Enhancing the methane steam reforming catalytic performance of Ni monolithic catalysts: Via Ni-Re surface alloying

Article abstract of DOI:10.1039/c9cy02539a

High cell density Ni monolithic catalysts coated with a series of Ni-Re bimetallic surface layers (Re 0-38 at%) were synthesized and examined for application in methane steam reforming (MSR) under a steam-To-carbon ratio of 1.36 and a gas hourly space vel

Full text of DOI:10.1039/c9cy02539a

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Enhancing the methane steam reforming catalytic
performance of Ni monolithic catalysts via Ni–Re
surface alloying†
Cite this: DOI: 10.1039/c9cy02539a
a Toshiyuki Hirano,a Hirotaka Kunieda,b Yuta Harab and Yasushi Miyatac
*
Ya Xu,
High cell density Ni monolithic catalysts coated with a series of Ni–Re bimetallic surface layers (Re 0–38
at%) were synthesized and examined for application in methane steam reforming (MSR) under a steam-to-
carbon ratio of 1.36 and a gas hourly space velocity of 6400 h−1 in the temperature range of 773–1173 K.
The influence of mass and heat transport was verified to be minor below 973 K under the present
experimental conditions. The catalytic activities of the Ni monolithic catalysts were effectively improved in
the presence of a Ni–Re bimetallic surface layer coating, and the stability in terms of the catalyst activity
was significantly enhanced below 973 K upon increasing the Re content to >24 at%. The Ni–Re surface
layer was analyzed by using in situ X-ray absorption fine structure spectroscopy, synchrotron radiation
X-ray diffraction, scanning electron microscopy, and density functional theory calculations. These results
revealed that a Ni–Re alloy consisting of a Ni(Re) solid solution phase and a Re phase was formed during
hydrogen pre-reduction and the subsequent MSR reaction. The Re atoms in the Ni–Re alloy promoted
hydrogen adsorption on the surface, thereby suppressing the oxidation of the surface Ni atoms during
MSR, and enhancing the stability of the monolithic catalysts.
Received 16th December 2019,
Accepted 8th March 2020
DOI: 10.1039/c9cy02539a
rsc.li/catalysis
catalyst exhibited a good catalytic activity for MSR, and carbon
deposition was barely detected at 1073 K over 8000 h under
1. Introduction
Hydrogen is becoming increasingly important as a clean
energy carrier as it can be efficiently used to produce
electricity in fuel cells without the formation of substances
harmful to the environment.1–4 The main method for
hydrogen production is hydrocarbon steam reforming,
including methane steam reforming (MSR),5 for which porous
oxide-supported Ni-based pelleted catalysts have been widely
used on an industrial scale.6,7 However, their use is limited in
the context of fuel cell and small scale on-site hydrogen
production applications, as rapid heat and mass transfer
abilities are required.2,8,9 In contrast, metallic monolithic or
honeycomb catalysts have several attractive features, such as
rapid heat and mass transport and a low pressure drop per
catalyst volume, thereby rendering them suitable for fuel cell
applications.8,10,11
low steam-to-carbon ratio (S/C = 1.34) and gas hourly space
velocity (GHSV = 335 h−1) conditions. The catalytic activities of
this monolithic catalyst were further improved by increasing
the cell density of the monolith and by performing steam
pretreatment at an appropriate temperature prior to carrying
out the reaction.13,14 Our results therefore demonstrated the
potential of Ni monolithic catalysts for application in small-
scale hydrogen production facilities. However, the catalytic
performance at temperatures <1073
K still remained
insufficient, and so further studies were required.
It has been reported that the formation of an alloy/bimetallic
catalyst is often an efficient means to enhance the catalytic
activity of oxide-supported pelleted catalysts.15,16 For example,
the addition of a small quantity of noble metals, such as Pt, Pd,
or Rh, to the Ni catalyst can significantly improve its catalytic
performance for methane reforming, due to an enhancement in
the catalyst reducibility via alloying and/or the hydrogen
spillover effect.17–19 In addition to the noble metals, it was
reported that Ni–Re bimetallic catalysts supported on ZSM-5
zeolites or oxides enhanced the catalytic performance in the
steam reforming of gasoline,20,21 hydrogenation of carboxylic
acids and esters,22 selective deoxygenation of m-cresol to
toluene,23 reductive amination of monoethanolamine,24,25 and
methane dry reforming.26,27 These results suggest the possibility
of enhancing the catalytic activity in hydrogen production
We previously reported the development of a metallic
monolithic catalyst for MSR using a pure Ni thin foil.12 This
a National Institute for Materials Science, 1-2-1 Sengen, Tsukuba, Ibaraki 305-
0047, Japan. E-mail: xu.ya@nims.go.jp
b Hiroshima Co. Ltd., 41 Torashinden, Odaka, Midori, Nagoya 459-8001, Japan
c Nagoya Municipal Industrial Research Institute, 4-41, 6-3 Atsuta, Nagoya 456-
0058, Japan
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c9cy02539a
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
reactions through the preparation of Ni–Re bimetallic catalysts.
However, to the best of our knowledge, no reports exist on the
catalytic performance of Ni–Re bimetallic catalysts for MSR.
Furthermore, very few studies have reported the detailed
characterization of Ni–Re bimetallic catalysts, and the role of Re
in the enhancement of the catalytic performance remains
unclear.
Thus, we herein aim to enhance the catalytic performance
of Ni monolithic catalysts in MSR by the formation of an
active Ni–Re alloy layer on the catalyst surface, and elucidate
the role of Re in the catalytic performance of Ni–Re
bimetallic catalysts. The Ni–Re alloy layer is synthesized
directly on the surface of Ni monoliths without the use of an
oxide support, which should maintain the high thermal
conductivity of the monolithic structure and also prevent the
peeling of the catalyst layer from the monolithic substrate.
The Ni–Re-coated monolithic catalysts are characterized by in
situ X-ray absorption fine structure spectroscopy (XAFS),
powder X-ray diffraction using a synchrotron radiation source
(SR-XRD), temperature-programmed hydrogen reduction
(TPR), scanning electron microscopy (SEM), and first-
principles calculations based on the density functional theory
(DFT) approach.
(STP) min−1) and N2 (5 mL (STP) min−1) (STP: standard
temperature and pressure, 273 K and 0.1 MPa; all gas
volumes are presented as STP values). Following this
reduction process, a N2 flow was introduced to flush the
remaining H2 from the system. Subsequently, methane (10
mL min−1) and steam (13.6 mL min−1) were introduced into
the reactor with a flow of N2 (30 mL min−1). The S/C was 1.36
and the GHSV was 6400 h−1. Two types of MSR tests were
then performed, namely stepwise heating and isothermal
tests. For the stepwise heating test, the reaction temperature
was increased in stages from 773 to 1173 K at intervals of 50
K, maintaining each temperature for 30 min, and the activity
was measured at the end of each stage. For the isothermal
tests, the reaction was carried out at 873 K for 20 h, followed
by the reaction at 973 K for 20 h using the same catalyst.
During the tests, the compositions of the outlet gases were
measured by two on-line gas chromatographs equipped with
thermal conductivity detectors (GC; GL Science, GC323). One
column packed with zeolites (Molecular Sieve 13X) was used
for separating H2, N2, CH4, and CO, and the other packed
with a porous polymer adsorbent (Porapak Q) was used to
separate CO2 and H2O. The total flow rates of the outlet gases
were measured using a flow meter (Mesa Laboratories,
Defender530). The mass balance of carbon with a deviation
less than 5% was obtained during the reactions. The space
2. Experimental
2.1 Preparation of the Ni monolithic catalysts bearing a Ni–
time yield of H2 (STY-H2) and selectivities toward CO (SCO
)
and CO2 (SCO ) were calculated as follows. Here the STY-H2 is
2
Re surface layer
defined as the amount of produced H2 per geometrical
surface area of the monolith in unit time (mol m−2 s−1).
A Ni monolithic structure with a cell density of 2300 cpsi was
prepared using pure flat and corrugated Ni foils of 30 μm
thickness. The monolithic structure measured
8 mm in
STY-H2 = FH /A
(1)
(2)
(3)
2
diameter and 10 mm in height, and the weight of a single
monolith was ∼0.56 g. Details of the fabrication process and
the geometry of the monolithic structure were given in our
previous work.13,14 A Ni–Re surface layer was coated on the
monolithic structure using the following sequential
impregnation method. After washing the monolithic structure
with dilute hydrochloric acid (3 vol%), impregnation was carried
out using an aqueous solution of NiIJNO3)2·6H2O (50 wt%) and
subsequent drying at 393 K for >6 h. Subsequently, the
monolith bearing the loaded surface NiIJNO3)2·6H2O was
impregnated with an aqueous solution of NH4ReO4 (10 wt%),
followed by drying at 393 K for >6 h. Finally, the monolithic
structure was calcined at 773 K for 4 h in air. The quantity of
loaded surface Ni was 5 wt% of the monolithic structure, and
the quantities of loaded surface Re were 0.5, 5, and 10 wt% (i.e.,
Ni5, Ni5Re0.5, Ni5Re5, and Ni5Re10, corresponding to the
surface atomic concentrations of Ni-0, 3, 24, 38 at% Re,
respectively). For comparison, a Re (5 wt%) catalyst supported
on Al2O3 powder (JRC-ALO-6, BET surface area 180 m2 g−1) was
also synthesized using the impregnation method.
