Effects of Ru addition to Pd/Al2O3 catalysts on methanol steam reforming reaction: A mechanistic study

Article abstract of DOI:10.1016/j.apcata.2018.12.035

In this study, Pd/Al₂O₃ and Pd-Ru/Al₂O₃ catalysts were used to investigate the mechanism of the methanol steam reforming (MSR) reaction and the factors affecting the reaction activity. First, the MSR reaction activity was evaluated using Pd/Al₂O₃ and Pd-Ru/Al₂O₃ catalysts. The MSR reaction activities increased with the addition of Ru. In addition, FT-IR analysis was performed to investigate the MSR reaction mechanism. The reactions with the Pd/Al₂O₃ and Pd-Ru/Al₂O₃ catalysts proceeded through the same mechanism. After methanol was adsorbed on the catalyst surface as methoxy and formate species, it was converted into CO through decomposition and reaction with water, and then desorbed from the catalyst surface. TEM analysis revealed that the addition of Ru to the Pd/Al₂O₃ catalyst resulted in small particle sizes with highly dispersed Pd on the catalyst surface, which increased the conversion to CO. Lastly, analysis of adsorption characteristics showed that the addition of Ru enhanced the desorption rate of adsorbed CO species to promote the MSR reaction activity by weakening the adsorption intensity of CO. The desorption of CO adsorption species was the rate-determining step of the MSR reaction.

Full text of DOI:10.1016/j.apcata.2018.12.035

AppliedCatalysisA,General572(2019)115–123
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Effects of Ru addition to Pd/Al2O3 catalysts on methanol steam reforming
reaction: A mechanistic study
Geo Jong Kima, Min Su Kima, Ji-Young Byunb, Sung Chang Honga,⁎
a Department of Environmental Energy Engineering, Graduate School of Kyonggi University, 94-6 San, Iui-dong, Youngtong-ku, Suwon-si, Gyeonggi-do 443-760, Republic
of Korea
b Center for Materials Architecturing, Korea Institute of Science and Technology, Seoul 136-791, Republic of Korea
A R T I C L E I N F O
A B S T R A C T
Keywords:
Carbon monoxide
Oxidation
Room temperature
Metallic Pd
Valence state
Particle size
In this study, Pd/Al2O3 and Pd-Ru/Al2O3 catalysts were used to investigate the mechanism of the methanol
steam reforming (MSR) reaction and the factors affecting the reaction activity. First, the MSR reaction activity
was evaluated using Pd/Al2O3 and Pd-Ru/Al2O3 catalysts. The MSR reaction activities increased with the ad-
dition of Ru. In addition, FT-IR analysis was performed to investigate the MSR reaction mechanism. The reac-
tions with the Pd/Al2O3 and Pd-Ru/Al2O3 catalysts proceeded through the same mechanism. After methanol was
adsorbed on the catalyst surface as methoxy and formate species, it was converted into CO through decom-
position and reaction with water, and then desorbed from the catalyst surface. TEM analysis revealed that the
addition of Ru to the Pd/Al2O3 catalyst resulted in small particle sizes with highly dispersed Pd on the catalyst
surface, which increased the conversion to CO. Lastly, analysis of adsorption characteristics showed that the
addition of Ru enhanced the desorption rate of adsorbed CO species to promote the MSR reaction activity by
weakening the adsorption intensity of CO. The desorption of CO adsorption species was the rate-determining
step of the MSR reaction.
1. Introduction
hydrogen production. In general, Cu-based catalysts are used to pro-
duce hydrogen from methanol because of their excellent activity and
Currently, there is an urgent need to find eco-friendly energy re-
sources to address the environmental pollution problems caused by the
use of fossil fuels. As an ideal fuel, there is increasing interest in ob-
taining energy from hydrogen because water is the only waste product.
However, owing to the intrinsic properties of hydrogen, it is not an
ideal energy medium. According to adsorption and decomposition
studies on catalyst surfaces, methanol can be considered a hydrogen
carrier, as it has a high content of hydrogen and can be easily decom-
posed without breaking CeC bonds [1–3]. Methanol has also attracted
attention as one of the most promising energy carriers owing to its high
power density and eco-friendly properties [4,5]. Hydrogen can be
produced directly from methanol via the following reactions: metha-
nolysis (CH3OH → CO + 2H2) and methanol steam reforming (MSR)
(CH3OH + H2O → CO2 + 3H2). As methanol is one of the most
common synthetic chemicals and can provide chemical information
about more complex carbon compounds, understanding the reaction
mechanisms and properties of methanol on the surface of catalysts has
attracted considerable attention.
selectivity [6–9]. However, there are several problems associated with
Cu-based catalysts, such as requiring a pretreatment process in the
hydrogen atmosphere for several hours and coagulation at tempera-
tures above 300 °C owing to low thermal stability [10–12]. For this
reason, Pd-based catalysts have recently been investigated to replace
Cu-based catalysts for the production of hydrogen from methanol
[10,13–15]. Previous studies have examined the mechanisms of me-
thanol degradation on metal surfaces and transition metal catalysts
[16,17]. The MSR reaction mechanism of Pd-based catalysts was first
proposed by Iwasa et al. [16], who reported that the intermediate
species formed on the catalyst surface varies depending on the metallic
Pd and Pd alloys present, with the reactions proceeding via different
reaction pathways. In addition, Easwar et al. [17] reported that the
methanol conversion rate was proportional to the H2 chemisorption
amount and that the rate-determining step (RDS) was dependent on Pd
when Pd/CeO2 and Pd/ZnO catalysts were compared. In addition, the
selectivity for CO2 depended on the reaction pathway, and the reaction
pathway was determined by the intermediate species adsorbed on the
catalyst surface. However, there are few studies on step experiments to
The methanolysis and MSR reactions are effective methods for
⁎ Corresponding author.
