Steam reforming of aromatic hydrocarbon at low temperature in electric field

Article abstract of DOI:10.1016/j.apcata.2019.01.011

Toluene steam reforming was conducted over Ni-supported La_0.7Sr_0.3AlO_3-δ (LSAO) perovskite oxide at 473 K in an electric field, and high toluene conversion was achieved even at such low temperature. Steam reforming is generally done at higher temperatures because of the endothermic nature of this reaction and coke prevention, which requires the contribution of heat energy in great amounts. Reportedly, water activation is particularly difficult to achieve at lower temperatures. Results of kinetic analyses such as Arrhenius plots and partial pressure dependence have revealed that imposing an electric field promotes water activation over Ni/LSAO catalyst. According to steady state isotopic kinetic analysis (SSITKA) and temperature programmed desorption infrared (TPD-IR) measurements, such water activation was accelerated via a redox mechanism using surface lattice oxygen of LSAO. The stimulated mobile lattice oxygen oxidized the adsorbed toluene and formed coke precursor in the electric field. The electric field achieved a lowered reaction temperature for efficient hydrogen production and coke suppression with redox property of LSAO support during toluene steam reforming.

Full text of DOI:10.1016/j.apcata.2019.01.011

Accepted Manuscript
Title: Steam reforming of aromatic hydrocarbon at low
temperature in electric field
Authors: Kent Takise, Ayaka Sato, Keisuke Muraguchi,
Shuhei Ogo, Yasushi Sekine
PII:
S0926-860X(19)30020-1
DOI:
Reference:
https://doi.org/10.1016/j.apcata.2019.01.011
APCATA 16951
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
15 September 2018
2 December 2018
15 January 2019
Please cite this article as: Takise K, Sato A, Muraguchi K, Ogo S, Sekine Y, Steam
reforming of aromatic hydrocarbon at low temperature in electric field, Applied
Catalysis A, General (2019), https://doi.org/10.1016/j.apcata.2019.01.011
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Title
Steam reforming of aromatic hydrocarbon at low temperature in electric field
Authors
Kent Takisea, Ayaka Satoa, Keisuke Muraguchia, Shuhei Ogoa, Yasushi Sekinea*
Affiliations
a Waseda University, Department of Applied Chemistry, 3-4-1, Okubo, Shinjuku, Tokyo, 169-8555
Japan
* Corresponding author: Yasushi SEKINE, Ph.D., Prof., Waseda University
+81-3-5286-3114 (Tel/Fax), ysekine@waseda.jp
Graphical abstract
2.5
SR
ER
2.0
1.5
1.0
0.5
0.0
-0.5
Electric field
Ea = 26.1 kJ mol-1
Imposing electric field to Ni/La0.7Sr0.3AlO2.85
catalyst decreased the apparent activation
energy drastically on steam reforming of
Heated catalyst
Ea = 78.9 kJ mol-1
toluene.
1.3
1.4
1.5
1.6
1.7
1.8
1/T / 103 K-1
Highlights
 Catalytic toluene steam reforming was investigated over Ni/La0.7Sr0.3AlO3-
catalyst.
 Reaction was promoted by an electric field drastically even at 473 K.
 Low apparent activation energy of 26.1 kJ/mol was observed in the electric
field.
 SSITKA and TPD-IR revealed the electric field promoted redox of surface
lattice oxygen.
1

 Electric field achieved high activity and low carbon deposition at low reaction
temperature
Abstract
Toluene steam reforming was conducted over Ni-supported La0.7Sr0.3AlO3-(LSAO)
perovskite oxide at 473 K in an electric field, and high toluene conversion was achieved even at
such low temperature. Steam reforming is generally done at higher temperatures because of the
endothermic nature of this reaction and coke prevention, which requires the contribution of heat
energy in great amounts. Reportedly, water activation is particularly difficult to achieve at lower
temperatures. Results of kinetic analyses such as Arrhenius plots and partial pressure dependence
have revealed that imposing an electric field promotes water activation over Ni/LSAO catalyst.
According to steady state isotopic kinetic analysis (SSITKA) and temperature programmed
desorption infrared (TPD-IR) measurements, such water activation was accelerated via a redox
mechanism using surface lattice oxygen of LSAO. The stimulated mobile lattice oxygen oxidized
the adsorbed toluene and formed coke precursor in the electric field. The electric field achieved a
lowered reaction temperature for efficient hydrogen production and coke suppression with redox
property of LSAO support during toluene steam reforming.
Keywords
Coke suppression; Hydrogen production; Ni catalyst; Perovskite-type oxide; Redox mechanism;
Toluene steam reforming
Main text
1. Introduction
Hydrogen production has been attracting much attention because hydrogen is demanded for use
in various chemical industries and is anticipated for use as clean secondary energy for combustion
or fuel cell utilization. Currently, the typical method used for producing hydrogen is catalytic steam
reforming using light hydrocarbons such as methane or naphtha [1,2], but steam reforming of
aromatic hydrocarbons is also investigated extensively because it enables the use of biomass or
brown-coal-derived materials [3–19]. These aromatic hydrocarbons are less reactive and tend to
form coke during such catalytic reactions, deriving coverage of active sites and plugging of the
reactor in the worst case. Certainly, suppressing coke formation and lowering the reaction
temperature are extremely important for the steam reforming of aromatic hydrocarbons.
To investigate coke formation suppression, many studies have been conducted to assess the
inhibition of coke deposition during steam reforming. Dilution of an active metal such as Ni might
be an effective means of suppressing coke by-production [20–23]. For instance, Anjaneyulu et al.
reported that using a synthesized Ni-Zn alloy supported Al2O3 catalyst improved coke resistance
2

