Ni-based bimetallic nano-catalysts anchored on BaZr0.4Ce0.4Y0.1Yb0.1O3?δfor internal steam reforming of methane in a low-temperature proton-conducting ceramic fuel cell

Article abstract of DOI:10.1039/d0ta11359j

This study reports the catalytic performance of 8Ni (8 wt% Ni) and 6Ni2M (6 wt% Ni, 2 wt% M (M: Co, Cu, Rh)) anchored on BaZr_0.4Ce_0.4Y_0.1Yb_0.1O_3?δ, an anode backbone material of proton-conducting ceramic fuel cells (PCFCs), for steam reforming of methane at low temperatures (350-550 °C). Results show that all catalysts have coherent structural properties and form bimetallic alloys. Their catalytic activities are evaluated at various temperatures, steam-to-carbon ratios, and gas flow rates. It is shown that 6Ni2Rh has the highest catalytic activity under all operating conditions. 6Ni2Rh and 6Ni2Co exhibit higher methane conversion and hydrogen yield than 8Ni even at low steam-to-carbon ratios. Their high activities make them less dependent on gas flow rate. They show higher resistance to carbon formation and maintain their catalytic activities during long-term operation. 6Ni2Rh and 6Ni2Co can respond to diverse operating conditions of direct methane-fueled PCFCs while maintaining high catalytic activity and stability.

Full text of DOI:10.1039/d0ta11359j

Volume 9
Number 10
1
4 March 2021
Pages 5871–6592
Journal of
Materials Chemistry A
Materials for energy and sustainability
rsc.li/materials-a
ISSN 2050-7488
PAPER
Jongsup Hong et al.
Ni-based bimetallic nano-catalysts anchored on
BaZr0.4Ce0.4Y0.1Yb0.1O3−δ for internal steam reforming
of methane in a low-temperature proton-conducting
ceramic fuel cell

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
Ni-based bimetallic nano-catalysts anchored on
BaZr Ce Y Yb O for internal steam
reforming of methane in a low-temperature
0
.4
0.4 0.1
0.1 3ꢀd
Cite this: J. Mater. Chem. A, 2021, 9,
139
6
proton-conducting ceramic fuel cell†
Kyungpyo Hong,
and Jongsup Hong
Stephanie Nadya Sutanto,
*
Jeong A. Lee
This study reports the catalytic performance of 8Ni (8 wt% Ni) and 6Ni2M (6 wt% Ni, 2 wt% M (M: Co, Cu, Rh))
anchored on BaZr0.4Ce0.4
Y
0.1Yb0.1
O
3ꢀd, an anode backbone material of proton-conducting ceramic fuel
ꢁ
cells (PCFCs), for steam reforming of methane at low temperatures (350–550 C). Results show that all
catalysts have coherent structural properties and form bimetallic alloys. Their catalytic activities are
evaluated at various temperatures, steam-to-carbon ratios, and gas flow rates. It is shown that 6Ni2Rh
has the highest catalytic activity under all operating conditions. 6Ni2Rh and 6Ni2Co exhibit higher
methane conversion and hydrogen yield than 8Ni even at low steam-to-carbon ratios. Their high
activities make them less dependent on gas flow rate. They show higher resistance to carbon formation
and maintain their catalytic activities during long-term operation. 6Ni2Rh and 6Ni2Co can respond to
diverse operating conditions of direct methane-fueled PCFCs while maintaining high catalytic activity
and stability.
Received 21st November 2020
Accepted 10th February 2021
DOI: 10.1039/d0ta11359j
rsc.li/materials-a
overall cost by reducing the energy, system complexity, and
costs required to obtain hydrogen from an external reformer
1
. Introduction
A direct methane-fueled proton-conducting ceramic fuel cell installed outside a PCFC stack. PCFCs have been shown to
(
PCFC) has been highlighted as an alternative energy tech- tolerate carbon coking and impurity poisoning at intermediate
ꢁ
nology to tackle the global energy and environmental chal- temperatures (500–600 C) with direct internal steam reforming
13
lenges. Ceramic fuel cells generally operate at high temperature of methane. However, the sluggish kinetics for internal steam
ꢁ
(
>700 C) and have been gaining attention for the past few years reforming of methane and insufficient hydrogen supply at low
1–4
due to their high efficiency and impurity resistance. While temperature may prevent optimum power production. Given
they are feasible for small- to large-scale applications, several that steam reforming of methane is strongly endothermic (i.e.,
0
ꢀ1
technical issues should be resolved in order to commercialize its heat of reaction is DH298K ¼ +206 kJ mol ), lowering the
these ceramic fuel cells. Particularly, they require high temperature decreases its equilibrium constant and kinetic
production costs and are susceptible to performance degrada- rates, which is highly important in determining fuel cell oper-
14
tion due to their high operating temperature. This makes it ating conditions, methane conversion, and hydrogen yield.
crucially important to decrease the operating temperature of Therefore, a novel catalyst is required in the PCFC anode to
4–6
ꢁ
ceramic fuel cells. In terms of low temperature (350–550 C) overcome such a barrier, ensuring the best performance at low
operation, PCFCs theoretically work better than solid oxide fuel temperature.
cells (SOFCs) mainly because they rely on proton conduction
Addition of highly active catalysts to the existing anodes is
which generally has a lower activation energy than oxygen ion crucial to facilitating an efficient internal steam reforming of
5,7–12
conduction of SOFCs.
ceramic fuel cells, PCFCs need to support internal hydrocarbon (BaZr0.4Ce0.4
e.g., CH
) reforming in their anodes at such low temperatures. gated to overcome the performance issues arising at low
The internal steam reforming of methane can decrease the temperature while ensuring sufficient durability when being
In the meantime, to take advantage of methane in PCFCs at low temperature. Recently, the Ni-BZCYYb
3ꢀd) anode has been extensively investi-
Y
0.1Yb0.1O
(
4
15
exposed to hydrocarbons. Ni supported by BZCYYb may
function as an active catalyst for internal steam reforming of
School of Mechanical Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu,
ꢁ
7,16,17
Seoul 03722, South Korea. E-mail: jongsup.hong@yonsei.ac.kr; Fax: +82 2 362 methane at 600 C or higher.
Thanks to its cost competi-
2
†
1
736; Tel: +82 2 2123 4465
tiveness, conventional metal catalysts for industrial steam
reforming of methane are largely based on nickel as an active
Electronic supplementary information (ESI) available. See DOI:
0.1039/d0ta11359j
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. A, 2021, 9, 6139–6151 | 6139