E-mail address: schong@kyonggi.ac.kr (S.C. Hong).
https://doi.org/10.1016/j.apcata.2018.12.035
Received 3 August 2018; Received in revised form 17 December 2018; Accepted 28 December 2018
Availableonline29December2018
0926-860X/©2018ElsevierB.V.Allrightsreserved.

G.J. Kim et al.
AppliedCatalysisA,General572(2019)115–123
clearly determine the RDS of methanol conversion and the adsorption
characteristics. In addition, various density functional theory studies on
methanol decomposition mechanisms have recently been reported
were recorded on a JEM-2100 F microscope (JEOL Co.) operating at
200 kV. Samples for the FE-TEM measurements were prepared by sus-
pending ultrasonically treated catalyst powder in ethanol and placing a
drop of the suspension on the Cu grid.
[18,19].
Jiang
et
al.
[19]
reported
that
the
CH3OH−CH2OH−CHOH−CHO−CO pathway was the most likely
dehydrogenation pathway, where the high energy barrier of CH3OH
dissociation was the RDS of the total dehydrogenation reaction. How-
ever, desorption from the catalyst surface is an important factor in the
catalytic reaction and may affect the overall reaction rate.
X-ray diffraction (XRD) analysis was performed using an X'Pert PRO
MRD diffractometer (PANalytical Co.) with a Cu Ka (λ = 1.5056 Å)
radiation source and the X-ray generator operating at 30 kW. XRD
patterns were measured at a scanning speed of 6°/min in the 2θ range of
10–90°.
The catalysts were confirmed in terms of their dispersion by CO
chemisorption at 25 °C. The catalyst sample, which was activated in a
10% H2/N2 gas flow at 250 °C for 30 min, was cooled to 25 °C and sa-
turated with pulses of 10% CO/N2 gas.
The temperature-programmed reaction analyses were performed
using an Autochem 2920 analyzer (Micromeritics). The temperature-
programmed reduction (TPR) of H2 was analyzed using 10% H2/N2 and
0.3 g of catalyst at a total flow rate of 50 cc/min. All catalysts were
pretreated by injecting 5% O2/He 50 cc/min at 300 °C and then cooling
to 60 °C. After, The catalysts were then treated with 10% H2/N2 at 60 °C
for 0.5 h. The catalyst was placed in dilute H2, and the consumption of
Thus, the purpose of this study was to identify the factors affecting
the RDS of the catalytic reaction mechanism and to examine the effects
of Ru addition on the MSR reaction characteristics by investigating the
MSR reaction mechanism on a Pd/Al2O3 catalyst. To achieve this, we
evaluated the MSR reaction characteristics, used FT-IR analysis to un-
derstand the reaction mechanisms, and applied adsorption analysis to
propose factors affecting the RDS. The results of this study revealed key
factors for the MSR and methanol decomposition reactions.
2. Experimental
H
2 was monitored using a TCD in the Autochem 2920 while increasing
2.1. Catalyst preparation
the temperature to 800 °C at a rate of 10 °C/min.
The temperature-programmed desorption (TPD) of CO and CH3OH
was performed with 0.3 g of catalyst at total flow of 50 cc/min. All
catalysts were pretreated before the TPD analysis. The catalysts were
then treated with 1% CO/Ar or 1500 ppm CH3OH/N2 at 60 °C for 0.5 h.
The absorbed CO or CH3OH was purged with Ar for 1 h before starting
the TPD analysis. During the TPD analysis, the quantities of CO (m/
e = 28), CO2 (m/e = 44), CH4 (m/e = 16), and H2 (m/e = 2) were
continuously monitored using a quadrupole mass spectrometer (QMS
422) while the temperature was increased to 800 °C at a rate of 10 °C/
min.
The temperature-programmed surface reaction (TPSR) analysis was
performed with 0.3 g of catalyst using a fixed bed reactor and the QMS.
Before the TPSR analysis, all the catalysts were pretreated in Ar at
300 °C for 0.5 h and then cooled to 100 °C. During injection of CH3OH,
H2O, and N2 at a ratio of 1:2.5:4 (vol%), the temperature was increased
up to 500 °C at a rate of 10 °C/min. The generated gases were monitored
using the QMS.
The Pd/Al2O3, Ru/Al2O3 and Pd-Ru/Al2O3 catalysts used in this
study were prepared by the wet-impregnation method. First, using Pd
(OH)2 (Aldrich Co.) as a precursor, 1 wt% Pd (based on the weight of
Al2O3) was dissolved in distilled water. Second, Al2O3 was slowly added
to the Pd solution and then stirred for 1 h. The mixed solution was then
evaporated using a rotary vacuum evaporator. Thereafter, the solid was
dried in an oven at 103 °C for 24 h. Finally, the dried sample was cal-
cined at 400 °C for 4 h and then reduced in 30% H2/N2 at 600 °C for 1 h
to prepare the Pd/Al2O3 catalyst.
To prepare the Pd-Ru/Al2O3 catalyst, 0.9 wt% Pd and 0.1 wt% Ru
(based on the weight of Al2O3) were dissolved in distilled water. RuNO
(NO3)3 (Alfa Aesar Co.) was used as the Ru precursor and the sub-
sequent process was the same as that used for the Pd/Al2O3 catalyst.
2.2. Catalytic activity measurement
The MSR activity tests were conducted at a steam to methanol (S/C)
ratio of 2.5 and a space velocity of 10,000 h−1 in a fixed bed reactor.
