Microporous and Mesoporous Materials 262 (2018) 122–132

Contents lists available at ScienceDirect

Microporous and Mesoporous Materials

journal homepage: www.elsevier.com/locate/micromeso

Combined steam and CO2 reforming of methane for syngas production over T
carbon-resistant boron-promoted Ni/SBA-15 catalysts
Tan Ji Sianga, Thong L.M. Phamb, Nguyen Van Cuongc, Pham T.T. Phuongd,
Nguyen Huu Huy Phuce, Quang Duc Truongf, Dai-Viet N. Voa,g,∗
a
Faculty of Chemical & Natural Resources Engineering, University Malaysia Pahang, Lebuhraya Tun Razak, Gambang 26300, Pahang, Malaysia
b
Institute of Research and Development, Duy Tan University, 03 Quang Trung, Danang, Viet Nam
c
Faculty of Chemical Engineering, Industrial University of Ho Chi Minh City, 12 Nguyen Van Bao St., Go Vap, Ho Chi Minh City 7000, Viet Nam
d
Institute of Chemical Technology, Vietnam Academy of Science and Technology, 1 Mac Dinh Chi Str., Dist.1, Ho Chi Minh City, Viet Nam
e
Department of Electrical and Electronic Information Engineering, Toyohashi University of Technology, 1-1 Hibarigaoka, Tempaku, Toyohashi, Aichi 441-8580, Japan
f
Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Katahira 2-1-1, Aoba-Ku, Sendai 980-8577, Japan
g
Centre of Excellence for Advanced Research in Fluid Flow, Universiti Malaysia Pahang, 26300 Gambang, Kuantan, Pahang, Malaysia

A R T I C L E I N F O A B S T R A C T

Keywords: The unpromoted and B-promoted 10%Ni/SBA-15 catalysts synthesized via sequential incipient wetness im-
Boron promoter pregnation approach were assessed for combined steam and CO2 reforming of methane (CSCRM) at various
Bi-reforming of methane reaction temperatures of 973–1073 K and stoichiometric feed composition. An expected and noteworthy drop in
Mesoporous silica mean NiO crystallite size and BET surface area with boron promotion from 1% to 5%B loading could be due to
Ni/SBA-15 catalyst
the agglomeration of B2O3 particles and deboration reaction during calcination and hence blocking mesopores of
Synthetic gas
SBA-15 support at elevated B composition. The complete NiO reduction to metallic Ni0 form was achieved during
H2 activation and the reduction temperature of NiO phase was shifted towards higher temperature with B-
addition owing to enhancing interaction between the acidic B2O3 and basic NiO phases. For all reaction tem-
perature employed, 3%B appeared to be the optimal promoter loading in terms of reactant conversions and 3%B-
10%Ni/SBA-15 catalyst revealed the greatest H2 yield (69.4%) at 1073 K. In addition, CH4 and CO2 conversions
were enhanced about 23.2% and 32.4%, correspondingly with rising reaction temperature from 973 to 1073 K.
Ratio of H2 to CO varied from 1.26 to 2.71 and the desired H2/CO ratio of about 2 favored for Fischer-Tropsch
synthesis was achieved on 3%B-10%Ni/SBA-15 sample at 973 K. Boron promoter suppressed graphitic carbon
formation and the amount of carbonaceous deposition was reduced about 4 times. Noticeably, 3%B-10%Ni/SBA-
15 was also resilient to metallic Ni0 re-oxidation throughout CSCRM.

1. Introduction essential chemicals, namely, methyl tert-butyl ether (MTBE), dimethyl

The growing concerns towards the depleting non-renewable petro-
leum resources and the excessive release of anthropogenic CO2 green-
house gas from the industrial combustion and residential consumption
of fossil fuels have led to rising interests in exploring an alternative,
sustainable and clean energy as well as diversifying energy sources [1].
In fact, reducing CO2 emission, replacing petroleum-based energy with
less carbon-intensive fossil energy sources and decreasing the currently
significant dependency on non-renewable energy are an essential and
urgent task. Amongst alternative energies, syngas consisting of H2 and
CO has appeared as a promising and flexible building block for the
following downstream production such as synthetic long-chained hy-
drocarbons via Fischer-Tropsch synthesis (FTS) [2,3] and other
ether (DME) and methanol in petrochemical industries [4].
In general, the conventional steam reforming of methane, SRM (cf.
Eq. (1)) is currently employed in industry for syngas production.
However, this approach yields undesirable ratio of H2 to CO greater
than 3 and substantially releases a large amount of unfavorable CO2
gaseous by-product [2,3]. Although dry reforming of methane (DRM,
see Eq. (2)) could transform unwanted CO2 greenhouse gas to a value-
added syngas, the resulting H2/CO ratio of less than unity makes this
process inappropriate for FTS and methanol production requiring a
stoichiometric H2/CO ratio of 2 [5,6]. Thus, the implementation of this
approach could impose auxiliary separation and purification steps in
the downstream process for adjusting H2/CO ratio to around 2 [7] and
hence increasing capital cost.

∗
Corresponding author. Faculty of Chemical & Natural Resources Engineering, University Malaysia Pahang, Lebuhraya Tun Razak, Gambang 26300, Pahang, Malaysia.
E-mail address: vietvo@ump.edu.my (D.-V.N. Vo).

