Role of lattice oxygen in oxidative steam reforming of toluene as a tar model compound over Ni/La0.8Sr0.2AlO3 catalyst

Article abstract of DOI:10.1039/c5cy00412h

Catalytic steam reforming of tar with toluene as a model compound for production of synthesis gas (H₂ and CO) was studied using Ni/LaAlO₃, Ni/La_0.8Sr_0.2AlO₃, Ni/La₂O₃, and Ni/α-Al₂O₃ catalysts prepared using a wet impregnation method. The Ni/La_0.8Sr_0.2AlO₃ catalyst demonstrated the most superior catalytic performance in terms of both catalytic activity and coke resistance in the steam reforming of toluene. The presence of gas phase oxygen enhanced the catalytic performance of all four catalysts, with the extent of improvement being the greatest over the Ni/La_0.8Sr_0.2AlO₃ catalyst. Catalyst characterization by X-ray diffraction (XRD), temperature-programmed reduction (TPR), X-ray photoelectron spectroscopy (XPS) and temperature-programmed desorption of oxygen (TPD-O₂) revealed that the superior catalytic performance of the Ni/La_0.8Sr_0.2AlO₃ catalyst was a result of lattice distortion caused by strontium doping, which produced a higher concentration of oxygen vacancies on the catalyst surface. This lowered the activation energy of the migration of lattice oxygen, enhancing the mobility of lattice oxygen species, and also improved the adsorption abilities of gas phase oxygen species. Mobile lattice oxygen species (O_lattice) favored the direct partial oxidation of toluene, whereas gas-phase oxygen possessed stronger oxidative abilities and favored the complete oxidation of toluene. Both mobile lattice oxygen and gas phase oxygen species actively suppressed coke formation and oxidized coke deposited on the catalyst surface, conferring coking resistance.

Full text of DOI:10.1039/c5cy00412h

View Article Online
View Journal
Catalysis
Sc ience &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: U. Oemar, M. L.
Ang, Y. C. Chin, K. Hidajat and S. -. Kawi, Catal. Sci. Technol., 2015, DOI: 10.1039/C5CY00412H.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/catalysis

Page 1 of 14
Catalysis Science & Technology
Dynamic Article Links ►
Catalysis Science and
Technology
View Article Online
DOI: 10.1039/C5CY00412H
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
ARTICLE TYPE
Role of lattice oxygen in oxidative steam reforming of toluene as a tar
model compound over Ni/La Sr AlO catalyst
0
.8
0.2
3
*
Usman Oemar, Ming Li Ang, Yin Chee Chin, Kus Hidajat, and Sibudjing Kawi
Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX
5
DOI: 10.1039/b000000x
Catalytic steam reforming of tar with toluene as a model compound for production of synthesis gas (H2
and CO) were studied using Ni/LaAlO , Ni/La0.8Sr0.2AlO , NL, and Ni/αꢀAl O catalysts prepared using a
3
3
2
3
wet impregnation method. Ni/La0.8Sr0.2AlO3 catalysts demonstrated the most superior catalytic
performance in terms of both catalytic activity and coke resistance in the steam reforming of toluene. The
presence of gas phase oxygen enhanced the catalytic performances of all four catalysts, with the extent of
improvement being the greatest over Ni/La0.8Sr0.2AlO3 catalyst. Catalyst characterization by xꢀray
diffraction (XRD), temperatureꢀprogrammed reduction (TPR), xꢀray photoelectron spectroscopy (XPS)
1
1
2
0
5
0
and temperatureꢀprogrammed desorption of oxygen (TPDꢀO ) revealed that the superior catalytic
2
performance of Ni/La0.8Sr0.2AlO catalyst was a result of lattice distortion caused by strontium doping,
3
that formed a higher concentration of oxygen vacancies on the catalyst surface. This lowered the
activation energy of the migration of lattice oxygen, enhancing the mobility of lattice oxygen species, and
also improved the adsorption abilities of gas phase oxygen species. Mobile lattice oxygen species (Olattice
)
favored the direct partial oxidation of toluene, whereas gasꢀphase oxygen possessed stronger oxidative
abilities and favored the complete oxidation of toluene. Both mobile lattice oxygen and gas phase oxygen
species actively suppressed coke formation and oxidized coke deposited on the catalyst surface,
conferring coking resistance.
Introduction


m
2 
0
298
C H + nH O → nCO+ n+
H , ꢀH > 0
Biomass constitutes living species such as plants and animals that
are alive or were alive some time ago. While conventional fossil
fuels require millions of years to develop, biomass can readily
reproduce. For this reason, it is considered renewable . In
addition, biomass is said to be a “carbonꢀneutral” fuel as a
majority of biomass take in CO from the atmosphere for growth
and reproduction. This CO is subsequently released when the
biomass is burnt. As such, biomass gasification is an increasingly
popular area of research for the generation of energy or
production of syngas for use in the synthesis of FischerꢀTropsch
n
m
2


2
4
5
5
6
5
0
5
0
2
3
3
4
5
0
5
0
Reaction temperatures can be reduced through the use of
catalysts. Various metal catalysts have been investigated such as
1
6
ꢀ17
18ꢀ21
6ꢀ10, 19, 22ꢀ23
25ꢀ26
Ni , Co
, Fe
, noble metal
for steam reforming
2
of biomass tar using toluene or benzene as the model compound.
Recently, Li et al. published a review on catalyst development
2
27
for biomass tar reforming using either toluene, benzene,
naphthalene, or phenol as model compounds. They concluded
that Rh, Ni and Co are promising materials as catalyst for
biomass tar reforming. Ni and Co are much cheaper than Rh,
however, the improvement on catalytic activity and stability are
needed to obtain the similar catalytic performance with the Rh
catalyst.
It has been reported that coke formation can be minimized in
various ways using catalyst modification. One of the
modifications is to introduce oxygen property into the catalyst
1
ꢀ4
chemicals and methanol
.
However, tar formation as a byꢀproduct of biomass gasification
remains a problem. Tar is a thick, black and highly viscous liquid
comprising a complex mixture of 1ꢀ to 5ꢀring aromatic, as well as
3
complex polyꢀaromatic hydrocarbons . The condensation of tar in
the lowꢀtemperature zones of gasifiers results in plugging and
fouling. Typically, the amount of tar is reduced by steam
reforming at high temperatures which cracks tar, producing
2
8ꢀ30
using few ways, such as forming perovskite structure
promoting with elements having redox property
,
1
lighter molecules such H2 and CO . Steam reforming of tar
31ꢀ32
or oxygen
1, 5
(
C H ) is illustrated by the following equation
:
n
x
33ꢀ34
29
storage property
. Urasaki et al. investigated the coking
resistance as well as catalytic activities of several Ni/perovskite
This journal is © The Royal Society of Chemistry [year]
[journal], [year], [vol], 00–00 | 1