The feed gas steam mixture consisted of CH3OH, H2O, and N2 with a
CH3OH:H2O:N2 ratio of 1:2.5:4 (vol%). In addition, we injected
13.5 vol. % CH3OH in total gas. The experimental apparatus consisted
of a catalytic reactor, a digital bubble flow meter, and mass flow con-
trollers. The MSR reactor apparatus comprised a quartz tube (inner
diameter: 8 mm; height: 650 mm) and a catalytic bed filled with quartz
wool. The reactor was supplied with a CH3OH/steam mixture (1:2.5,
vol%) and a carrier gas, and their flow rates were controlled using a
micro liquid pump (JASCO, MINICHEMI PUMP). The inlet gas supply
pipe was made of stainless steel (size: 1/4′) and covered with a heating
band at 180 °C for preheating N2 gas. The methanol/stem mixture
supply pipe was made of stainless steel (size: 1/16′) and covered with a
quartz tube wound with a nichrome wire at 300 °C for vapor produc-
tion. The product gases were analyzed using a gas chromatograph
(6890 N Agilent Co.) equipped with 6 Å molecular sieve and Porapak-Q
columns and thermal conductivity detectors (TCDs).
In situ diffuse reflectance infrared spectroscopy (DRIFTS) analysis
was performed using a Nicolet iS10 spectrometer (Thermo Fisher)
equipped with a Diffuse Reflectance (DR) 400 accessory for solid re-
flectance analysis. A CaF2 window was used as a plate for the DR
measurements and the spectra were collected using a mercury cadmium
telluride detector. The sample was preprocessed with Ar at a flow rate
of 50 cc/min at 300 °C for 1 h. To collect the spectra of the catalysts, a
single-beam spectrum of the preprocessed sample was measured as a
background, and all analyses were performed via auto scanning at a
resolution of 8 cm−1
.
3. Results and discussion
3.1. Catalytic activity
In this study, we investigated changes in catalyst characteristics
when Ru was added to a Pd/Al2O3 catalyst and examined the effects of
the characteristics on the MSR reaction. Accordingly, we investigated
the MSR reaction activities of Pd/Al2O3, Ru/Al2O3, and Pd-Ru/Al2O3
catalysts at a space velocity of 10,000 h−1 and an S/C of 2.5, and the
results are shown in Fig. 1 (a)–(d). As shown in Fig. 1 (a), the methanol
conversion rate of the catalysts decreased in the following order: Pd-
Ru/Al2O3 > Pd/Al2O3 > Ru/Al2O3. In particular, the methanol con-
version rate of Ru/Al2O3 at 400 °C was 65%, and this catalyst showed
very low reaction activity. Thus, the MSR reaction characteristics of Ru
itself were low. In contrast, as the Pd/Al2O3 catalyst showed an ex-
cellent methanol conversion rate of 90% or more at 325 °C or
2.3. Catalyst characterization
The specific surface areas and pore sizes of the catalysts were in-
vestigated using an ASAP 2010C analyzer (Micromeritics). The specific
surface area was calculated using the Brunauer–Emmett–Teller (BET)
equation, and the average pore size was calculated using the Barrett-
Joyner-Halenda (BJH) method. Each sample was analyzed in a vacuum
state at 300 °C for 2 h after the gas was removed.
Field-emission transmission electron microscopy (FE-TEM) images
116

G.J. Kim et al.
AppliedCatalysisA,General572(2019)115–123
Fig. 1. Methanol steam reforming performance of Pd/Al2O3, Ru/Al2O3, and Pd-Ru/Al2O3 catalysts: (a) methanol conversion of catalysts, outlet gas concentration (b)
Pd/Al2O3, (c) Ru/Al2O3, and (d) Pd-Ru/Al2O3.
highermethanol. In addition, the Pd-Ru/Al2O3 catalyst, in which Ru
was added to the Pd/Al2O3 catalyst, exhibited a higher conversion rate
than the Pd/Al2O3 catalyst in all temperature ranges. As the Pd-Ru/
Al2O3 catalyst showed a 5% higher methanol conversion rate than the
Pd/Al2O3 catalyst at temperatures of 325 °C or lower, the reaction ac-
tivity was increased by the addition of Ru. In other words, as the MSR
reaction activity of the Ru/Al2O3 catalyst was very low, the Ru added to
the Pd/Al2O3 catalyst acted as a promoter rather than an active site.
Thus, the properties of the Pd active site and the reaction characteristics
were changed to promote the MSR reaction rate.
Fig. 1(b)–(d) shows the concentrations of the gases produced by the
MSR reaction with the Pd/Al2O3, Ru/Al2O3, and Pd-Ru/Al2O3 catalysts.
H2 and CO were produced as products with all catalysts, and trace
amounts of CH4 were detected at a temperature of 400 °C. As only CO
and H2 were produced, these results indicate that the decomposition
reaction proceeded according to Eq. (1). In addition, the production of
CO2 was not observed with the Pd/Al2O3 and Pd-Ru/Al2O3 catalysts
despite the injection of water, suggesting that the water gas shift re-
action (Eq. (2)) did not proceed. Finally, in comparison with the be-
havior at 350 °C, the concentration of H2 and CO decreased simulta-
neously with the production of trace amounts of CH4 at 400 °C. Meunier
et al. [19] reported that methanation occurred during the semi-hy-
drogenation reaction of acetylene on the Pd-Zn/CeO2 catalyst. In this
study, the observed production of CH4 together with the decreased
production of CO and H2 suggested that the methanation reaction (Eq.
(3)) occurred, in which produced CO was converted into CH4 through
reaction with H2.
CH3OH → CO + 2H2
CO + H2O → CO2 + H2
CO + 3H2 → CH4 + H2O
(1)
(2)
(3)
3.2. Characterization of Pd/Al2O3 and Pd-Ru/Al2O3 catalysts
FE-TEM images of the Pd/Al2O3 and Pd-Ru/Al2O3 catalysts are
shown in Fig. 2. The addition of Ru to the Pd/Al2O3 catalyst affected the
particle size. In the case of the Pd/Al2O3 catalyst, the average particle
size of the active particles was 5.5 nm. In contrast, the average particle
size of the active particles in the Pd-Ru/Al2O3 catalyst was 3.1 nm. The
active particle size of the catalyst with added Ru was smaller than that
of the Pd/Al2O3 catalyst. Huang et al. [20] compared two catalysts by
depositing Pd and PdRu on a mesoporous silica support and found that
the dispersion of Pd was increased by the formation of a Pd-Ru alloy
when Ru was added to Pd. Ma et al. [21] compared the catalysts ob-
tained by depositing Pd and PdRu on a carbon support. The active
particle diameter of Pd (3.4 nm) decreased to 2.8 nm when Ru was
added. In general, Pd is known to aggregate easily with heat treatment.