View Article Online
Journal of Materials Chemistry A
Paper
18–22
metal.
Besides, nickel has a relatively high methane obtained from this study suggest a candidate catalyst with high
conversion efficiency and good H selectivity at high tempera- potential and feasibility as a PCFC anode material.
2
ꢁ
23
ture (>700 C). However, despite many advantages of the Ni
catalyst, its catalytic activity is substantially reduced at low
temperature, and its use is restricted due to catalyst oxidation
2
. Experimental
3,14,24,25
and carbon deposition.
To overcome these problems, it is 2.1. Catalyst preparation
necessary to alloy Ni with a promoter metal that can enhance
low-temperature reforming kinetics and resistance to carbon
deposition. For example, many studies of hydrocarbon utiliza-
tion have reported that the formation of a Ni–Co or Ni–Cu alloy
reduces the carbon formation on the catalyst surface and
Monometallic Ni and Ni-based bimetallic catalysts were
prepared by using the wet impregnation method and subse-
quent high-temperature reduction process. The powder of the
support material BZCYYb (Kceracell, average size: 0.8 mm) was
mixed with ethanol used as a solvent to avoid secondary phase
formation. The volumetric ratio of the support to the solvent
was 3 : 7. The dispersant (HypermerTM KD-6, Croda) was mixed
14,24,26,27
improves catalytic performance.
The Co contained in the
Ni–Co alloy catalyst adsorbs O* and OH*, preventing carbon
14,28
formation on its surface and improving catalytic activity.
The Ni–Co alloy catalyst uses oxygen surface diffusion to oxidize
with 3 wt% support to disperse it uniformly in the solvent
42
3 2 2 3 3
mixture. Metal nitrates including Ni(NO ) $6H O, Rh(NO ) -
29,30
the sulfur adsorbed on its surface, ensuring sulfur tolerance.
$
xH O, Cu(NO ) $3H O, and Co(NO ) $6H O (Sigma-Aldrich)
2 3 2 2 3 2 2
On the other hand, the Cu contained in the Ni–Cu alloy catalyst
highly accelerates the water gas shi reaction, producing more
were used to prepare metal precursors. The metal loading of
all catalysts was xed at 8 wt% with a Ni to M (i.e., Co, Cu, Rh)
weight ratio of 3 to 1. These catalysts are named 8Ni (mono-
metallic catalyst) and 6Ni2M (bimetallic catalyst). The metal
precursor was mixed with BZCYYb in the solvent and stirred for
14,24
hydrogen, and suppresses coke deposition.
It also has sulfur
resistance because it weakens sulfur binding while reducing the
31
adsorption energy of atomic sulfur. Bimetallic Ni–Rh catalysts
have more carbon coking resistance and sulfur tolerance than
4
hours at room temperature. Then, the solvent was removed by
32–36
the monometallic Ni catalyst.
Moreover, alloying nickel with
ꢁ
drying at 100 C overnight. Subsequently, the mixtures were
calcined at 400 C in air for 1 hour in a furnace. To make an
noble metals such as Rh has been proven to improve the cata-
ꢁ
14
2
lytic activity and selectivity towards H . The abovementioned
alloy structure, the as-prepared (unreduced) catalysts were
metal promoters (i.e., Co, Cu, Rh) are applicable to internal
steam reforming of methane in the PCFC anode at low
temperature since these metals have the potential of oxygen
affinity, carbon and sulfur resistance, and reforming of hydro-
ꢁ
heated from room temperature to 700 C with a heating ramp of
ꢁ
ꢀ1
1
0 C min while being reduced by using a mixed gas (10% H
2
in Ar (molar basis)) of 100 sccm and maintained for 2 hours
under such environment.
24,29–31,36–38
carbons and oxygenates.
With commercialization in
mind, although the cost efficiency from a material point of view
may decrease by adding a promoter catalyst to pure Ni, Ni-based
2.2. Catalyst characterization
bimetallic alloys anchored on BZCYYb can be an effective The structural properties and alloy formation of the 8Ni and
option to guarantee sufficient internal steam reforming of 6Ni2M catalysts were examined by Brunauer–Emmett–Teller
methane at low temperature with substantial durability. In (BET), CO pulse chemisorption, X-ray diffraction (XRD), trans-
addition, the process of applying these additional catalysts to Ni mission electron microscopy (TEM), and temperature-
at the anode of PCFCs is simple without complex programmed reduction (TPR) methods. The BET analysis was
manufacturing procedures using an inltration method in conducted using a Quadrasorb SI (Quantachrome Instruments).
which a solution of metal precursor is injected into the sintered Prior to the BET analysis, each catalyst sample was thermally
39–41
ꢁ
anode.
effect of Ni-based bimetallic alloys anchored on BZCYYb on moisture and other contaminants adsorbed on the catalyst
internal steam reforming of methane at low temperature.
surface. Subsequently, the specic surface area of the pretreated
However, to date, there exist no attempts to study the pretreated in a vacuum for 15 hours at 250 C to remove
This study aims to examine the catalytic activity and dura- samples was measured by N2 adsorption at 77 K. The XRD
bility of Ni-based bimetallic alloys anchored on BZCYYb for patterns of the catalysts were recorded by using a D8 Advance
internal steam reforming of methane at low temperature, and (Bruker) using Cu Ka radiation to obtain the crystal structure.
suggests a novel catalyst to overcome the aforementioned The crystal structure of the catalysts was scanned with a step
ꢁ
ꢀ1
ꢁ
challenges. In this study, additional active metal catalysts such size of 0.02 s in the 2q range ¼ 25 to 80 . The TPR experiment
as Co, Cu, and Rh are introduced into Ni-BZCYYb, which forms was performed using 0.1 g of the unreduced catalyst sample
Ni-based bimetallic alloys anchored on BZCYYb. Their micro- placed in a U-tube reactor using an Autochem 2920 (Micro-
structural and surface characteristics are examined by the meritics) which includes a thermal conductivity detector (TCD).
Brunauer–Emmett–Teller method (BET), CO pulse chemisorp- Inert gas He of 50 sccm was injected into the reactor at room
ꢁ
tion, X-ray diffraction (XRD), temperature programmed reduc- temperature. Continuously, the sample was heated up to 200 C
ꢁ
tion (TPR), and transmission electron microscopy (TEM). Their for 1 hour and then cooled below 30 C to obtain a clear solid
catalytic activity and long-term stability for low-temperature surface of the catalysts. Aer pretreatment, the gas was changed
steam reforming of methane are investigated by product gas to an active gas (10% H in Ar (molar basis)) of 50 sccm, and
2
ꢁ
analysis using gas chromatography, which elucidates the effect then the temperature was raised up to 700 C at a heating rate of
ꢁ
ꢀ1
of the additional metal catalyst alloyed with Ni. The results 10
C
min
while measuring hydrogen consumption
6140 | J. Mater. Chem. A, 2021, 9, 6139–6151
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry A
calculated by TCD. The CO pulse chemisorption experiment was respectively. The gas hourly space velocity, GHSV, can be
carried out aer the TPR experiment. Aer decreasing the expressed using eqn (3). Eqn (4) represents the degradation rate
ꢁ
sample temperature to 35 C, He gas was used to purge the gas of active catalysts during a long-term stability test, which indi-
adsorbed on the sample surface. Subsequently, an active gas cates the degree of carbon resistance.
(
10% CO in He (molar basis)) in the loop was supplied and
Q
CO þ QCO2
CH4;conv
¼
ꢂ 100
(1)
(2)
pulsed repeatedly until adsorbed CO molecules saturated the
catalyst surface. TEM analysis was performed to examine the
structure and particle size of the catalysts, and energy dispersive
X-ray spectrometry (EDS) was also conducted to obtain the
information on chemical composition using a Talos F200X
F
CH
4
Q 2
H
H
Y 2
¼ 2F
ꢂ 100
CH4 þ FH O
2
(
FEI).
Ftotal ꢂ 60
GHSV ¼
(3)
(4)
Vol
2
.3. Catalyst activity test
CH4;conv;0 ꢀ CH4;conv;t
CH4;conv;0
Degradation ¼
ꢂ 100
The catalytic activity of the catalysts for steam reforming of
methane was evaluated under various operating conditions
including temperature, steam-to-carbon (S/C) ratio, and feed
gas ow rate. The test was carried out by using a packed-bed
quartz tube reactor (Din ¼ 1.1 cm) in a furnace containing
a K-type thermocouple at atmospheric pressure, as shown in
Fig. 1. To increase the accuracy of the experiment, 0.3 g of the
unreduced catalyst was placed in the vicinity of the thermo-
couple inside the furnace. The height of catalysts in the reactor
where CH4,conv: methane conversion [%]; YH2: hydrogen yield
ꢀ
1
[
%]; GHSV: gas hourly space velocity [h ]; Degradation: the
degree of methane conversion reduction compared to its initial
value; Q : the volumetric ow rate of species i at the outlet
sccm]; F : the volumetric ow rate of species i at the inlet
sccm]; Vol: catalyst volume [cm ]; CH4,conv,0: the methane
i
[
[
i
3
conversion at the initial time; CH4,conv,t: the methane conver-
sion at time t. Given the methane conversion expressed by
a reaction rate coefficient, k, in an Arrhenius form and the
3
was nearly 0.21 cm with a catalyst volume of 0.2 cm . The ow
rate of gases including H , CH , and Ar was controlled using
2
4
i
partial pressure of reactants, P ,
a mass ow controller (MKS Type 1179A). The partial pressure
of steam in the humidier was monitored using the saturation
temperature to control the steam concentration in the feed gas.
The composition of gas products exiting the reactor was
measured by gas chromatography (Agilent 7890B) equipped
with a dual TCD. The dual TCD was connected to two columns
ꢀ
ꢀ
À
ꢁ
Á
a
Pref PH2O Pref
À
ꢁ
Á
b
ꢂ
k P
CH4
lnðCH4convÞ ¼ ln
F
CH
4
À
ꢁ
Á
ꢂ
ꢀ
ꢂ
a
P
CH4
P
ref
Ea
R
1
À
ꢁ
Á
¼
ln
ꢀ
þ b ln PH O
P
ref
2
X
CH
T
4
ꢀ
ꢂ
A ꢂ 60
Vol ꢂ GHSV
(
HP-PLOT 5A: H , CO, and HP-PLOT-Q: CH , CO ) which sepa-
2 4 2
þ ln
rate the mixed gases, respectively.
The methane conversion, CH4,conv, and hydrogen yield, YH2,
where A: pre-exponential factor [sccm]; E : activation energy [kJ
a
are calculated using the dry gas composition analyzed from gas
chromatography and can be expressed using eqn (1) and (2),
ꢀ1
ꢀ1 ꢀ1
mol ]; R: universal gas constant [J mol
K ]; T: temperature
Fig. 1 Schematic diagram of the catalytic activity test facility equipped with online gas chromatography.
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. A, 2021, 9, 6139–6151 | 6141