SCO = FCO/(FCO + FCO ) × 100
2
SCO = FCO /(FCO + FCO ) × 100
2
2
2
where FCH , FH , FCO, and FCO are the outlet flow rates of
4
2
2
CH4, H2, CO, and CO2, respectively (mol s−1). A is the total
geometrical surface area in the monolithic structure (m2).
Details regarding evaluation of the catalytic performance are
described in our previous report.28 The turnover frequency
(TOF) (s−1) (defined as the number of H2 molecules formed
per second per Ni active site) was calculated for the
isothermal tests. The Ni active sites were determined based
on the surface area measurement of the reacted monoliths
using Kr or N2 adsorption and surface composition analysis
using X-ray photoelectron spectroscopy (XPS). Because it was
confirmed that the surface of the substrate Ni foil used in
the present study has a strong texture of (001)<100>,12 we
assumed the surface of the surface layer as Ni (100) for
simplicity, and Ni atoms serve as the active sites for the
calculation of TOF.
2.2 Catalytic reaction
Assuming that the monolithic catalyst is an ideal plug
flow reactor operated in the steady-state, and methane steam
reforming (MSR) (eqn (4)) and water gas shift reaction (WGS)
MSR tests were carried out over the monolithic catalyst in a
conventional fixed-bed flow reactor. Prior to the reaction, the
catalyst was reduced at 703 K for 1 h by a flow of H2 (30 mL
(eqn (5)) occur in the reactor, the reaction rates of MSR (rSR
)
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Catalysis Science & Technology
Paper
and WGS (rSF) (mol m−3 s−1) were calculated using the
where Vt is the total volume of the monolithic catalyst, FCH ,0
4
following expressions (eqn (6) and (7)).29,30
and FH O,0 are the inlet flow rates of CH4 and H2O,
2
respectively (mol s−1), and FCH ,m, FH O,m, FH ,m, FCO,m, and
4
2
2
CH4 + H2O ⇄ 3H2 + CO ΔH298 = 206 kJ mol−1
CO + H2O ⇄ H2 + CO2 ΔH298 = −41 kJ mol−1
(4)
(5)
FCO ,m are the measured rate of each component at the outlet.
2
The effects of mass and heat transfer on the reaction were
evaluated by estimating the difference between the bulk fluid
concentration of CH4 (CAb
, ) and the surface
mol m−3
ꢀ
ꢁ
ꢁ
concentration of CH4 (CAs, mol m−3) on the surface of the
monolithic channel using eqn (13), and the difference
between the temperature of the bulk fluid (Tb) and the
channel surface temperature (Ts) using eqn (14).32
3
PCOPH
KP;SR
2
rSR ¼ kSR PCH ·PH O
−
(6)
(7)
4
2
ꢀ
PCO ·PH
2
2
rS F ¼ kS F PCO·PH O
−
2
KP;S F
kmtam(CAb − CAs) = rSR
(13)
(14)
where kSR and kSF are the apparent rate constants of MSR
and WGS (mol m−3 s−1 atm−2), respectively, which are
temperature dependent, i.e. defined by the Arrhenius
equation. KP,SR and KP,SF are the equilibrium constants of
MSR and WGS, respectively. Pi is the partial pressure of
component i (atm), i = CH4, H2O, H2, CO, CO2. The rate of
each component based on the volume of monolithic catalyst
(V) can be expressed using the following equations (eqn (8)–
(12)) according to the mass balance of each component
during the tests.30,31
hpam(Ts − Tb) = (−rSR)(−ΔHSR
)
where kmt is the effective mass-transfer coefficient of CH4
(defined as kmt = k0mt/yfA, yfA reflects the influence of MSR on
the diffusion in the boundary film on the surface of channels
(the definition of yfA is shown in Table S1†), k0mt is the mass-
transfer coefficient of CH4, am is the geometric surface area
per volume of the monolith (m2 m−3), rSR is the reaction rate
of CH4 at the inlet of honeycomb catalysts, hp is the heat-
transfer coefficient of CH4, and ΔHSR is the reaction heat of
MSR. The mass-transfer coefficient k0mt and heat-transfer
coefficient hp of CH4 were estimated by calculating the
following dimensionless numbers: Sherwood number Sh (Sh
= k0mtdh/DAm), Reynolds number Re (Re = dhuρ/μ), Schmidt
number Sc (Sc = μ/ρDAm), Graetz number GZ (GZ = ReScdh/L),
and Prandtl number Pr (Pr = cpμ/λf).10,33–35 The definitions of
the involved parameters and the detailed results are shown
in the ESI.†
The Mears criterion was used to assess the mass and heat
transport limitations.36,37 Only the interphase transport
between the gas and coated layer was considered since the
coating layer is very thin (<1 μm) which was confirmed by
TEM observation of the cross-section of the monolith (not
shown in this paper), and the intraparticle and interparticle
transports in the coated layer were negligible. The details are
shown in the ESI.†
d FCH
dV
4
¼ −rSR
(8)
(9)
d F
dV
H2O
¼ −rSR − rS F
¼ 3rSR þ rS F
d FH
dV
2
(10)
(11)
(12)
d FCO
dV
¼ rSR − rS F
d FCO
dV
2
¼ rS F
As the methane conversions are generally larger than 5%
in this study, we treated the reactor as an integral reactor by
integrating the above equations using the Runge–Kutta
method, with the following boundary conditions. The kSR and
kSF were determined by minimizing the difference between
the calculated and the measured flow rates of each
component. A maximum difference of less than 6% was
obtained in the determination of kSR and kSF for all samples.
At V = 0 (corresponding to the entrance of the monolithic
structure),
2.3 Catalyst characterization
In situ XAFS experiments during hydrogen reduction were
carried out at the BL5S1 Beam-line of the Aichi Synchrotron
Radiation Center.38 The storage ring was operated at 1.2 GeV
with a high energy resolution (E/ΔE = 7000). The size of the
beam spot at the sample position was equal to 0.4 × 0.14
mm2. A reaction cell unit for the in situ XAFS measurements
allows heating to 1173 K with flows of H2 and N2. The flow
rates of H2 and N2 were individually controlled using digital
mass flow meters. Prior to introduction into the reaction cell,
the above species were mixed in an evaporator heated to 423
K. Two powder samples were collected from the surfaces of
the Ni5 and Ni5Re10 monolithic catalysts after calcination
for XFAS measurements. They were then ground and mixed
FCH = FCH ,0, FH O = FH O,0, FH = FCO = FCO = 0
4
4
2
2
2
2
At V = Vt (corresponding to the exit of the monolithic
structure),
F
= F
, F
= F , F = F
H2O,m H2
, FCO
H2,m
CH4
CH4,m H2O
= FCO,m, F
= F
CO2,m
CO2
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
with boron nitride powder, and finally molded into disk 3. Results and discussion
plates with a diameter of 7 mm and a thickness of 0.5 mm.
3.1 Catalytic properties of the Ni monolithic catalysts bearing
The Quick XAFS spectra were measured in transmission
mode during the hydrogen reduction process. The XAFS
spectra of pure NiO, Re, ReO2, ReO3, Re5O7 (Sigma-Aldrich
Co.), and Ni foil were measured at 298 K as references. In
addition, the XAFS spectrum of a Ni foil reference sample
was recorded simultaneously with each measurement for
energy calibration. The XAFS data were analyzed using
ATHENA and Artemis software.39
The crystal structures of the surface Ni–Re powder
samples before and after MSR were determined by SR-XRD
measurements at the NIMS Beam-line BL15XU of SPring-8.40
The wavelength of the incident X-ray beam was 0.06201 nm,
which corresponded to the Mo absorption K-edge of 20 000
eV. The XRD profiles were qualitatively analyzed by using the
PDXL2 software package (Rigaku Co., Japan).
The Brunauer–Emmett–Teller (BET) specific surface areas
of the monoliths before and after MSR were measured using
a surface area analyzer (Micromeritics, ASAP 2020). For the
monolithic structure before impregnation, Kr adsorption was
used due to the small surface area. After impregnation, N2
adsorption was used for sample analysis.