El Hawa et al. [22] reported that the formation of a Pd-Ru alloy could
improve the thermal stability of Pd. In this study, a Pd-Ru alloy was
formed when Ru was added and the thermal stability was increased to
suppress the aggregation of Pd and decrease the particle size of Pd.
117

G.J. Kim et al.
AppliedCatalysisA,General572(2019)115–123
Fig. 2. TEM images of (a) 1%Pd/Al2O3 and (b) 0.9%Pd-0.1%Ru/Al2O3 catalysts.
Table 1 shows the average particle sizes determined from the TEM
images and the basic physical properties of the catalysts. The surface
areas of the Pd/Al2O3 and Pd-Ru/Al2O3 catalysts calculated by BET
analysis were 139 and 132 m2/g, respectively. Thus, the surface areas of
these catalysts were similar, and there was no significant variation in
the pore sizes and pore volumes.
observed at 40°, 47°, and 68°, corresponding to metallic Pd [23,24]. The
intensities of the peaks attributed to metallic Pd were larger for the Pd/
Al2O3 catalyst than the Pd-Ru/Al2O3 catalyst. These differences were
caused by the formation of a Pd-Ru alloy when Ru was added to the Pd/
Al2O3 catalyst. These results were consistent with the results of the TEM
analysis.
Fig. 3 shows XRD patterns of the Pd/Al2O3, Pd-Ru/Al2O3, and Al2O3
catalysts. For all three samples, strong peaks were observed at 20°, 32°,
38°, 46°, and 66.8°, which correspond to r-Al2O3, the support. In con-
trast, for the Pd/Al2O3 and Pd-Ru/Al2O3 catalysts, peaks were also
A H2-TPR analysis was performed to evaluate the reducibility of the
catalyst, and the results are shown in Fig. 4. In the case of the Pd/Al2O3
catalyst, a decrease peak occurred owing to desorption of H2 adsorbed
on the surface of the catalyst at temperatures of 50–100 °C. Further, H2
Table 1
The physical properties of Pd/Al2O3 and Pd-Ru/Al2O3 catalysts.
Catalyst
Chemical compositiona
Active particle size (nm)
Metal dispersion (%)c
SBET
Total pore volume
(cm3/g)
Mean pore size
(nm)
(m2/g)
Pd (%)
Ru (%)
1%Pd/Al2O3
0.9%Pd-0.1%Ru/Al2O3
1.01
0.89
–
5.2b
3.5b
7.8c
5.7c
14.4
19.3
139
132
0.068
0.065
1.98
1.98
0.09
a
Determined by ICP-OES analysis; other values are nominal values.
Determined by TEM analysis.
Determined by CP chemisorption analysis.
b
c
118

G.J. Kim et al.
AppliedCatalysisA,General572(2019)115–123
hydrogen consumption by a spillover from Pd to the support material.
Also, they reported that high-temperature H2 consumption is due to
interactions between well-dispersed PdOx species. As a result of the
preceding CO-Chemisorption and TEM analyzes when Ru is added,
active metal is highly dispersed. For this reason, a larger amount of
water consumption has occurred by spillover in the Pd-Ru catalyst.
According to the H2-TPR analysis, the H2 desorption peaks occur-
ring at 50–100 °C occur owing to the desorption of H2 adsorbed on Pd
during the pretreatment process. Baylet et al. [26] observed the con-
sumption and emission of H2 during the H2-TPR analysis of a Pd/Al2O3
catalyst by heating and cooling twice in the range of 0–80 °C. They
reported that the emission of H2 occurred during heating at 68 °C and
absorption of H2 occurred during cooling. Pd absorbed H2 at low
temperatures to form a PdH phase (H2 + 2Pd ↔ 2PdH). The Pd-Ru/
Al2O3 catalyst showed H2 desorption peaks smaller than those of the
Pd/Al2O3 catalyst because the number of Pd sites that could absorb H2
was decreased by the formation of the Ru-Pd alloy. These findings were
consistent with the TEM and XRD results. In addition, the increase in
the H2 consumption peaks in the temperature ranges of 100–200 °C and
350–600 °C and the shift to low temperatures were caused by an in-
crease in reducibility owing to increased H2 consumption and the for-
mation of the Pd-Ru alloy occurring at the RuO site obtained by addi-
tion of Ru.
Fig. 3. XRD patterns of Al2O3, 1%Pd/Al2O3, and 0.9%Pd-0.1%Ru/Al2O3 cata-
lysts.
3.3. Mechanistic study of the MSR reaction using Pd/Al2O3 and Pd-Ru/
Al2O3 catalysts
Understanding the characteristics of reactions on catalyst surfaces
through mechanistic studies is important for identifying factors that
affect the RDS and promote the reaction activity. In this study, DRIFT
analysis was performed using FT-IR to investigate the characteristics of
the reactions on the surfaces of the Pd/Al2O3 and Pd-Ru/Al2O3 cata-
lysts.
All DRIFT analyses were performed after impurities were removed
from the catalyst surfaces by injecting Ar at 400 °C for 30 min before the
experiments. First, to investigate the adsorption characteristics of me-
thanol on the catalyst surfaces, changes in the surface of the Pd/Al2O3
catalyst were observed by injecting methanol at a temperature of 300 °C
for 20 min. Adsorption peaks caused by the injection of methanol have
been reported in many papers [14,17,27,28]. Ranganathan et al. [17]
performed an FT-IR analysis of Pd/ZnO and Pd/CeO2 catalysts and
reported that CeH peaks were observed at 2934, 2843, and 2818 cm−1
upon the injection of methanol. Haghofer et al. [14] performed an FT-
IR analysis to investigate the MSR characteristics of a Pd/Ga2O3 cata-
lyst. As the MSR reaction proceeded, a linear CO bond was observed to
be absorbed on Pd (2056 cm−1). Adsorbed CO species in a bridge-
bonded form (1968 cm−1
(1920 cm−1) were also observed.