Catalysis Science & Technology
Page 2 of 14
View Article Online
DOI: 10.1039/C5CY00412H
catalysts in the steam reforming of methane. Of which,
Ni/LaAlO and Ni/SrTiO demonstrated high catalytic activities.
3
3
Lattice oxygen in or on the support are believed to have played an
active role in oxidizing CH adsorbed onto the nickel catalyst
x
0
3
5
0
5
0
5
0
5
0
5
0
5
surface. Stathopoulos et al. also shed light on the high redox
property of La0.8Sr0.2FeO3ꢀx catalyst support for the deNOx
reaction by H and CH . Cu and Fe are popular elements having
6
0
Experimental
Catalyst synthesis
2
4
reductionꢀoxidation (redox) property which aids in the reduction
Ni catalysts supported on La0.8Sr0.2AlO , LaAlO , αꢀAl O and
3
3
2
3
1
8,31
of carbon deposition. Tomishige et al.
reported the use of Fe
La O3 were prepared using conventional wet impregnation
2
1
1
2
2
3
3
4
4
5
5
in NiꢀFe/αꢀAl O catalyst in steam reforming of biomass tar and
method with Ni loading amount of 10wt%. The molar ratio of
2
3
found that addition of Fe to Ni/αꢀAl O catalyst enhanced 65 La:Sr of 8:2 in La0.8Sr0.2AlO was chosen based on our previous
2
3
3
8
catalytic activity by formation of NiꢀFe alloy and catalytic
results . Catalyst supports La0.8Sr0.2AlO3 and LaAlO3 were
prepared by a citric acid complex method. Firstly, metal nitrate
precursors La(NO ) .6H O, Sr(NO3)2 and Al(NO ) .9H O
stability due to the presence of Fe on the catalyst surface which
6
increases the oxygen storage/vacancy of the catalyst. Kawi et al.
3
3
2
3 3
2
reported that the addition of CeO to the Ni/CaOꢀAl O catalyst
2
2
3
(SigmaꢀAldrich) were dissolved in water. Thereafter, an
resulted in less carbon formation compared to the unpromoted 70 equimolar amount of citric acid C H O .H O (SigmaꢀAldrich) to
6
8
7
2
catalysts. Another option for catalyst modification is using basic
metal was added to the solution. The mixture was then
evaporated at 55°C while stirring until it was dried. The dried
powder was stored at 100°C overnight until a dome shape was
formed. The sample was then crushed and calcined at 400°C for
30min, followed by further calcinations at 850°C for 7h.
La0.8Sr0.2AlO3, LaAlO3, La O3 and αꢀAl O3 were then
8
ꢀ10, 35ꢀ42
8ꢀ9
metal as promoter for the catalyst
. Our previous study
showed that substitution of small amount of Sr with La in the
La0.8Sr0.2Ni0.8Fe0.2O catalyst showed good catalytic activity and
3
stability, especially at low amount of steam (Steam/Carbon=1)
due to property of Sr in LSNFO catalyst to adsorb and desorb
more water at higher temperature than La. The synergetic effect
of Sr and La helps to enhance the catalyst stability. Choong et
75
2
2
impregnated with
a
solution comprising nickel nitrate,
Ni(NO ) .6H O (SigmaꢀAldrich). Likewise, the solution was
3
2
2
41
al. also reported that addition of CaO to Ni catalyst promoted
water adsorption and provided abundance of adsorbed OH groups
to facilitate the CꢀC break in Ni catalyst, resulting in higher
ethanol conversion in steam reforming of ethanol. Viinikainen et
evaporated at 55°C while stirring until it was dried. The dried
powder was then stored at 100℃ overnight. Finally, the samples
were calcined at 850°C for 7h. The prepared catalysts are denoted
as NLSAO, NLAO, NA, and NL for Ni/La0.8Sr0.2AlO3,
Ni/LaAlO , Ni/αꢀAl O , and NL.
80
42
al. also reported that higher basicity of catalyst is a favorable
property for tar decomposition resulting in enhancement of
catalyst stability. The last option to reduce coke deposition is by
operating the reaction at high reaction temperatures and excess
3
2
3
85
Catalyst Characterization
X-ray Diffraction (XRD)
4
3
steamꢀtoꢀcarbon ratio .
5,28,44ꢀ45
Sekine et al.
revealed that partially Srꢀsubstituted
Xꢀray diffraction was carried out using monochromatised Cu
Kα radiation at 40kV and 30mA (XRDꢀ6000; Shimadzu
Corporation) for both fresh and reduced catalysts. The reduced
catalyst support (Ni/La0.7Sr AlO3ꢀδ) demonstrated higher
0.3
catalytic activity and lower amounts of carbon deposition in the
steam reforming of toluene as opposed to Ni/LaAlO catalyst.
3
90
catalysts were treated in pure H gas at 700°C for 30min. The
2
This was attributed to the lattice distortion caused by partial Sr
doping, resulting in the increase in amount of oxygen vacancies,
which enhanced the mobility of lattice oxygen species. These
samples were placed on a clear plastic slide and scanned from 2θ
of 10° to 80° at a rate of 2°/min. The receiving slit was 0.3mm,
whereas the divergent and scattering slits were set to be 1mm.
18
were elucidated from XPS, XRD and transient isotopic H O
2
analyses. Indeed, lattice oxygen is known to be capable of
playing a significant role in promoting the catalytic activities of
perovskite catalysts, but little research has uncovered the actual
role of oxygen in these catalysts.
95
Hydrogen temperature-programmed reduction (H -TPR)
2
Temperatureꢀprogrammed reduction (TPRꢀH ) was conducted
2
using Thermo Scientific TPDRO 1100 TPR/TPD system with a
thermal conductivity detector (TCD). 50mg of fresh catalyst was
used in the experiment. Prior to analysis, the sample was preꢀ
In this work, the role of lattice oxygen in the steam reforming
of toluene would be elucidated through a series of comprehensive
catalyst characterization methods. In addition, it can be postulated
that the presence of oxygen vacancies formed by Srꢀdoping might
enhance the effectiveness of minute amounts of gasꢀphase oxygen
in the feed introduced to the toluene steam reforming reaction.
Varied reforming temperatures and oxygen gas concentrations
might also affect the roles of lattice and gasꢀphase oxygen. As an
1
00 treated with helium gas at 300°C for 1h to remove impurities, and
then heated from room temperature at 10°C/min to 1000°C with a
5
% H /N mixture. The TPRꢀH profiles were generated through
2 2 2
the monitoring of the changes in the signal of the temperature
conductivity detector (TCD) against temperature.
1
05
4
4ꢀ46
extension of the studies by Sekine et al.
, this study aims to
X-ray Photon Spectroscopy (XPS)
reveal the actual roles of lattice and gasꢀphase oxygen in the
steam reforming of toluene under varied oxygen gas
concentrations.
Xꢀray photoelectron spectroscopy (XPS) was conducted in a
spectrophotometer (AXIS Ultra DLD; Kratos Analytical) using
monoꢀchromatized Al Kα xꢀray source with 1486.71eV photons,
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 2

Page 3 of 14
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C5CY00412H
at 15kV and 5mA, and with a constant dwell time of 100min.
Prior to the analysis, the catalyst was exꢀsitu reduced at 700°C
molar ratio and GHSV 460,000 ml/g.h. Helium was selected as
the carrier gas for the steam and toluene feed. The product gas
exiting the reactor was connected to a condenser that was held at
5°C in to condense and remove unreacted toluene and water.
under H for 30 mins which is the same as the reduction condition
2
during reaction. The sample was then mounted on the standard
5
sample stub using doubleꢀsided adhesive tapes. The resulting 60 Nonꢀcondensable product gases were analyzed using gas
XPS peaks were fitted using XPSPeak fitting software. Binding
energies were calibrated by referencing the C1 (CꢀC) bonds to
chromatography. For reaction with oxygen, the flowrate of
helium was adjusted so that the total flowrate of helium and
oxygen is still 120 mL/min.
2
84.5eV.
The gasꢀchromatograph (HP 6890; Agilent) was equipped with
a 15” column (I.D. 2.1mm) packed with 60/80 mesh Carboxenꢀ
65
1
0
Temperature programmed desorption of oxygen (TPD–O2)
1
000 packing (Carboxenꢀ1000; SigmaꢀAldrich) and a thermal
Temperature programmed desorption of oxygen (TPD–O2)
was conducted using Quantachrome ChemBET 3000 TPR/TPD
system with a thermal conductivity detector (TCD). Prior to
conductivity detector (TCD). The gas chromatograph was
programmed to collect gas samples at 15ꢀminute intervals. The
chromatogram obtained for each reaction run showed peaks for
analysis, 100mg of each fresh catalyst was packed in a Uꢀtube, 70 H , CO and CO product species. The area of each peak was
2
2
1
5
in-situ reduced at 700°C for 30min using pure H gas, and flushed
calculated using the integral function of the offline software
coupled to the gas chromatograph. Based on the peak areas, the
volume percentage of each product gases was obtained using the
calibration curves that were previously determined. The
2
with oxygen at 300°C for 100min. The samples were then heated
at a rate of 10°C/min to 1000°C in an atmosphere of helium gas.
Desorption analyses were carried out by monitoring the signal
readings of the thermal conductivity detector against the 75 individual volumetric flow rate of the product gases was obtained
2
0
temperature of the sample. The TPD profile was recorded by a
based on their volume percentage and the total volumetric flow
microcomputer.
nCO + nCO2 + nCH4
XToluene (%) =
x100
7
*nT ,in
TGA-DTA
rate of the product gases. The conversion of toluene was
expressed in term of carbon conversion and calculated using the
following formula:
The amounts of carbon deposition on spent catalysts were
evaluated using a thermogravimetric analysis (TGA) (DTGꢀ60
TG/DTA; Shimadzu). Around 10mg of spent catalysts were
placed on an alumina pan under atmospheric pressure and heated
at 10°C/min to 1000°C. The change in mass of the spent catalysts
against temperature would be recorded. In addition, the
differential peak temperature would be recorded to reveal the
nature of the carbonaceous compounds present in the spent
catalysts.
2
5
8
0
where n is molar flow rate of each gas.
Results and discussion
3
0
XRD - Catalyst structure analysis
Xꢀray diffraction was carried out for freshly calcined and
reduced NLSAO, NLAO, NA and NL catalysts and depicted in
Figures 1a and 1b. All XRD profiles were referenced with the Xꢀ
ray Powder Diffraction Database from the International Centre
for Diffraction Data (ICDD) for characterization of the phases
85
Field Emission Scanning Electron Microscopy (FESEM)
3
5
The morphologies of the spent catalysts were observed using a
field emission scanning electron microscope (FESEM) (JSMꢀ
present. In Ni/Al
O
3
catalyst, small NiAl
O
4
spinel phase is
700F; JEOL). A small amount of each spent catalyst was thinly 90 observed in addition to NiO peak. The presence of NiAl O peak
2 4
2
2
6
dispersed on a copper stub, and thereafter coated with a layer of
platinum using Ptꢀsputtering that was carried out for 60s at
20mA. Prepared samples were then analyzed with an electron
beam of 5 – 15kV, using a magnification of 25,000X.
shows strong metalꢀsupport interaction in NA catalyst, resulting
0
in enhanced difficulty for complete reduction of nickel to the Ni
47
4
0
metallic state , even though the Al O is αꢀAl O which is very
2 3 2 3
stable at high temperature. The NiO peak is observed strongly in
the XRD profile showing that this catalyst has quite big size of
NiO particles. In contrast, the NiO peak is hardly not observed in
fresh NL catalyst. Instead, LaNiO perovskite and La NiO spinel
95
Evaluation of Catalytic Activities
3
2
4
Catalytic activity tests were conducted at atmospheric pressure
using a 60 cm fixed bed quartz tube reactor (I.D. 4mm) which
phases are observed well, showing that Ni has interaction
between Ni and La . Both NLSAO and NLAO fresh catalysts
was enclosed within a programmable furnace. 30mg of catalyst 100 show similar XRD patterns corresponding to the perovskite
4
5
5
5
0
5
O
2 3
was utilized for each reaction, and held in the middle of the fixedꢀ
bed reactor using quartz wool. Prior to the catalytic reaction at
structure.
A comparison of the (110) perovskite peak for both freshly
calcined NLSAO and NLAO catalysts reveals that the peak is
6
50°C, the catalyst was reduced in 30 mL/min H at 600°C for 30
2
mins, followed by purging in 120 mL/min helium while the
observed at 2θ of 33.62° for NLAO catalyst, but a lower angle at
temperature was increased to the reaction temperature. The feed 105 2θ of 33.54° for NLSAO catalyst. This shift is better illustrated in
of toluene and water was directed through a preꢀheating coil held
at a temperature of 300°C prior to entering the reactor to vaporize
the inlet feed. The feed components, steam and toluene were
pumped into the reactor at a steam/carbon (S/C) ratio of 1:1
the magnified (110) peaks in Figure 1c. Furthermore, the full
width at half maximum (FWHM) of both (110) perovskite peaks
is calculated to be 0.54° for NLSAO catalyst as opposed to 0.40°
for NLAO catalyst. The slight shift in peak angle and broader
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 3