View Article Online
Journal of Materials Chemistry A
Paper
Table 1 Catalytic activity and long-term stability test conditions
while maintaining the temperature and S/C ratio constant at
ꢁ
5
00 C and 1, respectively. The long-term stability was evaluated
Temperature Steam-to-
Gas ow rate [sccm]
(GHSV [h ])
ꢁ
ꢀ
1
by using harsh conditions (temperature: 500 C, gas composi-
tion: methane with 9.1% steam, gas ow rate: 50 sccm) which
induce carbon formation.
ꢁ
Test type
[ C]
carbon ratio
Temperature
Steam-to-
carbon ratio
350–550
500
S/C ¼ 2
100 (30 000)
S/C ¼ 2, 1, 0.5 100 (30 000)
3
. Results and discussion
Gas ow rate 500
Stability 500
S/C ¼ 1
150 (45 000),
00 (30 000), 50 (15 000)
50 (15 000)
1
3.1. Catalyst characterization
S/C ¼ 0.1
The coherent structural properties of the 8Ni and 6Ni2M cata-
lysts, as summarized in Table 2, imply the reliability of the
synthesis method and provide the same basis for catalytic
[
K]; XCH4: mole fraction of methane in the feed gas; a: the order performance evaluation. The specic surface area of the cata-
of reaction with respect to PCH4; b: the order of reaction with lysts was examined by BET analysis. The bare support BZCYYb
2
ꢀ1
respect to PH2O. Note that the rst term on the right hand side is has a specic surface area of 27 m
g , while the 8Ni and
2
ꢀ1
maintained constant throughout the catalytic activity test. The 6Ni2M catalysts have specic surface areas of 8–9 m g . The
effect of the second, third, and fourth term on the right hand lower specic surface area of the catalysts is attributed to the
side will be elucidated in Section 3.2.1, 3.2.2, and 3.2.3, high metal loading amount (8 wt%) and the pore blockage by
20
respectively.
impregnation. If larger catalyst particles are deposited on the
The effects of temperature, S/C ratio, and feed gas ow rate support, it may induce substantial pore blockage and lower gas
on the catalytic activity were investigated, followed by a long- diffusion rates, particularly at the anode of a full cell. Never-
term stability test. Prior to the test, the catalyst samples were theless, given that the specic surface areas of these catalysts
subjected to a high-temperature reduction process forming an show little deviation, it can be argued that their catalytic activ-
in situ alloy. The operating conditions for the activity test and ities can be compared on the same basis of surface sites. This
long-term stability test conditions considered in this study are argument is further strengthened by the surface information of
summarized in Table 1. The carbon formation regime in equi- the catalysts for the metal formed on the support, obtained
librium under these operating conditions was conrmed from through CO pulse chemisorption experiments. As evidenced in
the ternary diagram shown in Fig. S1.† In equilibrium, the S/C Table 2, the metallic particle diameter and the metallic surface
2
ratio of 2 makes the reaction environment free of carbon area are in the range of 23–38 nm and 1.4–2.2 m
M
/gcat,
formation (Fig. S2(a)†), but it falls in the carbon formation respectively, for these catalysts. The metal dispersion is in the
ꢁ
regime at the S/C ratio less than 1 at 500 C (Fig. S2(b)†). The range of 2.7–4.5%. All these metallic structural indices show
equilibrium values under given operating conditions were similar values with respect to various metal species impreg-
calculated by using the soware NASA CEA (at T, P ¼ constant). nated on the support, which indicates the reproducibility and
Throughout the catalytic activity test, the molar fraction of reliability of catalyst synthesis. Along with the consistent
methane and operating pressure were maintained constant at specic surface area, these metallic structural properties with
2
0% and 1 atm, respectively. The effect of temperature was small deviation provide the same ground on which catalytic
elucidated by changing the temperature from 350 C to 550 C activities can be investigated.
ꢁ
ꢁ
with a feed gas ow rate of 100 sccm and an S/C ratio of 2. The
In addition to the coherent structural properties, all bime-
effect of the S/C ratio was examined by changing it from 2 to 0.5 tallic catalysts form an alloyed state between Ni and promoter M
while maintaining the gas ow rate and temperature constant at (M: Co, Cu, Rh) metal. Given that the performance of the Ni-
ꢁ
1
00 sccm and 500 C, respectively. The effect of the feed gas ow based bimetallic catalysts relies heavily on the formation of
rate was investigated by changing it from 150 sccm to 50 sccm an alloyed state between Ni and promoter M metal, XRD anal-
ysis is crucial to determining whether the alloy is properly
formed during the fabrication process. Fig. 2 shows the overall
Table 2 Structural properties (i.e., specific surface area, metallic XRD patterns of the bare support BZCYYb, 8Ni and 6Ni2M
particle diameter, metallic surface area, and metal dispersion) of 8Ni
and 6Ni2M catalysts
ꢁ
catalysts. There is no peak in BZCYYb between 43 and 45 ,
whereas the catalysts have a metal peak in this peak range,
which is highlighted and magnied using a red dotted box in
Fig. 2. It can be evidenced that the XRD peaks of the 6Ni2M
catalysts are located at a lower degree than those of the 8Ni
catalyst. To elucidate this further, the monometallic (8M)
catalysts were fabricated in the same manner described above
and were characterized under the same conditions to compare
with the 8Ni and 6Ni2M catalysts (refer to Fig. S3†). Each of the
metal peaks observed in Fig. 2 is compared with the mono-
metallic (8M) peak, as can be seen in Fig. S3.† It can be evi-
denced that the metal peak in the bimetallic (6Ni2M) catalysts
Specic
Metal
a
b
surface area
Metallic particle Metallic surface dispersion
2
ꢀ1
b
b
2
Sample [m g
]
diameter [nm] area [m
M
/gcat] [%]
BZCYYb 27
—
27
32
38
23
—
—
8
6
6
6
Ni
9
2.0
1.7
1.4
2.2
3.8
3.1
2.7
4.5
Ni2Co 8
Ni2Cu 9
Ni2Rh 9
a
b
BET. CO pulse chemisorption.
6142 | J. Mater. Chem. A, 2021, 9, 6139–6151
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 2 X-ray diffraction patterns of BZCYYb ( ), 8Ni ( ), and 6Ni2M (M:
peaks: Ni, Ni–Co alloy, Ni–Cu alloy, Ni–Rh alloy.
Co,
Cu,
Rh) catalysts. Support peak: BZCYYb(4411); metal
shis towards the middle of the two metal peaks observed in the 6Ni2Co catalyst is an alloy and is well supported on BZCYYb. In
corresponding monometallic (8Ni and 8M) catalysts, which the case of the 6Ni2Cu catalyst, Fig. S4(c)† shows the same trend
conrms the successful alloy formation. Specically, as shown as that for the 6Ni2Co catalyst. As shown in Fig. S4(d),† the
ꢁ
in Fig. S3(a),† the alloyed metal peak of 6Ni2Co appears at 44.6
6Ni2Rh catalyst has the same tendency of uniformity as that of
ꢁ
in between the 8Ni metal peak at 44.7 (Ni(111), PDF 04-001- other bimetallic catalysts (6Ni2Co and 6Ni2Cu). The size of the
ꢁ
3
8
156) and 8Co metal peak at 44.4 (Co(111), PDF 04-006- alloyed metal particles is smaller (less than about 20 nm) than
43–45
ꢁ
067).
Likewise, the alloy peak of 6Ni2Cu is located at 44.4
that of other catalysts. Overall, the metal particle sizes of the
ꢁ
ꢁ
which is between 8Ni (44.7 ) and 8Cu (43.4 , Cu(111), PDF 04- catalysts obtained from the TEM images are similar to the
46,47
003-5318) metal peaks, as shown in Fig. S3(b).†
In the same metallic particle diameter provided in Table 2.
The alloy formation enhances the reducibility of the 6Ni2M
4.3 between that of 8Ni (44.7 ) and that of 8Rh (41.2 , Rh(111), catalysts compared to that of the 8Ni catalyst. Fig. 4 shows the
way, Fig. S3(c)† shows that the alloy peak of 6Ni2Rh exists at
ꢁ
ꢁ
ꢁ
4
48
PDF 04-016-1279). Therefore, considering the peak location of results of TPR experiments that examine the reducibility of the
the alloyed metal catalysts, it can be discussed that all bime- 8Ni and 6Ni2M catalysts. The 8Ni catalyst has a reduction
tallic (6Ni2M) catalysts formed alloys, which is further demon- temperature range of 220–340 C in which Ni shows a peak at
ꢁ
ꢁ
strated by TEM analysis as follows.
338 C, which is associated with the reduction of relatively free
The structure analysis of all catalysts by TEM conrms again NiO particles. The broad peak at higher temperature (340–500
ꢁ
20,49
their alloy formation and demonstrates nano-sphere particles
C) results from a strong interaction of Ni with the support.
anchored on the BZCYYb support. EDS mapping was performed The support (BZCYYb) located at the bottom shows no reduc-
to analyze the chemical composition of the metal distribution tion. When a bimetallic catalyst is formed, the degree of
and alloy structure. As shown in Fig. 3, the 8Ni and 6Ni2M reduction changes due to the interaction between alloyed
44,50
catalysts are in the form of nano-sphere particles of size less metals.
It can be observed that the reduction temperature of
than about 30 nm. In addition, EDS mapping analysis conrms the 6Ni2M catalysts is located at a lower temperature than that
that the metal particles are anchored on BZCYYb consisting of of the 8Ni catalyst. In addition, since the 6Ni2M catalysts
large and heavy atoms (Ba, Zr, Ce, Y, Yb, O). Moreover, nickel formed an alloy as veried by the results of XRD, they show one
27,51
and promoter metals (M: Co, Cu, Rh) are found to be located at stage reduction like the reduction prole of 8Ni.
To elucidate
the same location, which claries the fact that two different further the effect of alloy formation on the reducibility, the
metals were successfully alloyed. Fig. S4† provides extra infor- monometallic (8M) catalysts were characterized under the same
mation which helps to check the size and uniformity of the conditions to compare with the 8Ni and 6Ni2M catalysts, as
metal particles in more detail. In Fig. S4(a),† the Ni particles of shown in Fig. S5.† It can be seen that the reducibility of the
the 8Ni catalyst are generally 10–30 nm in size and are evenly 6Ni2M catalysts is improved compared to that of the 8Ni cata-
spread on the support. As shown in Fig. S4(b),† 6Ni2Co has lyst, attributed to the alloyed structure of the 6Ni2M catalysts. In
ꢁ
a metal particle size and distribution similar to 8Ni. In addition, the reduction prole of 8Co in Fig. S5(a),† the rst peak (249 C)
the positions of Ni and Co are the same and they are arranged is the reduction step from Co
on the support elements. Thus, it can be inferred that the (318 C) is the reduction step from CoO to Co. In the case of
3
O
4
to CoO, and the second peak
ꢁ
52
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. A, 2021, 9, 6139–6151 | 6143