The surface morphology was analyzed using a scanning
electron microscope (SEM; JEOL, JSM-7000F) coupled with an
X-ray energy dispersive spectroscopy (EDS) system. The
surface chemical states and compositions of the monolithic
catalysts were measured using an X-ray photoelectron
a Ni–Re surface layer
The initial catalytic performances of the Ni monolithic catalysts
bearing a Ni–Re surface coating were firstly investigated at
various temperatures using stepwise heating tests. For
comparison, the catalytic performance of Re/Al2O3 as a catalyst
for MSR was also examined under the same conditions. Since
the BET surface area decreased significantly during the stepwise
tests (as shown in Table 1), it was difficult to estimate the
number of active sites at each temperature during the stepwise
tests. We use the space-time yield of hydrogen (STY-H2) to
evaluate the initial activities of the monolithic catalysts. Fig. 1
shows the STY-H2 of the monolithic catalysts during the
stepwise test. For the Ni monolithic catalyst without a surface
coating (denoted as Ni monolith in the figure), the values of
STY-H2 were much lower at temperatures below 973 K, although
they gradually increased with the temperature. In contrast, in
the presence of a surface coating, the STY-H2 increased
significantly over the entire temperature range. Furthermore,
the values of the STY-H2 increased with the increase of Re
addition among the surface coated samples. The amount of
increase was small below 973 K, and became larger at
temperatures above 973 K. In addition, the Re/Al2O3 catalyst
exhibited an extremely low activity below 1023 K, suggesting
that in the Ni–Re catalysts, Ni acts as the active site for MSR
rather than Re. The above results indicate that the initial
catalytic activity of the Ni monolithic catalysts was significantly
enhanced at all temperatures by the presence of a surface
coating. Furthermore, the addition of Re effectively increased
the initial activities of the monolithic catalysts, especially at
high temperatures above 973 K.
The time dependences of the catalytic performances of
these monolithic catalysts were then examined using
isothermal tests. More specifically, Fig. 2a shows the turnover
frequency (TOF) as a function of time on stream at 873 K. As
indicated, the TOF decreased slightly at the beginning, and
approached a constant value of ∼8 s−1 after 20 h for the
Ni5Re0.5, Ni5Re5, and Ni5Re10 samples, while it continually
decreased from 12.5 to 3.5 s−1 after 20 h for the Ni5 sample,
thereby indicating that the addition of Re enhanced the
catalytic stability at 873 K. After being reacted at 873 K for 20
h, the same sample was used to examine the catalytic
performance at 973 K. Fig. 2b shows the TOF as a function of
time on stream at 973 K. As shown, the TOFs of all the Ni–Re
samples were initially comparable (35–45 s−1), whereas that
of the Ni5 sample was significantly lower (∼9 s−1). The time
dependences of TOF changed significantly with the quantity
of added Re. For the Ni5 sample, the TOF decreased in the
initial few hours, prior to stabilizing at ∼4 s−1. For the
Ni5Re0.5 sample, the TOF decreased continuously from 36 to
6 s−1 over 20 h, while for the Ni5Re5 sample, it decreased
from 44 to 20 s−1, and for the Ni5Re10 sample, from 44 to 35
s−1 over the same period. These results indicate that
increasing the amount of Re addition effectively enhanced
spectrometer (XPS; ULVAC-PHI, Quantera SXM) with
a
monochromatic Al Kα (1486.7 eV) X-ray source.
The adsorption energies of hydrogen on the various sites
of the crystal surfaces of Ni and Re, i.e. Ni (111), Ni (011), Ni
(001), and Re (001), were calculated using the PHASE
program package (https://azuma.nims.go.jp) for performing
first-principles electronic calculations based on DFT. A slab
consisting of 63 atoms in seven atomic layers (each layer = 3
× 3 atoms) was used for the calculations of Ni (111) and Re
(001). The coordinates of the atoms in the bottom three
layers were fixed. A smaller slab consisting of 16 atoms in
four atomic layers (each layer = 2 × 2 atoms) was used for the
calculations of Ni (011) and Ni (001), and the coordinates of
the atoms in the bottom two layers were fixed in these cases.
A vacuum layer of 10 Å was employed for separation of the
slab. Brillouin-zone integration was performed using the
Monkhorst–Pack method41 with 4 × 4 × 1 k-point sampling
mesh. The cut-off energy for the wave function was 400 eV,
and the cut-off energy for the charge density was 1600 eV.
The adsorption energy of one hydrogen adsorbed on the slab
surface (Eads) was estimated as follows:
Eads = EH/slab − Eslab − 1/2EH
(15)
2
where EH/slab is the total energy of the slab surface with the
adsorbed hydrogen atom, Eslab is the total energy of the bare
slab surface, and EH is the total energy of the isolated
2
hydrogen molecule.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Catalysis Science & Technology
Paper
Table 1 The specific surface area of samples before and after reaction
BET specific surface areaa (m2 g−1
)
After calcination
(773 K 5 h)
After hydrogen
reduction (703 K 1 h)
After isothermal test
(873 K 20 h + 973 K 20 h)
After stepwise test
(773–1173 K, 0.5 h at each given temp.)
Samples
Ni5
0.7
2.7
2.7
3.0
0.113
0.224
0.634
1.161
0.084
0.115
0.157
0.137
0.011
0.024
0.040
0.067
Ni5Re0.5
Ni5Re5
Ni5Re10
a
The samples after calcination were measured using N2 adsorption, and the others were measured using Kr adsorption.
the catalytic stability of the monolithic catalyst even at high
temperatures above 973 K.
tendency was observed for the CO2 selectivity, thereby
suggesting that the addition of Re preferentially promoted
the MSR reaction.
Fig. 3a and b show the selectivities toward CO and CO2 as
a function of time on stream at 873 K. As indicated, both
were in the range of 40–60% for all the samples, indicating
that the MSR and WGS transformations occurred to the same
extent at this temperature. However, the time dependence of
the selectivities toward CO and CO2 differed between the Ni
and Ni–Re samples. More specifically, for the Ni5 sample, the
CO selectivity decreased slightly with time, while that to CO2
increased slightly. In contrast, for the Ni–Re samples, the CO
selectivity was generally constant or increased slightly with
time, and the CO2 selectivity was generally constant or
decreased slightly during the test.
The differences in these selectivities became more
pronounced at 973 K, as shown in Fig. 3c and d. For the Ni5
sample, the CO selectivity decreased slightly from the initial
value of 60% to a stable value of ∼47% after 6 h, while the
CO2 selectivity showed the opposite tendency, i.e., it
increased from an initial value of 40% to a stable value of
∼53% after 6 h. In contrast, the initial CO selectivity was
∼80% for all the Ni–Re samples. In the case of the Ni5Re0.5
sample, the CO selectivity gradually decreased with time on
stream, reaching 60% after 20 h, while a less pronounced
decrease was observed for the Ni5Re5 sample, and almost no
decrease occurred in the case of Ni5Re10. The opposite
3.2 The effects of mass and heat transfer on the reaction rates
The difference between the bulk fluid concentration and the
channel surface concentration of CH4 was estimated for all
the samples in the whole test temperature range, for
evaluating the effect of mass transfer on reaction rates.
Fig. 4a shows the changes of CH4 concentration between the
bulk fluid and surface at the inlet of the monolith during the
stepwise tests. The concentration change of CH4 was less
than 1% between the bulk and surface at 973 K for all the
samples. The Mears criterion for interphase mass transport
(in order for the reaction rate rSR to not deviate from the
value without the mass transport effect by more than 5%, the
inequality, rSRrpn/(CAbkmt) < 0.15 should be satisfied for the
spherical particle catalysts, where n is the reaction order
number and rp is the catalytic particle radius)37 was applied
to the monolithic catalyst, and can be expressed as rSRdhn/
(CAbkmt) < 0.2 (the details are shown in the ESI†). The value
of the left side of the inequality was smaller than 0.025 at
973 K and below for all the samples by approximating as a
first-order reaction (n = 1), which was much smaller than the
right side of the inequality, i.e. the Mears criterion was
satisfied at 973 K in the present study. These results
confirmed that the effect of mass transport on the reaction
rate was minor, and can be negligible below 973 K under the
present test conditions. In addition, as shown in Fig. 4a, the
concentration change of CH4 was less than 2% even at
temperatures above 973 K, suggesting that the effect of mass
transfer was very little even at 1173 K under the present
experimental conditions.
The difference between the temperatures of the bulk fluid
and the channel surface was estimated for all the samples
during the stepwise tests, for evaluating the effect of heat
transfer on reaction rates. Fig. 4b shows the changes of
temperature between the bulk fluid and surface at the inlet
of the monolith during the stepwise tests. The temperature
difference between the bulk fluid and the channel surface
was less than 7 K at 973 K and below for all the samples,
suggesting that the effect of heat transfer was small at 973 K
and below. The Mears criterion for interphase heat transport
Fig.
1 Space-time yield of hydrogen (STY-H2) of the monolithic
catalysts and the referred 5 wt% Re/Al2O3 catalyst during the stepwise
heating test. CH4/steam/N2 = 10/13.6/30 mL min−1, S/C = 1.36, GHSV
= 6400 h−1
.
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Fig. 2 Catalytic activity as a function of time on stream at 873 K (a) and 973 K (b). CH4/steam/N2 = 10/13.6/30 mL min−1, S/C = 1.36, GHSV = 6400 h−1
.
Fig. 3 CO and CO2 selectivities as a function of time on stream at 873 K (a and b) and at 973 K (c and d). CH4/steam/N2 = 10/13.6/30 mL min−1
S/C = 1.36, GHSV = 6400 h−1
,
.
(in order for the reaction rate rSR to not deviate from the
value without the mass transport effect by more than 5%,
|ΔHSRrSRrpEa/(hpRTb2)| < 0.15 should be satisfied for the
spherical catalytic particle, where Ea is the apparent
activation energy for MSR)36,37 was applied to the monolithic
structural catalyst, and can be expressed as |ΔHSRrSRdhEa/
(hpRTb2)| < 0.2 (the details are shown in the ESI†). The value
of the left side was 0.091 at 973 K for the Ni5Re10 sample,
which was smaller than the value of the right side,
confirming that the Mears criterion was satisfied at 973 K.