during ethanol steam reforming [20]. In this case, Zn played important roles in dilution of Ni metal
and in the enhancement of hydrogen spillover to form adsorbed water to suppress coking [20].
Similarly, the presence of alkali or magnesia-based catalyst can enhance steam adsorption to depress
coke formation [21,24]. Additionally, lattice oxygen in/on support oxides can contribute to coke
resistance [25–35]. Marinho et al. stated that the formed coke was calcined by transferred oxygen
from perovskite-type oxide, LaNiO3/CeSiO3 catalyst under water presence during ethanol steam
reforming [25]. In an earlier study, we demonstrated that lattice oxygen of La0.7Sr0.3AlO2.85 (LSAO),
a perovskite-type oxide, drastically decreased the amount of coke formation during toluene steam
reforming [36–41]. Furthermore, such lattice oxygen promoted toluene decomposition with a redox
mechanism. Consequently, LSAO support achieved both high activity and stability during steam
reforming of aromatic hydrocarbons.
Lowering the reaction temperature of steam reforming is beneficial for removing coke
oxidatively and for achieving high-efficiency hydrogen production, simultaneously utilizing low-
grade heat. Various approaches have been inspected for lowering the reaction temperature, for
example auto thermal reforming (ATR) [42–45]. Electric field addition to the catalyst bed can also
lower the reaction temperature of endothermic reactions such as steam reforming [46–49]. Manabe
et al. reported that steam reforming of methane in an electric field proceeded with high activity even
at temperatures lower than 473 K [50]. In this reaction, the adsorbed water promoted proton transfer
and a collisional reaction between accelerated protons and methane over Pd-supported CeO2 [50,51].
Vayenas et al. suggested that electric current promoted oxygen mobility in metal oxides and their
catalytic oxidation through NEMCA (Non-Faradaic Electrochemical Modification of Catalytic
Activity) [53-55]. Accordingly, ion transfer or activation might be promoted in an electric field
during a steam reforming reaction. For this study, toluene steam reforming was conducted at 473 K
in an electric field over Ni-supported LSAO catalyst, which showed high activity and stability
because of redox ability: lattice oxygen mobility.
2. Experimental
2.1 Catalyst preparation
Perovskite-type oxide, La0.7Sr0.3AlO3- (LSAO) for catalyst support was synthesized using the
citric acid complex method. The crystalline structure was confirmed by XRD as shown in
supporting information figure S1, and we could observe only for LSAO perovskite peaks and no
other peaks for impurity phases were confirmed. Earlier reports explained additional details related
to the citric acid complex method [36, 38, 40, 41]. After powdered LSAO was impregnated with a
solution of nickel (II) nitrate (Kanto Chemical Co. Inc.), the LSAO was soaked in 25 mL of water
for 2 h. Then Ni nitrate was added with 15 mL of water and was dissolved for 2 h using an evaporator.
Concentration of nickel nitrate was adjusted for each loading amount i.e., 1, 3 and 5 wt% of Ni.
Later, it was dried and calcined at 773 K for 2 h. The obtained catalyst size was adjusted to 355–
500 m with sieving.
3

2.2 Activity tests with/without the electric field
Catalyst activity tests were conducted using a tubular reactor in which 200 mg of catalyst was
charged into a fixed bed as shown in supporting information figure S2. Two stainless steel electrodes
were set contiguously on the upper side and bottom side of the fixed catalysts to charge an electric
field. The same setup was used for steam reforming without the electric field. Toluene steam
reforming was conducted for 60 min in each condition after pre-reduction at 873 K. For reaction
with an electric field, 3 mA of constant current was applied between the two electrodes. The reaction
gas composition was C7H8 : H2O : Ar = 3 : 42 : 155 (mL min-1). The steam to carbon ratio (S/C)
was 2.0. Also, the weight flow ratio was 3.14 g-cat h mol-1. We confirmed that the condition is in
the kinetic region and that diffusion is not a rate determining factor. Results demonstrate that the
effect of mass transportation was negligible in such a condition. The product gas of the reaction was
measured using GC-FID (GC-8A; Shimadzu Corp.) and GC-TCD (GC-8A; Shimadzu Corp.).
Toluene conversion and hydrogen yield were defined with a toluene feed rate (mmol s-1) and the
following formation rate of products: rCO (mmol s-1) for CO, rCH4 (mmol s-1) for CH4, rCO2 (mmol
s-1) for CO2, and rH2 (mmol s-1) for hydrogen.
Toluene conversion (%) = (rCO + rCH4 + rCO2) / (C7H8 feed rate × 7) × 100 eq. 1
Toluene steam reaction is described as C7H8 + 7H2O  7CO + 11H2
And sequential water gas shift reaction is described as CO + H2O  CO2 + H2
By combining these two reactions, the total reaction can be described as
C7H8 + 14H2O  7CO2 + 18H2
So, the hydrogen yield can be calculated as follows;
Hydrogen yield (%) = rH2 / (C7H8 feed rate × 18) × 100
eq. 2
After reaction for 60 min, the amount of deposited coke was measured with a gas chromatograph
using temperature programmed oxidation (TPO) measurements to estimate CO or CO2 from coke.
The temperature was increased from 298 K to 1173 K at 10 K min-1 for the gas composition of O2 :
He = 10 : 90 (mL min-1). The resultant ratio of produced gas components and coke formation
amounts was confirmed based on the carbon balance, which was almost 1.
2.3 Catalytic structure analysis
The obtained perovskite-type structure of LSAO support was confirmed from X-ray diffraction
measurements (XRD: Ultima III; Rigaku Corp.) using Cu Kα X-ray radiation of 40 kV at 40 mA.
The average particle diameter of the supported Ni metal was estimated using CO pulse
measurements (BELCAT II; MicrotracBEL Corp.). Furthermore, the obtained particle diameter was
confirmed with images and EDX mapping results obtained using scanning transmission electron
microscopy (STEM; HF-2210; Hitachi Ltd.).
2.4 Kinetic analysis of Ni supported LSAO catalyst
4