View Article Online
Journal of Materials Chemistry A
Paper
Fig. 3 TEM and EDS mapping images of (a) 8Ni; (b) 6Ni2Co; (c) 6Ni2Cu; (d) 6Ni2Rh.
6
Ni2Co, since Ni and Co are alloyed with each other, the reforming of methane at low temperature were evaluated. To be
ꢁ
temperature of the reduction peak is at 325 C which is between applicable in the anode of PCFCs, catalysts should have high
that of 8Ni and 8Co.
peaks at 257 C and 299 C. The peak at lower temperature is the cient hydrogen reformed internally from methane and steam.
reduction of the copper oxide cluster, and the other peak at In this regard, experiments were performed under conditions of
higher temperature is associated with bulk copper.
Affected by the interaction of both Ni and Cu, the reduction ratios (0.5–2), and gas ow rates (50–150 sccm), all of which
4
4,50
In Fig. S5(b),† 8Cu has two reduction activity and high hydrogen yield so that they can supply suffi-
ꢁ
ꢁ
47,53–55
ꢁ
various operating temperatures (350–550 C), steam-to-carbon
ꢁ
46
peak of 6Ni2Cu is located at 332 C. Fig. S5(c)† shows that the are feasible operating conditions for PCFC unit-cells.
Rh in the 6Ni2Rh catalyst signicantly improves the reducibility
3.2.1. Effect of temperature. The 6Ni2Rh catalyst shows
56,57
of Ni through the synergy effect of the alloy.
The peak at superior activity with substantial hydrogen yield at all temper-
ꢁ
1
39 C in the 8Rh reduction prole represents the reduction of atures compared to 8Ni and other 6Ni2M catalysts, attributed to
ꢁ
Rh O . The peak of 6Ni2Rh at 240 C is due to the interplay of its low activation energy for the overall reaction. The rst
2
3
Rh and Ni.
activity test was carried out to elucidate the effect of tempera-
ture (350–550 C) at a ow rate of 100 sccm and an S/C ratio of 2,
ꢁ
which falls in a carbon-free eld at equilibrium. As shown in
Fig. 5(a), the order of methane conversion and hydrogen yield is
3
.2. Catalytic activity test
Based on the conrmed alloy structure with coherent nano-
scale particles, the catalytic activities of 6Ni2M for steam
6Ni2Rh > 8Ni > 6Ni2Co > 6Ni2Cu in all temperature regions. All
6144 | J. Mater. Chem. A, 2021, 9, 6139–6151
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 4 Temperature-programmed reduction profiles of BZCYYb ( ), 8Ni ( ) and 6Ni2M (M:
Co,
Cu,
Rh) catalysts.
catalysts tend to decrease methane conversion and hydrogen adsorbed to cobalt contained in 6Ni2Co at this S/C ratio, which
yield as the temperature is lowered, during which the reduction makes the surface of 6Ni2Co poisoned by O*, as reported by
28
rate of 6Ni2Rh is substantially smaller than that of other cata- Jones et al. It can be inferred that O* poisoning on the 6Ni2Co
lysts. Particularly, 6Ni2Rh exhibits catalytic activity even below surface hinders the dissociative adsorption of methane on its
ꢁ
4
00 C where 8Ni and other 6Ni2M catalysts are inactive. It surface compared to the 8Ni catalyst. Therefore, the methane
maintains not only methane conversion, but also hydrogen conversion and hydrogen yield of 6Ni2Co could be lower than
yield which is critical for PCFC operation. This implies that those of 8Ni under steam rich conditions, which is further
PCFCs may potentially expand their operating regime below elucidated in the following section. In the case of 6Ni2Cu, the
ꢁ
4
00 C, with the aid of the Ni–Rh bimetallic alloy catalyst methane conversion and hydrogen yield are signicantly lower
fabricated in their anode. Such a dependence on temperature is than those of other catalysts due to its highest activation energy.
further explained by their activation energy, as shown in Bian et al. reported that methane is adsorbed predominantly on
Fig. 5(b). It can be observed that the order of activation energy is the copper surface since the copper contained in 6Ni2Cu has
ꢀ
1
opposite to that of methane conversion: 6Ni2Rh (39.3 kJ mol
)
not only lower surface energy than nickel, but also slow kinetics
ꢀ1
ꢀ1
<
8Ni (E
a
(1) ¼ 68.6 kJ mol , E (2) ¼ 150.5 kJ mol ) < 6Ni2Co of methane dissociation, which suppresses the methane acti-
a
ꢀ
1
ꢀ1
37
(E
a
(1) ¼ 92.9 kJ mol , E
a
(2) ¼ 160.1 kJ mol ) < 6Ni2Cu (E
a
(1) ¼ vation. An et al. also reported that the activation energy for
ꢀ1
ꢀ1
110.5 kJ mol , E
a
(2) ¼ 174.8 kJ mol ). The activation energy of methane dissociation on the (111)Cu/Ni surface is about 1.3
59
6
Ni2Rh is 1.7–4.4 times less than that of 8Ni and other 6Ni2M times larger than that on the (111)Ni surface. In this study, the
catalysts. In particular, its activation energy is consistently low activation energy of 6Ni2Cu is also approximately 1.3 times
throughout the temperature, while that of other catalysts larger than that of 8Ni, from which it can be inferred that Ni–Cu
ꢁ
increases substantially below 500 C (i.e., E (1) to E (2) shown in alloy formation reduces the methane dissociation rate and
a
a
Fig. 5(b)). The low activation energy of 6Ni2Rh supports its hence methane activation.
catalytic activity, especially in the low temperature regime. Note
Based on the results, it can be discussed that further catalytic
that the calculated activation energy for the dissociation of activity testing should be needed at other S/C ratios (i.e., partial
methane (i.e., rate-limiting step) on the (111) Ni surface, re- pressure of steam) at low temperature. The operating temper-
5
8
ꢀ1
ꢁ
ported by Nikolla et al., is in the range of 70–125 kJ mol , into ature of 500 C seems appropriate for the following study given
which the activation energy of 8Ni falls. It can be inferred that that relatively sufficient or measurable catalytic activities are
ꢁ
the Ni–Rh bimetallic alloy formation contributes to lowering observed at around 500 C below which the activation energies
the activation energy of the rate-limiting step of the Ni mono- of 8Ni, 6Ni2Co, and 6Ni2Cu increase. Indeed, a number of
metallic catalyst, promoting methane conversion and hydrogen previous studies related to low temperature SOFCs and
5
,8,13,15,60
ꢁ
ꢁ
yield in the low temperature regime.
PCFCs
have been performed at 500 C. In this regard,
The 6Ni2Co and 6Ni2Cu catalysts exhibit comparable or the next experiments were conducted at 500 C with various S/C
lower catalytic activity in comparison with the 8Ni catalyst due ratios, as follows.
to their high activation energy for the onset of steam reforming.
The methane conversion and hydrogen yield of 6Ni2Co and test was performed to verify the catalytic activity of 6Ni2M
Ni2Cu are lower than those of 8Ni, as shown in Fig. 5(a), due to catalysts at different S/C ratios (i.e., 2, 1, 0.5). From the
their high activation energy evidenced in Fig. 5(b). In the case of perspective of PCFC system integration, a low S/C ratio is pref-
Ni2Co, O* and OH* produced from H O could be strongly erable due to low thermal energy requirement for steam supply,
3.2.2. Effect of steam-to-carbon ratio. The second activity
6
6
2
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. A, 2021, 9, 6139–6151 | 6145