We also confirmed that the Mears criterion for heat
interphase transport was satisfied for other the samples at
973 K and below. These results suggest that the catalytic tests
carried out at temperatures below 973 K were mainly in the
kinetic regime. In contrast, the temperature difference
between the bulk fluid and the channel surface increased
sharply with increasing temperature above 973 K for Ni5Re5
and Ni5Re10 samples, and reached a value of 14 K at 1073 K
(Fig. 4b), indicating that the effect of heat transfer could not
be negligible at high temperatures above 973 K.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Catalysis Science & Technology
Paper
Fig. 4 (a) The changes of CH4 concentration between the fluid bulk and surface, and (b) the changes of temperature between the fluid bulk and
surface, at the inlet of the monolith during the stepwise tests. CH4/steam/N2 = 10/13.6/30 mL min−1, S/C = 1.36, GHSV = 6400 h−1
.
Therefore, the apparent activation energies for MSR was
estimated by utilizing the reaction rate constants below 973
K. Fig. 5 shows the Arrhenius plot of the reaction rate
constants in the temperature range of 773–973 K for the
monolithic catalysts. A good linear relationship was obtained
for all the samples, suggesting that the reaction rates were in
the kinetic regime. The calculated apparent activation
energies and pre-exponential factors are listed in Table 2. The
apparent activation energy was similar (within a narrow range
of 65–68 kJ mol−1) for both Ni and Ni–Re surface coated
samples, suggesting that the Ni atoms in both catalysts are
active sites for MSR. The values of the apparent activation
energy obtained in this study were relatively smaller than that
of Ni honeycomb catalysts we previously reported (78 kJ
mol−1),13 and those for MSR over supported Ni catalysts
reported by many researchers (95–240 kJ mol−1).42–45
On the other hand, the Arrhenius plot of the reaction rate
constants of WGS in the temperature range of 773–973 K for
the samples deviated greatly from the linear relationship,
and a meaningful apparent activation energy for WGS could
not be obtained. Probably this is because the reaction rate of
WGS was much larger than that of SR, and it can easily reach
equilibrium under the present experimental conditions.
3.3 Characterization of the catalysts before the reaction
The SEM images given in Fig. 6a–d show the surface
morphologies of the Ni5, Ni5Re0.5, Ni5Re5, and Ni5Re10
samples after calcination at 773 K for 5 h. All the samples
exhibited
a similar surface morphology consisting of
numerous fine particles smaller than 100 nm in diameter.
The BET specific surface areas of the samples after
calcination are listed in Table 1, with values ranging from
0.61 to 3.02 m2 g−1, which are >60 times larger than that of
the Ni monolith structure without coating, which was
confirmed to be ∼0.01 m2 g−1 in our previous study.14
XPS measurements revealed that both Ni and Re in the
surface layer were in their oxide states after calcination. The
Re 4f and Ni 2p3/2 spectra of the Ni5Re0.5 and Ni5Re10
samples in the calcination state are shown in Fig. 7. In the
case of the Re 4f spectra, a 4f7/2 peak at ∼46 eV and a 4f5/2
peak at ∼48 eV were detected in both samples. These peaks
corresponded to Re2O7 and ReO3.46,47 For the Ni 2p3/2
spectra, a peak at ∼854 eV with a shoulder peak at ∼856 eV
(accompanied by a satellite peak at ∼861.5 eV) was detected,
which corresponds to the spectra of NiO.48–52
The SEM images given in Fig. 6e–h show the surface
morphologies of the Ni5, Ni5Re0.5, Ni5Re5, and Ni5Re10
samples after H2 reduction at 703 K for 1 h. A surface
morphology consisting of numerous nanoparticles was
observed in all cases, which was similar to that observed for
the calcination state. In this case, the particle sizes ranged
from 100–200 nm, which were slightly larger than those of
the corresponding calcinated samples.
Table 2 Apparent activation energies and pre-exponential factors for
MSR over various monolithic catalysts (stepwise tests; CH4/steam/N2
10/13.6/30 mL min−1, S/C = 1.36, GHSV = 6400 h−1
=
)
Preexponential factor
Sample
Ea (kJ mol−1
)
(mol m−3 atm−2
)
Ni5
65.6 0.5
67.2 1.9
66.8 2.5
68.3 3.4
15.3
15.7
15.6
15.9
Ni5Re0.5
Ni5Re5
Ni5Re10
Fig. 5 Arrhenius plot of the apparent reaction rate constants for MSR
in the temperature range of 773–973 K for the monolithic catalysts.
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Fig. 6 SEM secondary electron images of the monolithic catalysts after calcination (a–d), after hydrogen reduction at 703 K for 1 h (e–h), and after
MSR at 873 K over 20 h followed by 20 h at 973 K (i–l).
Fig. 7 Re 4f and Ni 2p3/2 XPS spectra of Ni5Re0.5 (a and b) and Ni5Re10 (c and d) after calcination, after hydrogen reduction at 703 K for 1 h, and
after MSR reaction at 873 K for 20 h followed at 973 K for 20 h.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Catalysis Science & Technology
Paper
¯
XPS measurements were then carried out to identify the
chemical states of Ni and Re on the sample surfaces after
reduction. Fig. 7 shows the Re 4f and Ni 2p3/2 spectra of the
Ni5Re0.5 and Ni5Re10 samples. For the Re 4f spectra, a 4f7/2
peak at ∼40.5 eV and a 4f5/2 peak at ∼43.0 eV were detected
in both samples, corresponding to metallic Re.53 In the case
of the Ni 2p3/2 spectrum, a peak at ∼852.6 eV accompanied
by a small satellite peak at ∼856 eV was detected, which was
attributed to metallic Ni.49,52,54 These results indicate that
both Re and Ni oxides were reduced to the metallic state by
the hydrogen reduction reaction.
space group symmetry Fm3m) Ni phase. For the Ni5Re10
sample, in addition to the fcc-Ni phase, an hcp (hexagonal
close-packed structure, with space group symmetry P63/mmc)
Re phase was also detected, indicating the coexistence of Ni
and Re phases. An enlarged view of the 311 peak of the Ni
phase in Fig. 8a shows that the peaks of the Ni5Re0.5 and
Ni5Re10 samples shifted to lower values of 2θ compared to
those of Ni5, which indicates that the lattice constants of the
Ni phases in Ni5Re0.5 and Ni5Re10 were greater than that in
Ni5. The lattice constants of the Ni phases were calculated for
all the samples and plotted as
a function of the Re
The formation of a Ni–Re bond in the reduced samples
was implied by XAFS measurements. Fig. 8 shows the Fourier
transforms (FT) of the k3-weighted Ni K-edge and Re L3-edge
XAFS spectra of the Ni–Re samples after reduction at 703 K
for 1 h. The results obtained for the reference samples, pure
Ni foil, NiO, Re, ReO2, ReO3, and Re2O7, are also shown in
this figure. In the case of the pure Ni foil, a peak at 2.16 Å
concentration (Fig. 9b). More specifically, the lattice constant
of the Ni phase in the Ni5 sample was 3.524 Å, and this
increased to 3.543 Å in the Ni5Re0.5 sample, corresponding
to a 3.1 at% Re content. At higher Re concentrations above
3.1 at%, this value remained relatively unchanged. It is
therefore supposed that the increase in the lattice constant
below 3.1 at% Re was due to the formation of a Ni(Re) solid
solution phase through the substitution of Ni by Re in the Ni
phase, as Re has a larger atomic radius than Ni. The relatively
stable lattice constants at the higher Re concentrations is
supposed to be due to the formation of two phases, Ni(Re)
and Re, because of the limits of solid solubility of Re in Ni at
703 K, i.e., ∼5 at%, as indicated in the Ni–Re phase
diagram.56
From the enlarged view of Ni (311) in Fig. 9a, it is
apparent that the diffraction peaks of the Ni phase became
broader with the addition of Re, which indicates that the full-
width at half maximum (FWHM) of the spectra increased.
The crystallite size was then estimated based on the FWHM
using the Williamson–Hall method,57 as outlined in Table 3.
More specifically, the crystallite size of the Ni(Re) solid
solution phase in the Ni–Re samples was significantly smaller
than that of the Ni phase in the pure Ni sample (Ni5), thereby
suggesting that Re suppressed the coarsening of Ni(Re)
crystallites during the hydrogen reduction process. In
addition, as shown in Table 3, the crystallite sizes of the Re
phase were significantly smaller than those of the Ni(Re)
phase in Ni5Re0.5 and Ni5Re10, suggesting that the grain
growth of the Re phase was slower than that of the Ni phase.