Arrhenius plots were obtained over 5 wt% Ni/LSAO catalyst to evaluate the apparent active
energy of toluene steam reforming in each condition: (1) at 473 K with the electric field and (2) at
698 K without the electric field. Reaction rates were evaluated in terms of kinetic control. The
reaction rates were defined from the formation rate of CO, CO2, and CH4 in the same manner as
that used for activity tests. In the case of reaction with an electric field, catalyst temperatures were
increased with Joule heat by charged electricity. Thermocouples detected the temperature on the
catalyst. Those data were then used to ascertain the temperature rise.
To elucidate reaction mechanisms, partial pressure dependences of the reaction rates were
confirmed for toluene and steam pressure in each reaction condition: (1) at 473 K with the electric
field and (2) at 698 K without the electric field. The feed gas compositions were modified as C7H8 :
H2O : Ar = (1, 2, 3) : 42 : (157, 156, 155) (mL min-1) and C7H8 : H2O : Ar = 3 : (30, 42, 54) : (167,
155, 143) (mL min-1), respectively, for toluene and steam pressure. Reaction rates were calculated
from the formation rates of CO, CO2, and CH4 analyzed using GC-FID (GC-8A; Shimadzu Corp.).
2.5 Evaluation of the lattice oxygen release rate with steady state isotopic transient kinetic analysis
(SSITKA)
Lattice oxygen release rates were estimated over the 5 wt% Ni/LSAO catalyst, in the following
three conditions: (1) at 473 K with the electric field, (2) at 698 K without the electric field and (3)
at 473 K without the electric field. Detailed procedures for SSITKA have been described elsewhere
[25, 27–29]. For this experiment, the first reaction was conducted with gas including isotopic water
i.e. 1.5%C7H8 : 6.3%H216O : 14.7%H218O : 5%Ar : 72.5%He vol% (total flow rate: 200 mL min-1)
to replace lattice oxygen in/on the LSAO support by 18O. After helium purging treatment for 120
min, sequential reactions were conducted in the following gas composition: 1.5%C7H8 : 21%H216O :
5%Ar : 72.5%He vol% (total flow rate: 200 mL min-1). Products of the second reaction were
measured using a quadrupole mass spectrometer (Q-Mass, HPR20; Hiden Analytical Ltd.). The
observed signals of m/e were 2(H2), 4(He), 15(CH4), 18(H2O), 20(H218O), 28(CO), 30(C18O),
40(Ar), 44(CO2), 46(C16O18O), 48(C18O2), 78(C6H6), and 91(C7H8). We found the effect of Ar2+
(m/z = 20) signal to H218O is negligible [38–41]. The lattice oxygen release rate was defined as
formation rates of 30(C18O), 46(C16O18O), and 48(C18O2).
2.6 Temperature programmed desorption infrared (TPD-IR) measurements in an electric field
Diffuse reflectance infrared spectrometry was applied to observe adsorbed species on Ni/LSAO
catalyst using a Fourier transform infrared spectrometer (FT/IR-6100; Jasco Corp.). Additionally,
the measurement apparatus was assembled for applying the electric field on the samples as described
elsewhere in the literature [51]. Sieved catalyst was used in this experiment. Two stainless steel
electrodes were set on the catalyst. An earlier report described the temperature program scheme for
this measurement [38]. After pre-reduction at 873 K and background measurement at 323 K, diluted
toluene gas was flowed into a measurement chamber for 30 min with 2 mL min-1 of the toluene feed
5