https://doi.org/10.1016/j.micromeso.2017.11.028
Received 31 July 2017; Received in revised form 10 November 2017; Accepted 15 November 2017
Available online 22 November 2017
1387-1811/ © 2017 Elsevier Inc. All rights reserved.
T.J. Siang et al.
o −1
CH4 + H2 O → CO + 3H2 (ΔH298K = + 206 kJ mol ) (1)
o −1
CH4 + CO2 → 2CO + 2H2 (ΔH298K = + 247 kJ mol ) (2)
From the above reasons, the combined steam and CO2 reforming of
methane, abbreviated as CSCRM (see Eq. (3)), also known as bi-re-
forming of methane, has recently appeared as a promising technique
since it is capable of generating a green and sustainable energy source
from biogas, a non-fossil fuel resource mainly consisting of methane,
carbon dioxide and water [8]. In fact, CSCRM is reportedly a pro-
spective substitution of other common reforming processes owing to its
high catalytic stability in the coexistence of CO2 and H2O oxidizing
reactants and flexible adjustment of H2/CO ratios by the manipulation
of feedstock composition [9,10].
o −1
3CH4 + 2H2 O + CO2 → 8H2 + 4CO (ΔH298K = + 712 kJ mol ) (3)
Nevertheless, the bibliographic knowledge about CSCRM reaction is
still little-known as reported in literature due to the complexity of this
reaction involving multiple main reactions, namely, SRM (see Eq. (1))
and DRM (see Eq. (2)) as well as several parallel reactions including
non-coke (viz. reverse water-gas shift reaction given in Eq. (4)) and coke
forming reactions (see Eqs. (5)–(7)).
o −1
CO2 + H2 ⇄ H2 O + CO (ΔH298K = + 41 kJ mol ) (4)
o −1
2CO ⇄ CO2 + C (ΔH298K = − 172 kJ mol ) (5)
o −1
CH4 → 2H2 + C (ΔH298K = + 75 kJ mol ) (6)
o
CO + H2 ⇄ H2 O + C (ΔH298 −1
K = − 131 kJ mol ) (7)
In general, from an industrial standpoint, Ni-based catalyst is re-
cognized as one of the most attractive candidates for the large-scale
industrial application of reforming processes since this catalyst type is
cost-effective, abundant availability and relatively high catalytic ac-
tivity comparable to precious metals [2,3,10]. However, Ni-based cat-
alysts could encounter challenging barriers of coke formation and Ni
metal sintering resulting in rapid catalyst deactivation at high reaction
temperature. In the research of CSCRM reaction over Ni-Ce/Al2O3
catalyst, Koo et al. [11] reported that high CH4 and CO2 conversions of
about 90.9% and 69.2%, respectively were achieved at 1023 K with the
feed composition, CH4:CO2: H2O:N2 of 1.0:0.4:0.8:1.0. However, gra-
phitic carbon formation was still evident even with the utilization of
redox CeO2 promoter widely reported as carbon removal agent [2,3].
The same behavior was also observed by Park et al. for spent La-Ni/
MgAl2O4 catalyst after CSCRM reaction at 1173 K and 5 atm with CH4/
H2O/CO2/N2 of 1/1.2/0.4/0.3 [12]. Thus, exploring a new catalyst
system resistant to carbon deposition by the implementation of suitable
combination of support and promoter is crucial in CSCRM.
In recent years, mesoporous siliceous SBA-15 support has drawn
significant attention from both academics and industry for wide-ran-
ging heterogeneous catalytic reactions, namely, methane steam re-
forming [13,14], methane dry reforming [15,16] and methane cracking
[17] due to its substantially large specific surface area, thick framework
walls with ordered mesoporous structure and high thermal stability
[18]. The confinement effect of highly ordered two-dimensional hex-
agonal SBA-15 support could anchor metal nanoparticles inside the
mesoporous silica channels, inhibit metal sintering and hence resistance
to carbon deposition [7,19].
Apart from utilizing rare-earth metal oxides as redox promoters for
elimination of carbonaceous species [2,3], by using first principles
density functional theory (DFT) computations, Saeys' group found that
boron promoter may bind to the octahedral sites of the first Ni (1 1 1)
subsurface layer preferred for carbon adsorption and thus suppressing
surface carbon diffusion for forming graphene islands [20–22]. How-
ever, after meticulously exploring published literature, there were no
reportedly preceding experimental studies regarding synergetic influ-
ence of boron promoter and SBA-15 support on CSCRM. Therefore, the
123
Microporous and Mesoporous Materials 262 (2018) 122–132
goal of this project was to scrutinize the influence of B-promoter on the
physicochemical properties of 10%Ni/SBA-15 catalyst and examine
impact of reaction temperature on CSCRM performance.
2. Experimental
2.1. Synthesis of catalyst
Hydrothermal preparation technique was used for the synthesis of
siliceous SBA-15 support. An accurately measured quantity of Pluronic
P-123 triblock copolymer (EO20PO70EO20, triblock-poly(ethylene
glycol)-block-poly(propylene glycol)-block-poly(ethylene glycol) sup-
plied by Sigma-Aldrich Chemicals) was mixed with HCl solution (pH ≃
1) for 2 h at 313 K in order to completely dissolve the P-123 template.
The tetraethyl orthosilicate (TEOS, procured from Merck Millipore)
silica precursor was then added to resulting crystal clear solution and
vigorously mixed at similar temperature about 24 h. Afterward, the
resulting mixture was subjected to hydrothermal treatment in an au-
toclave (possessing an inner Teflon-lined container) for 24 h at 373 K
and the achieving milky slurry was filtered as well as rinsed with
purified water to pH value ≃ 4. The obtained white solid powder was
further dried at 323 K for 24 h and calcined in furnace at 823 K
(ramping rate: 2 K min−1) with flowing air for 5 h to yield the meso-
porous SBA-15 support.
X%B-10%Ni/SBA-15 (X%: 0-5 wt%) catalysts were synthesized
using a sequential incipient wetness impregnation (SIWI) approach
employing as-prepared SBA-15 support, Ni(NO3)2.6H2O metal pre-
cursor and boric acid (H3BO3) promoter precursor solutions (supplied
by Sigma-Aldrich Chemicals). Accurately balanced amounts of Ni
(NO3)2.6H2O solution and original SBA-15 support were mixed and
thoroughly stirred at 333 K for 3 h in a rotary evaporator (BÜCHI
Rotavapor R-200) at vacuum conditions.
The mixture was further dehydrated by heating up to 373 K for 24 h
before being air-calcined for 5 h at 1073 K with a heating rate of
2 K min−1 to produce an unpromoted 10%Ni/SBA-15 catalyst. The
incipient wetness impregnation approach between abovementioned
unpromoted 10%Ni/SBA-15 catalyst and different calculated quantities
of H3BO3 aqueous solution was further conducted using the same
aforementioned experimental procedure in order to synthesize B-pro-
moted catalysts.
2.2. Characterization of catalyst
The physical attributes (viz., Brunauer-Emmett-Teller (BET) surface
area, average pore diameter and total pore volume) of unpromoted, B-
promoted Ni-based catalysts and SBA-15 support were measured in
Micromeritics ASAP-2010 equipment using N2 adsorption-desorption
isotherms data obtained at 77 K. Before conducting any BET measure-
ments, the specimen was pretreated at 573 K for about 1 h in N2 flow to
remove traced moisture and possibly volatile contamination on sample
surface. The attenuated total reflectance Fourier-transform infrared
(ATR-FTIR) spectroscopy of support and catalysts was recorded within
400–4000 cm−1 with spectral resolution of 4 cm−1 and scan number of
100 on a Thermo Fisher Scientific Nicolet iS5 FTIR spectrometer at-
tached with iD7 ATR component.
The low-angle X-ray diffraction (XRD) experiments for fresh SBA-15
support and catalysts were conducted on a Philips X′ Pert MPD (3 kW)
diffractometer employing Cu Kα as radiation source (with a wave-
length, λ of 1.5405 Å). All samples were scanned from 2θ of 0.5° to 5°
with a step size of 0.005°. In addition, the Rigaku Miniflex II system
with a Cu monochromatic X-ray radiation (λ = 1.5418 Å) was utilized
for wide-angle XRD measurements. This system was operated at 15 mA
and 30 kV. Fairly small scan speed (1° min−1) and step size (0.02°) were
applied for all samples to obtain high-resolution patterns during 2θ
scanning range from 3° to 80°. In order to calculate average NiO crys-
tallite size, d(NiO), Scherrer equation given in Eq. (8) was employed
T.J. Siang et al. Microporous and Mesoporous Materials 262 (2018) 122–132

