Chemical Engineering Journal 491 (2024) 151865

Contents lists available at ScienceDirect

Chemical Engineering Journal

journal homepage: www.elsevier.com/locate/cej

In situ exsolved NiFe nanoparticles in Ni-doped Sr0.9Ti0.3Fe0.63Ni0.07O3‑δ

anode with a three-dimensionally ordered macroporous structure for solid
oxide fuel cells fueled by alkanes
Yuying Jiang a, 1, Jiamei Liu a, b, 1, Bo Cheng a, Xinyue Dang a, Huaiqiang Su c, *, Yani Hua a, *,
Zhan Gao a, *
a
School of Chemical Engineering and Technology, Xi’an Jiaotong University, Xi’an, Shaanxi 710049, People’s Republic of China
b
Instrument Analysis Center of Xi’an Jiaotong University, Xi’an, Shaanxi 710049, People’s Republic of China
c
Shaanxi Coal Based Special Fuel Research Institute Co., Ltd, Xi’an, Shaanxi 710100, People’s Republic of China

A R T I C L E I N F O A B S T R A C T

Keywords: In this work, A-site deficient Ni-doped Sr0.9Ti0.3Fe0.63Ni0.07O3‑δ perovskite anode with ordered macroporous
Solid oxide fuel cells structures (STFN-3DOM) is designed to facilitate in-situ Ni-Fe exsolution for solid oxide fuel cells fueled by al­
Anode kanes, e.g., methane and propane. The large specific surface area and A-site deficient of STFN-3DOM provide
In situ exsolution
more exsolved Ni-Fe active sites and enhance the reaction kinetics of alkane reforming. Quasi-in-situ XPS is used
Three-Dimensionally Ordered Macroporous
to reveal the changes in the chemical state of elements during the exsolution and propane reforming process on
Structure
A-site deficient the STFN-3DOM surface. The synergistic effect of A-site deficient, three-dimensional ordered macroporous
structure, and in-situ Ni-Fe exsolution promotes the La0.8Sr0.2Ga0.83Mg0.17O3‑δ (LSGM) electrolyte-supported
SOFC single cell to deliver advanced performance with both wet H2 (1.33 W cm− 2) and wet C3H8 (0.838 W
cm− 2) at 850 ◦ C, exhibiting satisfactory long-time stability in CH4 and redox stability. Density functional theory
calculation confirms the low activation energy of STFN-3DOM for the rate-limiting step of C3H8* dissociation.

1. Introduction reactor for co-production of ethylene and electricity from ethane,

achieving a peak power density (Pmax) of 0.09 W cm− 2 under wet H2
Solid oxide fuel cells (SOFCs) are becoming the most promising de­ while the conversion of ethane came to 41.5 % at 800℃. Cimenti et al.
vices for combined heat and power generation with their unique high [13] tested the catalytic activity of La0.75Sr0.25Cr0.5Mn0.5O3− δ (LSCM) in
efficiency, fuel flexibility, and environmentally friendly. Hydrocarbons, direct methanol and ethanol SOFC, and the Pmax was 0.021 W cm− 2 and
e.g., natural gas and diesel, are used as fuels that provide advantages for 0.039 W cm− 2, at 800℃ respectively. Long et al. [14] evaluated the CO
stationary cogeneration and mobile auxiliary power units (APUs). tolerance of La0.5Sr0.5Fe0.9Mo0.1O3–δ under SOFC operating conditions,
However, hydrocarbon fuels result in serious carbon coking issues when the single cell showed a polarization impedance (RP) of 1.099 Ω cm2 at
using the traditional Ni-YSZ anode. Additionally, the Ni-YSZ anode 750℃. However, satisfactory performance cannot be obtained only by
suffers from Ni-agglomeration and poor re-dox stability. During the past doping. The in situ exsolution strategy has been used to improve the
decade, many mixed ion–electron conductors (MIECs) have been anode performance fueled by hydrocarbons [15]. The selected active
exploited as anodes for SOFC, improving the carbon coking while the metal elements (Mn+) doped in the B site of perovskite anode material
electrocatalytic performance is not satisfactory [1–5].ABO3 perovskite will be reduced to M0 under the reducing condition and finally form
materials have attracted the attention of most researchers because of reversible active nanoparticles (NPs) [16–19], which have firmly
their good conductivity, controllable composition, and structural sta­ bonded to the supporting and inhibiting agglomeration [20–24]. Barnett
bility [6–11]. Doped perovskite materials have performed well in hy­ et al. [25] discovered that Ru in La0.8Sr0.2Cr1-xRuxO3-δ will be reduced
drocarbon fuel internal reforming. Li et al. [12] established the Cu and form Ru islands on the surface, which can improve the electro­
substitution La0.6Sr0.4Fe0.8Nb0.1O3− δ (LSFNCu) anode, used it in SOFC catalytic activity of the anode, strengthen H2 adsorption kinetics, and