)
and
a
hollow/bridge-bonded form
As shown in Fig. 5, peaks were observed at 2944, 2906, 2843, 2820,
2053, 1920, 1600, 1390, and 1374 cm−1 upon injection of methanol.
The peaks at 2944 and 2843 cm−1, corresponding to CeH stretching
(bridged), were caused by the adsorption of methoxy species and the
peak at 2906 cm−1 was caused by the adsorption of formate species.
The peak at 2820 cm−1 was also caused by the adsorption of methoxy
species [17]. The peaks observed at 2053 and 1920 cm−1 corresponded
to CO adsorption species linearly adsorbed on the surface of Pd
(2053 cm−1) and CO adsorption species in the hollow/bridge-bonded
form (1920 cm−1) [29]. Finally, the adsorption peaks at 1600, 1390,
and 1374 cm−1 were caused by formate species adsorbed in various
forms on the catalyst surface. Thus, CO adsorption species, methoxy
species in the form of CeH, and formate species were adsorbed on the
surface of the catalyst upon injection of methanol.
Fig. 4. Temperature-programed reduction of H2 over 1%Pd/Al2O3 and 0.9%Pd-
0.1%Ru/Al2O3 catalysts.
consumption peaks were observed at temperatures in the range of
100–200 °C and 400–600 °C. In contrast, in the case of the Pd-Ru/Al2O3
catalyst, decrease peaks, which were smaller than those observed for
Pd/Al2O3, were observed in the temperature range of 50–100 °C and a
H2 consumption peak, which was larger than that observed for Pd/
Al2O3, was observed in the temperature range of 100–200 °C. In addi-
tion, in the range of 350–600 °C, H2 consumption peaks were observed
at 450 and 500 °C, similar to those observed for Pd/Al2O3. The hy-
drogen consumption peak at 450 °C and 500 °C is the peak due to spill
over from Pd and Pd-Ru to the support material. Zheng at al. [25] re-
ported H2 consumption peaks at high temperatures (> 300 °C) were
observed for Pd/Al2O3 catalysts, and were either assigned to the
When the adsorption characteristics of the catalyst surface were
examined over time, formate species (peaks at 1600, 1390, and
1374 cm−1) were formed in the first 2 min after the injection of me-
thanol. In comparison, relatively few methoxy species (peaks at 2944,
119

G.J. Kim et al.
AppliedCatalysisA,General572(2019)115–123
Fig. 7. Dynamic changes in the in situ FT-IR spectra for H2O purging after
methanol adsorption on Pd/Al2O3 at 300 °C as a function of time (inlet gas: 3%
H2O mixed with Ar). [MeOH ads: (a) 20 min, H2O des: (b) 1 min, (c) 2 min, (d)
3 min, (e) 5 min, (f) 10 min, (g) 15 min, (h) 20 min].
Fig. 5. Dynamic changes in the in situ FT-IR spectra for methanol adsorption on
Pd/Al2O3 at 300 °C as a function of time (inlet gas: 1500 ppm methanol mixed
with N2). [MeOH ads: (a) 1 min, (b) 2 min, (c) 3 min, (d) 4 min, (e) 6 min, (f)
10 min, (g) 15 min, (h) 20 min].
According to the results in Fig. 6, the methoxy species peak dis-
appeared and then the CO adsorption species peak disappeared, which
indicated that the methoxy species was converted into an adsorbed CO
species that was then desorbed. In addition, as the peaks corresponding
to the formate species did not change, even following injection of Ar,
the mechanism for the methanol decomposition reaction was identified
to involve methanol adsorbed on the catalyst surface as the methoxy
species and only the methoxy species was converted into the CO ad-
sorption species. Finally, the CO adsorption species was desorbed from
the surface of the catalyst. According to the results shown in Fig. 5, the
peaks corresponding CO, which was the final stage of the reaction on
the catalyst surface, increased after 3 min and then did not change
further. In contrast, the peaks corresponding to the methoxy species
continued to increase up to 20 min. As the desorption rate of CO was
slower than the conversion rate of methoxy species to CO, the methoxy
species was not converted into CO and thus was accumulated on the
catalyst surface.
Fig. 7 shows the changes in the catalyst surface caused by injecting
H2O 20 min after methanol was injected and absorbed on the catalyst
surface. As a result, the peaks corresponding to the methoxy species at
2944, 2843, and 2820 cm−1 tended to decrease rapidly after water was
injected. In contrast, the peaks corresponding to the formate species at
2906, 1600, 1390, and 1374 cm−1 decreased sharply within 1 min of
the injection of water and then gradually decreased over 10 min. In
addition, the peaks corresponding to CO at 2053 and 1920 cm−1 de-
creased sharply within 1 min of the injection of water and then gra-
dually decreased over 10 min. Finally, the growth of a peak at 1640
cm−1 was observed 10 min after the injection of water. This peak
corresponded to water and was thought to appear due to the injection
of water [30,31].
According to the results shown in Fig. 6, the methoxy and CO ab-
sorption species were all desorbed within 3 min of injecting Ar and
were no longer observed on the catalyst surface. In contrast, the results
in Fig. 7 showed that the methoxy species was rapidly desorbed, but the
CO absorption peak was observed up to 10 min after the injection of
water. According to the results in Fig. 6, there was no change in the
peaks corresponding to the formate species following the injection of
Ar. However, the results in Fig. 7 shows that the peaks corresponding to
the formate species gradually decreased when water was injected. In
other words, the formate species reacted with the injected water to be
converted into CO, which was absorbed on the catalyst and then des-
orbed. In addition, it was found that the peak at 1640 cm−1
Fig. 6. Dynamic changes in the in situ FT-IR spectra for Ar purging after me-
thanol adsorption on Pd/Al2O3 at 300 °C as a function of time (inlet gas: Ar).