[23]. activity. The loaded catalyst was initially reduced in-situ with 50%H2/
N2 mixture (60 mL min−1) for 2 h at 1073 K and further purged with
0.94λ
d (NiO ) =
(8) 60 mL min−1 of N2 for 30 min before swapping to the mixture of
B × cos θ
gaseous reactants for CSCRM reaction. The Agilent 6890 Series GC
where B, λ and θ represent the corresponding line broadening at half equipment installed with capillary column (HP-PLOT Q) and TCD de-
the maximum intensity, X-ray wavelength and Bragg angle. By em- tector was used to analyze the gaseous effluent composition from the
ploying the relative molar volumes of NiO and metallic Ni0 phases, the outlet of reactor. Cold trap and drierite adsorbent bed were installed on-
particle size of metallic Ni0, d(Ni0) after the H2 reduction of NiO par- line to condense and adsorb the unreacted steam in effluent gases be-
ticles could be approximately computed from Eq. (9) [22,24]. fore GC analysis.
d (Ni 0) = 0.84d (NiO) The CSCRM reaction's catalytic performance was assessed based on
(9)
CH4 conversion ( XCH4 ), CO2 conversion ( XCO2 ), H2 yield (YH2 ), CO yield
The AutoChem II-2920 system was employed to conduct the H2 (YCO ) and H2/CO ratio as shown in Eqs. (10)–(14), respectively.
temperature-programmed reduction (H2-TPR) measurement. For each In Out
measurement, around 0.1 g of sample was sandwiched and mounted at FCH 4
− FCH4
XCH4 (%) = In
× 100%
the centre of a quartz U-tube by quartz wool. The pre-treatment was FCH 4 (10)
carried out in flowing He inert gas (50 mL min−1) at 373 K about 0.5 h
In Out
for the elimination of moisture prior to H2 reduction. The specimen FCO 2
− FCO 2
XCO2 (%) = In
× 100%
underwent reduction step with a temperature range between 373 and FCO 2 (11)
1173 K using ramping rate of 10 K min−1 in 10%H2/Ar (50 mL min−1)
mixture. Raman spectroscopy analysis of fresh and spent catalysts as 2FHOut
2
YH2 (%) = In
× 100%
well as SBA-15 support was performed on a JASCO NRS-3100 Raman 4FCH 4
+ 2FHIn2 O (12)
spectrometer unit with a green laser of 532 nm and laser power of
Out
below 5 mW during measurement in ambient air. FCO
YCO (%) = × 100%
The Carl Zeiss AG - EVO® 50 Series instrument controlled by a In
FCH 4
In
+ FCO 2 (13)
SmartSEM software was used for conducting scanning electron micro-
scopy (SEM) measurements for fresh and spent catalysts. The specimens and
were coated on platinum plate using a BAL-TEC SCD 005 Sputter Coater FHOut
2
to enhance the sharpness of SEM images. Transmission electron mi- Ratio of H2 to CO = Out
FCO (14)
croscopy (TEM) measurements were also carried out for selective fresh
and spent catalysts by a FEI Tecnai G2 F20 TEM unit. Prior to TEM test, where F In is inlet molar flow rate and F Out is outlet molar flow rate (mol
the solid sample was suspended in ethanol solution for 1 h to minimize s−1).
particle agglomeration and single drop of this slurry was positioned on
a copper gridiron layered with permeable carbon film. 3. Results and discussion
Temperature-programmed oxidation (TPO) measurements were
employed to identify and quantify deposited carbon on spent catalyst 3.1. Physicochemical attributes of catalysts
surface. The TPO tests were conducted for unpromoted and selective
promoted catalysts after CSCRM reaction on a TGA Q500 unit (TA 3.1.1. Nitrogen adsorption studies
Instruments). The specimen was heated up to 373 K from ambient The summary of textural attributes for SBA-15 support, unpromoted
temperature and the temperature was kept constant for 30 min to en- and boron-promoted Ni-based catalysts with different B loadings is
sure the thorough removal of moisture and volatile compounds. About given in Table 1. SBA-15 support possessed a high BET surface area of
100 mL min−1 of purging 20%O2/N2 mixture was introduced to the 780.1 m2 g−1 comparable with other studies [25,26]. A noticeable and
system whilst temperature was grown to 1023 K with controlled inevitable decrease in textural properties including BET surface area
heating rate of 10 K min−1. Specimen was held isothermally for and total pore volume with NiO and B2O3 addition during impregnation
30 min at 1023 K in the former purging gaseous mixture to confirm the was observed for unpromoted and promoted catalysts. However, in
completed oxidation of deposited carbon before being cooled down to comparison between SBA-15 support and catalyst, average pore dia-
room temperature. meter of SBA-15 support (6.94 nm) did not change appreciably with
NiO and B2O3 addition. Therefore, the substantial reduction of BET
2.3. CSCRM reaction surface area and slight drop in mean pore diameter of catalysts com-
pared with SBA-15 support could indicate that the loaded NiO and B2O3
CSCRM reaction was conducted at 973–1073 K and stoichiometric particles may reside on the external support surface.
feed composition (FCH4 : FH2 O : FCO2 = 3: 2: 1 with F being molar flow BET surface area for 10%Ni/SBA-15 catalyst (408.1 m2 g−1)
rates (mol s−1)) under atmospheric pressure in a fixed-bed reactor dropped significantly with increasing boron loading to 5.4 m2 g−1 (5%
(quartz tube with outer diameter of 3/8 inch and length of 17 inch). B loading) as seen in Table 1. This observation could be due to the
The quartz wool was used to mount about 0.1 g of X%B-10%Ni/SBA-15 agglomeration of B2O3 and NiO particles plugging the mesoporous
catalysts at reactor centre placed axially in a divided LT furnace which structure of SBA-15 support at high B loading. Eswaramoorthi and Dalai
was continuously monitored by a temperature controller during time- also observed the reduction in BET surface area with B-addition and
on-stream. The purified water precisely fed by KellyMed KL-602 syringe attributed it to the deboration reaction during calcination (cf. Eq. (15))
pump to the top of reactor was vaporized and mixed with other gaseous and hence blockage of mesopores [27].
reactants (viz., CH4 and CO2) previously diluted with N2 inert gas. The
2H3 BO3 → B2 O3 + 3H2 O (15)
Alicat mass flow controllers were utilized for precisely regulating the
flow rates of all gases. The N2 diluent gas was employed for the purpose In addition, the considerably low BET surface area of 5%B-10%Ni/
of material balance and ensuring the unchanged overall flow rate of SBA-15 catalyst (5.4 m2 g−1) was most likely owing to the collapse of
gaseous mixture (60 mL min−1) for each run. The small average par- thin pore walls of SBA-15 support in extremely acidic environment with
ticle size ranging of 100–140 μm and fast gas hourly space velocity high boric acid concentration [28]. In fact, the total pore volume also
(GHSV = 36 l gcat−1 h−1) were used to neglect the existence of external experienced a pronounced decline with rising B loading (cf. Table 1)
and internal transport resistances for achieving the intrinsic catalytic further confirming the loss in BET surface area.

124
T.J. Siang et al. Microporous and Mesoporous Materials 262 (2018) 122–132

Table 1
Textural attributes for unpromoted, B-promoted 10%Ni/SBA-15 catalysts and SBA-15 support.

Sample Average BET surface area Total pore volume, Vp (cm3 g−1)a Average pore diameter, Dp Average crystallite size of Average crystallite size of
(m2 g−1) (nm)b NiO, d(NiO)c (nm) Ni0, d(Ni0)d (nm)

SBA-15 support 780.1 ± 8.07 1.35 ± 6.77 × 10−3 6.94 ± 0.03 – –

10%Ni/SBA-15 493.7 ± 3.74 0.97 ± 4.86 × 10−3 6.69 ± 0.03 15.6 13.1
1%B-10%Ni/SBA- 408.1 ± 4.35 0.68 ± 3.41 × 10−3 6.68 ± 0.03 20.4 17.1
15
3%B-10%Ni/SBA- 213.2 ± 2.35 0.35 ± 1.75 × 10−3 6.52 ± 0.03 20.5 17.2
15
5%B-10%Ni/SBA- 5.4 ± 0.14 0.01 ± 5.01 × 10−5 5.84 ± 0.02 21.4 18.0
15

a
Total pore volume was achieved from p/p0 = 0.99.
b
Average pore diameter was computed from Barret-Joyner-Halenda desorption approach.
c
Average NiO crystal size was estimated using Scherrer equation for the highest NiO peak (2θ = 43.54°).
d
Average crystallite size of Ni0 was estimated from Eq. (9).

Fig. 1. N2 physisorption curves for unpromoted, B-promoted Ni/SBA-15 samples and

SBA-15 support.
Fig. 2. Pore size distribution for unpromoted, B-promoted Ni/SBA-15 samples and SBA-
15 support.
The N2 physisorption isotherms of SBA-15 support, unpromoted and
B-promoted catalysts are displayed in Fig. 1 to assess the meso-structure
B2O3 addition reduced significantly peak intensity and shifted peak
of catalysts. According to the IUPAC classification, the isotherm curves
location to smaller pore size region. These observations could be as-
of SBA-15 support, 10%Ni/SBA-15, 1%B-10%Ni/SBA-15 and 3%B-10%
cribed to the partial blockage of primary mesopores associated with the
Ni/SBA-15 catalysts could be categorized as type IV isotherm curve
agglomeration of NiO and B2O3 particles on support surface [30]. In
possessing a H1-type hysteresis loop which is a typical feature of me-
addition, the absence of main peak within 4–10 nm for 5%B-10%Ni/
soporous material structure with cylinder-shaped channels [29,30]. In
SBA-15 catalyst could be caused by collapsed SBA-15 pore walls during
addition, at p/p0 ranging from 0.50 to 0.85 in these isotherm curves,
impregnation in acidic environment induced by high boric acid con-
the detected sharp inflection corresponded to capillary N2 condensation
centration.
within uniform mesopores indicating the existence of large mesopores
[31].
As seen in Fig. 1, the similar N2 adsorption-desorption curves for
3.1.2. FTIR analysis
SBA-15 support, 10%Ni/SBA-15, 1%B-10%Ni/SBA-15 and 3%B-10%
The FTIR spectra for unpromoted, B-promoted 10%Ni/SBA-15 cat-
Ni/SBA-15 catalysts could suggest that the meso-structure in SBA-15
alysts and bare SBA-15 support are depicted in Fig. 3. FTIR spectrum for
support was most likely retained after the addition of NiO and B2O3
bare SBA-15 support (see Fig. 3 (a)) illustrates that the significant and
particles. However, the isotherms of 5%B-10%Ni/SBA-15 catalyst
broad absorption band ranging from 1000 to 1300 cm−1 and possessing
showed the completely different pattern without the existence of H1-
a hump at about 1225 cm−1 was induced by the asymmetric stretching
type hysteresis loop further corroborating the collapse of mesoporous
(AS) types of Si-O-Si vibrational moiety [32]. Indeed, the AS motion of
SBA-15 structure at high boron loading.
Si-O-Si bond was composed of two main vibrational modes, namely,
The pore size distribution for SBA-15 support, unpromoted and B-
AS1 and AS2 modes. In the first AS1 mode, two adjacent oxygen atoms
promoted catalysts (see Fig. 2) was interpreted from the desorption
reportedly perform the AS motion in phase with the middle silicon atom
branches of N2 isotherms using Barret-Joyner-Halenda (BJH) approach.
while the AS2 mode belongs to the out-of-phase movement for two
Apart from 5%B-10%Ni/SBA-15 catalyst, SBA-15 support and catalysts
adjacent oxygen atoms with the middle silicon atom [32]. Particularly,
possessed narrow pore size distribution with mesopores ranging from 4
the absorbance peaks located at 1225 and 1059 cm−1 were caused by
to 10 nm. However, as seen in Fig. 2, the pore size distribution peak of
the corresponding longitudinal-optic (LO) and transverse-optic (TO)
SBA-15 support was noticeably widened with NiO addition and it was
vibrational modes of AS1 stretching of Si-O-Si bond [32,33]. In addi-
further expanded with increasing boron loadings. Additionally, NiO and
tion, the detected bands at about 450 and 814 cm−1 could be indicative