Catalysis Science & Technology
Page 4 of 14
View Article Online
DOI: 10.1039/C5CY00412H
(
110) perovskite peak for NLSAO catalyst demonstrated the
Fig. 1c Magnified (1 1 0) LaAlO
3
perovskite xꢀray diffraction peaks of
20
3 3
freshly calcined Ni/La0.8Sr0.2AlO and Ni/LaAlO
distortion of the structure of perovskite due to the partial
substitution with Sr, which can cause structural defects, resulting
4
8
in oxygen nonstoichiometry i.e. lattice oxygen . This structural
distortion could lead to greater lattice oxygen of NLSAO catalyst
as compared to the relatively stable NLAO catalyst. This
enhanced lattice oxygen mobility of NLSAO catalyst could
confer higher redox abilities, thereby resulting in improved
catalytic activity and stability.
TPR – catalyst reducibility analysis
5
As elucidated from Figure 2, the TPR profiles of NLAO and
NLSAO catalysts both possessed a single peak at 457°C and
402°C, respectively. These reduction peaks correspond to the
2
3
3
4
4
5
5
5
0
5
0
5
0
5
2+
0 49ꢀ50
reduction of bulk NiO (Ni ) to Ni
. The partial substitution
1
0
of lanthanum with strontium resulted in a shift of the reduction
peak to a lower temperature. This is coherent with literature,
3
+
which reported that the substitution of an Aꢀsite cation (La ) by
2+
another cation of lower valence (Sr ) could significantly reduce
the stability of the perovskite structure, thereby increasing the
51ꢀ52
mobility of lattice oxygen
. This facilitates its removal as
water during reduction, therefore enhancing the reducibility of
53
the perovskite . In order to maintain the stability of the
perovskite structure after such substitution, charge compensation
is required. This could be achieved through either through the
existence of Ni ions possessing valence states higher than 3+, or
through the formation of an oxygen deficit perovskite, in the form
53
ABO3–δ
.
The TPR profile of NL catalyst possessed 3 peaks at 390°C,
14.34°C and 611.1°C respectively. The low temperature peak at
90°C can be attributed to the reduction of Ni oxide, whereas the
5
3
higher temperature peaks at 514°C and 611°C were reported to be
due to the reduction of LaNiO and partial reduction of La NiO
Fig 1a XRD of fresh catalysts [α: LaAlO
3
, β: NiO, δ: La
2 3 2 4
O , ζ: La NiO ,
3
2
4
ε: LaNiO , κ: NiAl , γ: La(OH)
3
2
O
4
3
, λ: Al O ]
2 3
54
respectively . Since the reduction peak at 514°C is the major
peak, this suggests that LaNiO existed as the main phase. This
3
observation is in agreement with XRD of fresh NL catalyst which
shows the presence of LaNiO and La NiO .
3
2
4
Finally, the TPR profile of NA catalyst demonstrated a strong
single peak at 425°C and a small shoulder peak at higher
temperature. This single peak at lower temperature corresponds
2+
0 49ꢀ50
to the reduction of bulk NiO (Ni ) to Ni
, meanwhile the
small shoulder peak at higher temperature can be attributed to the
presence of NiAl O phase, which can be confirmed by XRD
2
4
data. The presence of weak NiAl O4 peak could indicate the
2
presence of big Ni particle, which can be observed clearly from
XRD of reduced NA catalyst.
15
Fig. 1b XRD of reduced catalysts, 2θ = 10° to 2θ = 80° [α: LaAlO
3
, η: Ni,
δ: La , κ: NiAl , λ: Al
2
O
3
2
O
4
2 3
O ]
Fig. 2 TPR profiles of various catalysts
60
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 4