(non-phase
shift
corrected)
was
observed,
which
corresponded to the nearest neighbor (NN) Ni–Ni bond (2.49
Å).2,3,13 However, this peak shifted slightly to 2.11 Å for the
Ni5Re0.5, Ni5Re5, and Ni5Re10 samples (Fig. 8a). In the case
of the pure Re sample, a peak at 2.33 Å (non-phase shift
corrected) was observed, corresponding to the NN Re–Re
bond (2.74 Å),55 and this peak shifted to 2.06 Å for the
Ni5Re5 and Ni5Re10 samples (Fig. 8b). These peak shifts in
both the Ni K-edge and the Re L3-edge spectra were
attributed to close interactions between the Ni and Re atoms,
thereby suggesting the formation of Ni–Re bonds during the
hydrogen reduction reaction. In addition to the peak
corresponding to the Ni–Re bond, a peak at 1.30 Å was also
detected, which corresponded to the Re–O bond in Re2O7.
This species are artifacts which likely arose from the partial
oxidation of Re to Re2O7 during the mortar milling required
for XAFS sample preparation.
SR-XRD measurements were then performed on the
surface layer of the monolithic catalysts, and Fig. 9a shows
the SR-XRD profiles of the Ni5, Ni5Re0.5, and Ni5Re10
samples. For the Ni5 and Ni5Re0.5 samples, all diffraction
peaks were indexed as a single fcc (face-centered cubic, with
Fig. 8 Fourier transform of k3-weighted Ni K-edge (a) and Re L3-edge (b) XAFS spectra of Ni5Re0.5, Ni5Re5, and Ni5Re10 samples after hydrogen
reduction at 703 K for 1 h.
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Fig. 9 (a) SR-XRD peak profiles of Ni5, Ni5Re0.5, and Ni5Re10 samples after hydrogen reduction at 703 K for 1 h, with an enlarged view of the Ni
(311) peak (λ = 0.06201 nm); (b) calculated lattice constant of the Ni(Re) phase in Ni5, Ni5Re0.5, Ni5Re5, and Ni5Re10 samples after hydrogen
reduction at 703 K for 1 h.
Table 3 Lattice constants and crystallite sizes of Ni(Re) and Re phases formed on the surface of the monolith before (after hydrogen reduction at 703 K
for 1 h) and after the MSR reaction at 873 K for 20 h followed at 973 K for 20 h
Lattice constant of the Ni phase (Å)
Crystallite size of Ni and Re phases (Å)
Before reaction
After MSR reaction
Ni
Before
After MSR
reaction
Samples
Ni
Re
Re
reaction
Ni5
3.52411(2)
3.54284(3)
3.54385(3)
3.54307(3)
399(28)
106(11)
99(5)
Ni5Re0.5
Ni5Re5
Ni5Re10
3.533410IJ3)
301(46)
438(7)
21(6)
19(2)
3.54444(5)
83(8)
22(2)
3.4 Characterization of the catalysts after MSR
2p3/2 peaks at ∼854 and 856 eV corresponding to NiO and
NiIJOH)2 were also observed for both samples. These results
suggest that the oxidation of Re might proceed slower
compared to that of Ni during the MSR reaction.
SR-XRD measurements revealed that the Ni–Re alloy phase
was maintained even after MSR. More specifically, Fig. 10
shows the SR-XRD profiles of the Ni5Re0.5 and Ni5Re10
samples after the MSR reaction at 873 K for 20 h followed by
the reaction at 973 K for 20 h. For comparison, the profile of
The SEM images presented in Fig. 6i–l show the surface
morphologies of the Ni5, Ni5Re0.5, Ni5Re5, and Ni5Re10
samples after MSR at 873 K over 20 h followed by 20 h at 973
K. As indicated, the surface morphology changed significantly
from that after hydrogen reduction for all the samples, and
no nanoparticles were observed after the reaction, indicating
that nanoparticle agglomeration occurred during MSR. No
obvious carbon deposition was observed on the surface of the
samples after the reaction. The BET specific surface areas of
the samples after the reaction are listed in Table 1, with
values ranging from 0.084 to 0.157 m2 g−1 being calculated.
These values are significantly smaller than those calculated
prior to the reaction, further suggesting that nanoparticle
agglomeration took place during the reaction.
XPS measurements also revealed that only a small degree
of Re oxidation took place. This contrasts to the case of Ni,
where significant oxidation was observed during the MSR
reaction. Fig. 7 shows the Re 4f and Ni 2p3/2 spectra of the
Ni5Re0.5 and Ni5Re10 samples after the MSR reaction. In the
case of the Re 4f spectra, the main peaks were a 4f7/2 peak at
∼40.5 eV and a 4f5/2 peak at ∼43.0 eV, which correspond to
metallic Re, and were accompanied by some less intense
peaks at ∼45 and 48 eV, which correspond to Re2O7. In
contrast, in addition to the Ni 2p3/2 peak observed at ∼852.6
eV, which was attributed to metallic Ni, relatively strong Ni
Fig. 10 SR-XRD peak profiles of Ni5Re0.5, and Ni5Re10 samples after
the MSR reaction at 873 K for 20 h and 973 K for 20 h (λ = 0.06201
nm). The enlarged view of the marked area shows the (311) peak of the
Ni phase.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Catalysis Science & Technology
Paper
the Ni5 sample before the reaction (after the hydrogen
reduction at 703 K for 1 h) is also shown. After the reaction,
the main diffraction peaks observed for the Ni5Re0.5 sample
were assigned to a single fcc-Ni(Re) phase, whereas the main
peaks originating from Ni5Re10 were assigned to two phases
of fcc-Ni(Re) and hcp-Re. An enlarged view of the 311
diffraction peak of the Ni phase is shown in Fig. 10, and
upon comparison with the corresponding peak prior to the
reaction (2θ = 33.75°, as shown in the enlarged view of Ni
(311) in Fig. 9a), it is apparent that no obvious shift in the
peak position occurred in Ni5Re10, while a clear shift to
higher 2θ values occurred in Ni5Re0.5. The calculated lattice
constants of the two samples after the reaction (Table 3)
revealed that the lattice constant of the Ni(Re) phase in
Ni5Re10 remained almost unchanged, while that of the
Ni(Re) phase in Ni5Re0.5 became smaller over the course of
the reaction. This decrease was likely due to a decrease in the
Re concentration of the Ni(Re) phase, which would be caused
by diffusion of the Re atoms from the surface Ni–Re alloy
layer into the Ni foil substrate during the reaction. In
contrast, in the case of Ni5Re10, which consisted of Re and
Ni(Re) phases, the Re concentration in the Ni–Re alloy phase
was maintained by the diffusion of Re atoms from the Re
phase to the Ni(Re) phase. The crystallite sizes of the Ni(Re)
phase in both the Ni5Re0.5 and Ni5Re10 samples increased
significantly, whereas that of the Re phase in the Ni5Re10
sample only slightly increased after MSR (Table 3), suggesting
that the grain growth of the Ni(Re) phase occurred more
easily than for the Re phase during the MSR reaction. Some
weak signals corresponding to the ReO3 phase were
confirmed in the Ni5Re10 sample after the reaction (Fig. 10),
further confirming that a degree of oxidation took place for
Re during the MSR process. Some of the Re oxides might
evaporate during the reaction because of their high vapor
pressures,58 which can also contribute to the decrease in the
Re concentration of the Ni(Re) phase.
Fig. 11 TPR profiles of Ni5, Ni5Re0.5, Ni5Re5 and Ni5Re10 samples
after calcination at 773 K for 5 h. For reference, TPR profiles of pure
ReO2, ReO3, and Re2O7 powders are also shown.
Ni5Re0.5, however, the main peak shifted to
a lower
temperature, i.e., ∼680 K, and further shifts to ∼625 K were
observed for Ni5Re5 and Ni5Re10. These signals likely
correspond to the reduction of NiO, NiIJOH)2, ReO3, and
Re2O7, as the reduction temperatures of ReO3 and Re2O7 are
close to the temperature of this main peak. These results
therefore suggest that the addition of Re significantly
increased the reducibility of NiO.
To elucidate the effect of Re on the reduction rate of Ni
oxides, in situ XAFS measurements were carried out for the
Ni5 and Ni5Re10 samples during reduction at 703 K. The
obtained results clearly revealed that Re significantly
promoted the reduction of Ni oxides. Fig. 12a shows the Ni
K-edge X-ray absorption near edge structure (XANES) spectral
changes for the Ni5 sample, while Fig. 12b shows the Ni
K-edge XANES spectral changes for Ni5Re10 during the
reduction. The Ni K-edge XANES spectra of pure NiO and the
Ni foil were measured at 298 K, and are plotted in Fig. 12b
for reference, showing that the relative intensity of the peak
at 8346 eV was significantly higher for NiO than that for the
Ni foil. Prior to commencing the reduction, the Ni K-edge
XANES spectra of both samples corresponded to that of NiO.
However, as the reduction progressed, clear spectral changes
were observed. More specifically, in the case of the Ni5
sample, although the intensity of the peak at 8346 eV showed
no change in the initial 2 min after starting the H2 flow, a
gradual decrease was observed as the reduction progressed
(Fig. 12a). In contrast, for the Ni5Re10 sample, the intensity
of this peak decreased sharply to a level close to that of the
metallic Ni foil within the initial 2 min of the reaction, and
no further changes were obtained upon increasing the
reduction time (Fig. 12b). These results revealed that NiO was
reduced to metallic Ni more rapidly in the Ni5Re10 sample
than in the Ni5 sample.