[MeOH ads: (a) 20 min, Ar des: (b) 1 min, (c) 2 min, (d) 3 min, (e) 5 min, (f)
10 min, (g) 15 min, (h) 20 min].
2843, and 2820 cm−1) and CO absorption species (peak at 1920 cm−1
)
were formed. Subsequently, peaks at 2053 and 1920 cm−1 corre-
sponding to CO absorption species were formed from 4 min after the
injection of methanol. In addition, the growth of methoxy species was
observed over time after 4 min, but no changes in the peaks of CO
absorption species and formate were observed. Fig. 6 shows the changes
in the catalyst surface 30 min after the injection of methanol gas was
stopped, with Ar injected 20 min after the injection of methanol. It was
found that the peaks corresponding to the methoxy species at 2944,
2843, and 2820 cm−1 disappeared rapidly within 1 min of injecting Ar.
In addition, the peaks corresponding to CO absorption species at 2053
and 1920 cm−1 disappeared completely within 3 min of injecting Ar,
indicating that this species was desorbed from the surface of the cata-
lyst. In contrast, the peaks at 2906, 1600, 1390, and 1374 cm−1 cor-
responding to the formate species did not change, even after Ar was
injected, which showed that the formate species was not desorbed from
the catalyst surface.
120

G.J. Kim et al.
AppliedCatalysisA,General572(2019)115–123
Fig. 8. Dynamic changes in the in situ FT-IR spectra for the methanol surface
reaction on Pd/Al2O3 and Pd-Ru/Al2O3 at 300 °C as a function of time (inlet
gas: 1500 ppm methanol, H2O mixed with N2).
corresponding to water was observed after 10 min because water was
adsorbed on the catalyst surface once all formate species that could
react were consumed. Thus, the formate species was found to react with
water. The FT-IR analysis revealed that methoxy and formate species
were formed as methanol intermediates and were converted into ad-
sorbed CO species, which were then desorbed from the catalyst surface.
It was also found that the rate of CO desorption from the catalyst sur-
face was slower than the rate of conversion from the intermediate
species to CO. These results confirmed that the RDS of the entire re-
action was the desorption of CO.
Fig. 9. Temperature-programed desorption of CO over 1%Pd/Al2O3 and 0.9%
Pd-0.1%Ru/Al2O3 catalysts.
3.4. TPD and TPSR analyses
Fig. 8 shows the changes in the catalyst surface under various
conditions, such as 20 min after the injection of methanol, 20 min after
desorption under Ar, and 20 min after injection of H2O for the Pd/Al2O3
and Pd-Ru/Al2O3 catalysts. The Pd-Ru/Al2O3 catalyst, which showed
excellent reaction activity, exhibited the same overall reaction pathway
as the Pd/Al2O3 catalyst. However, a difference between the Pd/Al2O3
catalyst and the Pd-Ru/Al2O3 catalyst was observed when methanol
was injected for 20 min. In the case of the Pd-Ru/Al2O3 catalyst, the
peaks at 2053 and 1920 cm−1 corresponding to the adsorption of CO
were larger than those for the Pd/Al2O3 catalyst, whereas the peaks at
2906, 2843, and 2820 cm−1 corresponding to the intermediate
methoxy species were smaller than those for Pd/Al2O3 catalyst. These
results indicated that the Pd particle size decreased with the addition of
Ru, as confirmed through FE-TEM analysis, and the number of Pd sites
of for conversion to and adsorption of CO increased. Thus, increased
adsorption of CO occurred on the surface of catalyst with added Ru. In
addition, the decrease in the methoxy species peaks may be caused by
two phenomena. First, the increase in the number of sites for conversion
to CO increased the conversion of methoxy species to CO, resulting in a
decrease in the methoxy species peaks. Second, the decrease could be
caused by an increase in the desorption rate of the CO adsorption
species from the catalyst surface, which was identified as the RDS of the
entire catalytic reaction. Fig. 6 shows that an increase in the methoxy
species peaks was caused by the slow desorption of CO adsorption
species in the final stage of the reaction. In other words, if the deso-
rption rate of the CO adsorption species increased, the size of the
methoxy species peaks could be decreased. In this study, TPD analyses
were conducted to investigate how the addition of Ru affected the
amount and intensity of CO adsorption, as described in the next section.
A CO-TPD analysis were performed to identify the effect of Ru ad-
dition on the adsorption characteristics of CO, and the results are shown
in Fig. 9. The Pd/Al2O3 catalyst showed CO desorption peaks at 100 and
285 °C, whereas the Pd-Ru/Al2O3 catalyst with added Ru showed CO
desorption peaks at 100, 220, and 260 °C. The CO peak at 100 °C cor-
responded to the desorption of linearly-bonded CO species adsorbed at
Pd and Pd-Ru sites, whereas the CO peaks at 220, 260, and 285 °C
corresponded to the desorption of Bridged CO species [32,33]. The
results in Fig. 9 shows that the CO adsorption amount of the Pd-Ru/
Al2O3 catalyst with added Ru was greater than that of the Pd/Al2O3
catalyst, and that the CO desorption peak area of the Pd-Ru/Al2O3
catalyst was 1.29 times larger than that of the Pd/Al2O3 catalyst. The
amount of CO adsorbed on the Pd-Ru/Al2O3 catalyst was greater than
that on the Pd/Al2O3 catalyst. As revealed in the TEM images, such an
increase was caused by the decrease in the Pd particle size caused by
the formation of a Pd-Ru alloy and the increase in the number of CO
adsorption sites caused by the high dispersion of Pd on the catalyst
surface. In addition, the CO-TPD results revealed that CO desorption
peaks of the Pd-Ru/Al2O3 catalyst with added Ru were observed at
lower temperatures than those of the Pd/Al2O3 catalyst. Hartmann
et al. [34] reported that the CO adsorption intensity on Pd in which a
PdRu surface alloy was formed was weakened by approximately
30–40% compared with the CO adsorption intensity of CO on the bare
Pd surface. In addition, Liu et al. [35] reported that a PdRu/C catalyst
exhibited excellent activity for the formic acid electro-oxidation test
because the adsorption intensity of CO was decreased. In this study, as a
Pd-Ru alloy was formed by the addition of Ru, the desorption peak of
CO on the Pd-Ru/Al2O3 catalyst was observed at a lower temperature
than that on the Pd/Al2O3 catalyst, indicating that the CO adsorption
intensity was weakened. As indicated by the FT-IR results, the RDS of
the entire reaction was the desorption of CO adsorption species. Fig. 1
121

G.J. Kim et al.