Page 5 of 14
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C5CY00412H
XPS of reduced catalysts
4
5
The concentration of surface oxygen vacancies for both
reduced NLSAO and NLAO catalysts were calculated and
tabulated in Table 1. Evidently, strontium substitution in
La0.8Sr0.2AlO3 catalyst support yielded a catalyst with higher
concentrations of surface oxygen vacancies. This conferred
NLSAO catalyst with better oxygen adsorption abilities.
Furthermore, this also enhanced the mobility of lattice oxygen in
NLSAO catalyst to partake in oxidation as surface oxygen
The O 1s XPS profiles of reduced NLSAO, NLAO, NA and
NL catalysts are presented in Figure 3. The XPS O 1s peak was
5
5
reported to have a binding energy of 529.9eV . Based on curve
fitting analyses, the O 1s XPS profiles of reduced NL and
Ni/Al O catalysts both possessed a single peak at similar binding
5
0
5
0
5
0
2
3
50
energies of 528.1eV. Both of these peaks could be attributed to
2ꢀ
the O oxidation state.
Doublet peaks are present in the O 1s XPS spectra of both
reduced NLSAO and NLAO catalyst. The peak with lower
binding energy is attributed to lattice oxygen (Olattice), whereas the
5
species .
1
1
2
2
3
peak with higher binding energy is attributed to adsorbed oxygen 55 Table 1 Peak binding energies and concentration of surface oxygen
5
vacancies (δ) on reduced NLSAO and NLAO
species onto the catalyst surface (O
) . These could include
adsorbed
ꢀ
2ꢀ
adsorbed oxygen (O ), hydroxyl (OH ), carbonate (CO ), and
2
3
Binding energy of peak/ eV Concentration of surface
5
1
Catalyst
water (H O) species . The presence of these additional peaks
2
O
lattice
O
adsorbed
oxygen vacancies (δ)
indicates that the perovskite based catalysts possessed oxygen
vacancies that were not present in NL or Ni/Al O .
Reduced NLSAO 525.62
Reduced NLAO 525.99
528.12
528.20
2.16
1.61
2
3
The binding energies of both peaks of reduced NLSAO and
NLAO catalysts are recorded in Table 1. Evident from Table 1,
the binding energy of the lattice oxygen peak of reduced NLSAO
catalyst is significantly lower than that of reduced NLAO
catalyst. This could be explained by the partial substitution of
lanthanum with strontium that is of lower valence. This oxidation
Temperature-Programmed Desorption of Oxygen (TPD-O2)
TPDꢀO2 was carried out for reduced NLSAO, NLAO,
Ni/Al O and NL catalysts, in order to determine the extent of
2 3
gasꢀphase oxygen uptake and the temperature of desorption of
6
6
7
7
8
0
5
0
5
0
3
+
2+
adsorbed oxygen species. The TPDꢀO results are illustrated in
state imbalance between lanthanum (La ) and strontium (Sr )
sites were reported to increase the amount of oxygen vacancies,
thereby increasing the mobility of lattice oxygen species,
allowing them to partake in oxidation as surface oxygen
2
Figure 4. It can be observed that reduced NLSAO and NLAO
catalysts demonstrated the largest extent of oxygen desorption
between the temperatures of 100°C and 1000°C, while no
obvious peaks are observed for NL and NA catalysts, indicating
that there is no lattice oxygen in these both catalysts, which is in
agreement with XPS result. The presence of oxygen desorption
peak shows the presence of oxygen vacancies in these perovskite
oxide catalysts, thereby allowing a greater extent of lattice and
gas phase oxygen uptake onto the surface of these catalysts.
Therefore, a larger amount of oxygen can also be desorbed
eventually at higher temperatures. In addition, from the
desorption peak formations illustrated in Figure 4, it was evident
that a majority of the oxygen on both NLSAO and NLAO
catalysts desorbed at temperatures between 600°C to 800°C,
which coincides with typical reaction temperatures used for the
steam reforming of toluene. Therefore, oxygen would be
expected to play a positive role in enhancing the catalytic
performances of NLSAO and NLAO catalysts in the steam
reforming of toluene.
5
1,53
species
. The binding energy of the adsorbed oxygen species
of NLSAO catalyst is also lower than that of reduced NLAO
catalyst.
Fig. 3 XPS O 1s profiles of various reduced catalysts
3
5
Assuming the number of electrons recorded by XPS analysis is
proportional to the number of atoms at a given state, the
percentage atomic concentrations of each atom could be
calculated. For the perovskite catalysts, the ratio of Oadsorbed and
Olattice species could be determined based on curve fitting
analyses using XPSPeak software to calculate the concentration
of surface oxygen vacancies (δ). For the defective perovskite
4
0
5
present in the form of ABO3ꢀδ, O/(A+B) is equivalent to (3–δ)/2 .
Therefore,
2
[
Olattice
Sr
This journal is © The Royal Society of Chemistry [year]
]
δ
=
[
La
]
+
[
]
+
[
Al
]
Journal Name, [year], [vol], 00–00 |
5

Catalysis Science & Technology
Page 6 of 14
View Article Online
DOI: 10.1039/C5CY00412H
35
Catalytic performances of prepared catalysts in absence of
gas-phase oxygen
The catalytic performance of prepared catalysts in steam
reforming of toluene in terms of toluene conversion is shown in
Figure 5a and H yield in Figure 5b. The catalytic activities of
2
4
4
5
5
6
6
7
7
0
5
0
5
0
5
0
5
those catalysts decrease in the following order: NL> NLSAO>
NLAO> NA. Even though NL catalyst demonstrates the highest
catalytic conversion, it has very high carbon deposition as the
pressure of reactor system increased until the flow of He carrier
gas could not pass through the catalyst bed and then the reaction
was stopped. In contrast, the other three catalysts demonstrate
relatively stable carbon conversions throughout the entire
duration of the reaction runs.
2
Fig. 4 TPDꢀO profiles of various reduced catalysts
Table 2 tabulates the average results of the catalytic
performance in terms of carbon conversion, production rate, and
H /CO. It can be seen in Table 2 that the H /CO ratios over
2
2
The TPDꢀO2 profile of reduced NLSAO catalyst clearly
demonstrated two desorption peaks, one at a lower temperature of
around 330°C and another at approximately 710°C. The lower
temperature desorption peak is characteristic of oxygen species
NLAO, NA and NL are comparable at 3.03, 3.58 and 3.83
respectively, but that over NLSAO catalyst is significantly higher
5
0
5
0
5
0
at 8.35. This enhancement of H /CO ratio could be attributed to
2
5
6ꢀ57
the Sr substitution of the NLAO perovskite based catalyst. The
ideal steam reforming of toluene reaction with steam/carbon ratio
of 1 should follow the reaction C H + 7H O → 7CO+11H ,
that were released from surface oxygen vacancies
. This is
indicative that surface oxygen vacancies were present on the
surface of reduced NLSAO catalyst. In contrast, reduced NLAO
catalyst also appeared to possess such a low temperature peak at a
higher temperature of around 500°C, but the peak was less
obvious. This may be indicative that there exists a lower
concentration of such surface oxygen vacancies on the surface of
reduced NLAO catalyst.
7
8
2
2
1
1
2
2
3
which result in H /CO ratio of 11/7. However, in reality, the
2
water gas shift (WGS) reaction always occur simultaneously (CO
+
H O ꢀ H + CO). The combination of these two reaction will
2 2
give H /CO more than 11/7, depending on the contribution of
2
8ꢀ10
WGS reaction. Furthermore, our previous study
shows that
combination of La O and SrO is very good in water activation
2
3
The higher temperature desorption peak of both reduced
NLSAO and NLAO catalysts correspond to the migration of
which provides more hydrogen to the whole reaction. Hence, the
5
9
high H /CO ratio such as 8.35 is possible to be achieved.
2
lattice oxygen in the bulk perovskite catalysts . Comparing the
oxygen desorption profiles of reduced NLSAO and NLAO
catalysts, the strontium doped perovskite catalyst possessed a
slightly lower lattice oxygen desorption peak temperature of
approximately 710°C, as opposed to a desorption peak of
approximately 730°C for the NLAO catalyst. These shifts of both
adsorbed oxygen and lattice oxygen desorption peaks of reduced
NLSAO catalyst to a lower temperature as compared to NLAO
catalyst are consistent with XPS results, where binding energies
of both the adsorbed oxygen species and lattice oxygen species of
NLSAO catalyst were found to be lower than that of reduced
NLAO catalyst. Evidently, the partial substitution of lanthanum
with strontium that resulted in oxidation state imbalance between
Furthermore, carbon formation rate on NLSAO catalyst (which is
explained later) is the lowest among other catalysts even though
the conversion is quite high due to the presence of lattice oxygen
which can oxidize the carbon species into CO and/or CO2.
In addition, Table 2 also shows the average %ꢀmolar of
product gases produced from the reaction. The H %ꢀmolar of
2
NLSAO catalyst is slightly higher than the one of NLAO catalyst,
but the CO %ꢀvolume of NLSAO catalyst is much lower than
NLAO catalyst. The low CO %ꢀmolar from NLSAO catalyst is
due to very good water activation of combination between La O
2
3
and SrO as well as high contribution of WGS reaction, which can
be confirmed with the high CO %ꢀmolar. This result shows that
2
3
+
2+
WGS reaction has high contribution in this reforming reaction.
lanthanum (La ) and strontium (Sr ) sites, creating oxygen
vacancies that enhanced both the lattice oxygen and surface
oxygen desorption abilities of reduced NLSAO catalyst.
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 6