3.5 Reducibility of the Ni–Re surface layers
TPR measurements were carried out for the Ni5, Ni5Re0.5,
Ni5Re5, and Ni5Re10 samples after calcination at 773 K for 5
h, and the corresponding TPR profiles are shown in Fig. 11.
For reference, TPR measurements were carried out for the
pure Re oxide powders (ReO2, ReO3, and Re2O7), and the TPR
profiles are also shown in Fig. 11. More specifically, ReO2
exhibited a reduction peak at ∼760 K, while ReO3 gave a
reduction peak at ∼680 K, and Re2O7 at 600 K, the latter of
which was accompanied by a broad weak peak at ∼630 K;
these observations correspond to the following reducibility
sequence: Re2O7 > ReO3 > ReO2.
For the Ni5 sample, a predominant peak was observed at
∼770 K, accompanied by several smaller peaks at ∼625 and
680 K. The main signal corresponded to the reduction of
NiO, while the smaller peaks corresponded to the reduction
of NiIJOH)2 which might be formed when the sample was
exposed to air before TPR measurement.59,60 In the case of
Fig. 12c and d show the corresponding FTs of the Ni
K-edge XAFS spectra during the reduction process. As shown,
in the initial 2 min, the FT profile of the Ni5 sample was
similar to that of pure NiO for the Ni5 sample, i.e., exhibiting
two peaks at 1.66 and 2.59 Å (non-phase shift corrected),
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Fig. 12 Evolution of Ni K-edge XANES spectra of Ni5 (a) and Ni5Re10 (b) obtained from the in situ XAFS measurement during hydrogen reduction
at 703 K. The corresponding Fourier transforms of the k3-weighted Ni K-edge XAFS spectra are shown in (c) Ni5 and (d) Ni5Re10. The results of
the Ni foil and pure NiO are shown as reference.
which corresponded to the NN Ni–O bond (2.08 Å) and the
second shell Ni–Ni bond (2.95 Å), respectively.2,3,13 Upon
further increasing the reduction time, the peak at 2.16 Å
(non-phase shift corrected) corresponding to the NN Ni–Ni
bond (2.49 Å) became stronger, while those at 1.66 and 2.59
Å became weaker. In contrast, in the initial minute of the
reaction, the Ni5Re10 sample exhibited a similar FT profile
to that of pure NiO, but rapidly changed to give an FT profile
similar to that of the metallic Ni foil in the second minute,
i.e., only the peak at 2.16 Å corresponding to the NN Ni–Ni
bond was identified. The reduction rate of NiO was then
estimated by a linear combination fitting of the normalized
intensity profiles of the XAFS spectra during the reduction
using Athena software, and the results are presented in
Fig. 13 as a function of the reduction time. This clearly
demonstrates that the Ni oxides formed in the Ni5Re10
sample were reduced more rapidly than those in the pure Ni
sample.
content to >24 at% (Ni5Re5 and Ni5Re10). SR-XRD
measurement revealed that a Ni–Re alloy consisting of a
Ni(Re) solid solution phase and a Re phase was formed in
Ni5Re5 and Ni5Re10 samples during hydrogen reduction and
the subsequent MSR reaction. In situ XAFS measurement
revealed that the addition of Re significantly accelerated the
reduction of Ni oxides to metallic Ni, and formed a Ni–Re
bond in the surface layer.
3.6 Effect of Re on the reducibility and catalytic performance
The above results revealed that the catalytic activities of the
Ni monolithic catalysts were effectively improved in the
presence of a Ni–Re bimetallic surface layer coating; in
particular the stability in terms of the catalyst activity at 873
and 973 K was significantly enhanced upon increasing the Re
Fig. 13 Reduction rates of NiO in Ni5 and Ni5Re10 samples
determined by a linear combination fitting of the normalized intensity
profiles of the in situ XAFS spectra during the hydrogen reduction.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Catalysis Science & Technology
Paper
The adsorption energy of hydrogen on the Re and Ni
surfaces was then calculated to explain the effect of Re on
the reducibility of the Ni–Re catalyst. Table 4 shows the
adsorption energy of one hydrogen atom on various possible
sites of the Ni (111), Ni (011), Ni (001), and Re (001) surfaces,
i.e., top, fcc, hcp, 4F (four-fold hollow), bridge, hole, and
pseudo (pseudo three-fold hollow) sites.
A schematic
representation of these sites is shown in Fig. S1 of the ESI†
and in ref. 61. The adsorption energies of a hydrogen atom
on the hole and hcp sites of Re (001) were smaller than those
on all the sites of Ni (111), Ni (011), and Ni (001). These
results revealed that hydrogen preferentially adsorbed onto
the Re surface as opposed to the Ni surface.
Based on these results, the effect of Re on the reducibility
and catalytic performance of the Ni–Re-coated monolithic
catalysts was analyzed as illustrated in Fig. 14. More
specifically, a surface layer of Ni and Re oxide particles was
formed on the monolithic foil surface by the sequential
impregnation and calcination at 773 K (Fig. 14a). During the
hydrogen reduction reaction at 703 K, the Re oxides were
reduced preferentially over the Ni oxides as the majority of
Re oxides (mainly ReO3) are located on the surface of NiO
and NiIJOH)2 due to the sequential impregnation method
employed herein; furthermore, as shown by the TPR
measurements depicted in Fig. 11, the Re oxides, ReO3 and
Re2O7, were more easily reduced than NiO. The reduced
metallic Re adsorbed hydrogen more easily than Ni, as
revealed by the above first-principles calculation results of
the adsorption energies on the Ni and Re surfaces (Table 4).
It is considered that the hydrogen adsorbed on the metallic
Re particles is transported to the surface of the neighboring
Ni oxide particles through hydrogen spillover,62 thereby
accelerating the reduction of the Ni oxides (Fig. 13).
Subsequently, the reduced Ni and Re species form a Ni–Re
alloy consisting of a Ni(Re) solid solution phase containing a
high at% of Re, in addition to the formation of a Re phase
during the hydrogen reduction process (Fig. 14b).
As shown in Fig. 10, the crystalline structure of the Ni(Re)
phase in the Ni5Re0.5 sample remained unchanged after the
MSR reaction at 873 and 973 K. Furthermore, a two-phase
structure of the Re and Ni(Re) phase was retained in the
Ni5Re10 sample. The Re atoms in the Re phase and the
Ni(Re) phase are proposed to play a role in increasing the
hydrogen concentration on the surface, since Re atoms can
adsorb the produced hydrogen more easily than Ni atoms
which was proved by the DFT calculations (Table 4). This
thereby suppressed the oxidation of the Ni atoms on the
Fig. 14 Schematic representation of the formation of the Ni–Re
surface alloy layer during hydrogen reduction and methane steam
reforming.
surface (Fig. 14c). Because the deactivation of Ni monolithic
catalysts mainly occurs due to the oxidation of surface Ni by
steam during MSR,13 suppression of this oxidation led to an
enhancement in the reactivity and stability. In the case of the
Ni5Re0.5 sample, only a single Ni(Re) phase with a low
concentration of Re (∼3 at%) was formed (Fig. 9). The
concentration of Re might further be lowered during the
reaction at higher temperatures (973 K) because of the
increased inner diffusion of Re and evaporation of the Re
oxides formed during the reaction. This low concentration of
Re on the surface layer was not sufficient to maintain a rich
hydrogen atmosphere on the surface, and therefore could not
retain a high reactivity and stability, especially at high
temperatures. In contrast, for the Ni5Re10 sample in which
two phases of Ni(Re) and Re with a relative high Re
concentration (∼38 at% Re) were formed, a rich hydrogen
atmosphere was easily formed on the surface, thereby
suppressing the oxidation of the Ni atoms and resulting in a
higher stability and reactivity, particularly at high
temperature (973 K) (Fig. 2).
The dependency of selectivity on Re concentration can
also be interpreted by considering that the Re atoms adsorb
the produced hydrogen more easily than Ni atoms. That is,
in the case of the Ni5Re10 sample with high Re
concentration, more hydrogen was adsorbed on the surface
by the Re atoms, resulting in
a
higher hydrogen
concentration near the catalyst surface. This may promote
Table 4 Adsorption energies (eV) of one hydrogen atom on various possible sites of the Ni (111), Ni (001)*, Ni (110)*, and Re (001) surfaces (*by using a
slab consisting of 16 atoms in four atomic layers (each layer = 2 × 2 atoms))
Top
Fcc
Hcp
Hole
4F
Bridge
Long bridge
Short bridge
Pseudo
Ni (111)
Ni (001)*
Ni (011)*
Re (001)
−0.19
−0.10
−0.19
−0.01
−0.74
—
—
−0.69
—
—
−0.78
—
—
—
−0.82
—
−0.74
−0.57
—
−0.59
—
—
—
—
—
—
−0.61
−0.36
—
−0.49
−0.58
−0.61
—
—
—
—
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
the reverse reaction of WGS (eqn (5)), which can led to the
decrease of selectivity to CO2 with the increase of Re
concentration (Fig. 3).
grateful to Drs. M. Tanaka and Y. Katsuya at NIMS for their
help with SR-XRD experiments. The authors would like to
thank Ms. T. Harimoto at NIMS for her help with catalytic
experiments, Dr. H. Ohata at NIMS for his help with XPS
experiments, and Dr. J. Nara at NIMS and Mr. H. X. Jin at
Shandong University for their help with the PHASE software
package for DFT calculations.