125
T.J. Siang et al. Microporous and Mesoporous Materials 262 (2018) 122–132

Fig. 3. FTIR patterns for (a) SBA-15 support, (b) 10%Ni/SBA-15, (c) 1%B-10%Ni/SBA- Fig. 4. Low-angle X-ray diffractograms for fresh (a) SBA-15 support, (b) 10%Ni/SBA-15,
15, (d) 3%B-10%Ni/SBA-15 and (e) 5%B-10%Ni/SBA-15 catalysts. (c) 1%B-10%Ni/SBA-15, (d) 3%B-10%Ni/SBA-15 and (e) 5%B-10%Ni/SBA-15 catalysts.

of asymmetric and symmetric stretching vibrations of Si-O bond in this

order [34]. Additionally, the small peak detected at 965 cm−1 for the
bare SBA-15 support was probably owing to Si-OH group vibration
[35]. The appearance of these typical bonds, viz., Si-O-Si, Si-O and Si-
OH could be indicative of the presence of mesoporous SBA-15 support
in agreement with other studies [34]. These aforementioned char-
acteristic bands were also observed for both unpromoted and B-pro-
moted Ni/SBA-15 catalysts regardless of B loadings (see Fig. 3(b)–(e)).
As seen in Fig. 3(c)–(e), for B-promoted catalysts, the main band
detected in the high wavenumber of 1280–1530 cm−1 was possibly a
result of the asymmetric stretching relaxation belonging to B-O bonds of
the trigonal borate anions (BO33−) in various borate groups [36,37]. A
low intensity peak located at around 670 cm−1 was also attributed to B-
O-B bending vibrations [37,38]. The detection of above-mentioned B-O
and B-O-B vibrations could suggest vitreous B2O3 phase formation on
the surface of B-promoted catalysts. Rada et al. also reported that the
glassy B2O3 form normally exists in the boroxol rings (B3O6), which are
interconnected by individual [BO3] groups [36]. Moreover, the in-
tensity of both B-O and B-O-B bands gradually strengthens with an
increase in B content (see Fig. 3(c)–(e)) rationally owing to the rising
amount of B2O3 formation.
Fig. 5. XRD patterns for (a) SBA-15 support, (b) 10%Ni/SBA-15, (c) 1%B-10%Ni/SBA-15,
(d) 3%B-10%Ni/SBA-15 and (e) 5%B-10%Ni/SBA-15 samples.

3.1.3. XRD measurements

Fig. 4 illustrates low-angle X-ray diffractograms for SBA-15-sup-
ported Ni catalysts and fresh SBA-15 support. The low-angle XRD pat-
tern for bare SBA-15 support (see Fig. 4 (a)) exhibited an intensive main
diffraction peak (2θ of around 0.89°) and two small peaks (at 2θ of
2.27° and 2.64° as seen in inset of Fig. 4) indexed as the corresponding
(100), (110) and (200) reflections of two dimensional, ordered hex-
agonal meso-structure associated with p6mm hexagonal symmetry
[39,40].
As seen in Fig. 4 (b)-(e), all promoted and unpromoted catalysts
possessed the main (100) peak located at 2θ = 0.89° with similar in-
tensity. However, the intensity of (110) and (200) peaks decreased with
rising B loadings and these characteristic peaks completely disappeared
in 5%B-10%Ni/SBA-15 catalyst. Thus, this behavior could indicate that
the incorporation of acidic B2O3 promoter may distort or degrade the
pore walls inside SBA-15 support.
Fig. 5 displays the wide-angle X-ray diffractograms for fresh X%B-
10%Ni/SBA-15 catalysts and SBA-15 support. All XRD patterns were
assessed by employing database from Joint Committee on Power Dif-
fraction Standards (JCPDS) [41]. XRD pattern for SBA-15 support was
also used for comparison and facilitation of XRD analysis. A similar
broad diffraction peak was clearly observed for all samples at 2θ of
around 22.16° assigned to the presence of amorphous silica framework
in SBA-15 support (JCPDS card No. 29-0085) [42]. In addition, the five
sharp and high peaks identified at 2θ of 37.46°, 43.54°, 63.03°, 75.14°
and 79.52° were assigned to the (1 1 1), (2 0 0), (2 2 0), (3 1 1) and (2 2
2) NiO crystal planes (JCPDS card No. 47-1049), respectively for X%B-
10%Ni/SBA-15 catalysts (see Fig. 5 (b)-(e)) [43]. Additionally, there
were no typical peaks corresponding to any boron oxide phases de-
tected. This could be due to the fine dispersion of boron oxide particles
which in turn possessed a substantially tiny crystallite size below the
XRD detection capability. In fact, Fouskas et al. also observed the same
behavior for boron-modified Ni/Al2O3 catalysts during XRD measure-
ments and suggested the formation of amorphous glassy boron oxide
phases since boron oxides were reportedly difficult to be crystallized
[28].
The average crystallite size of NiO particles and metallic Ni0 parti-
cles for all catalysts was estimated using Scherrer equation and

126
T.J. Siang et al.
Fig. 6. H2-TPR profiles for (a) 10%Ni/SBA-15, (b) 1%B-10%Ni/SBA-15, (c) 3%B-10%Ni/
SBA-15 and (d) 5%B-10%Ni/SBA-15 samples.
summarized in Table 1. The NiO crystallite size somewhat increased
from 15.6 to 21.4 nm with rising B-loadings from 0 to 5%, respectively
and the same trend was evidenced for the average crystallite size of
metallic Ni0 phase. The increasing crystallite size with B-addition could
be due to the reducing composition of SBA-15 support at high boron
loading and hence growing metal agglomeration on support surface.
The mean crystallite size for NiO particles was significantly superior to
the mean pore diameter for catalysts and SBA-15 support (cf. Table 1).
Thus, this observation could further affirm that NiO particles were lo-
cated on support surface.
3.1.4. H2-TPR analyses of fresh catalysts
H2-TPR analyses for X%B-10%Ni/SBA-15 catalysts carried out to
investigate the reducibility of active metal oxide are shown in Fig. 6.
The process of NiO reduction to metallic Ni0 phase was reportedly a
direct process without intermediate form formation [44]. In addition,
boric oxide was allegedly stable and resistant to H2 reduction [45].
Thus, the two noteworthy peaks (P1 and P2) observed for all catalysts
were ascribed to NiO reduction to metallic Ni0 phase reliant on metal-
support interaction level in association with NiO crystallite size and its
location on mesoporous SBA-15 support [30]. Indeed, peak P1 (at low
reduction temperature) could belong to reducing process for large or
external NiO particles having a slight interaction with support whilst
the reducibility of NiO nanoparticles (positioned on support surface or
Fig. 7. SEM micrographs for fresh (a) 10%Ni/SBA-15 and (b) 3%B-10%Ni/SBA-15 catalysts.
127
Microporous and Mesoporous Materials 262 (2018) 122–132
within meso-channels of support) possessing a strong interaction be-
tween metal and support could account for the appearance of second
reduction peak P2 located at greater temperature [44,46].
Notably, both peaks (P1 and P2) appeared to be shifted towards
higher temperature region with B-addition and the reduction tem-
perature increased linearly with rising B loading (see Fig. S1 in sup-
plementary data) most likely due to the strengthening interaction be-
tween the basic NiO and acidic B2O3 phases [37,47]. In fact, Xu and
Saeys also found that boron promoter favorably adsorbs in the octa-
hedral sites of the top Ni (1 1 1) subsurface layer by using ab initio
density functional theory calculations further confirming the strong
interaction between B2O3 and NiO phases [20].
As seen in Fig. 6, the absence of H2 consumption peaks for both
promoted and unpromoted catalysts at beyond 1070 K could indicate
that all NiO particles were completely reduced in H2 pretreatment.
Thus, high reduction temperature (1073 K) with long holding time (2 h)
was implemented in this study for activating all catalysts during H2
reduction. In addition, the amount of H2-uptake (see Fig. S2 of sup-
plementary data) increased significantly from 0.11 mmol H2 g−1 (for
10%Ni/SBA-15 sample) to the highest value as 0.56 mmol H2 g−1 (3%
B-10%Ni/SBA-15 sample) because of the increasing electron density
donated by boron promoter to Ni metal oxide and hence facilitating H2
reduction. However, an appreciable decline in H2-uptake at higher B
loading beyond 3%B was observed as seen in Fig. S2. This behavior
could be ascribed to the surplus coexistence of B2O3 particles covering
NiO surfaces and preventing them from the access of gaseous H2 re-
ducing agent.
3.1.5. SEM and TEM measurements
To investigate the morphology of catalysts, SEM measurements
were conducted for unpromoted catalyst and representative B-pro-
moted catalyst. The SEM micrographs of fresh 10%Ni/SBA-15 and se-
lected 3%B-promoted 10%Ni/SBA-15 samples are depicted in Fig. 7.
The SEM micrograph of unpromoted catalyst (see Fig. 7 (a)) demon-
strates that SBA-15 support had somewhat uniform and small rope-
shaped particles and these nanoparticles were stacked together to form
a wheat-like macro-structure similar to other studies [48,49]. As shown
in Fig. 7 (b), there was no noticeable change in surface morphology and
significant alteration in macroscopic structure of 3%B-10%Ni/SBA-15
catalyst indicating structural integrity during B-addition.
The TEM measurements were also performed to examine the impact
of NiO and B2O3 addition on the morphology of catalysts. As seen in
Fig. 8 (a), the feature of SBA-15 support could be visibly observed with
the presence of hexagonal arrays and cylindrical mesopores in long-
itudinal ordered structure arrangement. Additionally, the clear and
dark thin lines corresponded to the empty mesopores and pore walls,
respectively in agreement with other studies [50]. As exhibited in Fig. 8
(b), the TEM micrograph of 10%Ni/SBA-15 catalyst reveals that small
T.J. Siang et al. Microporous and Mesoporous Materials 262 (2018) 122–132