Page 7 of 14
Catalysis Science & Technology
Dynamic Article Links ►
Catalysis Science and
Technology
View Article Online
DOI: 10.1039/C5CY00412H
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
ARTICLE TYPE
Table 2 Summary of catalytic activity of various catalysts at 650°C
Average production rate
Average
conversion
ꢀ1
ꢀ1
Average % molar
(mol.gcat .min )
H /CO
Ratio
2
Catalyst
(%)
H
2
CO
CH
4
CO
2
H
2
CO
CH
4
CO
2
NLSAO
NLAO
NA
20.03
12.84
5.31
501.6
235.9
65.1
60.3
78.9
0
0
0
192.2
82.9
66.51
59.32
49.32
65.02
8
0
0
0
25.49
20.84
36.82
17.66
8.35
3.03
3.58
3.83
19.84
13.86
17.21
18.3
48.6
NL
35.51
831.6
220.1
1.4
225.9
0.11
conversion and H yield, respectively for all prepared catalysts. In
2
the presence of 5 ml/min (223 ꢁmol/min) of pure gasꢀphase
oxygen, both catalytic conversions and H2 yield in steam
reforming of toluene over all four catalysts increase. Gasꢀphase
oxygen gas in the feed is found to have a positive role in
enhancing the catalytic activities of all four catalysts in the steam
reforming of toluene. Similar to catalytic performance in the
absence of gasꢀphase oxygen, catalytic activities of the four
catalysts decrease in the following order: NL> NLSAO> NLAO>
NA. Compared to Figure 5a and 5b, Figures 6a and 6b also reveal
that catalytic activities over all four catalysts are more stable in
the steam reforming reaction in the presence of 5ml/min of gasꢀ
phase oxygen. Likewise, it is apparent that amongst the four
catalysts, NL catalyst demonstrates the lowest catalytic stability
as evidenced by reactor blockage around 7.5h of steam reforming
reaction.
2
3
3
4
5
0
5
0
5
1
0
A comparison of Tables 2 and 3 reveals that in steam
reforming of toluene over NLSAO, NLAO, NA and NL catalysts
in the presence of 5ml/min gasꢀphase oxygen, the following
trends for are observed as compared to the corresponding
reactions in the absence of gasꢀphase oxygen:
1
5
2
Fig. 5 (a) Carbon conversion and (b) H yield of various catalysts at
6
50°C. Reaction condition: toluene 188 ꢁmol/min; steam 1316 ꢁmol/min;
He 5357 ꢁmol/min; W = 30 mg; reaction temperature 650°C; GHSV
1
. The average production rates of H , CO and CO over all four
2 2
4
60,000 ml/g.h
45
catalysts increase.
2
. This resulted in a decrease in H /CO ratio over all four
catalysts in the presence of gasꢀphase oxygen.
2
2
0
Catalytic performances of prepared catalysts in presence of
gas-phase oxygen
Specifically for NLSAO catalyst, the CO %ꢀmolar in the presence
of oxygen is more than twice higher than the one in the absence
of oxygen while the H and CO %ꢀmolar in the presence of
Figure 6a and 6b show the catalytic activity in terms of toluene
50
2
2
oxygen are only slightly lower than those in the absence of
oxygen. This result indicates that the oxygen helps to oxidize the
deposited carbon to CO.
This journal is © The Royal Society of Chemistry [year]
[journal], [year], [vol], 00–00 | 7

Catalysis Science & Technology
Page 8 of 14
View Article Online
DOI: 10.1039/C5CY00412H
Figure 6 (a) Carbon conversion and (b) H
2
yield of various catalysts at
5
0
5
650°C. Reaction condition: toluene 188 ꢁmol/min; steam 1316 ꢁmol/min;
2
O 223 ꢁmol/min; He 5134 ꢁmol/min; W = 30 mg; reaction temperature
650°C; GHSV 460,000 ml/g.h
=
1
1
Table 3 Summary of catalytic activity of various catalysts in the presence of O
2
at 650°C
Average production rate
ꢀ1
ꢀ1
Average %molar
Average
conversion (%)
(mol.gcat .min )
2
H /CO
Ratio
Catalyst
H
2
CO
CH
4
CO
2
H
2
CO
CH
4
CO
2
NLSAO
NLAO
NA
67.5
31.65
25.93
75.12
1197.3
348.7
373.7
152
0
0
0
476.8
246.9
207.1
413.5
58.47
46.64
37.47
58.97
18.25
20.33
22.90
22.25
0
0
0
23.28
33.03
39.63
17.92
3.22
2.29
1.64
2.66
195.8
119.7
513.2
NL
1360.3
19.9
0.86
20
25
30
35
Catalytic performance of NLSAO catalyst at various O flow
rates
2
The results of the steam reforming reactions over NLSAO
catalyst at 650°C in the presence of 5ml/min, 7.5ml/min,
1
0ml/min and 12.5ml/min gasꢀphase oxygen in the feed stream
are illustrated in Figures 7a and 7b. It can be seen that as oxygen
flow rate increases from 5ml/min to 12.5ml/min, the percentage
carbon conversion and H yield also increases. Therefore, overall
2
catalytic activity increases as the oxygen flow rate was increases.
In particular, this increase in catalytic activity is the most
significant when O2 flow rate increases from 5ml/min to
7
7
.5ml/min. Thereafter, as O2 flow rate further increases from
.5ml/min through 12.5ml/min, the increment in catalytic activity
is marginal.
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 8

Page 9 of 14
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C5CY00412H
deposited carbon into
3
types, i.e. amorphous carbon,
4
4
5
0
5
0
encapsulating carbon, and whisker/graphitic carbon, which has
DTA peak at around 330, 460, and 650°C, respectively. The
presence of whisker carbon is not a major issue as it can grow on
top of the metal, hence, it will not affect the catalytic
performance. In contrast, encapsulating carbon can cover the
metal active sites, causing catalyst deactivation. According to
these categories, the carbon on NLSAO and NLAO catalysts
could be whisker carbon, which can be seen as carbon nanotubes
in FESEM. Moreover, the peak is located at lower temperature
compared to reaction temperature (650°C), which can indicate
that the carbon can be easily reformed during reaction.
Fig. 7 (a) Carbon conversion and (b) H
various O flow rate. Reaction condition: toluene 188 ꢁmol/min; steam
316 ꢁmol/min; O 223ꢀ557.5 ꢁmol/min; He 4799.5ꢀ5134 ꢁmol/min; W
2
yield of NLSAO catalyst at
2
1
2
5
= 30 mg; reaction temperature = 650°C; GHSV 460,000 ml/g.h
Thermogravimetric Analysis (TGA)
Fig. 8 TGAꢀDTA profiles of (a) spent NLSAO catalyst and (b) spent
Thermogravimetric analyses were conducted on spent
NLSAO, NLAO, NL and NA catalysts after 8h of steam
reforming of toluene at 650°C in the absence and presence of
NLAO catalyst after 8 h reaction without oxygen
1
1
2
0
55
In the presence of gasꢀphase oxygen, the amount of carbon
formed over all four spent catalysts demonstrated a reduction.
The extent of carbon deposition over the spent catalysts followed
the order: NL> NA> NLAO> NLSAO. Evidently, the coking
resistance of the NLSAO and NLAO perovskite catalysts
5
ml/min gasꢀphase oxygen in the feed. The results are
summarized in Table 4. The amount of carbon deposition on
spent catalysts after 8h of steam reforming reaction in the absence
of gasꢀphase oxygen followed the order: NL> NLAO> NA>
5
NLSAO. The low amounts of carbon deposition over NA catalyst 60 exceeded that of NA catalyst in the presence of gasꢀphase oxygen
could be related to low catalytic activities of the NA catalyst,
whereas the high extent of carbon deposition over NL catalyst
demonstrates that despite its high catalytic activity, it was not
capable of conferring sufficient coke resistance in the steam
reforming of toluene. Therefore, the rate of production of coke
via the Boudouard reaction and toluene and methane
decomposition reactions far exceeded the rate at which carbon
was removed.
in the feed. The extents of reduction of the amounts of coke in the
presence of gasꢀphase oxygen were calculated to be 75.4%,
63.1%, 11.2% and 30.8% for spent NLSAO, NLAO, NA and NL
catalysts respectively.
0
65
The extent of reduction was the greatest for the perovskite
based catalysts, due to the presence of oxygen vacancies that is
characteristic of perovskite catalysts. These oxygen vacancies
allowed more oxygen species to be adsorbed onto the catalyst
surface to partake in the oxidation of coke. The larger amounts of
oxygen vacancies formed by the strontium substitution in
NLSAO catalyst resulted in its greater resistance to coke
formation as compared to NLAO catalyst. Evidently, oxygen
vacancies present in perovskite catalysts enhanced the
effectiveness of oxygen.
The results of thermogravimetric analyses of spent NLSAO
catalyst after toluene steam reforming at 650°C in the presence of
ml/min, 7.5ml/min, 10ml/min and 12.5ml/min gasꢀphase oxygen
are presented in Table 5. As oxygen flow rate was increased, the
carbon deposited on the spent catalyst surface after 100min of
70
2
5
Table 4 Amount of carbon deposition on spent catalysts after 8h reaction
with and without 5ml/min gas phase oxygen in feed.
Coke formation
(mg gꢀcat h )
Coke formation (%wt)
ꢀ1 ꢀ1
Catalyst
Without O
12.82
19.00
13.37
33.68
2
With O
5.02
8.41
12.03
23.30
2
Without O With O
2
2
75
NLSAO
NLAO
NA
16.02
23.75
16.71
42.10
6.27
10.51
15.04
29.13
5
NL
Comparing the perovskite catalysts, the amount of carbon 80 reaction decreased significantly. This elucidates that gasꢀphase
deposition over NLSAO catalyst was lower than that over NLAO
oxygen played a positive and significant role in conferring coking
resistance in the toluene steam reforming reaction.
3
0
catalyst due to strontium substitution of the LaAlO perovskite,
3
which led to greater mobility of lattice oxygen for the oxidation
of carbon formed during the steam reforming of toluene. In
addition to the amount of coke formation, Figure 8 shows the
DTA profiles of coke produced from NLSAO and NLAO
catalysts. The position of the peak can indicate the type of carbon
and difficultness to be oxidized. The NLSAO catalyst shows two
peaks of DTA profiles at 525°C and 605°C, while the NLAO
3
5
5
8
catalyst only show one peak at 552°C. Wang et al. classified the
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 9