View Article Online
Journal of Materials Chemistry A
Paper
Fig.
6
Effect of steam-to-carbon ratio on steam reforming of
methane by 8Ni and 6Ni2M catalysts. (a) CH
rium, CH : 20%, H O: 40%, Ar: 40% (H O/CH
O: 20%, Ar: 60% (H O/CH : 20%, H
¼ 1); CH
O/CH
¼ 0.5)) and hydrogen yield ( ); (b) the order of reaction with
respect to PH2O 8Ni, 6Ni2Co, 6Ni2Cu, 6Ni2Rh). Variable: S/C
4
conversion (
¼ 2); CH
O: 10%, Ar: 70%
equilib-
Fig. 5 Effect of temperature on steam reforming of methane by 8Ni
4
2
2
4
4
: 20%,
and 6Ni2M catalysts under carbon-formation-free conditions. (a) CH
4
H
2
2
4
4
2
conversion ( equilibrium, 8Ni, 6Ni2Co, 6Ni2Cu, 6Ni2Rh) and
hydrogen yield ( 8Ni, 6Ni2Co, 6Ni2Cu, 6Ni2Rh); (b) activation
(H
2
4
(
energy ( 8Ni, 6Ni2Co, 6Ni2Cu, 6Ni2Rh). Variable: temperature
ratio ¼ 2, 1, 0.5. Fixed conditions: pressure ¼ 1 atm, temperature ¼
ꢁ
¼
350–550 C. Fixed conditions: pressure ¼ 1 atm, S/C ratio ¼ 2 (CH
4
:
ꢁ
5
00 C, gas flow rate ¼ 100 sccm.
2
0%, H
2
O: 40%, Ar: 40% (H
2
O/CH
4
¼ 2)), gas flow rate ¼ 100 sccm.
concentration given its resistance to oxide formation. Fig. 6(a)
shows that methane conversion and hydrogen yield of 6Ni2Rh
are higher than those of 8Ni, as the S/C ratio is raised. Partic-
ularly, 6Ni2Rh's activity is much stronger than that of 8Ni at the
S/C ratio of 2. The change in 6Ni2Rh's methane conversion and
hydrogen yield is more evident when the S/C ratio is raised from
3
2
resulting in a high energy conversion efficiency. In addition,
the support BZCYYb may undergo phase change when being
exposed to high partial pressure of steam, making it necessary
42
to reduce the steam concentration in its anode. In this regard,
the activity test was conducted by decreasing the S/C ratio to
elucidate its effect on the catalytic performance. The other
1
to 2. This is further evidenced by its order of reaction with
ꢁ
operating conditions including a temperature of 500 C and
respect to PH O, as shown in Fig. 6(b). It can be observed that
2
a gas ow rate of 100 sccm were maintained constant. Note that
eqn (2) implies that the hydrogen yield varies with the hydrogen
production rate, which is dependent on methane conversion,
and the amount of hydrogen-containing reactants (i.e., CH
O), dictated by the S/C ratio. Their relative variation is
examined in the following.
6Ni2Rh has a b(2) higher than b(1), whereas those of 8Ni change
in the opposite way. The high activity of 6Ni2Rh provides
a reaction rate as high as the increase of reactants, which results
in a high hydrogen yield and b along with an increase of the S/C
ratio. On the other hand, the increase of 8Ni's reaction rate does
not match the increase of S/C ratio, in particular from 1 to 2,
making the hydrogen yield and b to decrease above the S/C ratio
4
and
H
2
The 6Ni2Rh catalyst maintains a high methane conversion
and hydrogen yield at all S/C ratios, in particular at a high steam
6146 | J. Mater. Chem. A, 2021, 9, 6139–6151
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry A
of 1. It can be inferred that such a high partial pressure of steam gas ow rates (i.e., 150, 100, 50 sccm) to examine the effect of
forms an oxidizing environment, where the oxidation of 8Ni the feed gas ow rate on their catalytic activity. Other conditions
6
1–63
occurs and lowers the activity.
However, the rhodium con- including the temperature and the S/C ratio were maintained
ꢁ
tained in 6Ni2Rh has the ability of hydrogen spill-over, which constant at 500 C and 1, respectively.
inhibits the formation of metal-oxide and maintains its metallic
The methane conversion and hydrogen yield of 6Ni2Rh are
64
form even at a high S/C ratio. This feature corresponds to the maintained at high values even at a large gas ow rate, and
reducibility of 6Ni2Rh and 8Ni, as observed in Fig. 4. 6Ni2Rh those of 6Ni2Co approach 6Ni2Rh's activity, as the gas ow rate
has a higher reducibility than 8Ni, providing oxidation resis- is lowered. Fig. 7(a) shows that the methane conversion and
tance and maintaining the metallic form in its surface.
hydrogen yield of all catalysts increase with a decrease of the gas
The 6Ni2Co catalyst shows higher catalytic activity than 8Ni, ow rate, during which the order of catalytic activities is 6Ni2Rh
as the S/C ratio is lowered, indicating its feasibility as > 6Ni2Co > 8Ni > 6Ni2Cu. As indicated by eqn (1) and (2),
a reforming catalyst in the anode of PCFCs. Fig. 6(a) shows that lowering the feed gas ow rate increases the methane conver-
20
the methane conversion and hydrogen yield of 6Ni2Co peak at sion and hydrogen yield for a given reaction rate. On the other
the S/C ratio of 1 and maintain a higher value than those of 8Ni, hand, their response to changes in the gas ow rate varies from
as the S/C ratio is further lowered to 0.5. As evidenced in catalyst to catalyst, attributed to different catalytic activities or
Fig. 6(b), 6Ni2Co's order of reaction with respect to PH O reaction rates. The high catalytic activity of 6Ni2Rh, evidenced
2
changes the sign of its slope (from the negative to the positive)
at around the S/C ratio of 1 and becomes larger than that of 8Ni,
when the S/C ratio is lowered from 2 to 1. This is explained by its
high oxygen affinity and resistance to methane adsorption at
28,65
high partial pressure of steam (e.g., the S/C ratio of 2),
explained in Section 3.2.1. In contrast, at the S/C ratios of 1 and
.5, the catalytic activity of 6Ni2Co is enhanced signicantly due
as
0
to overcoming the surface poisoning by O* and resistance to
methane adsorption. 6Ni2Co's ability to reform methane and
produce hydrogen becomes evident at low partial pressure of
66
steam. Its methane conversion and hydrogen yield at the S/C
ratios of 1 and 0.5 are comparable to those of 6Ni2Rh, con-

