel particles were fully surrounded by a fatty acid shell and uplifted from support,
Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
6
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
Fig. 5 e TEM images of reduced 0.25BAN/A catalyst: (a) TEM image and (b) HR-TEM image with lattice fringes.
706
Hydrogen desorbed (A. U.)
Hydrogen consumption (A. U.)
0BAN/A 0BAN/A 691 0.1BAN/A
683
0.25BAN/A 691 0.5BAN/A
696 489
0.1BAN/A
0.25BAN/A
0.5BAN/A
1BAN/A
1BAN/A
0
200
400
600
800
1000
100 200 300 400 500 600 700 800 900
Temperature (oC) Fig. 6 e TPR profiles of calcined XBAN/A catalysts.
resulting in more opportunity to migrate and combine with other metal particles nearby. In a similar manner in this work, high dispersion of nickel species could be achieved when medium amount of butyric acid was introduced.
Temperature (oC) Fig. 7 e H2-TPD profiles of reduced XBAN/A catalysts.
Table 2 e H2-TPD results of reduced XBAN/A catalysts. Catalyst
CH4-TPD result of reduced catalysts CH4-TPD experiments on reduced XBAN/A catalysts were performed and their corresponding profiles are shown in Fig. 8. Broad desorption peaks were observed for all the catalysts. For precise quantification, deconvolution of the CH4TPD profiles was conducted. According to the literature [31], the peak appearing below 700 C is known to be related to the desorption of methane from the reduced nickel species. On the other hand, methane desorption peak at around 800 C is related to the desorption of methane from alumina support [32]. In this work, therefore, deconvoluted desorption peak
Nickel Amount of desorbed hydrogen dispersionb (%) (mmol-H2/g)a Weak site Strong site Total (<300 C) (300e600 C)
0BAN/A 0.1BAN/A 0.25BAN/A 0.5BAN/A 1BAN/A
18 26 26 28 33
38 33 75 64 35
56 59 101 92 68
7.2 7.6 13.0 11.9 8.8
Measurement errors for the amount of desorbed hydrogen and nickel dispersion are ± 4 mmol-H2/g and ±0.3%, respectively. a Calculated from deconvoluted peak area of H2-TPD profiles in Fig. 7. b Calculated by assuming H/Niatom ¼ 1.
Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
7
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
Methane desorbed (A. U.)
0BAN/A
0.1BAN/A 0.25BAN/A
0.5BAN/A
1BAN/A
200
400
600
800
1000
Temperature (oC)
affected by butyric acid/Ni molar ratio. The amount of desorbed methane decreased in the order of 0.25BAN/A > 0.5BAN/ A > 1BAN/A > 0.1BAN/A > 0BAN/A. This trend was well consistent with the trend of nickel dispersion calculated in Section H2-TPD result of reduced catalysts (Table 2). In other words, a XBAN/A catalyst with higher nickel dispersion showed higher methane adsorption capacity. Qian et al. have reported that partially or fully dehydrogenated methane is desorbed from nickel/alumina catalyst below 700 C [32]. It has also been reported that dehydrogenated methane species are major reaction intermediates in the methane reforming [33,34]. From the above researches, it can be inferred that a nickel/alumina catalyst with high methane adsorption capacity will dissociate much amount of methane effectively in the reforming reaction. The dissociative adsorption of methane into dehydrogenated carbon species is known as the rate-determining step of reforming reactions [35e37]. Therefore, it is expected that the XBAN/A catalyst with higher methane adsorption capacity will show a better catalytic performance in the steam reforming of natural gas.
Fig. 8 e CH4-TPD profiles of reduced XBAN/A catalysts.
Catalytic activity in steam reforming of natural gas
area below 700 C was only considered for quantification. The amount of desorbed methane was calculate by integrating the deconvoluted peaks below 700 C as summarized in Table 3. It was found that the amount of desorbed methane was greatly
Table 3 e CH4-TPD results of reduced XBAN/A catalysts. Catalyst
0BAN/A 0.1BAN/A 0.25BAN/A 0.5BAN/A 1BAN/A
Amount of desorbed methane (mmol-CH4/g)a Weak site (<400 C)
Strong site (400e700 C)
Total
6.5 6.9 8.9 8.2 7.6
12.0 12.0 15.4 13.8 13.2
18.5 18.8 24.2 22.0 20.8
Measurement error for the amount of desorbed methane is ±0.2 mmol-CH4/g. a Calculated from deconvoluted peak area of CH4-TPD profiles in Fig. 8.