Fig. 8. TEM micrographs for fresh (a) SBA-15

support, (b) 10%Ni/SBA-15, (c) 3%B-10%Ni/
SBA-15 and (d) 5%B-10%Ni/SBA-15 samples.

NiO particles present on mesoporous support surface. However, the

crystallite size for NiO phase appeared to be larger with the addition of
B2O3 promoter (see Fig. 8 (c) and (d)). Whereas as depicted in Fig. 8 (c),
the meso-structure in SBA-15 support was still retained for 3%B-10%
Ni/SBA-15 catalyst, the disappearance of the ordered mesopores for 5%
B-10%Ni/SBA-15 catalyst (referring to Fig. 8 (d)) was observed and
possibly due to the damage of thin pore walls of SBA-15 support at
excessive boric acid environment in agreement with N2 adsorption-
desorption isotherms (see Fig. 1) and low-angle XRD (see Fig. 4) results.
As a result of collapsed pore walls in 5%B-10%Ni/SBA-15 catalyst, NiO
particles could migrate to outer support surface from porous channels
and hence aggregating there to form large NiO clusters (see Fig. 8 (d)).

3.2. Influence of reaction temperature on CSCRM performance

The impact of reaction temperature on catalytic performance for

CSCRM was investigated at stoichiometric feedstock composition (i.e.,
FCH4 : FH2 O : FCO2 = 3: 2: 1) and varying temperature from 973 K to
1073 K for 10 h on-stream. As seen in Fig. 9, regardless of boron
loading, both reactant conversions enhanced with a rise in reaction
temperature from 973 K to 1073 K rationally owing to the CSCRM
endothermic character [51]. In addition, 3%B loading appeared to be
the optimal promoter content in terms of reactant conversions irre-
spective of reaction temperature. For 3%B-10%Ni/SBA-15 catalyst, CH4
and CO2 conversions were improved appreciably about 23.2% and
32.4%, correspondingly with increasing reaction temperature within
973–1073 K. The enhancement in reactant conversions with rising B
loading from 1 to 3% was possibly owing to the suppression of de-
posited carbon on catalyst surface consistent with other studies [28]. In
the density functional theory study of carbon chemisorption on surface
of boron-promoted Ni (1 1 1), Xu and Saeys reported that B atoms could
reside on the octahedral sites of the first subsurface layer of active Ni
atoms and hence hinder the diffusion of adsorbed surface carbon (CxH1-
x with x ≤ 1 formed from dissociative CH4 adsorption [16,52]) into Ni
lattice [20,21]. As a result, it not only inhibits the bulk carbon and
graphene islands formation but also retains the on-surface carbon for
subsequent gasification by CO2 and H2O oxidizing agents. In addition,
Chen et al. also found that electron density on the surface of active Ni
Fig. 9. Impact of B loading and reaction temperature on (a) CO2 and (b) CH4 conversions
of unpromoted and promoted 10%Ni/SBA-15 catalysts with stoichiometric feed compo-
sition (FCH4: FH2 O: FCO2 = 3: 2: 1).
metal was increased due to electron donation from boron dopant [53].
Therefore, the enhancement of electron density on catalyst surface
could facilitate CO2 adsorption and hence increasing catalytic activity.
However, at greater B loading beyond 3%, a substantial drop in
reactant conversions was observed for all reaction temperature em-
ployed as seen in Fig. 9. This could be due to the tremendous decline in
BET surface area at high B loading (see Table 1). Additionally, the ex-
cess loading of B dopant could agglomerate on catalyst surface and
isolate the active Ni metal from the access of gaseous reactants, re-
sulting in the reduction in catalytic activity. In fact, the degree of in-
teraction between NiO particle and B2O3 dopant was highest at 5%B
loading as observed from H2-TPR measurements (cf. Fig. 6). The drop in
reactant conversions at surplus B loading could be also ascribed to the
difference of electronic environment possessed by Ni species depending