Catalysis Science & Technology
Page 10 of 14
View Article Online
DOI: 10.1039/C5CY00412H
Table 5 Carbon deposition and first differential peak temperature of spent
NLSAO catalyst in various amount of oxygen.
h
g
O
(
2
flow rate Coke formation Coke formation Differential peak
ꢀ1
ꢀ1
mlꢀmin )
(%wt)
9.09
6.63
5.34
2.67
(mg gꢀcat )
temperature (°C)
5
90.91
66.34
53.38
26.73
595.73
601.33
602.32
495.94
7
.5
1
0
1
2.5
Figure 9 (a) – (d): Spent NLSAO, NLAO, NL and NA catalysts after 8h
of steam reforming of toluene at 650°C in the absence of gas phase
oxygen. (e) – (h): Spent NLSAO, NLAO, NL and NA catalysts after 8h of
steam reforming of toluene at 650°C in the presence of 5ml/min gas phase
oxygen.
Field Scanning Electron Microscopy (FESEM)
5
0
5
0
FESEM measurement was conducted for all spent NLSAO,
NLAO, NL and NA catalysts after 8h of steam reforming of
toluene at 650°C in the absence and presence of 5ml/min gas
phase oxygen in the feed to elucidate the morphology of the
surface of spent catalysts and was depicted in Figure 9 (a) to (h).
For reaction in the absence of gas phase oxygen in the feed
stream in Figure 8 (a)–(d), it was apparent that spent NL catalyst
contained the largest surface concentration of carbon nanotubes.
This was also evident during the steam reforming reaction, where
a significant buildꢀup of pressure within the reaction tube was
experienced. This suggests that NL catalyst possessed high
activity. However, because of high carbon formation rate, there is
no space left for the carbon nanotubes to grow causing pressure
builtꢀup in the system and blocking of the gas flow. Spent NA
catalyst appeared to possess the lowest surface concentration of
carbon nanotubes. This could be attributed to the low catalytic
activity of NA catalyst, therefore smaller amounts of carbon was
formed as a side product via the Boudouard reaction and
deposited on the catalyst surface.
3
5
Perovskite catalysts prove to be a plausible solution to
overcoming the shortcomings of both NL and NA. They
conferred relatively high catalytic activities, and low to moderate
amounts of carbon deposition. This was due to the redox
1
1
2
4
4
5
5
6
6
0
5
0
5
0
5
5
capabilities of perovskite catalysts . Comparing the morphologies
of two spent perovskite catalysts after 8 hours of toluene steam
reforming without oxygen in the feed stream in Figure 8 (a) and
(b), it was apparent that spent NLSAO catalyst possessed a lower
concentration of carbon nanotubes on its surface as opposed to
spent NLAO catalyst. This was due to the doping of La by Sr
which is of lower valance, resulting in oxidation state imbalance
and thereby increasing the amount of oxygen vacancies in
NLSAO catalyst. This increased the mobility of lattice oxygen
species, allowing them to react more quickly with carbon
deposited on the catalyst surface during steam reforming of
toluene, and therefore reducing the observed surface
concentration of carbon nanotubes on the spent catalyst
morphology. All the above observations are consistent with TGA
analyses.
For spent NLSAO, NLAO, NA and NL catalysts after 8h of
toluene steam reforming in the presence of 5ml/min oxygen in the
feed stream, it was evident that the presence of oxygen enhanced
the coking resistance of all catalysts. Figures 8 (e)–(h) all showed
a substantially lower amount of carbon present on the spent
catalyst surface as opposed to the corresponding Figures 8 (a)–
a
b
d
f
2
5
(d). Gas phase oxygen played a positive role in supressing carbon
c
formation and removing deposited carbon in all four catalysts.
This result is inꢀline with carbon deposition measurement from
TGA.
Role of lattice oxygen in the catalytic steam reforming of
toluene
7
7
8
0
5
0
The NLSAO catalyst demonstrated higher catalytic activity in
terms of toluene conversion as well as H production rate in the
2
e
steam reforming of toluene at 650°C in the absence of gas phase
oxygen compared to NLAO catalyst. XRD, TPR and XPS
catalyst characterization analyses indicated that the partial
substitution of lanthanum with strontium that is of lower valence
in NLSAO catalyst resulted in oxidation state imbalance between
3
+
2+
lanthanum (La ) and strontium (Sr ) sites, thereby reducing the
stability of the perovskite structure and conversely increasing the
amount of oxygen vacancies in the perovskite catalyst. The
presence of oxygen vacancies lowered the activation energy of
the migration of lattice oxygen, enhancing the mobility of lattice
3
0
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 10

Page 11 of 14
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C5CY00412H
oxygen species. This enhanced lattice oxygen mobility of
NLSAO catalyst allowed more lattice oxygen species to partake 55 Comparing the overall increase in average percentage carbon
in oxidation as surface oxygen species, resulting in higher redox
abilities and consequently improved catalytic activity and
stability.
It was found that the strontium substituted NLSAO catalyst not
only demonstrated higher overall toluene conversion and 60 conversion increased from 20.0% to 67.5% (3.4 times) for
conversion and hydrogen yield over NLSAO and NLAO catalysts
in the presence of gas phase oxygen, both improvements were
greater for the reaction over NLSAO catalyst as opposed to that
over NLAO catalyst. In particular, average percentage carbon
5
0
5
percentage H2 yield as opposed to NLAO catalyst, but also
demonstrated lower CO yield. This resulted in a much higher
H /CO ratio over NLSAO catalyst (H /CO = 8.35) as compared
NLSAO catalyst, but only 12.8% to 31.6% (2.5 times) for NLAO
in the presence of gas phase oxygen. Whereas, the average
1
percentage H yield increased from 15.5% to 36.9% (2.4 times)
2
2
2
to NLAO catalyst (H /CO = 3.03). However, the overall average
for NLSAO catalyst, but only 7.3% to 10.8% (1.5 times) for
2
production rates of H , CO and CO over NLSAO catalyst were 65 NLAO in the presence of gas phase oxygen. Such phenomena can
2
2
all higher than that over NLAO catalyst. This is indicative that
the lattice oxygen of NLSAO catalyst that possessed enhanced
mobility was able to function as active oxygen partook in the
partial oxidation of toluene:
be explained by the presence of a greater amount of oxygen
vacancies present on the surface of NLSAO catalyst as opposed
to NLAO catalyst as elucidated from XPS analyses. This allowed
more gas phase oxygen to be adsorbed onto the oxygen vacancies
present on the NLSAO catalyst surface to partake in the oxidation
of toluene, thereby enhancing the effectiveness of the 5ml/min
gas phase oxygen introduced to the feed stream.
1
70
ꢊ
C H + 7O
→ 7COꢂꢃꢄꢄꢅꢆꢇ + 4H , ∆H = −824.18kJ/mol
ꢀ
ꢁ
ꢂꢃꢄꢄꢅꢆꢇ
ꢈ
ꢈꢉꢁ
Thereafter, some of the COlattice underwent the waterꢀgas shift
reaction,
Additionally, through thermogravimetric analysis (TGA), the
spent catalysts after 8h of steam reforming of toluene in the
presence of 5ml/min gas phase oxygen all demonstrated a
decrease in the amount of carbon formation as opposed to the
spent catalysts after the reforming reaction in the absence of gas
phase oxygen. This indicates that gas phase oxygen played a
positive role in the oxidation of carbon deposited on the catalyst
2
0
75
ꢊ
COꢂꢃꢄꢄꢅꢆꢇ + H O → COꢂꢃꢄꢄꢅꢆꢇO + Hꢈ,
∆H
= −41.2kJ/mol
ꢈ
ꢈꢉꢁ
The oxygen vacancies were regenerated by steam. These
mechanisms are consistent with transient response tests using
18
2
5
H2 O over NLAO and NLSAO catalysts conducted by Sekine et 80 surface,
45
ꢊ
al. .
Additionally, thermogravimetric analysis (TGA) analysis
C + 1⁄2 O → CO,
∆H
= −110.5kJ/mol
ꢈ
ꢈꢉꢁ
elucidated that the amounts of carbon formed over NLSAO
catalyst was lower than NLAO catalyst in the steam reforming of
toluene at 650°C for 8h. This is indicative that lattice oxygen with
In particular, the extents of reduction of the amounts of carbon
nanotubes in the presence of gas phase oxygen were calculated to
be 75.4%, 63.1%, 11.2% and 30.8% for spent NLSAO, NLAO,
3
0
enhanced mobility in NLSAO catalyst could also partake in the 85 NA and NL catalysts respectively, as opposed to the reaction in
partial oxidation of carbon deposited on the catalyst surface as
active oxygen species, increasing the overall coking resistance of
the absence of gas phase oxygen. The extent of reduction was the
greatest for the perovskite based catalysts, due to the presence of
oxygen vacancies. The larger amounts of oxygen vacancies
formed by the strontium substituted in NLSAO resulted in its
greater improvement in coking resistance in the presence of gas
phase oxygen as compared to NLAO catalyst.
the catalyst in the steam reforming of toluene.
ꢊ
C + Oꢂꢃꢄꢄꢅꢆꢇ → COꢂꢃꢄꢄꢅꢆꢇ
,
∆H
= −110.5kJ/mol
ꢈꢉꢁ
90
3
5
Role of gas phase oxygen in the catalytic steam reforming of
toluene
As the flow rate of gas phase oxygen was increased from
ml/min to 12.5ml/min at 650°C, the overall percentage carbon
5
conversion and H yield for the steam reforming of toluene over
NLSAO catalyst also increased. This increase in catalytic activity
In the presence of 5ml/min gas phase oxygen in the feed
stream, the overall percentage carbon conversion and hydrogen
yield over all four catalysts in the steam reforming of toluene at
2
95
was the most significant when O flow rate was increased from
4
0
2
5
ml/min to 7.5ml/min. Thereafter, as O flow rate was further
2
6
50°C increased, as opposed to the reaction in the absence of gas
increased from 7.5ml/min through 12.5ml/min, improvements in
catalytic activity were increasingly marginal. This may be
phase oxygen. This is indicative that the presence of gas phase
oxygen in the feed led to enhanced catalytic activities. Likewise,
gas phase oxygen undertook the partial oxidation of toluene:
1
00 indicative that 7.5ml/min O is optimal given the existing toluene
2
steam reforming conditions.
4
5
ꢊ
At low O flow rates, the surface oxygen vacancies on NLSAO
were unsaturated, therefore an increase in oxygen flow rates from
C H + 7⁄2 O → 7CO + 4H ,
∆H
= −824.18kJ/mol
2
ꢀ
ꢁ
ꢈ
ꢈ
ꢈꢉꢁ
5
ml/min to 7.5ml/min allowed more gas phase oxygen species to
Based on catalytic performance trends observed in the presence
of gas phase oxygen, it was apparent from these results that some
1
05 be adsorbed onto the oxygen vacancies present on the catalyst
surface, enhancing the oxidation of toluene. As O flow rates
of the adsorbed gas phase oxygen (O species) possessed strong
2
2
were continually increased, the oxygen vacancies present on the
catalyst surface became increasingly saturated, therefore any
increase in oxygen flow rates would only result in a less than
5
0
oxidative abilities, and were capable of completely oxidizing
some of the carbon monoxide species:
ꢊ
110 proportionate increase in catalytic activities at high oxygen flow
CO + 1⁄2 ꢋ → CO , ∆H
= −283kJ/mol
ꢈ
ꢈ
ꢈꢉꢁ
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 11