Fig. 9 shows the catalytic activities of XBAN/A catalysts in the steam reforming of natural gas at 550 C. The reaction was performed for 1000 min. It was observed that all the catalysts maintained initial catalytic activity with no significant deactivation. In commercial steam reforming process, nickelbased catalysts undergo activity drop caused by the growth of large nickel particles and the consequent reduction of active surface area [38]. Carbon formation on the catalyst surface is another problem which results in the loss of active sites by encapsulating the catalyst surface or blocking pores of the catalysts [39]. When considering the stable catalytic performance of XBAN/A catalysts, all the catalysts are believed to retain a strong resistance toward sintering and carbon deposition because of finely dispersed nickel particles on alumina support with well-developed mesopores. Fig. 10 shows the catalytic activity of XBAN/A catalysts in the steam reforming of natural gas plotted as a function of butyric acid (BA)/Ni molar ratio. Volcano-shaped trends were observed with respect to butyric acid/Ni molar ratio.
70
(a)
70
(b) 65
65
60
60 55
55 0BAN/A 0.1BAN/A 0.25BAN/A 0.5BAN/A 1BAN/A
50
0BAN/A 0.1BAN/A 0.25BAN/A 0.5BAN/A 1BAN/A
50
45 0
200
400
600
800
1000 0
200
400
600
800
45 1000
Fig. 9 e Catalytic performances with time on stream: (a) Natural gas conversion and (b) hydrogen yield. Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
8
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
72 66 70 64 68
62
66
60
Fig. 11 shows the correlations between hydrogen yield and nickel dispersion, and between hydrogen yield and methane adsorption capacity of XBAN/A catalysts. A distinctive feature is that hydrogen yield increased with increasing nickel dispersion and with increasing methane adsorption capacity. On the basis of H2-TPD and CH4-TPD analyses, it is believed that an optimal butyric acid/Ni molar ratio was required for the highest nickel dispersion and methane adsorption capacity, and in turn, for the highest hydrogen yield in the steam reforming of natural gas.
64
58
62
56
60
54
Conclusions
52
A series of butyric acid (BA)-assisted nickel/alumina (XBAN/A, X ¼ 0, 0.1, 0.25, 0.5, and 1) catalysts were characterized and evaluated in order to find an optimal butyric acid/Ni molar ratio (X) in the steam reforming of natural gas. It was found that characteristics of XBAN/A catalysts were greatly affected by the butyric acid/Ni molar ratio. All the prepared catalysts showed a mesoporous structure. In addition, all the calcined catalysts retained a nickel aluminate phase. Nickel dispersion of XBAN/A catalysts was improved when an appropriate amount of butyric acid was introduced. This was due to the dispersing effect of butyric acid in thermal treatment step. A XBAN/A catalyst with high nickel dispersion also showed high methane adsorption capacity. Although all the XBAN/A catalysts showed a stable catalytic performance in the steam reforming of natural gas, high catalytic performance was obtained when butyric acid was introduced. Natural gas conversion and hydrogen yield increased with increasing nickel dispersion and with increasing methane adsorption capacity. Among the catalysts tested, 0.25BAN/A catalyst with the highest nickel dispersion and methane adsorption capacity showed the best catalytic performance in terms of natural gas conversion and hydrogen yield.
58 0
0.2
0.4
0.6
0.8
1.0
Butyric acid (BA)/Ni molar ratio Fig. 10 e Average catalytic performances during a 1000min reaction over XBAN/A catalysts, plotted as a function of butyric acid (BA)/Ni molar ratio.
26
20 0.25BAN/A
24
18 0.5BAN/A
16 14
22
1BAN/A 0.25BAN/A
0BAN/A 0.1BAN/A
12
18 0.5BAN/A
16
10
14
1BAN/A
8 6
0BAN/A
20
12
0.1BAN/A
10 59
60
61
62
63
64
65
66
Acknowledgements Fig. 11 e Correlations between hydrogen yield and nickel dispersion, and between hydrogen yield and methane adsorption capacity of XBAN/A catalysts.
0.25BAN/A catalyst treated with medium amount of butyric acid exhibited the best catalytic performance. This result was due to high methane adsorption capacity of 0.25BAN/A catalyst caused by high nickel dispersion. According to our previous studies [40,41], a nickel-based catalyst with high nickel dispersion retains a lot of active sites for steam reforming. Small nickel crystallite in finely dispersed nickelbased catalysts usually contains high proportion of step sites. The step sites are known to facilitate dehydrogenation of methane and ethane, which is the rate-determining step of steam reforming reaction. Thus, it is evident that the XBAN/A catalyst with high nickel dispersion and high methane adsorption capacity showed high catalytic performance in the steam reforming of natural gas.