4. Conclusions
We herein reported the coating of a high cell density Ni
monolithic catalyst with a bimetallic Ni–Re layer through a
sequential impregnation method to enhance the catalytic
performance for methane steam reforming (MSR). Both
before and after hydrogen reduction and MSR, the Ni–Re
References
1 S. D. Angeli, G. Monteleone, A. Giaconia and A. A.
Lemonidou, State-of-the-Art Catalysts for CH4 Steam
Reforming at Low Temperature, Int. J. Hydrogen Energy,
2014, 39, 1979–1997.
2 D. L. Li, I. Atake, T. Shishido, Y. Oumi, T. Sano and K.
Takehira, Self-Regenerative Activity of Ni/Mg(Al)O Catalysts
with Trace Ru During Daily Start-up and Shut-Down Operation
of CH4 Steam Reforming, J. Catal., 2007, 250, 299–312.
3 M. Ferrandon, A. J. Kropf and T. Krause, Bimetallic Ni-Rh
Catalysts with Low Amounts of Rh for the Steam and
Autothermal Reforming of N-Butane for Fuel Cell
Applications, Appl. Catal., A, 2010, 379, 121–128.
4 J. R. Rostrup-Nielsen, Conversion of Hydrocarbons and
Alcohols for Fuel Cells, Phys. Chem. Chem. Phys., 2001, 3,
283–288.
5 J. R. Rostrupnielsen, L. J. Christiansen and J. H. B. Hansen,
Activity of Steam Reforming Catalysts - Role and Assessment,
Appl. Catal., 1988, 43, 287–303.
6 H. F. Rase, Handbook of Commercial Catalysts- Heterogeneous
Catalysts, CRC Press, U. S. A., 2000, pp. 403–418.
7 J. R. Rostrup-Nielsen, Steam Reforming, in Handbook of
Heterogeneous Catalysis, ed. G. Ertl, H. Knozinger, F. Schuth
and J. Weitkamp, WILEY-VCH Verlag GmbH & Co. kGaA,
Germany, 2008, vol. 6, pp. 2882–2904.
8 R. J. Farrauto, Y. Liu, W. Ruettinger, O. Ilinich, L. Shore and
T. Giroux, Precious Metal Catalysts Supported on Ceramic
and Metal Monolithic Structures for the Hydrogen Economy,
Catal. Rev.: Sci. Eng., 2007, 49, 141–196.
9 T. Miyata, M. Shiraga, D. L. Li, I. Atake, T. Shishido, Y.
Oumi, T. Sano and K. Takehira, Promoting Effect of Ru on
Ni/Mg(Al)O Catalysts in Dss-Like Operation of CH4 Steam
Reforming, Catal. Commun., 2007, 8, 447–451.
surface layer was analyzed by
a range of analytical
techniques. It was found that the catalytic performance of
the Ni monolithic catalyst was significantly improved by
coating with a Ni–Re bimetallic surface layer, and the catalyst
stability at 873 and 973 K was remarkably enhanced upon
increasing the Re content of the Ni–Re surface layer. A Ni–Re
alloy nanoparticle surface layer was formed on the Ni
monolithic catalyst during hydrogen reduction at 703 K. This
alloy consisted of a single Ni(Re) solid solution phase in the
case of the Ni5Re0.5 sample, and two phases (i.e., Ni(Re) and
Re) in the Ni5Re5 and Ni5Re10 samples. The crystal structure
of the Ni–Re alloy remained unchanged during the MSR
reaction, although agglomeration of the Ni–Re nanoparticles
occurred. Furthermore, in situ X-ray absorption fine structure
spectroscopy measurements showed that Re significantly
accelerated the reduction of Ni oxides to metallic Ni.
Moreover,
density
functional
theory
calculations
lower
demonstrated that the Re (001) surface has
a
adsorption energy for hydrogen compared to the Ni (111),
(110), and (100) surfaces. These results reveal that the Re
atoms adsorb hydrogen more easily during MSR compared to
the Ni atoms, and this adsorbed hydrogen inhibits the
oxidation of the vicinal Ni atoms while enhancing the overall
catalytic stability of the Ni monolithic catalysts. These results
are of importance as they could lead to novel and more
efficient monolithic catalysts for the MSR process, which is
the main industrial route to hydrogen production for use as
a renewable fuel.
Conflicts of interest
There are no conflicts to declare.
10 M. Flytzani-Stephanopoulos, G. E. Voecks and T. Charng,
Modeling of Heat-Transfer in Nonadiabatic Monolith
Reactors and Experimental Comparisons of Metal Monoliths
with Packed-Beds, Chem. Eng. Sci., 1986, 41, 1203–1212.
11 A. Cybulski and J. A. Moulijn, Monoliths in Heterogeneous
Catalysis, Catal. Rev.: Sci. Eng., 1994, 36, 179–270.
Acknowledgements
This work was supported by the Aichi Science & Technology
Foundation, Japan. XAFS experiments were performed at the
BL5S1 beam-line of the Aichi Synchrotron Radiation Center,
Aichi Science
&
Technology Foundation (Proposal No.
12 Y. Hiramitsu, M. Demura, Y. Xu, M. Yoshida and T. Hirano,
Catalytic Properties of Pure Ni Honeycomb Catalysts for
Methane Steam Reforming, Appl. Catal., A, 2015, 507,
162–168.
13 Y. Xu, T. Harimoto, T. Hirano, H. Ohata, H. Kunieda, Y. Hara
and Y. Miyata, Catalytic Performance of a High-Cell-Density
Ni Honeycomb Catalyst for Methane Steam Reforming, Int. J.
Hydrogen Energy, 2018, 43, 15975–15984.
201803006). The authors are grateful to Drs. T. Hiromoto and
K. Takahama at the Aichi Synchrotron Radiation Center for
their help with XAFS experiments. SR-XRD experiments were
performed at the Spring-8 facility under the approval of the
NIMS Synchrotron X-ray Station (Proposal No. 2018A4907).
The DFT calculations in this study were performed on
Numerical Materials Simulator at NIMS. The authors are
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Catalysis Science & Technology
Paper
14 Y. Xu, T. Harimoto, L. S. Wang, T. Hirano, H. Kunieda, Y.
Hara and Y. Miyata, Effect of Steam and Hydrogen
Treatments on the Catalytic Activity of Pure Ni Honeycomb
for Methane Steam Reforming, Chem. Eng. Process.,
2018, 129, 63–70.
15 S. De, J. G. Zhang, R. Luque and N. Yan, Ni-Based Bimetallic
Heterogeneous Catalysts for Energy and Environmental
Applications, Energy Environ. Sci., 2016, 9, 3314–3347.
16 V. Ponec and G. C. Bond, Catalysis by Metals and Alloys -
Prologue, Catalysis by Metals and Alloys, 1995, vol. 95, pp. 1–
5.
17 Y. G. Chen, O. Yamazaki, K. Tomishige and K. Fujimoto,
Noble Metal Promoted Ni0.03Mg0.97O Solid Solution Catalysts
for the Reforming of CH4 with CO2, Catal. Lett., 1996, 39,
91–95.
18 K. Tomishige, S. Kanazawa, M. Sato, K. Ikushima and K.
Kunimori, Catalyst Design of Pt-Modified Ni/Al2O3 Catalyst
with Flat Temperature Profile in Methane Reforming with
CO2 and O2, Catal. Lett., 2002, 84, 69–74.
19 S. Roy, S. Hariharan and A. K. Tiwari, Pt-Ni Subsurface Alloy
Catalysts: An Improved Performance toward CH4
Dissociation, J. Phys. Chem. C, 2018, 122, 10857–10870.
20 L. S. Wang, K. Murata and M. Inaba, Control of the Product
Ratio of CO2/(CO+CO2) and Inhibition of Catalyst
Deactivation for Steam Reforming of Gasoline to Produce
Hydrogen, Appl. Catal., B, 2004, 48, 243–248.
21 L. S. Wang, K. Murata, Y. Matsumura and M. Inaba, Lower-
Temperature Catalytic Performance of Bimetallic Ni-Re/Al2O3
Catalyst for Gasoline Reforming to Produce Hydrogen with
the Inhibition of Methane Formation, Energy Fuels, 2006, 20,
1377–1381.
22 K. T. Liu, et al., Supported Nickel-Rhenium Catalysts for
Selective Hydrogenation of Methyl Esters to Alcohols, Chem.
Commun., 2017, 53, 9761–9764.
23 F. F. Yang, D. Liu, H. Wang, X. Liu, J. Y. Han, Q. F. Ge and
X. L. Zhu, Geometric and Electronic Effects of Bimetallic Ni-
Re Catalysts for Selective Deoxygenation of M-Cresol to
Toluene, J. Catal., 2017, 349, 84–97.
29 B. A. Haberman and J. B. Young, Three-Dimensional
Simulation of Chemically Reacting Gas Flows in the Porous
Support Structure of an Integrated-Planar Solid Oxide Fuel
Cell, Int. J. Heat Mass Transfer, 2004, 47, 3617–3629.