Catalysis Science & Technology
Page 12 of 14
View Article Online
DOI: 10.1039/C5CY00412H
rates.
In addition, thermogravimetric analysis (TGA) revealed that as
55 1. P. Basu, Biomass Gasification and Pyrolysis: Practical Design and
Theory, Elsevier Inc., 2010.
2
3
4
5
.
.
.
.
T. Nordgreen, V. Nemanova, K. Engvall, and K. Sjöström, Fuel
012, 95, 71ꢀ78.
H. de Lasa, E. Salaices, J. Mazumder, and R. Lucky, Chem. Rev.
2011, 111, 5404ꢀ5433.
the flow rates of gas phase oxygen were increased from 5ml/min
to 12.5ml/min, the extent of carbon deposition on spent NLSAO
catalyst after the toluene steam reforming reaction decreased.
This is consistent with the preceding postulation that gas phase
oxygen oxidized carbon deposited on the catalyst surface,
resulting in enhanced coke resistance. In excess of gas phase
oxygen, more of the carbon deposited would be oxidized, leading
to lower amounts of carbon deposition.
2
5
60
G.W. Huber, S. Iborra, and A. Corma Chem. Rev. 2006, 106, 4044ꢀ
4
098.
D. Mukai, S. Tochiya, Y. Murai, M. Imori, T. Hashimoto, Y. Sugiura
and Y. Sekine, Appl. Catal. A 2012, 453, 60ꢀ70.
6
7
7
5
0
5
6. J. Ashok and S. Kawi, Int. J. Hydr. Energ. 2013, 38, 13938ꢀ13949.
7. U. Oemar, P.S. Ang, K. Hidajat, and S. Kawi, Int. J. Hydr. Energ.
10
2
013, 38, 5525ꢀ5534.
8
.
U. Oemar, M.L. Ang, W.F. Hee, K. Hidajat, and S. Kawi, Appl.
Catal. B 2014, 148ꢀ149, 231ꢀ242.
Conclusion
9. U. Oemar, M.L. Ang, K. Hidajat, and S. Kawi, AIChE J. 2014, 60,
190ꢀ4198.
4
Four
Niꢀbased
catalysts:
Ni/LaAlO3
(NLAO),
1
0. U. Oemar, M.L. Ang, K. Hidajat, and S. Kawi, RSC Adv. 2015, 5,
Ni/La0.8Sr0.2AlO (NLSAO), Ni/La O (NL), and Ni/Al O (NA)
3
2
3
2
3
17834ꢀ17842.
1
2
2
3
3
4
5
0
5
0
5
0
were developed for the steam reforming of tar using toluene as a
model compound. NLSAO catalyst demonstrated the best
catalytic performance in terms of both catalytic activity, as well
as coke resistance. As evident from XRD, TPRꢀH2 and XPS
catalyst characterization analyses, the partial replacement of
11. T. Kimura, T. Miyazawa, J. Nishikawa, S. Kado, K. Okumura, T.
Miyao, S. Naito, K. Kunimori, and K. Tomishige, Appl. Catal. B
2
006, 68, 160ꢀ170.
1
2. Z.A.ElꢀRub, E.A. Bramer, and G. Brem, Ind. Eng. Chem. Res. 2004,
43, 6911ꢀ6919.
13. D. Li, M. Koike, J. Chen, Y. Nakagawa, and K. Tomishige, Int. J.
Hydr. Energ. 39 (2014) 10959ꢀ10970.
3+
2+
lanthanum (La ) with strontium that is of lower valence (Sr )
resulted in oxidation state imbalance in NLSAO catalyst, thereby
reducing the structural stability of the perovskite oxide catalyst,
and increasing the number of oxygen vacancies present in the
80
1
4. C. Phuhiran, T. Takarada, and S. Chaiklangmuang, S., Int. J. Hydr.
Energ. 2014, 39, 3649ꢀ3656.
1
5. J. Ashok, Y. Kathiraser, M.L. Ang, S. Kawi, Appl. Catal. B 2015,
1
72ꢀ173, 116ꢀ128.
catalyst. The presence of oxygen vacancies in NLSAO catalyst 85 16. M.J. Pérez Zurita, M. Bartolini, T. Righi, G. Vitale, P. Pereira
lowered the activation energy of the migration of lattice oxygen
Olattice), enhancing their mobility and allowing them to function
as surface oxygen species. For the steam reforming of toluene at
50°C, mobile lattice oxygen partook in the direct partial
oxidation of toluene, forming COlattice and H2.
In addition, the formation of oxygen vacancies also allowed
Almao, Fuel 2015, 154, 71ꢀ79.
7. C.P.B. Quitete, R.C.P. Bittencourt, M.M.V.M. Souza, Appl. Catal. A
1
1
(
2
014, 478, 234ꢀ240.
8. L. Wang, Y. Hisada, M. Koike, D. Li, H. Watanabe, Y. Nakagawa,
and K. Tomishige, Appl. Catal. B 2012, 121ꢀ122, 95ꢀ104.
19. M. Koike, Y. Hisada, L. Wang, D. Li, H. Watanabe, Y. Nakagawa,
6
9
9
0
5
and K. Tomishige, Appl. Catal. B 2013, 140ꢀ141, 652ꢀ662.
2
0. C. Soongprasit, D.D. AhtꢀOng, and D. Atong, Adv. Mat. Res. 2013,
47, 690ꢀ693.
more gas phase oxygen species (O ) introduced in the feed stream
2
7
to be adsorbed onto the catalyst surface of NLSAO catalyst to
partake in the oxidation of toluene. This enhanced the
improvement in catalytic activity of NLSAO catalyst in the
presence of gas phase oxygen as compared to undoped NLAO
catalyst. Gas phase oxygen species possessed strong oxidative
abilities and were capable of completely oxidizing some of the
21. K. Soongprasit, D. AhtꢀOng, V. Sricharoenchaikul, and D. Atong,
Top. Catal. 2013, 56, 339ꢀ344.
22. N. Laosiripojana, W. Sutthisripok, S. Charojrochkul, S.
Assabumrungrat, Fuel Process. Technol. 2014, 127, 26ꢀ32.
3. A. Laobuthee, C. Veranitisagul, W. Wattanathana, N. Koonsaeng, N.
Laosiripojana, Renew. Energ. 2015, 74, 133ꢀ138.
4. J.R. Kastner, S. Mani, A. Juneja, Fuel Process. Technol. 2015, 130,
31ꢀ37.
2
1
1
1
1
1
00
2
feed toluene gas to form CO . As elucidated from TGA spent
2
catalyst characterization techniques, both mobile lattice and gas
phase oxygen species played positive roles in oxidizing deposited
carbon on the catalyst surface during the steam reforming of
toluene.
25. K. Tomishige, T. Miyazawa, M. Asadullah, S.ꢀI. Ito, K. Kunimori,
Green Chem. 2003, 5, 399ꢀ403.
05 26. D. Mukai, Y. Murai, T. Higo, S. Ogo, Y. Sugiura, and Y. Sekine,
Appl. Catal. A 2014, 471, 157ꢀ164.
7. D. Li, M. Tamura, Y. Nakagawa, K. Tomishige, Bioresources
Technol. 2015, 178, 53ꢀ64.
8. Y. Sugiura, D. Mukai, Y. Murai, S. Tochiya and Y. Sekine, Int. J.
Hydr. Energ. 2013, 38, 7822ꢀ7829.
9. K. Urasaki, Y. Sekine, S. Kawabe, E. Kikuchi and M. Matsukata,
Appl. Catal. A 2005, 286, 23ꢀ29.
2
2
2
3
10
Acknowledgements
45
The authors acknowledge financial support from National
University of Singapore and NEAꢀETRP (Project No. 1002 114,
Research Grant No. 279ꢀ000ꢀ333ꢀ490). Usman Oemar sincerely
thanks Dr. Yasotha Kathiraser and Dr. Jangam Ashok for
technical support and discussions.
0. V. Stathopoulos, V. Belessi, T. Bakas, S. Neophytides, N. Costa, P.
Pomonis and A. Efstathiou, Appl. Catal. B 2009, 93, 1ꢀ11.
15 31. M. Koike, C. Ishikawa, D. Li, L. Wang, Y. Nakagawa, and K.
Tomishige, Fuel 2013, 103, 122ꢀ129.
32. L. Wang, D. Li, M. Koike, S. Koso, Y. Nakagawa, Y. Xu, and K.
Tomishige, Appl. Catal. A 2011, 392, 248ꢀ255.
3
3
3
3. J. Nishikawa, K. Nakamura, M. Asadullah, T. Miyazawa, K.
Kunimori, and K. Tomishige, Catal. Today 2008, 131, 146ꢀ155.
4. C. Li, D. Hirabayashi, and K. Suzuki, Appl. Catal. B 2009, 88, 351ꢀ
50
Notes and references
20
Department of Chemical and Biomolecular Engineering, National
University of Singapore, 4 Engineering Drive 4, Singapore 117585. Fax:
3
60.
5. J.S. Lisboa, D.C.R.M. Santos, F.B. Passos, and F.B. Noronha, Catal.
Today 2005, 101, 15ꢀ21.
+
65ꢀ6779 1936; Tel: +65ꢀ6516 6312; Eꢀmail: chekawis@nus.edu.sg
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 12