This research was supported by C1 Gas Refinery Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (2016M3D3A1A01913252). This research was also supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (MSIP) (NRF-2016R1A5A1009592).
references
[1] Dincer I, Acar C. Innovation in hydrogen production. Int J Hydrogen Energy 2017;42:14843e64. [2] Momirlana M, Veziroglu TN. The properties of hydrogen as fuel tomorrow in sustainable energy system for a cleaner planet. Int J Hydrogen Energy 2005;30:795e802. [3] El-Melih AM, Iovine L, Shoaibi AA, Gupta AK. Production of hydrogen from hydrogen sulfide in presence of methane. Int J Hydrogen Energy 2017;42:4764e73.
Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e9
[4] Palma V, Martino M, Meloni E, Ricca A. Novel structured catalysts configuration for intensification of steam reforming of methane. Int J Hydrogen Energy 2017;42(3):1629e38. [5] Haryanto A, Fernando S, Murali N, Adhikari S. Current status of hydrogen production by steam reforming of ethanol: a review. Energy Fuels 2005;19:2098e106. [6] Geissler K, Newson E, Vogel F, Truong T-B, Hottinger P, Wokaun A. Autothermal methanol reforming for hydrogen production in fuel cell applications. Phys Chem Chem Phys 2001;3:289e93. [7] Gillan C, Fowles M, French S, Jackson SD. Ethane steam reforming over a platinum/alumina catalyst: effect of sulfur poisoning. Ind Eng Chem Res 2013;52:13350e6. [8] Matsumura Y, Nakamori T. Steam reforming of methane over nickel catalysts at low reaction temperature. Appl Catal A Gen 2004;258:107e14. [9] Hoshi N, Nakamura M, Kida K. Structural effects on the oxidation of formic acid on the high index planes of palladium. Electrochem Commun 2007;9(2):279e82. [10] Ali S, Al-Marri MJ, Abdelmoneim AG, Jumar A, Khader MM. Catalytic evaluation of nickel nanoparticles in methane steam reforming. Int J Hydrogen Energy 2016;41(48):22876e85. [11] Evans SE, Staniforth JZ, Darton RJ, Ormerod RM. A nickel doped perovskite catalyst for reforming methane rich biogas with minimal carbon deposition. Green Chem 2014;16:4587e94. [12] Shin G, Yun J, Yu S. Thermal design of methane steam reformer with low-temperature non-reactive heat source for high efficiency engine-hybrid stationary fuel cell system. Int J Hydrogen Energy 2017;42(21):14697e707. [13] Christensen KO, Chen D, Lødeng R, Holmen A. Effect of supports and Ni crystal size on carbon formation and sintering during steam methane reforming. Appl Catal A Gen 2006;314:9e22. [14] Zhu J, Peng X, Yao L, Tong D, Hu C. CO2 reforming of methane over Mg-promoted Ni/SiO2 catalysts: the influence of Mg precursors and impregnation sequences. Catal Sci Technol 2012;2:529e37. [15] Bang Y, Han SJ, Yoo J, Choi JH, Kang KH, Song JH, et al. Hydrogen production by steam reforming of liquefied natural gas (LNG) over trimethylbenzene-assisted ordered mesoporous nickel-alumina catalyst. Int J Hydrogen Energy 2013;38:8751e8. [16] Mo L, Saw ET, Du Y, Borgna A, Ang ML, Kathiraser Y, et al. Highly dispersed supported metal catalysts prepared via insitu self-assembled core-shell precursor route. Int J Hydrogen Energy 2015;40:13388e98. [17] Mourdikoudis S, Liz-Marzan LM. Oleylamine in nanoparticle synthesis. Chem Mater 2013;25:1465e76. [18] Pei Z-F, Ponec V. On the intermediates of the acetic acid reactions on oxides: an IR Study. Appl Surf Sci 1996;103:171e82. [19] Wu N, Fu L, Su M, Aslam M, Wong KC, Dravid VP. Interaction of fatty acid monolayers with cobalt nanoparticles. Nano Lett 2004;4(2):383e6. [20] Morris SM, Fulvio PF, Jaroniec M. Ordered mesoporous alumina-supported metal oxides. J Am Chem Soc 2008;130:15210e6. [21] Gobara HM, Gomaa II MM. Electrical properties of Ni/silica gel and Pt/alumina catalysts in relation to catalytic activity. Pet Sci Technol 2009;27:1572e91. [22] Bang Y, Seo JG, Youn MH, Song IK. Hydrogen production by steam reforming of liquefied natural gas (LNG) over mesoporous Ni-Al2O3 aerogel catalyst prepared by a singlestep epoxide-driven sol-gel method. Int J Hydrogen Energy 2012;37(2):1436e43. [23] Nazemi MK, Sheibani S, Rashchi F, Gonzalez-DelaCruz VM, Caballero A. Preparation of nanostructured nickel aluminate