128
T.J. Siang et al.
Fig. 10. Impact of B loading and reaction temperature on (a) CO yield and (b) H2 yield of
unpromoted and promoted 10%Ni/SBA-15 catalysts with stoichiometric feed composition
(FCH4: FH2 O: FCO2 = 3: 2: 1).
on the amount of boron loading [28]. Xu and Saeys revealed that the
activation energy for CH4 dissociation over the B-promoted Ni (111)
surface was slightly raised by 12 kJ mol−1 compared to that of un-
promoted Ni (111) surface [21]. They also found that the centre of the
3d states for the Ni (111) surface of B-promoted catalysts was reduced
about 0.26 eV and hence weakening the interaction between anti-
bonding C-H orbital of CH4 molecule and Ni 3d electrons. This behavior
could result in a greater activation barrier for C-H bond cleavage. In-
terestingly, the relationship between reactant conversions vs. B loading
in Fig. 9 was parallel to the trend of H2-uptake vs. B content (see Fig. S2
in supplementary data). Therefore, the degree of H2 reduction de-
pending on the amount of B loading could also affect the catalytic
performance and contribute to explain the behavior of reactant con-
versions against promoter loading. Indeed, 3%B-10%Ni/SBA-15 cata-
lyst possessed the highest H2-uptake during H2 reduction. Thus, it could
have the highest number of active Ni metallic sites resulting in the
greatest catalytic activity.
The impact of reaction temperature and boron loading on H2 and
CO yields for X%B- 10%Ni/SBA-15 catalysts is displayed in Fig. 10. For
all boron loadings, H2 and CO yields improved with growing tem-
perature range within 973–1073 K. The enhancing product yields with
temperature could be attributed to the endothermic nature of CSCRM
leading to growing H2 and CO formation rates. Additionally, the op-
timal product yields were obtained at boron loading of about 1–3%B. In
addition, the highest CO and H2 yields were around 57.8% and 69.4%
at 1073 K, respectively comparable with other studies using noble
catalysts [54].
Fig. 11 displays the impact of reaction temperature and B loading on
ratio of H2 to CO for both promoted and unpromoted catalysts. The
ratio of H2 to CO varied within 1.26–2.71 for all catalysts. Irrespective
of catalysts used, increasing reaction temperature from 973 to 1073 K
resulted in an increment in H2 to CO ratio. This observation could be
owing to the enhancement of endothermic methane steam reforming
reaction reportedly favored for generating H2-rich syngas (H2/CO ≥ 3)
[51]. In addition, H2/CO ratio was increased with the addition of B
promoter regardless of reaction temperature.
As seen in Fig. 11, the ratio of H2 to CO could be easily manipulated
within the range of 1.26–2.71 by changing reaction temperature and B
content. Remarkably, the ideal H2/CO ratio of about 2 could be
achieved over 3%B-10%Ni/SBA-15 sample at 973 K. This desirable H2/
CO ratio was appropriate for downstream Fischer-Tropsch synthesis to
129
Microporous and Mesoporous Materials 262 (2018) 122–132
Fig. 11. Impact of B loading and reaction temperature on H2/CO ratio for unpromoted
and promoted 10%Ni/SBA-15 catalysts with stoichiometric feed composition
(FCH4: FH2 O: FCO2 = 3: 2: 1 ).
generate long-chain hydrocarbons without the requirement of mod-
ification of feed composition [55].
3.3. Characterization of spent catalysts
Characterizing spent catalysts following CSCRM tests was also
conducted for examining morphology, crystalline structure and carbon
content as well as carbonaceous types on catalyst surface after CSCRM.
Since 3%B-10%Ni/SBA-15 sample seemed to be the optimal catalyst
with respect to reactant conversions and H2 yield at 1073 K, this used
catalyst was further selected for post-reaction characterization and the
resulting physicochemical attributes of it were compared with those of
spent unpromoted catalyst in this section.
3.3.1. Raman spectroscopy measurements
The type of carbon deposits present on the surface of used catalysts
was inspected by Raman measurements. Raman spectra of fresh SBA-15
support and catalysts are also performed for comparing with spent
catalysts and easing data interpretation. As seen in Fig. S3 (referring to
supplementary data), for fresh SBA-15 support and SBA-15 supported
catalysts, two typical peaks detected at Raman shift of about
487.3 cm−1 (P1) and 1068.7 cm−1 (P2) were attributed to the corre-
sponding cyclic tetrasiloxane rings and Si-OH stretching mode, typical
for SBA-15 support [56,57] in agreement with FTIR results (cf. Fig. 3).
As seen in Fig. S3 (supplementary data), the intensity of peak P1 for
promoted and unpromoted 10%Ni/SBA-15 catalysts was considerably
greater in comparison with original SBA-15 support. This could be as-
cribed to the presence of NiO phase exhibiting Ni-O bond oscillation at
the similar Raman band [58,59]. However, boric oxide phase was also
not identified for all B-promoted 10%Ni/SBA-15 catalysts probably
because of the vitreous B2O3 feature [60] consistent with the observa-
tion from XRD measurements (see Fig. 5).
As seen in Raman spectra of spent catalysts (cf. Fig. 12 (b)-(e)), both
peaks P1 and P2 representative for SBA-15 phase were also detected but
the intensity of the first peak P1 was substantially smaller than that of
fresh catalyst (see Fig. 12 (a)). As aforementioned, the characteristic
peaks of NiO phase and SBA-15 support were overlapped at Raman
band of peak P1. Thus, a drop in the intensity of peak P1 in comparison
with fresh catalyst could be ascribed to the absence of NiO phase, which
was previously reduced in H2 for activation to active Ni0 form prior to
CSCRM.
T.J. Siang et al.
Fig. 12. Raman spectra of fresh (a) 10%Ni/SBA-15, spent (b) 10%Ni/SBA-15, (c) 1%B-
10%Ni/SBA-15, (d) 3%B-10%Ni/SBA-15 and (e) 5%B-10%Ni/SBA-15 catalysts after
CSCRM with stoichiometric feed composition (FCH4: FH2 O: FCO2 = 3: 2: 1) and 1073 K.
In addition, as seen in Fig. 12 (b), two major absorption peaks were
only detected on used 10%Ni/SBA-15 catalyst at bands of 1340 and
1572 cm−1 for the corresponding D and G bands indicative of existing
heterogeneity of carbonaceous deposits on catalyst surface. In fact, the
D band is reportedly induced by defective sp3-bonded carbon atoms of
carbon nanofilament (CNF) or amorphous carbon whilst the G band
corresponds to the existence of an ordered carbon structure, viz., gra-
phitic carbon due to the carbon-carbon stretching vibrations of sp2
carbon (C=C) bonds [19,61].
3.3.2. XRD measurement of spent catalyst
The XRD analysis was also carried out for examining the crystalline
phases of both spent unpromoted and 3%B-promoted catalysts after
CSCRM reaction for 10 h on-stream at 1073 K and stoichiometric feed
composition. The apparently observed broad peak ranging from 15° to
30° with a small intense peak (located at 2θ = 26.63°) on the shoulder
(cf. Fig. 13 (a)) could be ascribed to the overlapped peaks between
amorphous SiO2 (2θ = 22.16°) phase and graphite with 2θ value of
26.63° (JCPDS card No. 75-1621) for used 10%Ni/SBA-15 sample [62].
In comparison with XRD patterns of spent promoted catalyst (see
Fig. 13 (b)), the similar bell-shaped curves of SiO2 peaks for both spent
promoted and unpromoted catalysts could suggest that the typical si-
liceous framework of mesoporous SBA-15 support was largely retained
with B-addition during CSCRM reaction in agreement with TEM images
(see Fig. 15 discussed later in section 3.3.3).
As seen in Fig. 13 (b), no characteristic peak corresponding to
graphitic carbon detected at the same position for the spent 3%B-10%
Ni/SBA-15 catalyst indicated that spent B-promoted catalyst surface
was free from graphite. The absence of graphitic carbon on B-promoted
catalyst could be a result of strong B adsorption on octahedral sites
belonging to subsurface layer of active Ni phase and hence inhibiting
reactive surface carbon diffusion into the bulk of Ni to form graphene
islands [28]. This result is in line with the observation from Raman
spectroscopy measurements (see Fig. 12).
Additionally, the characteristic peaks at 2θ = 44.58°, 51.86° and
76.38° for both spent unpromoted and promoted 10%Ni/SBA-15 cata-
lysts were attributed to various crystal planes of metallic Ni0 form,
namely, (1 1 1), (2 0 0) and (2 2 0), respectively (JCPDS card No. 04-
0850) [63]. No characteristic NiO peaks were observed in X-ray dif-
fractograms of both used promoted and unpromoted catalysts in
agreement with Raman analysis of spent catalysts (see Fig. 12). This
130
Microporous and Mesoporous Materials 262 (2018) 122–132
Fig. 13. XRD patterns of spent (a) 10%Ni/SBA-15 and (b) 3%B-10%Ni/SBA-15 catalysts
after CSCRM with stoichiometric feed composition (FCH4: FH2 O: FCO2 = 3: 2: 1) and
1073 K.
behavior could suggest that the active Ni0 particles were largely pre-
served on catalyst surface during CSCRM reaction and resisted to re-
oxidation for forming NiO particles possibly owing to the presence of
reducing gaseous products (viz., CO and H2) and the strong metal-
support interaction. The re-oxidation of metallic Ni0 phase to NiO form
during CSCRM reaction has been widely reported in literature [64] and
is considered as one of the main factors inducing severe catalytic de-
activation [65]. Thus, the resistance to re-oxidation is most likely an
outstanding attribute of the as-synthesized Ni/SBA-15 catalysts in this
study.
3.3.3. SEM and TEM analyses of spent catalysts
Fig. 14 shows SEM micrographs of both spent unpromoted and re-
presentative 3%B-10%Ni/SBA-15 catalysts achieved from CSCRM tests
at 1073 K. The wheat-like macroscopic structure of spent catalysts
visibly seemed to be mainly preserved. For spent unpromoted 10%Ni/
SBA-15 catalyst, carbon nanofilament was apparently observed and
surrounded on used catalyst surface after CSCRM reaction (see green
circles in Fig. 14 (a)). Indeed, CNF formation during CSCRM was also
reported in other studies [7,11]. However, as seen in Fig. 14 (b), the
catalyst surface of 3%B-10%Ni/SBA-15 appeared to be free from fila-
mentous carbon. This could indicate that there was no formation of CNF
species or CNF could be present on catalyst surface with a significantly
low amount inferior to the detection limit of SEM measurement. The
carbon resistance of B-promoted catalyst was possibly due to the pre-
ferential occupation of boron on the subsurface octahedral sites of Ni
atoms and hence blocking carbon diffusion for forming graphene is-
lands [20].
Fig. 15 displays the TEM images of spent unpromoted and re-
presentative 3%B-10%Ni/SBA-15 catalysts. In comparison between
spent promoted and unpromoted catalysts, the ordered hexagonal
structure and cylindrical pores belonging to the intrinsic nature of SBA-
15 support were largely retained with B-addition during CSCRM reac-
tion. Unlike SEM images (Fig. 14), CNF species were observed on the
surface of spent unpromoted and promoted catalysts. However, as seen
in Fig. 15 (a), graphitic carbon encapsulating Ni particles was visibly
observed for spent unpromoted catalyst (referring to green circle). This
could contribute to explain the lowest catalytic activity of unpromoted
catalyst for CSCRM reaction. In addition, as previously mentioned, the
absence of graphitic carbon (see Fig. 15 (b)) on used 3%B-10%Ni/SBA-
15 sample was assigned to the suppression of graphene islands
T.J. Siang et al. Microporous and Mesoporous Materials 262 (2018) 122–132