Page 13 of 14
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C5CY00412H
3
3
6. D. Świerczyński, S. Libs, C. Courson, and A. Kiennemann, Appl.
Catal. B 2007, 74, 211ꢀ222.
7. K. Nakamura, T. Miyazawa, T. Sakurai, T. Miyao, S. Naito, N.
Begum, K. Kunimori, and K. Tomishige, Appl. Catal. B 2009, 86,
36ꢀ44.
5
0
5
0
5
0
5
0
5
3
3
8. A. Carrero, J.A. Calles, and A.J. Vizcaino, Chem. Eng. J. 2010, 163,
3
95ꢀ402.
9. D. Li, L. Wang, M. Koike, Y. Nakagawa, and K. Tomishige, Appl.
Catal. B 2011, 102, 528ꢀ538.
1
1
2
2
3
3
4
4
40. F. Bimbela, D. Chen, J. Ruiz, L. Garcia, and J. Arauzo, Appl. Catal.
B 2012, 119ꢀ120, 1ꢀ12.
4
1. C.K.S. Choong, L. Huang, Z. Zhong, J. Lin, L. Hong, and L. Chen,
Appl. Catal. A 2011, 407, 155ꢀ162.
2. T. Viinikainen, H. Rönkkönen, H. Bradshaw, H. Stephenson, S.
Airaksinen, M. Reinikainen, P. Simell, and O. Krause, Appl. Catal. A
4
2
009, 362, 169ꢀ177.
4
4
4
3. D. Świerczyński, C. Courson, and A. Kiennemann, Chem. Eng.
Process. 2008, 47, 508ꢀ513.
4. D. Mukai, S. Tochiya, Y. Murai, M. Imori, Y. Sugiura, and Y.
Sekine, Appl. Catal. A 2013, 464–465, 78ꢀ86.
5. Y. Sekine, D. Mukai, Y. Murai, S. Tochiya, Y. Izutsu, K. Sekiguchi,
N. Hosomura, H. Arai, E. Kikuchi, Y. Sugiura, Appl. Catal. A 2013,
4
51, 160ꢀ167.
4
6. K. Takise, M. Imori, D. Mukai, S. Ogo, Y. Sugiura, Y. Sekine, Appl.
Catal. A 2015, 489, 155ꢀ161.
7. C. Li and Y.ꢀW. Chen, Thermochim. Acta 1995, 256, 457ꢀ465.
8. M.A. Peña and J.L.G. Fierro, Chem. Rev. 2001, 101, 1981ꢀ2017.
9. U. Oemar, K. Hidajat, and S. Kawi, Appl. Catal. A 2011, 402, 176ꢀ
4
4
4
1
87.
50. M.L. Ang, U. Oemar, E.T. Saw, L. Mo, B.H. Chia, and S. Kawi, ACS
Catal. 2014, 4, 3237ꢀ3248.
5
5
5
1. K. Sutthiumporn and S. Kawi, Int. J. Hydr. Energ. 2011, 36, 14435ꢀ
4446.
1
2. K. Sutthiumporn, T. Maneerung, Y. Kathiraser, S. Kawi, Int. J. Hydr.
Energ. 2012, 37, 11195ꢀ11207.
3. M. Goldwasser, M. Rivas, E. Pietri, M. PérezꢀZurita, M. Cubeiro, L.
Gingembre, L. Leclercq and G. Leclercq, Appl. Catal. A 2003, 255,
4
5ꢀ57.
5
4. E. Ruckenstein and Y. H. Hu, J. Catal. 1996, 161, 55ꢀ61.
55. D. N. Bangala, N. Abatzoglou and E. Chornet, AIChE J. 1998, 44,
27ꢀ936.
6. B. P. Barbero, J. A. Gamboa and L. E. Cadus, Appl. Catal. B 2006,
5, 21–30.
7. I. Rossetti and L. Forni, Appl. Catal. B 2001, 33, 345ꢀ352.
9
5
5
6
58. C. Wang, N. Sun, N. Zhao, W. Wei, J. Zhang, T. Zhao, Y. Sun, C.
Sun, H. Liu, C.E. Snape, ChemCatChem 2014, 6, 640ꢀ648.
5
9. S. Chen, H. Wang and Y. Liu, Int. J. Hydr. Energ. 2009, 34, 7995ꢀ
005.
8
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 13

Catalysis Science & Technology
Page 14 of 14
View Article Online
DOI: 10.1039/C5CY00412H
Faculty of Engineering
Department of Chemical and Biomolecular Engineering
Graphical Abstract
4
Engineering Drive 4, Singapore 119260 Tel: (65) 6516 6312 Fax: (65) 6779 1936
E-mail: chekawis@nus.edu.sg Website: http://www.nus.edu.sg