rate. Subsequently, argon gas was introduced to remove gas-state toluene before the temperature
rise. Without the electric field, the IR cell was heated and kept at 373, 473, 573, 673, 773, and 873
K for 10 min independently in Ar gas. For reaction with an electric field, the temperature was raised
to 373, 473, 573, and 673 K because of thermotolerance of the Teflon cell. Also, an electric field
was charged for 10 min at each temperature. After each heating operation, the cell was cooled to
323 K. Then the IR spectrum was measured. Figures presented in an earlier report present a
comprehensive view of the relevant spectra [38].
3. Results and Discussion
3.1 Catalytic activity and coke deposition during toluene steam reforming in an electric field
Toluene steam reforming was conducted over supported metal catalyst with 5 wt% of active
metal i.e. Ni, Co, Fe or Cu supported on LSAO, in a tubular reactor at 473 K with an electric field.
Figure 1 shows the time courses of the catalytic activities and hydrogen yields for these catalysts.
Results indicate that Ni/LSAO catalyst showed the highest activity among these catalysts in this
condition during toluene steam reforming. Consequently, toluene steam reforming was also
conducted over Ni/LSAO catalyst without an electric field at 473 K for better comparison. However,
Ni/LSAO showed no activity at such a low reaction temperature. To elucidate the effects of the
electric field over Ni/LSAO catalyst, catalytic activity was investigated at each reaction temperature
from 473 K to 873 K. The results of activity testing at each reaction temperature with and without
the electric field are presented in Figure 2, in which filled plots and open plots respectively represent
the activities with (ER: reaction with an electric field) and without (SR: steam reforming without
an electric field) the electric field. In the case of ER, the catalyst bed was heated by Joule heating
of about 100 K, so filled plots are shifted to higher temperatures than the original furnace
temperature. The reaction temperature was measured using a thermocouple on the catalyst bed
inside of a tubular reactor. Results show that the catalytic activity of ER was much higher than that
of steam reforming at the same temperature: for example, ER, 44.7% at 573 K vs. SR, 1.3% at 573
K. The electric field effect is greater at temperatures lower than 773 K. We confirmed the waveform
of response voltage during electric field addition using an oscilloscope, and no plasma was found.
Also no emission spectra can be observed. Therefore, the electric field promotes some catalytic
reaction steps effectively at lower temperatures, at which toluene steam reforming does not
generally proceed.
Effects of Ni loading amounts on the coke deposition activity were investigated using LSAO
support in the electric field. Figure 3 and supporting information Tables S1 and S2 depict the
respective toluene conversion, hydrogen yields and other products including deposited coke over
three catalysts, i.e. 1, 3 and 5 wt%Ni/LSAO, during steam reforming with and without the electric
field. Results demonstrated that 5 wt%Ni/LSAO showed higher activity than other catalysts in both
cases. As for ER, the catalyst attained better stability. The catalytic activity was very low without
supported Ni metal as shown in supporting information figure S3. Table 1 shows the catalytic
6

activities and the amounts of formed coke over these Ni/LSAO catalysts at 473 K with the electric
field and at 698 K without the electric field because conversion showed similar values in these
conditions. In the case of ER, the amounts of formed coke were very small over 1, 3 and 5
wt%Ni/LSAO as shown in supporting information figure S4. These values were below 10 mg g-cat-1.
In the case of SR, the amounts of coke increased considerably from 39.2 mg g-cat-1 to 525.8 mg
g-cat-1 as the Ni loading amounts and catalytic activities increased. Steam reforming was promoted
with Ni loading, simultaneously causing coke by-production without the electric field. To confirm
coke suppression effects of the electric field, an activity test was conducted over 5 wt% Ni/LSAO
catalyst at 473 K for 60 min after toluene steam reforming at 698 K for 60 min. After all these
treatments, the amount of formed coke was decreased to 253.4 mg g-cat-1 compared to 525.8 mg g-
cat-1 as shown in supporting information figure S5. In the lower temperature region, it was described
that more reactive coke such as Cα and Cβ tended to deposit [56,57]. This experiment suggests that
some reactive coke is formed, but such coke can be removed by electric field charging. These results
revealed that the electric field promoted endothermic steam reforming at lower temperature, and
that it strongly suppressed coke formation during toluene steam reforming.
3.2 Active site specification and kinetic analysis for toluene steam reforming in an electric field
Catalytic activities were improved by the increase of Ni metal loading. Actually, Ni metals
played important roles as reactive sites for progression of the steam reforming reaction.
Consequently, the Ni-specific area dependency of activities was evaluated over 1, 3 and 5 wt%
Ni/LSAO catalysts with/without the electric field. Table 2 and supporting information figure S6
present turnover frequencies (TOF) of two kinds, as determined by the Ni specific surface area
(TOF-s) and Ni-support perimeter (TOF-p), STEM image of supported Ni is shown in supporting
information figure S7 and Ni particle size was evaluated by CO adsorption as shown in Table S3.
Regarding ER, the value of TOF-p was constant over each Ni catalyst. The toluene steam reforming
activity was determined by the Ni-support perimeter. However, both TOF-s and TOF-p values were
increased as the Ni particle diameter increased in the case of steam reforming without an electric
field. Generally, adsorbed water was regarded as activated on metal oxide supports rather than on
supported active metal. Additionally, it is difficult to activate water in the lower temperature region
[40]. Consequently, the water activation rate over LSAO support affects the total reaction rate of
steam reforming. According to reports of earlier studies, catalytic activity is affected strongly by
the Ni-support perimeter when the adsorbed water is dissociated rapidly, for example via redox
mechanism [37, 39-41]. The electric field possibly promoted water activation over Ni/LSAO
catalyst.
Kinetic analysis was performed over 5 wt% Ni/LSAO catalyst to elucidate the electric field
effects of more specifically. Figure 4 shows the resultant Arrhenius plots over Ni/LSAO catalyst
with and without the electric field. These results show that the apparent active energy was 26.1 kJ
mol-1 for ER and 78.9 kJ mol-1 for SR. The evaluation was conducted in a kinetic region, confirmed
7