rming their high catalytic activity at low partial pressure of
steam. Note that the high hydrogen yield of 6Ni2Co at the low S/
C ratio is particularly desirable for PCFC application. Its
product selectivity towards hydrogen makes it a promising
catalyst for the PCFC anode where sufficient hydrogen, as well
as internal methane reforming rate, is required. On the other
hand, 6Ni2Cu shows lower methane conversion and hydrogen
yield than 8Ni under all conditions, making its application to
the PCFC anode difficult.
Lowering the S/C ratio, desirable for PCFC operation, reveals
that both 6Ni2Rh and 6Ni2Co are promising catalysts for
internal reforming of methane in the PCFC anode. In particular,
they provide substantial methane conversion and hydrogen
ꢁ
yield at the S/C ratio of 1 (and at a temperature of 500 C).
Considering the relative cost of Co and Rh, the catalytic activity
of the former makes it attractive as a candidate material when
alloyed with Ni in the PCFC anode for direct hydrocarbon feed.
To verify further their catalytic activity and feasibility for PCFCs,
the effect of gas ow rate is elucidated at the S/C ratio of 1 and
ꢁ
temperature of 500 C as follows.
3.2.3. Effect of gas ow rate. The third catalytic activity test
was conducted by changing the feed gas ow rate. During PCFC
operation, the anode gas ow rate determines fuel utilization
Fig. 7 Effect of gas flow rate on steam reforming of methane by 8Ni
conversion ( equilibrium, 150 sccm,
00 sccm, 50 sccm) and hydrogen yield ( ); (b) the dependence on
the gas hourly space velocity, represented by its slope or time scale
needed for methane conversion ( 8Ni, 6Ni2Co, 6Ni2Cu,
Ni2Rh). Variable: gas flow rate ¼ 150, 100, 50 sccm. Fixed conditions:
(
dened as the ratio of fuel consumed to fuel supplied) along and 6Ni2M catalysts. (a) CH
4
1
with electrical current imposed on PCFC unit-cells. Controlling
the fuel utilization upon the variation of electrical demand is
critical in maintaining the electrochemical environment in the
anode, making it necessary to change accordingly the anode gas
6
ꢁ
pressure ¼ 1 atm, temperature ¼ 500 C, S/C ratio ¼ 1 (CH
4 2
: 20%, H O:

ow rate. In this regard, all catalysts were evaluated at various 20%, Ar: 60% (H
2
O/CH
4
¼ 1)).
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. A, 2021, 9, 6139–6151 | 6147

View Article Online
Journal of Materials Chemistry A
Paper
so far, results in a large reaction rate for steam reforming of 8Ni, 6Ni2Co, and 6Ni2Rh at a very low S/C ratio of 0.1, where
methane and hence high methane conversion and hydrogen carbon deposition occurs predominantly from a thermody-
yield at all gas ow rates. This makes 6Ni2Rh less dependent on namic point of view. Such harsh conditions were considered to
the variation of the gas ow rate, conrming its feasibility as the accelerate their degradation and investigate their stability
PCFC anode catalyst under diverse operating conditions. In the within a short period of time. As shown in Fig. 8, although all
meantime, 6Ni2Co provides a reaction rate as high as that of catalysts exhibit decreasing methane conversion rates with
6Ni2Rh, when the gas ow rate decreases to 50 sccm, as shown operating time, there are obvious discrepancies between cata-
in Fig. 7(a). It can be inferred that, with a sufficient residence lysts. 6Ni2Rh and 6Ni2Co show lower degradation rates of 3.3%
time (i.e., low gas ow rate), 6Ni2Co's catalytic activity is as and 10.5%, respectively, than Ni (24.8%), during the 30 h
strong as that of 6Ni2Rh, improving its applicability to the PCFC operations. The activity degradation of 6Ni2Co and 8Ni is
anode. Note that the high fuel utilization is desirable for substantial in the beginning of the experiment, while the
enhancing the energy conversion efficiency of a PCFC system, degradation of 6Ni2Co reduces to a small value with time. In
for which the feed gas ow rate needs to be reduced. In this contrast, the degradation of 8Ni proceeds continuously
sense, 6Ni2Co can be a promising catalyst, as good as 6Ni2Rh, throughout the operating time. Generally, nickel is known to be
for the PCFC anode.
susceptible to carbon deposition, but in this experiment, 8Ni
The high catalytic activity of 6Ni2Rh and 6Ni2Co makes them exhibits a relatively long active methane conversion period. This
less dependent on the gas hourly space velocity. Fig. 7(b) shows the phenomenon can be explicated by the PCFC anode support
13
dependence of catalysts on the gas hourly space velocity, in which material BZCYYb which has robust carbon resistance.
the slope represents the time scale required for methane conver- Furthermore, OH* which is generated by hydrolysis is spilled
5
sion. The higher the time scale or the slope, the lower the catalytic over, thereby removing carbon deposited over the Ni surface.
activity. It can be observed that 6Ni2Rh has the smallest time On top of such a support effect, in comparison with the
ꢀ
2
ꢀ2
scales (t(1) ¼ 1.19 ꢂ 10 s, t(2) ¼ 1.02 ꢂ 10 s), indicating its monometallic Ni catalyst, the Ni–Rh alloy has higher activation
high catalytic activity and low dependency on the gas hourly space energy of the C–C bond, which prevents carbon deposition on
ꢀ2
velocity. 6Ni2Co also has small time scales (t(1) ¼ 3.61 ꢂ 10 s, the catalyst surface, and lower C–O formation activation energy,
ꢀ
2
t(2) ¼ 4.89 ꢂ 10 s), which makes it less dependent on the gas which helps to remove carbon deposited on the catalyst surface,
32
hourly space velocity, in particular at the low gas ow rate region elucidated by computational analysis. Indeed, 6Ni2Rh shows
i.e., region (1)). This corresponds to the results shown in Fig. 7(a). less carbon deposition than 8Ni, as shown in Fig. S6.† More-
(
The high catalytic activity of 6Ni2Rh and 6Ni2Co results in a short over, the catalyst particles of 6Ni2Rh were not separated from
time needed for converting methane through steam reforming the BZCYYb, whereas those of 8Ni were detached from the
reactions. Based on such small time scales, the gas hourly space support along with the growth of carbon nanobers. This
velocity, which dictates the residence time, has a limited effect on feature further enhances 6Ni2Rh's resistance to carbon forma-
their methane conversion. This feature is substantially important tion on its surface, which explains its lower degradation rate
when accounting for diverse anode gas ow rates and providing
sufficient internal methane reforming and hydrogen production in
the PCFC anode. In contrast, the time scales for 8Ni (t(1) ¼ 8.87 ꢂ
ꢀ
2
ꢀ2
ꢀ2
1
¼
0
s, t(2) ¼ 9.48 ꢂ 10 s) and 6Ni2Cu (t(1) ¼ 7.56 ꢂ 10 s, t(2)
ꢀ2
6.92 ꢂ 10 s) are higher than those of 6Ni2Rh and 6Ni2Co at all
gas hourly space velocities, which implies their low catalytic
activities.
Based on the results shown above, 6Ni2Rh and 6Ni2Co are
promising candidates for the PCFC anode catalyst. The former
maintains the high methane conversion and hydrogen yield
under various operating conditions including temperature, S/C
ratio, and feed gas ow rate. In the meantime, the latter shows
comparable catalytic activity at low S/C ratio and low feed gas