30 V. Palma, A. Ricca, E. Meloni, M. Martino, M. Miccio and P.
Ciambelli, Experimental and Numerical Investigations on
Structured Catalysts for Methane Steam Reforming
Intensification, J. Cleaner Prod., 2016, 111, 217–230.
31 H. Ohashi and Y. Inagaki Catalytic Activity of Catalysts for
Steam Reforming Reaction, JAERI-Tech 2003-046, Japan
Atomic Energy Research Institute, Ibaraki, Japan, 2003.
32 K. Hashimoto, Reaction Engineering, Baifukan, Tokyo, Japan,
1993.
33 E. Tronconi and P. Forzatti, Adequacy of Lumped Parameter
Models for Scr Reactors with Monolith Structure, AIChE J.,
1992, 38, 201–210.
34 E. Tronconi and A. Beretta, The Role of Inter- and Intra-
Phase Mass Transfer in the Scr-DenoIJX) Reaction over
Catalysts of Different Shapes, Catal. Today, 1999, 52,
249–258.
35 R. D. Hawthorn, Reaction Research Needs 4. Broad Range of
Design Problems, Chem. Eng. Prog., 1974, 70, 41–43.
36 D. E. Mears, Diagnostic Criteria for Heat Transport
Limitations in Fixed Bed Reactors, J. Catal., 1971, 20, 127–131.
37 D. E. Mears, Tests for Transport Limitations in Experimental
Catalytic Reactors, Ind. Eng. Chem. Process Des. Dev.,
1971, 10, 541–547.
38 M. Tabuchi, H. Asakura, H. Morimoto, N. Watanabe and Y.
Takeda, Hard X-Ray XAFS Beamline, BLS1, at AichiSR,
J. Phys.: Conf. Ser., 2016, 712, 1–4.
39 B. Ravel and M. Newville, Athena, Artemis, Hephaestus: Data
Analysis for X-Ray Absorption Spectroscopy Using Ifeffit,
J. Synchrotron Radiat., 2005, 12, 537–541.
40 M. Tanaka, Y. Katsuya and A. Yamamoto, A New Large
Radius Imaging Plate Camera for High-Resolution and High-
Throughput Synchrotron X-Ray Powder Diffraction by
Multiexposure Method, Rev. Sci. Instrum., 2008, 79, 075106.
41 H. J. Monkhorst and J. D. Pack, Special Points for Brillouin-
Zone Integrations, Phys. Rev. B: Solid State, 1976, 13,
5188–5192.
24 L. Ma, K. J. Sun, M. Luo, L. Yan, Z. Jiang, A. H. Lu and Y. J.
Ding, Role of Reox Species in Ni-Re/Al2O3 Catalyst for
Amination of Monoethanolamine, J. Phys. Chem. C,
2018, 122, 23011–23025.
25 L. Ma, L. Yan, A. H. Lu and Y. J. Ding, Effect of Re Promoter
on the Structure and Catalytic Performance of Ni-Re/Al2O3
42 J. M. Wei and E. Iglesia, Isotopic and Kinetic Assessment of
the Mechanism of Reactions of CH4 with CO2 or H2O to
Form Synthesis Gas and Carbon on Nickel Catalysts,
J. Catal., 2004, 224, 370–383.
Catalysts
for
the
Reductive
Amination
of
43 K. Ahmed and K. Foger, Kinetics of Internal Steam
Reforming of Methane on Ni/YSZ-Based Anodes for Solid
Oxide Fuel Cells, Catal. Today, 2000, 63, 479–487.
44 E. Nikolla, J. Schwank and S. Linic, Comparative Study of
the Kinetics of Methane Steam Reforming on Supported Ni
and Sn/Ni Alloy Catalysts: The Impact of the Formation of Ni
Alloy on Chemistry, J. Catal., 2009, 263, 220–227.
45 J. G. Xu and G. F. Froment, Methane Steam Reforming,
Methanation and Water-Gas Shift 1. Intrinsic Kinetics, AIChE
J., 1989, 35, 88–96.
46 E. Broclawik, J. Haber and L. Ungier, Electronic-Structure of
Rhenium Oxides, J. Phys. Chem. Solids, 1981, 42, 203–208.
Monoethanolamine, RSC Adv., 2018, 8, 8152–8163.
26 B. A. Rosen, E. Gileadi and N. Eliaz, Electrodeposited Re-
Promoted Ni Foams as a Catalyst for the Dry Reforming of
Methane, Catal. Commun., 2016, 76, 23–28.
27 D. Zubenko, S. Singh and B. A. Rosen, Exsolution of Re-Alloy
Catalysts with Enhanced Stability for Methane Dry
Reforming, Appl. Catal., B, 2017, 209, 711–719.
28 Y. Xu, Y. Ma, M. Demura and T. Hirano, Enhanced Catalytic
Activity of Ni3Al Foils Towards Methane Steam Reforming by
Water Vapor and Hydrogen Pretreatments, Int. J. Hydrogen
Energy, 2016, 41, 7352–7362.
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
47 A. S. Fung, P. A. Tooley, M. J. Kelley, D. C. Koningsberger
and B. C. Gates, Cationic Trirhenium Rafts on Gamma-Al2O3
- Characterization by X-Ray Absorption-Spectroscopy, J. Phys.
Chem., 1991, 95, 225–234.
55 S. R. Bare, S. D. Kelly, B. Ravel, N. Greenlay, L. King and
G. E. Mickelson, Characterizing Industrial Catalysts Using In
Situ XAFS under Identical Conditions, Phys. Chem. Chem.
Phys., 2010, 12, 7702–7711.
48 C. L. Bianchi, M. G. Cattania and P. Villa, XPS
Characterization of Ni and Mo Oxides before and after In
situ Treatments, Appl. Surf. Sci., 1993, 70–71, 211–216.
49 Y. Xu, J. Sakurai, Y. Teraoka, A. Yoshigoe, M. Demura and T.
Hirano, Initial Oxidation Behavior of Ni3Al (210) Surface
Induced by Supersonic Oxygen Molecular Beam at Room
Temperature, Appl. Surf. Sci., 2017, 391, 18–23.
50 L. Salvati, L. E. Makovsky, J. M. Stencel, F. R. Brown and
D. M. Hercules, Surface Spectroscopic Study of Tungsten-
Alumina Catalysts Using X-Ray Photo-Electron, Ion-
Scattering, and Raman Spectroscopies, J. Phys. Chem.,
1981, 85, 3700–3707.
56 H. Okamoto, Ni-Re Phase Diagram, Asm Alloy Phase Diagrams
Database, ed. P. Villars, ASM International, OH, USA, 2016.
57 G. K. Williamson and W. H. Hall, X-Ray Line Broadening
from Filed Aluminium and Wolfram, Acta Metall. Mater.,
1953, 1, 22–31.
58 Y. F. Jia, L. Duan, D. Q. Zhang, J. Qiao, G. F. Dong, L. D.
Wang and Y. Qiu, Low-Temperature Evaporable Re2O7: An
Efficient P-Dopant for Oleds, J. Phys. Chem. C, 2013, 117,
13763–13769.
59 Y. Xu, M. Demura and T. Hirano, Effect of Alkali Leaching
on the Surface Structure of Ni(3)Al Catalyst, Appl. Surf. Sci.,
2008, 254, 5413–5420.
51 A. M. Venezia, R. Bertoncello and G. Deganello, X-Ray
Photoelectron-Spectroscopy Investigation of Pumice-Supported
Nickel-Catalysts, Surf. Interface Anal., 1995, 23, 239–247.
52 J. F. Moulder, W. F. Stickle, P. E. Sobol and K. D. Bomben,
Handbook of X-Ray Photoelectron Spectroscopy, Physical
Electronics USA, Inc., Chanhassen, Minnesota, U.S.A., 1995.
53 C. J. Powell, Recommended Auger Parameters for 42 Elemental
Solids, J. Electron Spectrosc. Relat. Phenom., 2012, 185, 1–3.
54 F. U. Hillebrecht, J. C. Fuggle, P. A. Bennett, Z. Zolnierek and
C. Freiburg, Electronic-Structure of Ni and Pd Alloys 2. X-Ray
Photoelectron Core-Level Spectra, Phys. Rev. B: Condens.
Matter Mater. Phys., 1983, 27, 2179–2193.
60 Y. Xu, J. Y. Yang, M. Demura, T. Hirano, Y. Matsushita, M.
Tanaka and Y. Katsuya, Catalytic Performance of Ni-Al
Nanoparticles Fabricated by Arc Plasma Evaporation for
Methanol Decomposition, Int. J. Hydrogen Energy, 2014, 39,
13156–13163.
61 X. B. Hao, B. J. Wang, Q. Wang, R. G. Zhang and D. B. Li,
Insight into Both Coverage and Surface Structure Dependent
Co Adsorption and Activation on Different Ni Surfaces from
DFT and Atomistic Thermodynamics, Phys. Chem. Chem.
Phys., 2016, 18, 17606–17618.
62 K. Fujimoto, Catalyst Design Based on Spillover Theory,
Stud. Surf. Sci. Catal., 1993, 77, 9–16.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020