by changing the contact time as shown in figure S8, and the Arrhenius plot for higher temperature
in a diluted condition is shown in figure S9. The apparent active energy decreased drastically by
electric field application, meaning that the electric field promoted a specific reaction path.
Subsequently, as shown in Figure 5, partial pressure dependences for toluene and steam were
analyzed over Ni/LSAO catalyst in each condition: ER at 473 K and SR at 698 K. The reaction rate
of toluene steam reforming (rSR) is defined as shown below.
rSR = k[C7H8]a[H2O]b
eq. 3
For steam reforming without the electric field, the reaction rate depended predominantly on the
steam partial pressure i.e., a = 0.23 and b = 0.93. As described above, the adsorbed water is difficult
to dissociate at a low temperature such as 698 K [37, 39–41]. Regarding ER, the reaction rate more
reliant on toluene partial pressure i.e., a = 0.95 and b = 0.29. Apparently, the electric field facilitated
dissociation of adsorbed water, even at a low temperature such as 473 K, which achieved
acceleration of steam reforming in the lower temperature region and coke resistance. This partial
pressure tendency was also observed when the redox mechanism was expressed over LSAO support
[37, 39–41]. It is typically affirmed that redox mechanism occurs at higher temperatures above
approximately 800 K without any special treatment such as an electric field [37, 40]. For this reason,
further experiments were conducted to clarify water activation mechanisms and the conduction
species over LSAO support in an electric field.
As for the energy balance for this reaction, energy balance was calculated with the following
equation;
Energy balance (%) = Converted toluene (mol sec-1) × ΔH for the reaction (kJ mol-1) / Input
power (W) × 1000
eq. 4
ΔH: Required enthalpy for C7H8 + 14H2O  7CO2 + 18H2 (828.6 kJ mol-1)
Accordingly, resulted energy balance was 18.3%. Therefore, there is a room for improvement in
status quo. Faraday efficiency was 29.4 (one electron promoted the conversion of 29.4 times of
toluene). This reaction was not electrochemical reaction, and promising for reducing required
electric power by controlling the bulk electron conductivity (to decrease wasted Joule heat).
3.3 Lattice oxygen redox property over LSAO support in an electric field
Steady state isotopic transient kinetic analysis (SSITKA) was applied over Ni/LSAO catalyst
during toluene steam reforming with and without the electric field because LSAO support was
regarded to accelerate water activation with its redox property. Figure 6 and Table 3 present lattice
oxygen release rates over 5 wt% Ni/LSAO catalyst in the following conditions: (a) at 473 K with
the electric field, (b) at 698 K without the electric field, and (c) at 473 K without the electric field.
Lattice oxygen release rates were estimated with isotopic 18O release after surficial purge treatment,
8

after which 18O exists only inside of LSAO lattice and defects [39–41]. The results of Figures 6(b)
and 6(c) showed that the lattice oxygen release rates were 0.59 and 0.03 μmol s-1 at 698 K and 473
K during steam reforming without the electric field. Regarding the ER, the lattice oxygen release
rate was 1.77 μmol s-1 at 473 K. The lattice oxygen release rate was remarkably faster with the
electric field. Even catalytic activities were similar between the two conditions: (a) at 473 K with
the electric field and (b) at 698 K without the electric field. Furthermore, this oxygen mobility was
not observed over the catalyst without redox ability such as Ni/LaAlO3, on which the electric field
was not stable and generated blue-white plasma with discharging. Consequently, results suggest
that the redox ability of LSAO support promoted water activation at low temperatures, and that
mobile oxygen ions and their vacancies play an important role in toluene steam reforming in an
electric field.
Water activation was accelerated with redox ability of LSAO in the electric field. However, it
remains unclear that toluene steam reforming proceeded with a redox mechanism. Therefore,
temperature programmed desorption infrared (TPD-IR) measurements were conducted over
Ni/LSAO with/without the electric field. Figure 7 shows results of TPD-IR measurements
conducted with and without the electric field, respectively. At low temperatures, typical ring C-C
vibration was detected at wavenumbers 1500 and 1600 cm-1 in the respective conditions [38,40].
Symmetric and asymmetric COO vibrations started to appear at 673 K in the case of steam
reforming without the electric field, as shown in Figure 7(B) [38,40]. By contrast, those peaks were
detected at a lower temperature, 473 K in the case of ER. In earlier studies, adsorbed toluene was
converted to oxygenate intermediates on LSAO support before sequential oxidation to CO or CO2
over Ni particles. Then, such formation of oxygenates facilitates decomposition of toluene [38,40].
These results suggest that the redox mechanism is predominant over Ni/LSAO catalyst in the
electric field, which is important for coke resistance and processes of steam reforming.
4. Conclusion
We investigated catalytic steam reforming of toluene as a model compound of aromatic
hydrocarbons at low temperatures in an electric field. Even at 473 K, Ni/LSAO showed high
catalytic activity during toluene steam reforming in an electric field. Without the electric field,
almost no activity was observed at 473 K. The coke deposition amount was 10 mg g-cat-1 for ER at
473 K and 525.8 mg g-cat-1 for SR at 698 K, although the toluene conversion was almost identical
for these two conditions. The formed coke can be removed by the imposition of an electric field
because more reactive coke such as Cα and Cβ tended to be produced in the low temperature region.
The apparent active energy was lowered from 78.9 kJ mol-1 to 26.1 kJ mol-1 by application of the
electric field. The steam partial pressure dependence changed simultaneously. The lattice oxygen
release rate was considerably higher. Oxygenate intermediates were formed at lower temperature in
the electric field. Therefore, we infer that the redox mechanism is predominant over Ni/LSAO
9