ow rate which are desirable operating conditions for
enhancing the efficiency of a PCFC system. Given their catalytic
activity and feasibility as the PCFC anode material, their
stability needs to be demonstrated, which is discussed in the
following.
4
. Carbon resistance and stability
Fig. 8 Change in methane conversion during steam reforming of
methane by 8Ni, 6Ni2Rh, and 6Ni2Co catalysts under high carbon
The 6Ni2Rh and 6Ni2Co catalysts show higher resistance to
carbon formation and stability than 8Ni, conrming again their
feasibility as the PCFC anode material. The stability test was
formation conditions. CH
4
conversion ( 8Ni,
6Ni2Co,
6Ni2Rh).
ꢁ
Fixed conditions: pressure ¼ 1 atm, temperature ¼ 500 C, S/C ratio ¼
conducted to elucidate the resistance to carbon deposition of 0.1 (CH
4
: 90.9%, H
2
O: 9.1% (H
2
O/CH
4
¼ 0.1)), gas flow rate ¼ 50 sccm.
6148 | J. Mater. Chem. A, 2021, 9, 6139–6151
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry A
than 8Ni. In the case of 6Ni2Co, the high oxygen affinity of the ratio improved 6Ni2Co's catalytic activity. Decreasing the steam
Ni–Co alloy, discussed above, may enable O* and OH* adsorbed concentration in the PCFC anode is desirable from the
on its surface to enhance the oxidation of carbon, thereby perspective of system efficiency and BZCYYb stability, which
restraining carbon deposition. Similar to the results of 6Ni2Rh, implies the feasibility of 6Ni2Co as a reforming catalyst in direct
Fig. S6† shows that the carbon deposition on the 6Ni2Co methane-fueled PCFC anodes. The high catalytic activity of
surface is less than that on 8Ni. The surface effect of 6Ni2Rh 6Ni2Rh and 6Ni2Co at low temperature and low steam-to-
and 6Ni2Co improves their carbon resistance and long-term carbon ratio makes them less dependent on the feed gas ow
stability, which enhances their feasibility as the direct rate and gas hourly space velocity. Maintaining the high cata-
hydrocarbon-fueled PCFC anode material. Given that a recent lytic activity at various anode gas ow rates is critical for
study demonstrated the feasibility of PCFCs based on internal accounting for the variation of electrical demand, which implies
67
dry reforming of methane, the enhanced carbon resistance the feasibility of 6Ni2Rh and 6Ni2Co as good reforming cata-
and long-term stability of 6Ni2Rh and 6Ni2Co can extend the lysts in PCFC anodes. They also showed higher resistance to
operating regime of PCFCs from steam reforming to dry carbon formation and stability in comparison with 8Ni, con-
reforming. However, long-term tests over 100 to 1000 hours may rming again their feasibility as catalysts in direct methane-
result in surface diffusion and Ostwald ripening of nano-scale fueled PCFC anodes.
catalyst particles, which induces particle growth and activity
degradation. This will be further examined by increasing the
6
. Author contributions
long-term test time. Their high catalytic activity and long-term
durability at low temperature, low S/C ratio, and low gas ow
rate need to be further demonstrated in a full-cell operation
Kyungpyo Hong: conceptualization, methodology, investiga-
tion, formal analysis, writing – original dra. Stephanie Nadya
Sutanto: methodology, investigation. JeongA Lee: methodology,
investigation. Jongsup Hong: supervision, writing – review &
editing, project administration.
68,69
with a kinetics study,
following study.
which will be performed in the
5. Conclusions
Conflicts of interest
In this study, we investigated Ni-based bimetallic alloy catalysts
anchored on BaZr0.4Ce0.4 3ꢀd (BZCYYb) which can be
directly applied to the proton-conducting ceramic fuel cell
PCFC) anode for internal steam reforming of methane at low
temperature. Maintaining the catalyst loading at 8 wt%, 8Ni,
Ni2Rh, 6Ni2Co, and 6Ni2Cu (the number in front of the
Y0.1Yb0.1O
The authors declare that they have no known competing

nancial interests or personal relationships that could have
(
appeared to inuence the work reported in this paper.
6
species name represents the metal loading in a mass basis) were Acknowledgements
considered as candidate materials. Characterization was carried
This research was nancially supported by the Energy Tech-
out to conrm the structural properties and the formation of
alloyed catalysts. The BET and CO pulse chemisorption analyses
showed the coherent structural properties (i.e., specic surface
area, metallic particle diameter, metallic surface area, and
metal dispersion) of all catalysts, providing the same basis for
a catalytic activity test. The XRD analysis conrmed the alloy
formation of 6Ni2M (M: Co, Cu, Rh) catalysts by observing the
peak of a 6Ni2M catalyst existing between the 8Ni and 8M
peaks. TEM image analysis conrmed again their alloy forma-
tion and demonstrated nano-sphere particles anchored on the
BZCYYb support. The TPR analysis revealed that the reducibility
of 6Ni2M catalysts was improved compared to that of 8Ni.
The catalytic activities and long-term stability of 8Ni and
nology Development Program of the Korea Institute of Energy
Technology Evaluation and Planning (KETEP), which granted

nancial resource from the Ministry of Trade, Industry &
Energy, Republic of Korea (no. 20173010032170), and a National
Research Foundation of Korea (NRF) grant funded by the Korea
government (MSIT) (2019R1C1C1005152).
References
1 B. C. Steele, Nature, 1999, 400, 619–621.
2 Y.-L. Huang, A. M. Hussain and E. D. Wachsman, Nano
Energy, 2018, 49, 186–192.
6
Ni2M catalysts were evaluated for the internal steam reforming
3 Y. Choi, E. C. Brown, S. M. Haile and W. Jung, Nano Energy,
2016, 23, 161–171.
4 Y. Chen, Y. Lin, Y. Zhang, S. Wang, D. Su, Z. Yang, M. Han
and F. Chen, Nano Energy, 2014, 8, 25–33.
5 C. Duan, J. Tong, M. Shang, S. Nikodemski, M. Sanders,
S. Ricote, A. Almansoori and R. O'Hayre, Science, 2015, 349,
1321.
6 K. Pei, Y. Zhou, K. Xu, Z. He, Y. Chen, W. Zhang, S. Yoo,
B. Zhao, W. Yuan and M. Liu, Nano Energy, 2020, 104704.
7 H. An, H.-W. Lee, B.-K. Kim, J.-W. Son, K. J. Yoon, H. Kim,
D. Shin, H.-I. Ji and J.-H. Lee, Nat. Energy, 2018, 3, 870–875.
of methane at low temperature under various conditions (e.g.,
operating temperature, steam-to-carbon ratio, and gas ow
rate). 6Ni2Rh exhibited the highest catalytic activity at all
temperatures given its low activation energy. The methane
conversion and hydrogen yield of 6Ni2Co and 6Ni2Cu were
lower than those of 8Ni due to their high activation energy for
the onset of steam reforming. 6Ni2Rh's catalytic activity
remained strong at various steam-to-carbon ratios. Its high
activity was more evident at a high steam concentration due to
its resistance to oxide formation. Lowering the steam-to-carbon
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. A, 2021, 9, 6139–6151 | 6149