AppliedCatalysisA,General572(2019)115–123
shows increases in the reaction activities with the addition of Ru. Such
increases were caused by the decrease in CO adsorption intensity owing
to the formation of the Pd-Ru alloy and the increase in CO adsorption
sites. These phenomena were confirmed by the results of the methanol-
TPD analysis shown in Fig. 10. The methanol-TPD analysis revealed the
formation of H2 (m/e = 2), CO (m/e = 28), CH4 (m/e = 16), and CO2
(m/e = 44). First, a CO2 peak was observed in the temperature range of
100–250 °C and this peak was larger for the Pd-Ru/Al2O3 catalyst than
for the Pd/Al2O3 catalyst at low temperatures. In addition, a CO2 peak
was observed in the temperature range of 300–500 °C. These results
were highly consistent with the hydrogen consumption peaks caused by
oxygen species present in the catalyst observed in the H2-TPR analysis
(Fig. 4). In other words, it was found that the CO2 peak was generated
by the oxidation reaction between oxygen in the catalyst and methanol
or intermediate species absorbed on the catalyst surface.
In terms of the H2 and CO results, in the case of the Pd-Ru/Al2O3
catalyst, the H2 and CO desorption peaks were observed at 218 °C. In
the case of the Pd/Al2O3 catalyst, the H2 and CO desorption peaks were
observed at 280 °C. In addition, it was found that the CO desorption
peak temperature observed in the CO-TPD analysis was similar to the
desorption peak temperatures of H2 and CO observed in the methanol-
TPD analysis for both catalysts. As shown in Fig. 10, H2 was generated
in the process of converting the methoxy species into CO. In other
words, as the CO adsorption species was desorbed from the catalyst
surface, the adsorbed methoxy species was converted into CO, H2 then
it is concluded that desorbed from the catalyst surface. The MSR re-
action was thought to proceed from the moment when the CO ad-
sorption species adsorbed on the catalyst surface was desorbed. The
formation of the Pd-Ru alloy with the addition of Ru weakened the
adsorption intensity of the CO adsorption species, thus promoting the
MSR reaction activity.
Lastly, Fig. 11 shows the results of the TPSR analysis. The TPSR
analysis revealed that the MSR reaction proceeded from 180 °C because
H2 and CO were generated at 180 °C with the Pd-Ru/Al2O3 catalyst.
However, as the Pd/Al2O3 catalyst only generated H2 and CO from
Fig. 10. Temperature-programed desorption of methanol over 1%Pd/Al2O3 and
0.9%Pd-0.1%Ru/Al2O3 catalysts.
Fig. 11. Temperature-programed methanol steam reforming over 1%Pd/Al2O3 and 0.9%Pd-0.1%Ru/Al2O3 catalysts.
122

G.J. Kim et al.
AppliedCatalysisA,General572(2019)115–123
Fig. 12. Reaction scheme of the methanol steam reforming over Pd-based catalysts.
220 °C, it was found that the MSR reaction proceeded at a higher
temperature. As shown by the results of the methanol-TPD analysis
(Fig. 10), the Pd-Ru/Al2O3 catalyst generated H2 and CO from 150 °C,
indicating that the MSR reaction proceeded from when the desorption
of the CO adsorption species began. According to the results of the
methanol-TPD and TPSR analyses, in the case of the Pd/Al2O3 catalyst,
the generation of H2 was observed from the same temperature. Fig. 12
shows a reaction scheme based on the results obtained in this study. In
the MSR reaction, injected methanol was adsorbed on the catalyst
surface as methoxy and formate species, and then converted through
decomposition and reaction with H2O into CO adsorption species,
which were subsequently desorbed from the catalyst surface. In addi-
tion, the RDS of the entire reaction was found to be the desorption of
CO adsorption species.
[3] L.J. Drake, K.L. Fujdala, A.T. Bell, T.D. Tilley, J. Catal. 230 (2005) 14–27.
[4] J. Agrell, H. Birgersson, M. Boutonnet, I. Melian-Cabreara, R.M. Navarro,
J.L.G. Fierro, J. Catal. 219 (2003) 389–403.
[5] M. Momirlan, T.N. Veziroglu, Renew. Sustain. Energy Rev. 6 (2002) 141–179.
[6] J. Agrell, M. Boutonnet, J.L. Fierro, Appl. Catal. A Gen. 253 (2003) 201–211.
[7] J. Agrell, M. Boutonnet, J.L. Fierro, Appl. Catal. A Gen. 253 (2003) 213–222.
[8] W. Tong, A. West, K. Cheung, K.M. Yu, S.C.E. Tsang, ACS Catal. 3 (2013)
1231–1244.
[9] D. Das, J. Llorca, M. Dominguez, S. Colussi, A. Trobarelli, A. Gayen, Int. J.
Hydrogen Energy 40 (2015) 10463–10479.