catalyst in the electric field. These results indicate that the electric field promoted activation of water
using lattice oxygen ion and its vacancy, which is important for oxidation of surficial reactive coke
and adsorbed toluene.
REFERENCES
[1] S. D. Angeli, G. Monteleone, A. Giaconia, A. A. Lemonidou, Int. J. Hydrogen Energy 39 (2014)
1979-1997.
[2] F. Melo, N. Morlanes, Catal. Today 107-108 (2005) 458-466.
[3] C. P. B. Quitete, R. C. P. Bittencourt, M. M. V. M. Souza, Appl. Catal. A: Gen. 478 (2014) 234-
240.
[4] D. Li, M. Tamura, Y. Nakagawa, K. Tomishige, Bioresour. Technol. 178 (2015) 53-64.
[5] I. Zamboni, C. Courson, A. Kiennemann, Appl. Catal. B: Environ. 203 (2017) 154-165.
[6] U. Oemar, M. L. Ang, W. F. K. Hijat, S. Kawi, Appl. Catal. B: Environ. 148-149 (2014) 231-
242.
[7] D. Li, M. Koike, J. Chen, Y. Nakagawa, K. Tomishige, Int. J. Hydrogen Energy 39 (2014)
10959-10970.
[8] W. Nabgan, T. A. T. Abdullah, R. Mat, B. Nabgan, Y. Gambo, S. Triwahyono, Int. J. Hydrogen
Energy (2016) 22922-22931.
[9] K. Polychronopoulou, J. L. G. Fierro, A. M. Efstashiou, J. Catal. 228 (2004) 417-432.
[10] N. Kaisalo, P. Simell, J. Lehtonen, Fuel 182 (2016) 696-703.
[11] J. Ashok, S. Kawi, Appl. Catal. A: Gen. 490 (2015) 24-35.
[12] J. Ashok, S. Kawi, ACS Catal. 4 (2014) 289-301.
[13] U. Oemar, M. L. Ang, Y. C. Chin, K. Hidajat, S. Kawi, Catal. Sci. Technol. 5 (2015) 3585-
3597.
[14] J. Ashok, Y. Kathiraser, M. L. Ang, S. Kawi, Catal. Sci. Technol. 5 (2015) 4398-4409.
[15] U. Oemar, P. S. Ang, K. Hidajat, S. Kawi, Int. J. Hydrogen Energy 38 (2013) 5525-5534.
[16] J. Ashok, S. Kawi, Int. J. Hydrogen Energy 38 (2013) 13938-13949.
[17] J. Ashok, Y. Kathiraser, M. L. Ang, S. Kawi, Appl. Catal. B: Environ. 172-173 (2015) 116-
128.
[18] U. Oemar, M. L. Ang, K. Hidajat, S. Kawi, RSC Adv. 5 (2015) 17834-17842.
[19] U. Oemar, Y. Kathiraser, M. L. Ang, K. Hidajat, S. Kawi, ChemCatChem 7 (2015) 3376-3385.
[20] C. Anjaneyulu, L. O. Costa, M. C. Ribeiro, R. C. Rabelo-Neto, L. V. Mattos, A. Venugopal, F.
B. Noronha, Appl. Catal. A: Gen. 519 (2016) 85-98.
[21] D. L. Trimm, Catal. Today 49 (1999) 3-10.
[22] J. R. Rostrup-Nielsen, J. Catal. 85 (1984) 31-43.
[23] L. Zhao, T. Han, H. Wang, L. Zhang, Y. Liu, Appl. Catal. B: Environ. 187 (2016) 19-29.
[24] J. R. Rostrup-Nielsen, J. Catal. 33 (1974) 184-201.
10