Fig. 14. SEM images of spent (a) 10%Ni/SBA-15 and (b) 3%B-10%Ni/SBA-15 catalysts after CSCRM with stoichiometric feed composition (FCH4: FH2 O: FCO2 = 3: 2: 1) and 1073 K.

formation reportedly prone to the production of unreactive and un-

desirable graphite [20,21]. The nonexistence of graphite on used 3%B-
10%Ni/SBA-15 catalyst surface observed by TEM image is also corro-
borated with results from Raman (see Fig. 12) and XRD (cf. Fig. 13)
measurements of spent promoted catalysts. Abatzoglou and Fauteux-
Lefebvre previously found that CNF would not result in significant
catalytic deterioration due to its reactiveness with oxidizing reactants
(i.e., CO2 and H2O) to be removed from catalyst surface whereas gra-
phitic carbon arising from methane cracking (cf. Eq. (6)) could en-
capsulate active Ni0 particles and subsequently prevent them from the
adsorption of gaseous reactants [66].

3.3.4. Temperature-programmed oxidation measurements

Although CNF and graphite on the surface of spent catalysts were
qualitatively detected by Raman, SEM and TEM analyses, TPO experi-
ments were further performed on used unpromoted and 3%B-promoted
catalysts to quantify the total amount of carbon deposits. As illustrated
in TPO profile of used 10%Ni/SBA-15 sample (see Fig. 16 (a)), two
intense peaks (denoted as P1 and P2) were detected at 613 K and 760 K,
respectively. The first oxidation peak P1 at low temperature was as-
Fig. 16. TPO envelopes for spent (a) 10%Ni/SBA-15 and (b) 3%B-10%Ni/SBA-15 cata-
signed to carbon nanofilament oxidation whereas the high-temperature lysts after CSCRM with stoichiometric feed composition (FCH4: FH2 O: FCO2 = 3: 2: 1) and
peak (P2) could belong to graphite gasification [67–69]. However, as 1073 K.
seen in Fig. 16 (b), only low temperature peak P1 was observed and it
was shifted to lower temperature of 455 K for 3%B-10%Ni/SBA-15
sample. Hence, this behavior could indicate that the promotion of fold lesser than that of unpromoted catalyst. The lower carbon de-
boron not only hindered graphitic carbon formation but also increased position for promoted catalyst was most likely due to the adsorption
the reactiveness of CNF species. Additionally, the weight loss measured competition of boron atoms and carbonaceous species on the octahedral
from TPO analysis for unpromoted and 3%B-promoted catalysts was sites of Ni atoms [21,22]. Moreover, CO2 adsorption on catalyst surface
about 6.4% and 1.5% in this order. Thus, 3%B-10%Ni/SBA-15 catalyst for coke gasification could also be enhanced reasonably due to in-
apparently exhibited a stronger inhibiting effect on carbon deposition creasing electron density on active Ni metal surface donated by boron
since the carbon content measured for promoted catalyst was about 4- promoter [20,53]. The inhibition of deposited carbon could also

Fig. 15. TEM micrographs of used (a) 10%Ni/SBA-15 and (b) 3%B-10%Ni/SBA-15 catalysts after CSCRM with stoichiometric feed composition (FCH4: FH2 O: FCO2 = 3: 2: 1) and 1073 K.