View Article Online
Journal of Materials Chemistry A
Paper
8
9
K. Bae, D. Y. Jang, H. J. Choi, D. Kim, J. Hong, B. K. Kim, 33 D. Kim, J. Kang, Y. Lee, N. Park, Y. Kim, S. Hong and
J. H. Lee, J. W. Son and J. H. Shim, Nat. Commun., 2017, 8,
4553.
D. Moon, Catal. Today, 2008, 136, 228–234.
34 M. Ferrandon, A. J. Kropf and T. Krause, Appl. Catal., A, 2010,
379, 121–128.
1
E. Fabbri, A. Depifanio, E. Dibartolomeo, S. Licoccia and
E. Traversa, Solid State Ionics, 2008, 179, 558–564.
35 J. Strohm, J. Zheng and C. Song, J. Catal., 2006, 238, 309–320.
10 K. Bae, H.-S. Noh, D. Y. Jang, J. Hong, H. Kim, K. J. Yoon, 36 S. L. Lakhapatri and M. A. Abraham, Catal. Sci. Technol.,
J.-H. Lee, B.-K. Kim, J. H. Shim and J.-W. Son, J. Mater.
Chem. A, 2016, 4, 6395–6403.
1 I.-H. Kim, D.-K. Lim, H. Bae, A. Bhardwaj, J.-Y. Park and
S.-J. Song, J. Mater. Chem. A, 2019, 7, 21321–21328.
2 T. Nakamura, S. Mizunuma, Y. Kimura, Y. Mikami,
2013, 3, 2755–2760.
37 Z. Bian, S. Das, M. H. Wai, P. Hongmanorom and S. Kawi,
ChemPhysChem, 2017, 18, 3117–3134.
38 Y. Yoon, H. Kim and J. Lee, J. Power Sources, 2017, 359, 450–
457.
1
1
K. Yamauchi, T. Kuroha, N. Taniguchi, Y. Tsuji, 39 S.-W. Kim, M. Park, H. Kim, K. J. Yoon, J.-W. Son, J.-H. Lee,
Y. Okuyama and K. Amezawa, J. Mater. Chem. A, 2018, 6,
5771–15780.
B.-K. Kim, J.-H. Lee and J. Hong, Appl. Catal., B, 2017, 200,
265–273.
1
13 C. Duan, R. J. Kee, H. Zhu, C. Karakaya, Y. Chen, S. Ricote, 40 K. J. Yoon, M. Biswas, H.-J. Kim, M. Park, J. Hong, H. Kim,
A. Jarry, E. J. Crumlin, D. Hook, R. Braun, N. P. Sullivan
and R. O'Hayre, Nature, 2018, 557, 217–222.
J.-W. Son, J.-H. Lee, B.-K. Kim and H.-W. Lee, Nano Energy,
2017, 36, 9–20.
1
1
1
1
1
1
2
2
2
2
2
2
2
4 H. Wu, V. La Parola, G. Pantaleo, F. Puleo, A. Venezia and 41 J. Shin, Y. J. Lee, A. Jan, S. M. Choi, M. Y. Park, S. Choi,
L. Liotta, Catalysts, 2013, 3, 563.
J. Y. Hwang, S. Hong, S. G. Park and H. J. Chang, Energy
Environ. Sci., 2020, 13, 4903–4920.
42 F. Boschini, A. Rulmont, R. Cloots and R. Moreno, Ceram.
Int., 2009, 35, 1007–1013.
5 S. Choi, C. J. Kucharczyk, Y. Liang, X. Zhang, I. Takeuchi,
H.-I. Ji and S. M. Haile, Nat. Energy, 2018, 3, 202–210.
6 W. Wang, C. Su, Y. Wu, R. Ran and Z. Shao, Chem. Rev., 2013,
1
13, 8104–8151.
7 C. Yang, J. Li, Y. Lin, J. Liu, F. Chen and M. Liu, Nano Energy,
015, 11, 704–710.
8 N. Laosiripojana,
S. Assabumrungrat, Fuel, 2006, 85, 323–332.
43 I. Luisetto, S. Tuti and E. Di Bartolomeo, Int. J. Hydrogen
Energy, 2012, 37, 15992–15999.
44 M. Lu, P. Lv, Z. Yuan and H. Li, Renewable Energy, 2013, 60,
522–528.
45 L. Zhao, T. Han, H. Wang, L. Zhang and Y. Liu, Appl. Catal.,
B, 2016, 187, 19–29.
46 A. Vizcaino, A. Carrero and J. Calles, Int. J. Hydrogen Energy,
2007, 32, 1450–1461.
47 E. T. Saw, U. Oemar, X. R. Tan, Y. Du, A. Borgna, K. Hidajat
and S. Kawi, J. Catal., 2014, 314, 32–46.
48 S. Alayoglu and B. Eichhorn, J. Am. Chem. Soc., 2008, 130,
17479–17486.
49 H.-S. Roh, K.-W. Jun and S.-E. Park, J. Ind. Eng. Chem., 2003,
9, 261–266.
2
W.
Sangtongkitcharoen
and
9 A. Effendi, Z.-G. Zhang, K. Hellgardt, K. Honda and
T. Yoshida, Catal. Today, 2002, 77, 181–189.
0 S. D. Angeli, L. Turchetti, G. Monteleone and
A. A. Lemonidou, Appl. Catal., B, 2016, 181, 34–46.
1 P. Ferreira-Aparicio, M. J. Benito and J. L. Sanz, Catal. Rev.:
Sci. Eng., 2005, 47, 491–588.
2 A. Iulianelli, S. Liguori, J. Wilcox and A. Basile, Catal. Rev.:
Sci. Eng., 2016, 58, 1–35.
3 P. Gangadharan, K. C. Kanchi and H. H. Lou, Chem. Eng. Res.
Des., 2012, 90, 1956–1968.
50 E. Pereira, N. Homs, S. Marti, J. Fierro and
P. Ramirezdelapiscina, J. Catal., 2008, 257, 206–214.
51 J. Rynkowski, T. Paryjczak, M. Lenik, M. Farbotko and
J. Goralski, J. Chem. Soc., Faraday Trans., 1995, 91, 3481–
3484.
4 S. De, J. Zhang, R. Luque and N. Yan, Energy Environ. Sci.,
2016, 9, 3314–3347.
5 Y. Chen, Y. Zhang, Y. Lin, Z. Yang, D. Su, M. Han and
F. Chen, Nano Energy, 2014, 10, 1–9.
6 M. Khzouz, J. Wood, B. Pollet and W. Bujalski, Int. J. 52 M.-S. Fan, A. Z. Abdullah and S. Bhatia, Appl. Catal., B, 2010,
Hydrogen Energy, 2013, 38, 1664–1675. 100, 365–377.
7 J. Zhang, H. Wang and A. Dalai, J. Catal., 2007, 249, 300–310. 53 Y. Li, Q. Fu and M. Flytzani-Stephanopoulos, Appl. Catal., B,
8 G. Jones, J. Jakobsen, S. Shim, J. Kleis, M. Andersson, 2000, 27, 179–191.
J. Rossmeisl, F. Abildpedersen, T. Bligaard, S. Helveg and 54 G. Avgouropoulos and T. Ioannides, Appl. Catal., B, 2006, 67,
B. Hinnemann, J. Catal., 2008, 259, 147–160. 1–11.
9 S. Misture, K. McDevitt, K. Glass, D. Edwards, J. Y. Howe, 55 J. Kugai, J. T. Miller, N. Guo and C. Song, J. Catal., 2011, 277,
2
2
2
K. Rector, H. He and S. Vogel, Catal. Sci. Technol., 2015, 5,
565–4574.
0 B. Saha, A. Khan, H. Ibrahim and R. Idem, Fuel, 2014, 120,
02–217.
1 Y. Choi, C. Compson, M.-C. Lin and M. Liu, J. Alloys Compd.,
007, 427, 25–29.
46–53.
4
56 A. Le Valant, N. Bion, F. Can, D. Duprez and F. Epron, Appl.
Catal., B, 2010, 97, 72–81.
57 J. Kugai, V. Subramani, C. Song, M. Engelhard and Y. Chin, J.
Catal., 2006, 238, 430–440.
58 E. Nikolla, J. Schwank and S. Linic, J. Catal., 2009, 263, 220–
227.
3
3
3
2
2
2 J. Guo, C. Xie, K. Lee, N. Guo, J. T. Miller, M. J. Janik and
C. Song, ACS Catal., 2011, 1, 574–582.
59 W. An, X. C. Zeng and C. H. Turner, J. Chem. Phys., 2009, 131,
174702.
6150 | J. Mater. Chem. A, 2021, 9, 6139–6151
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry A
60 Y. Chen, B. deGlee, Y. Tang, Z. Wang, B. Zhao, Y. Wei, 65 K. Hou and R. Hughes, Chem. Eng. J., 2001, 82, 311–328.
L. Zhang, S. Yoo, K. Pei and J. H. Kim, Nat. Energy, 2018, 3, 66 A. Koh, L. Chen, W. Keeleong, B. Johnson, T. Khimyak and
1
042–1050.
1 O. Yamazaki, K. Tomishige and K. Fujimoto, Appl. Catal., A, 67 B. Hua, N. Yan, M. Li, Y.-F. Sun, Y.-Q. Zhang, J. Li, T. Etsell,
996, 136, 49–56. P. Sarkar and J.-L. Luo, Adv. Mater., 2016, 28, 8922–8926.
2 A. Dicks, K. Pointon and A. Siddle, J. Power Sources, 2000, 86, 68 E. S. Hecht, G. K. Gupta, H. Zhu, A. M. Dean, R. J. Kee,
J. Lin, Int. J. Hydrogen Energy, 2007, 32, 725–730.
6
6
6
6
1
5
23–530.
3 J. Jeon, S. Nam and C. H. Ko, Catal. Today, 2018, 309, 140–
46.
4 D. Pakhare and J. Spivey, Chem. Soc. Rev., 2014, 43, 7813–
837.
L. Maier and O. Deutschmann, Appl. Catal., A, 2005, 295,
40–51.
69 L. Maier, B. Sch ¨a del, K. Herrera Delgado, S. Tischer and
O. Deutschmann, Top. Catal., 2011, 54, 845.
1
7
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. A, 2021, 9, 6139–6151 | 6151