[10] M. Kusche, F. Enzenberger, S. Bajus, H. Niedermeyer, A. Bosmann, A. Kaftan,
M. Laurin, J. Libuda, P. Wasserscheid, Angew. Chem. Int. 52 (2013) 5028–5032 ed..
[11] M. Kusche, F. Agel, N.N. Bhriain, A. Kaftan, M. Laurin, J. Libuda, P. Wasserscheid,
Chem. Sus. Chem. 7 (2014) 2516–2526.
[12] D.R. Palo, Chem. Rev. 107 (2007) 3992–4021.
[13] J. Kuc, M. Neumann, M. Armbruster, S. Yoon, Y. Zhang, R. Erni, A. Weidenkaff,
S.K. Matam, Catal. Sci. Technol. 6 (2016) 1455–1468.
[14] A. Haghofer, K. Fottinger, F. Girgsdies, D. Teschner, A. Knop-Gericke, R. Schlogl,
G. Rupprechter, J. Catal. 286 (2012) 13–21.
[15] K. Sun, W. Lu, M. Wang, X. Xu, Appl. Catal. A Gen. 268 (2004) 107–113.
[16] N. Takezawa, N. Iwasa, Catal. Today 36 (1997) 45–56.
[17] E.S. Ranganathan, S.K. Bej, L.T. Thompson, Appl. Catal. A Gen. 289 (2005)
153–162.
4. Conclusions
In this study, the MSR reaction activities of Pd/Al2O3 and Pd-Ru/
Al2O3 catalysts were evaluated. The MSR reaction activity was in-
creased by the addition of Ru. FT-IR analysis revealed that methanol
was adsorbed on the surfaces of the Pd/Al2O3 and Pd-Ru/Al2O3 cata-
lysts as methoxy and formate species and then converted through de-
composition and reaction with water into CO, which was desorbed from
the catalyst surface. It was found that the RDS of the entire reaction was
the desorption of the CO adsorption species. In addition, the TEM
analysis showed that the addition of Ru to the Pd/Al2O3 catalyst re-
sulted in smaller Pd particle sizes owing to the formation of a Pd-Ru
alloy, which was highly dispersed on the surface of the catalyst. Thus,
the CO adsorption amount was increased. The TPD analyses revealed
that the formation of the Pd-Ru alloy increased the desorption rate of
CO adsorption species. As the RDS of the MSR reaction was the deso-
rption of CO, the decrease in the CO adsorption intensity caused by the
formation of the Pd-Ru alloy enhanced the MSR reaction activity.
[18] Z.C. Kramer, X.K. Gu, D.D.Y. Zhou, W.X. Li, R.T. Skodje, J. Phys. Chem. C 118
(2014) 12364–12383.
[19] F. Meunier, M. Maffre, Y. Schuurman, S. Colussi, A. Trovarelli, Catalan J. Commun.
Cult. Stud. 105 (2018) 52–55.
[20] C. Huang, X. Yang, H. Yang, P. Huang, H. Song, S. Liao, Appl. Surf. Sci. 315 (2014)
138–143.
[21] L. Ma, H. He, A. Hsu, R. Chen, J. Power Sources 241 (2013) 696–702.
[22] H.W.A.E. Hawa, S.N. Pagileri, C.C. Morris, A. Harale, J. Douglas Way, J. Membr.
Sci. 466 (2014) 151–160.
[23] F. Braun, A.M. Tarditi, L.M. Cornaglia, J. Membr. Sci. 382 (2011) 252–261.
[24] S.K. Gade, M.K. Keeling, A.P. Davidson, O. Hatlevik, J. Douglas Way, Int. J.
Hydrogen Energy 34 (2009) 6484–6491.
[25] Q. Zheng, R. Farrauto, M. Deeba, Catalysts 5 (2015) 1797–1814.
[26] A. Baylet, P. Marecot, D. Duprez, P. Castellazzi, G. Groppi, P. Forzatti, Phys. Chem.
Chem. Phys. 13 (2011) 4607–4613.
[27] J. Toyir, P.R. de la Piscina, N. Homs, Int. J. Hydrogen Energy 40 (2015)
11261–11266.
[28] M.A. Larrubia Vargas, G. Busca, U. Costantino, F. Marmottini, P. Patrono, F. Pinzari,
G. Ramis, J. Mol. Catal. A Chem. 266 (2007) 188–197.
[29] N.M. Martin, M. Skoglundh, G. Smedler, A. Raj, D. Thompsett, P. Velin,
F.J. Martinez-Casado, Z. Matej, O. Balmes, P. Carlsson, J. Phys. Chem. C 121 (2017)
26321–26329.
Acknowledgements
[30] C. Busca, U. Costantino, F. Marmottini, T. Montanari, P. Patrono, F. Pinzari,
G. Ramis, Appl. Catal. A Gen. 310 (2006) 70–78.
[31] M.A. Larrubia Vargas, G. Busca, U. Costantino, F. Marmottini, T. Montanari,
P. Patrono, F. Pinzari, G. Ramis, J. Mol. Catal. A Chem. 266 (2007) 188–197.
[32] C. Dai, Y. Li, C. Ning, W. Zhang, X. Wang, C. Zhang, Appl. Catal. A:Gen. 545 (2017)
97–103.
This work was supported by the KIST Institutional Program (Project
No. 2E28020)
[33] O. Dulaurent, K. Chandes, C. Bouly, D. Bianchi, J. Catal. 188 (1999) 237–251.
[34] H. Hartmann, T. Diemant, J. Bansmann, R.J. Behm, Surf. Sci. 603 (2009)
1456–1466.
References
[1] Z. Jiang, B. Wang, T. Fang, Appl. Surf. Sci. 364 (2016) 613–619.
[2] G.W. Huber, J.W. Shabaker, J.A. Dumesic, Science 300 (2003) 2075–2077.
[35] Z. Liu, X. Zhang, S.W. Tay, J. Solid State Electrochem. 16 (2012) 545–550.
123