[25] A. L. A. Marinho, R. C. Rabelo-Neto, F. B. Noronha, L. V. Mattos, Appl. Catal. A: Gen. 520
(2016) 53-64.
[26] S. Hu, L. He, Y. Wang, S. Su, L. Jiang, Q. Chen, Q. Liu, H. Chi, J. Xiang, L. Sun, Int. J.
Hydrogen Energy 41 (2016) 17967-17975.
[27] S. Das, J. Ashok, Z. Bian, N. Dewangan, M. H. Wai, Y. Du, A. Borgna, K. Hidajat, S. Kawi,
Appl. Catal. B: Environ. 230 (2018) 220-236.
[28] F. B. Noronha, A. Shamsi, C. Taylor, E. C. Fendley, S. Stagg-Williams, D. E. Resasco, Catal.
Lett. 90 (2003) 13-21.
[29] V. Shanmugam, R. Zapf, S. Neuberg, V. Hessel, G. Kolb, Appl. Catal. B: Environ. 203 (2017)
859-869.
[30] U. Oemar, A. M. Li, K. Hidajat, S. Kawi, AIChE J. 60 (2014) 4190-4198.
[31] Z. Xiao, L. Li, C. Wu, G. Li, G. Liu, L. Wang, Catal. Lett. 146 (2016) 1780-1791.
[32] A. Arslan, T. Dogu, Int. J. Hydrogen Energy 38 (2016) 16752-16761.
[33] K. Sutthiumporn, T. Maneerung, Y. Kathiraser, S. Kawi, Int. J. Hydrogen Energy 37 (2012)
11195-11207.
[34] Y. Kathiraser, Z. Wang, S. Kawi, Environ. Sci. Technol. 47 (2013) 14510-14517.
[35] Z. Wang, U. Oemar, M. L. Ang, S. Kawi, J. Membrane Sci. 510 (2016) 417-425.
[36] Y. Sekine, D. Mukai, Y. Murai, S. Tochiya, Y. Izutsu, K. Sekiguchi, N. Hosomura, H. Arai, E.
Kikuchi, Y. Sugiura, Appl. Catal. A: Gen. 451 (2013) 160–167.
[37] D. Mukai, S. Tochiya, Y. Murai, M. Imori, T. Hashimoto, Y. Sugiura, Y. Sekine, Appl. Catal.
A: Gen. 453 (2013) 60–70.
[38] D. Mukai, Y. Murai, T. Higo, S. Tochiya, T. Hashimoto, Y. Sugiura, Yasushi Sekine, Appl.
Catal. A: Gen. 466 (2013) 190–197.
[39] K. Takise, M. Imori, D. Mukai, S. Ogo, Y. Sugiura, Y. Sekine, Appl. Catal. A: Gen. 489 (2015)
155–161.
[40] K. Takise, T. Higo, D. Mukai, S. Ogo, Y. Sugiura, Y. Sekine, Catal. Today 265 (2016) 111-
117.
[41] K. Takise, S. Manabe, K. Muraguchi, T. Higo, S. Ogo, Y. Sekine, Appl. Catal. A: Gen. 538
(2017) 181-189.
[42] M. Ortiz, A. Abad, L. F. Diego, F. Garcia-Labiano, P. Gayan, J. Adanez, Int. J.
Hydrogen Energy 36 (2011) 9663-9672.
[43] Y. Yan, Y. Cui, L. Zhang, L. Li, J. Zhang, Y. Chen, Q. Tang, C. Liu, Int. J. Hydrogen Energy
41 (2016) 13069-13076.
[44] W. Xie, J. yang, Q. Wang, L. Huang, N. Wang, Catal. Sci. Technol. 8 (2018) 3015-3024.
[45] S. Czernik, R. French, Int. J. Hydrogen Energy 39 (2014) 744-750.
[46] Y. Sekine, M. Haraguchi, M. Tomioka, M. Matsukata, E. Kikuchi, J. Phys. Chem. A, 114
(2010) 3824–3833.
[47] Y. Sekine, M. Haraguchi, M. Matsukata, E. Kikuchi, Catal. Today 171 (2011) 116-125.
11

[48] K. Oshima, T. Shinagawa, M. Haraguchi, Y. Sekine, Int. J. Hydrogen Energy 38 (2013) 3003-
3011.
[49] R. Inagaki, R. Manabe, Y. Hisai, Y. Kamite, T. Yabe, S. Ogo, Y. Sekine, Int. J. Hydrogen
Energy 43 (2018) 14310-14318.
[50] R. Manabe, S. Okada, R. Inagaki, K. Oshima, S. Ogo, Y. Sekine, Sci. Rep. 6 (2016) 38007.
[51] S. Okada, R. Manabe, R. Inagaki, S. Ogo, Y. Sekine, Catal. Today 307 (2018) 272-276.
[53] C. G. Vayenas, S. Bebelis, I. V. Yentekakis, H.-G. Lintz, Catal. Today 6 (1992) 303-438.
[54] I. V. Yentekakis, G. Moggridge, C. G. Vayenas, R. M. Lambert, J. Catal. 146 (1994) 292-305.
[55] A. D. Frantzis, S. Bebelis, C. G. Vayenas, Solid State Ionics 136-137 (2000) 863-872.
[56] D. L. Trimm, Catal. Today 37 (1997) 233-238.
[57] C. H. Bartholomew, Appl. Catal. A: Gen. 212 (2001) 17-60.
12

Captions to figures
Figure 1
Toluene conversion and hydrogen yield during toluene ER over metal-supported LSAO catalyst:
Ni/LSAO, Co/LSAO, Fe/LSAO, or Cu/LSAO. Metal loading amount: 5 wt%. Reaction conditions:
473 K reaction temperature; S/C = 2.0; W/F = 3.14 g-cat h mol-1; 3 mA current charged for filled
plots; no current charged for open plots.
Figure 2
Temperature dependence of catalytic activity on 5 wt% Ni/LSAO with or without the electric field.
Filled plots for ER and open plots for SR. Reaction conditions: 473–873 K reaction temperature;
S/C = 2.0; W/F = 3.14 g-cat h mol-1; 3 mA current charged.
Figure 3
Toluene conversion and hydrogen yield on Ni/LSAO catalyst, 1, 3 and 5 wt% Ni/LSAO, (A) for
ER at 473 K and (B) for SR at 698 K. Reaction conditions: 473 K reaction temperature; S/C = 2.0;
W/F = 3.14 g-cat h mol-1; 3 mA current applied for ER.
Figure 4
Arrhenius plots of 5 wt% Ni/LSAO catalyst for ER and SR. Reaction conditions: S/C = 2.0; W/F =
3.14 g-cat h mol-1; 3 mA current charged for ER.
Figure 5
Partial pressure dependences of (a) toluene and (b) steam on 5 wt% Ni/LSAO catalyst for ER at 473
K and SR at 698 K. Reaction conditions: S/C = 2.0; W/F = 3.14 g-cat h mol-1; 3 mA current charged
for ER.
Figure 6
Formation rate of 18O containing C1 products standing for lattice oxygen release on 5 wt% Ni/LSAO
in each condition: (a) 473 K for ER, (b) 698 K for SR. and (c) at 473 K for SR. Symbols: ■, C18O;
◆, C16O18O; ▲, C18O2; ○, H218O. Reaction conditions: S/C = 2.0; W/F = 3.14 g-cat h mol-1; 3 mA
current applied for ER.
Figure 7
Spectrum of TPD-IR measurement on 5 wt% Ni/LSAO (A) with the electric field charging (ER)
and (B) without the electric field (SR): 323–673 K temperature region; 3 mA applied current for
ER.
13