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33] [34]
[35] [36]
[37]
[38]
[39]
[40]
[41]
9
spinel powder from spent NiO/Al2O3 catalyst by mechanochemical synthesis. Adv Powder Technol 2012;23:833e8. Dimotakis ED, Pinnavaia TJ. New route to layered double hydroxides intercalated by organic anions: precursors to polyoxometalate-pillared derivatives. Inorg Chem 1990;29:2393e4. Zangouei M, Moghaddam AZ, Arasteg M. The influence of nickel loading on reducibility of NiO/Al2O3 catalysts synthesized by sol-gel method. Chem Eng Res Bull 2010;14:97e102. Zhang L, Zhang Y. An active and stable Ni/Al2O3 nanosheet catalyst for dry reforming of CH4. RSC Adv 2015;5:62173e8. Goula MA, Charisiou ND, Papageridis KN, Delimitis A, Pachatouridou E, Iliopoulou EF. Nickel on alumina catalysts for the production of hydrogen rich mixtures via the biogas dry reforming reaction: influence of the synthesis method. Int J Hydrogen Energy 2015;40(30):9183e200. Zhu X, Huo P, Zhang Y, Cheng D, Liu C. Structure and reactivity of plasma treated Ni/Al2O3 catalyst for CO2 reforming of methane. Appl Catal B Environ 2008;81:132e40. Youn MH, Seo JG, Cho KM, Jung JC, Kim H, La KW, et al. Effect of support on hydrogen production by auto-thermal reforming of ethanol over supported nickel catalysts. Korean J Chem Eng 2008;25:236e8. Velu S, Gangwal SK. Synthesis of alumina supported nickel nanoparticle catalysts and evaluation of nickel metal dispersions by temperature programmed desorption. Solid State Ionics 2006;177:803e11. Park S, Yoo J, Han SJ, Song JH, Lee EJ, Song IK. Steam reforming of liquefied natural gas (LNG) for hydrogen production over nickel-boron-alumina xerogel catalyst. Int J Hydrogen Energy 2017;42:15096e106. Qian L, Yan Z. Studies on adsorption and dissociation of methane and carbon dioxide on nickel catalyst. J Nat Gas Chem 2002;11:151e8. Hou K, Hughes R. The kinetics of methane steam reforming over a Ni/a-Al2O catalyst. Chem Eng J 2001;82:311e28. Wang C, Sun N, Wei W, Zhao Y. Carbon intermediates during CO2 reforming of methane over Ni-CaO-ZrO2 catalysts: a temperature-programmed surface reaction study. Int J Hydrogen Energy 2016;41(42):19014e24. Sehested J. Four challenges for nickel steam-reforming catalysts. Catal Today 2006;111:103e10. Bernardo CA, Alstrup I, Rostrup-Nielsen JR. Carbon deposition and methane steam reforming on silicasupported Ni-Cu catalysts. J Catal 1985;96:517e34. Munster P, Grabke HJ. Kinetics of the steam reforming of methane with iron, nickel, and iron-nickel alloys as catalysts. J Catal 1981;72:279e87. Sehested J, Gelten JAP, Remediakis IN, Bengaard H, Nørskov JK. Sintering of nickel steam-reforming catalysts: effects of temperature and steam and hydrogen pressures. J Catal 2004;223:432e43. Annesini MC, Piemonte V, Turchetti L. Carbon formation in the steam reforming process: a thermodynamic analysis based on the elemental composition. Chem Eng Trans 2007;11:21e6. Seo JG, Youn MH, Bang Y, Song IK. Hydrogen production by steam reforming of simulated liquefied natural gas (LNG) over mesoporous nickel-alumina (M¼Ni, Ce, La, Y, Cs, Fe, Co, and Mg) aerogel catalysts. Int J Hydrogen Energy 2011;36:3505e14. Bang Y, Park S, Han SJ, Yoo J, Song JH, Choi JH, et al. Hydrogen production by steam reforming of liquefied natural gas (LNG) over mesoporous Ni/Al2O3 catalyst prepared by an EDTA-assisted impregnation method. Appl Catal B Environ 2016;180:179e88.
Please cite this article in press as: Yoo J, et al., Hydrogen production by steam reforming of natural gas over butyric acid-assisted nickel/ alumina catalyst, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.09.148