131
T.J. Siang et al.
contribute to address the superior catalytic activity of B-promoted
catalyst to that of unpromoted catalyst.
4. Conclusions
The SIWI approach was implemented for preparing X%B-10%Ni/
SBA-15 catalysts and CSCRM tests were conducted at 973–1073 K and
stoichiometric feedstock. H2-TPR profiles revealed that the interaction
between basic NiO and acidic B2O3 phases was strengthened with in-
creasing boron loading, resulting in a linear shift of reduction tem-
peratures to higher temperature regions. Both promoted and un-
promoted catalysts possessed relatively uniform particles with small
rope-liked shape. Regardless of boron loadings, the catalytic activity
was improved with a growth in reaction temperature within
973–1073 K because of CSCRM endothermic nature. The optimal pro-
moter loading was about 3%B with respect to reactant conversions ir-
respective of reaction temperatures owing to the suppression of carbon
formation on catalyst surface. However, at an extreme B loading be-
yond 3%, a substantial drop in reactant conversions was observed for
all reaction temperature due to larger activation barrier for C-H bond
scission, significant decline in BET surface area and active metal ag-
glomeration. In addition, the greatest H2 yield was around 69.4% at
1073 K over 3%B-promoted catalyst. The H2/CO ratio could be easily
manipulated within the range of 1.26–2.71 appropriate for synthetic
long-chained hydrocarbon production from downstream Fischer-
Tropsch synthesis. Both graphitic and amorphous carbons were iden-
tified for unpromoted catalyst by Raman and TEM analyses. Noticeably,
the B-promoted catalyst could inhibit the formation of graphene is-
lands. Thus, it suppresses the formation of graphitic carbon and reduces
the quantity of carbonaceous deposition of about 4 times as evidenced
in the corresponding XRD and TPO measurements. Besides, TPO ana-
lysis indicated that the promotion of boron not only hindered graphitic
carbon formation but also increased the reactiveness of CNF species.
Acknowledgements
The financial aid from Universiti Malaysia Pahang (RDU170326,
UMP Research Grant Scheme) for conducting this research is gratefully
acknowledged.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://dx.
doi.org/10.1016/j.micromeso.2017.11.028.
References
[1] J.D. Figueroa, T. Fout, S. Plasynski, H. Mcllvried, R.D. Srivastava, Inter. J. Greenh.
Gas. Control 2 (2008) 9–20.
[2] M. Usman, W.M.A. Wan Daud, H.F. Abbas, Renew. Sustain. Energy Rev. 45 (2015)
710–744.
[3] B. Abdullah, N.A.A. Ghani, D.V.N. Vo, J. Clean. Prod. 162 (2017) 170–185.
[4] W. Wang, S. Wang, X. Ma, J. Gong, Chem. Soc. Rev. 40 (2011) 3703–3727.
[5] G.A. Olah, A. Goeppert, M. Czaun, G.K. Prakash, J. Am. Chem. Soc. 135 (2013)
648–650.
[6] N. Kumar, M. Shojaee, J.J. Spivey, Curr. Opin. Chem. Eng. 9 (2015) 8–15.
[7] K. Jabbour, P. Massiani, A. Davidson, S. Casale, N.E. Hassan, Appl. Catal. B Environ.
201 (2017) 527–542.
[8] G.A. Olah, A. Goeppert, M. Czaun, T. Mathew, R.B. May, G.S. Prakash, J. Am. Chem.
Soc. 137 (2015) 8720–8729.
[9] H.S. Roh, K.Y. Koo, J.H. Jeong, Y.T. Seo, D.J. Seo, Y.S. Seo, W.L. Yoon, S.B. Park,
Catal. Lett. 117 (2007) 85–90.
[10] J.D. Holladay, J. Hu, D.L. King, Y. Wang, Catal. Today 139 (2009) 244–260.
[11] K.Y. Koo, H.S. Roh, U.H. Jung, W.L. Yoon, Catal. Today 185 (2012) 126–130.
[12] J.E. Park, K.Y. Koo, U.H. Jung, J.H. Lee, H.S. Roh, W.L. Yoon, Int. J. Hydrogen
Energy 40 (2015) 13909–13917.
[13] H. Wan, X. Li, S. Ji, B. Huang, K. Wang, C. Li, J. Nat. Gas. Chem. 16 (2007)
139–147.
[14] V. Arcotumapathy, F.S. Alenazey, R.L. Al-otaibi, D.V.N. Vo, F.M. Alotaibi,
A.A. Adesina, Appl. Petrochem. Res. 5 (2015) 393–404.
132
Microporous and Mesoporous Materials 262 (2018) 122–132
[15] M.E. Galvez, A. Albarazi, P.D. Costa, Appl. Catal. A Gen. 504 (2015) 143–150.
[16] O. Omoregbe, H.T. Danh, C. Nguyen-Huy, H.D. Setiabudi, S.Z. Abidin, Q.D. Truong,
D.V.N. Vo, Int. J. Hydrogen Energy 42 (2017) 11283–11294.
[17] G. Gomez, J.A. Botas, D.P. Serrano, P. Pizarro, Catal. Today 277 (2016) 152–160.
[18] N. Yi, Y. Cao, Y. Su, W.L. Dai, H.Y. He, K.N. Fan, J. Catal. 230 (2005) 249–253.
[19] D. Liu, X.Y. Quek, H.H.A. Wah, G. Zeng, Y. Li, Y. Yang, Catal. Today 148 (2009)
243–250.
[20] J. Xu, M. Saeys, J. Catal. 242 (2006) 217–226.
[21] J. Xu, M. Saeys, Chem. Eng. Sci. 62 (2007) 5039–5041.
[22] J. Xu, L. Chen, K.F. Tan, A. Borgna, M. Saeys, J. Catal. 261 (2009) 158–165.
[23] A.L. Patterson, Phys. Rev. 56 (1939) 978–982.
[24] E.J. Baum, Chemical Property Estimation: Theory and Application, Lewis
Publishers, Boca Raton, 1998.
[25] S.W. Song, K. Hidajat, S. Kawi, Langmuir 21 (2005) 9568–9575.
[26] C.M. Yang, P.H. Liu, Y.F. Ho, C.Y. Chiu, K.J. Chao, Chem. Mater 15 (2003)
275–280.
[27] I. Eswaramoorthi, A.K. Dalai, Micro. meso. Mater 93 (2006) 1–11.
[28] A. Fouskas, M. Kollia, A. Kambolis, C. Papadopoulou, H. Matralis, Appl. Catal. A
Gen. 474 (2014) 125–134.
[29] S. Brunauer, L.S. Deming, W.E. Deming, E. Teller, J. Am. Chem. Soc. 62 (1940)
1723–1732.
[30] D. Li, L. Zeng, X. Li, X. Wang, H. Ma, S. Assabumrungrat, J. Gong, Appl. Catal. B
Environ. 176 (2015) 532–541.
[31] N. Wang, W. Chu, T. Zhang, X.S. Zhao, Inter. J. Hydrogen Energy 37 (2012) 19–30.
[32] S.N. Azizi, S. Ghasemi, E. Chiani, Electrochimica Acta 88 (2013) 463–472.
[33] J. Wang, B. Zou, M.A. El-Sayed, J. Mol. Struct. 508 (1999) 87–96.
[34] M. Pudukudy, Z. Yaakob, Z.S. Akmal, Appl. Surf. Sci. 330 (2015) 418–430.
[35] D. Rath, K.M. Parida, Ind. Eng. Chem. Res. 50 (2011) 2839–2849.
[36] S. Rada, M. Culea, E. Culea, J. Non-Cryst. Solids 354 (2008) 5491–5495.
[37] H. Chen, J. Tan, J. Cui, X. Yang, H. Zheng, Y. Zhu, Y. Li, Mol. Catal. 433 (2017)
346–353.
[38] S.C. Neumair, R. Kaindl, R.D. Hoffmann, H. Huppertz, Solid State Sci. 12 (2012)
229–235.
[39] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky,
Science 279 (1998) 548–552.
[40] S.N. Bukhari, C.Y. Chin, H.D. Setiabudi, D.-V.N. Vo, J. Environ. Chem. Eng. 5
(2017) 3122–3128.
[41] JCPDS Powder Diffraction File, International centre for diffraction data, PA:
Swarthmore, 2000.
[42] M.V. Landau, L. Vradman, X. Wang, L. Titelman, Micro. meso. Mater 78 (2005)
117–129.
[43] N. Mironova-Ulmane, A. Kuzmin, J. Grabis, I. Sildos, V.I. Voronin, I.F. Berger,
V.A. Kazantsev, Solid State Phenom. 168 (2011) 341–344.
[44] U. Oemar, Y. Kathiraser, L. Mo, X.K. Ho, S. Kawi, Catal. Sci. Technol. 6 (2016)
1173–1186.
[45] J. Zheng, Z. Xia, J. Li, W. Lai, X. Yi, B. Che, W. Fang, H. Wan, Catal. Commun. 21
(2012) 18–21.
[46] D. Kim, B.S. Kwak, B.K. Min, M. Kang, Appl. Surf. Sci. 332 (2015) 736–746.
[47] S. Zhu, X. Gao, Y. Zhu, Y. Zhu, H. Zheng, Y. Li, J. Catal. 303 (2013) 70–79.
[48] A.N. Kharat, S. Moosavikia, B.T. Jahromi, A. Badiei, J. Mol. Catal. A Chem. 348
(2011) 14–19.
[49] V. Brahmkhatri, A. Patel, Appl. Catal. A Gen. 403 (2011) 161–172.
[50] R. Sanz, G. Calleja, A. Arencibia, E. Sanz-Pérez, Micro. meso. Mater 158 (2012)
309–317.
[51] Ş. Özkara-Aydınoğlu, Inter. J. Hydrogen Energy 35 (2010) 12821–12828.
[52] S.Y. Foo, C.K. Cheng, T.H. Nguyen, A.A. Adesina, Ind. Eng. Chem. Res. 49 (2010)
10450–10458.
[53] X. Chen, S. Wang, J. Zhuang, M. Qiao, K. Fan, H. He, J. Catal. 227 (2004) 419–427.
[54] M.A. Soria, C. Mateos-Pedrero, A. Guerrero-Ruiz, I. Rodríguez-Ramos, Inter. J.
Hydrogen Energy 36 (2011) 15212–15220.
[55] W.J. Jang, D.W. Jeong, J.O. Shim, H.M. Kim, H.S. Roh, I.H. Son, S.J. Lee, Appl.
Energy 173 (2016) 80–91.
[56] C. Lin, K. Tao, H. Yu, D. Hua, S. Zhou, Catal. Sci. Technol. 4 (2014) 4010–4019.
[57] J.P. Thielemann, T. Ressler, A. Walter, G. Tzolova-Müller, C. Hess, Appl. Catal. A
Gen. 399 (2011) 28–34.
[58] F.F. de Sousa, H.S.A. de Sousa, A.C. Oliveira, M.C.C. Junior, A.P. Ayala, E.B. Barros,
B.C. Viana, J.M. Filho, A.C. Oliveira, Int. J. Hydrogen Energy 37 (2012) 3201–3212.
[59] M.B. Bahari, N.H.H. Phuc, B. Abdullah, F. Alenazey, D.V.N. Vo, J. Environ. Chem.
Eng. 4 (2016) 4830–4838.
[60] Y. Brik, M. Kacimi, F. Bozon-Verduraz, M. Ziyad, J. Catal. 211 (2002) 470–481.
[61] A. Zawadzki, J.D.A. Bellido, A.F. Lucrédio, E.M. Assaf, Fuel Process. Technol. 128
(2014) 432–440.
[62] H. Zhang, M. Li, P. Xiao, D. Liu, C.J. Zou, Chem. Eng. Technol. 36 (2013)
1701–1707.
[63] X.H. Huang, J.P. Tu, B. Zhang, C.Q. Zhang, Y. Li, Y.F. Yuan, H.M. Wu, J. Power
Sources 161 (2006) 541–544.
[64] H.S. Roh, K.Y. Koo, W.L. Yoon, Catal. Today 146 (2009) 71–75.
[65] C.H. Bartholomew, Appl. Catal. A Gen. 212 (2001) 17–60.
[66] N. Abatzoglou, C. Fauteux-Lefebvre, Wiley Interdiscip. Rev. Energy Environ 5
(2016) 169–187.
[67] D. González, O. Altin, S. Eser, A.B. Garcia, Mater. Chem. Phys. 101 (2007) 137–141.
[68] K. Chen, Z. Xue, H. Liu, A. Guo, Z. Wang, Fuel 113 (2013) 274–279.
[69] K. Selvarajah, N.H.H. Phuc, B. Abdullah, F. Alenazey, D.V.N. Vo, Res. Chem.
Intermed. 42 (2016) 269–288.