60
50
40
30
20
10
0
45
40
35
30
25
20
15
10
5
Co
Cu
Ni
Fe
Ni heat
Ni
Fe
Ni heat
Co
Cu
0
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
Time on stream / min
Time on stream / min
Figure 1
14

100
80
60
40
20
0
ER
SR
400
500
600
700
800
900
Catalyst-bed temperature / K
Figure 2
15

60
50
40
30
20
10
0
45
40
35
30
25
20
15
10
5
1wt%Ni/LSAO
3wt%Ni/LSAO
5wt%Ni/LSAO
1wt%Ni/LSAO
3wt%Ni/LSAO
5wt%Ni/LSAO
(A)
(A)
0
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
Time on stream / min
Time on stream / min
60
45
40
35
30
25
20
15
10
5
1wt%Ni/LSAO
1wt%Ni/LSAO
3wt%Ni/LSAO
5wt%Ni/LSAO
(B)
(B)
3wt%Ni/LSAO
5wt%Ni/LSAO
50
40
30
20
10
0
0
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
Time on stream / min
Time on stream / min
Figure 3
16

2.5
2.0
1.5
1.0
0.5
0.0
-0.5
ER
SR
Ea = 26.1 kJ mol-1
Ea = 78.9 kJ mol-1
1.3
1.4
1.5
1.6
1.7
1.8
1/T / 10-3 K-1
Figure 4
17

2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
2.1
2.0
1.9
1.8
1.7
1.6
1.5
1.4
(ER-a)
a = 0.95
(ER-b)
b = 0.29
-2.0
-1.8
-1.6
-1.4
-1.2
-5.4 -5.2 -5.0 -4.8 -4.6 -4.4 -4.2 -4.0
ln P (C H ) / -
ln P (H O) / -
2
7
8
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
2.1
2.0
1.9
1.8
1.7
1.6
1.5
1.4
(SR-a)
(SR-b)
a = 0.23
b = 0.93
-2.0
-1.8
-1.6
-1.4
-1.2
-5.4 -5.2 -5.0 -4.8 -4.6 -4.4 -4.2 -4.0
ln P (C H ) / -
ln P (H O) / -
2
7
8
Figure 5
18

2
1.5
1
2
1.5
1
2
1.5
1
(a)
(b)
(c)
0.5
0
0.5
0
0.5
0
0
20
40
60
80
100
0
20
40
60
80
100
0
20
40
60
80
100
Time on stream / sec
Time on stream / sec
Time on stream / sec
Figure 6
19

1480 cm-1
(A)
1400 cm-1
0.1
673 K
573 K
473 K
373 K
323 K
1600 cm-1
1500 cm-1
1800 1700 1600 1500 1400 1300 1200
Wavenumber / cm-1
1480 cm-1
(B)
0.1
873 K
773 K
673 K
573 K
473 K
1400 cm-1
373 K
323 K
1600 cm-1
1500 cm-1
1800 1700 1600 1500 1400 1300 1200
Wavenumber / cm-1
Figure 7
20

Table 1 Catalytic activity and amount of carbon deposition during steam reforming on Ni/LSAO catalysts.
Catalyst
Electric Field at 473 K
Heated at 698 K
carbon
deposition
/ mg g-cat-1
Toluene conv.
H2 yield
/ %
Toluene conv.
/ %
H2 yield
carbon deposition
/ %
/ %
/ mg g-cat-1
39.2
1wt%Ni/LSAO
3wt%Ni/LSAO
5wt%Ni/LSAO
10.9
29.1
44.7
14.0
26.7
33.3
7.5
9.1
10.0
7.2
25.3
42.9
5.5
26.3
40.0
322.8
525.8
Table 2 Ni structure and TOF determined by Ni specific surface area (TOF-s) and Ni-support perimeter (TOF-p)
Ni
diameter
particle
Ni-support
perimeter
Ni surface area
/ m2 g-1
TOFs / sec-1
TOFp / sec-1
Catalyst
/ nm
/ 108 m g-1
EF
heated
EF
heated
1wt%Ni/LSAO
3wt%Ni/LSAO
5wt%Ni/LSAO
8.9
9.5
9.9
0.56
1.58
2.55
1.27
3.32
5.16
0.41
0.39
0.37
0.27
0.34
0.36
9.29
9.44
9.34
6.15
8.20
8.95
Table 3 Release rates of lattice oxygen over Ni/LSAO in various conditions.
Olat release rate
Olat release amount
/ mol sec-1
/ mol
with EF (473 K)
without EF (698 K)
without EF (473 K)
1.77
0.59
0.03
42.9
14.3
0.8
21

22