* Corresponding authors.
E-mail addresses: [email protected] (H. Su), [email protected] (Y. Hua), [email protected] (Z. Gao).
1
These authors contributed equally to this paper.

https://doi.org/10.1016/j.cej.2024.151865
Received 13 January 2024; Received in revised form 2 April 2024; Accepted 1 May 2024
Available online 3 May 2024
1385-8947/© 2024 Elsevier B.V. All rights reserved.
Y. Jiang et al. Chemical Engineering Journal 491 (2024) 151865

enhance the performance of the anode. A recent study by Kim et al. [17] method using Methyl Methacrylate (99.8 %, TCL) and K2S2O8 (99 %,
also discovered that Pd-doped La0.6Sr0.4Co0.2Fe0.8O3-δ (LSCF) will be Adamas). Stoichiometric Sr(NO3)2 (99.9 %, Sigma-Aldrich), Fe
converted into the RP phase with the Co-Fe alloy nanoparticles exsolved. (NO3)3⋅9H2O (99.9 %, Sigma-Aldrich) and Ni(NO3)2⋅6H2O (99.9 %,
The metal nanoparticles produced by the exsolution path will form an Sigma-Aldrich) were dissolved in 10 mL of de-ionized water as liquid A,
active metal/oxide interface with the parent material, exhibiting high and Titanium(IV) butoxide (99 %, Acros) was dissolved in 10 mL of
activity, selectivity, and coking resistance [9,22,26]. Donazzi et al. [16] absolute methanol as liquid B. Then slowly dropped liquid A into liquid
explored the performance of Sr(Ti0.3Fe0.7Ru0.07)O3-δ (STFRu) anode in B under vigorous stirring to obtain STFN precursor solution. The PMMA
an ethanol/steam mixture. The cell reached a Pmax of 0.195 W cm− 2 at template was impregnated in the STFN precursor solution overnight and
800℃ with an ethanol/steam ratio of 3:1. Tsiotsias et al. [18] showed then dried naturally at 25 ◦ C and then held at 300 ◦ C for 3 h under an
that the composition of nanoparticles produced by in-situ exsolution can argon atmosphere to obtain orange-impregnated precursor powders.
be controlled. By changing the concentration of A-site-deficient in The orange-impregnated precursor powder was calcined in air at 650℃
La0.4Sr0.6Ti0.6Fe0.35Ni0.05O3-δ (LSTFN) supporting, Ni-Fe nanoclusters for 4 h to produce black STFN-3DOM powders with good phase and
with different element ratios were produced, showing better stability 3DOM structure. In order to draw a comparison, the STFN-Solid sample
than the commercial catalyst Ni/Al2O3. powders were fabricated utilizing a solid-state reaction technique. The
Currently, most oxide anodes are constructed in “disorderly stack­ SrCO3, Fe2O3, TiO2, and Ni(NO3)2⋅6H2O powders were weighed ac­
ing”, leading to a low density of active nanoparticles per unit volume cording to the stoichiometric ratio, blended in ethanol, and subjected to
and significant concentration polarization effects. Therefore, micro­ ball milling for 24 h. The mixture was dried and subjected to calcination
structural regulation is also a mainstream strategy to enhance the anode at 1150 ◦ C for a duration of 10 h under air atmospheric conditions to
performance of SOFCs [27–30]. Ordered mesoporous electrode mate­ obtain the STFN-Solid powders with 0.1 Sr deficient and 0.05 Sr
rials have been applied in SOFCs and achieved excellent performance. deficient.
Tarancon et al. [31] prepared Ni-Gd0.2Ce0.8O2-δ ordered mesoporous
anode with silica templates (KIT-6), and it showed lower interfacial 2.2. Fuel cell fabrication
polarization impedance. According to Li et al. [27], a Ni/Ce0.8Sm0.2O1.9
(SDC) anode with a three-dimensional ordered macroporous (3DOM) The single cell adopted the structure supported by LSGM electrolyte.
structure was achieved by polymethylmethacrylate (PMMA) as a tem­ To prepare LSGM electrolyte, LSGM (Paxair) green tapes were obtained
plate, the 3DOM structure not only brought larger specific surface area by tape casting method, which was cut into discs with a diameter of 13
but also provided more active lattice oxygen and stronger interfacial mm and then sintered freely at 1450℃ to obtain dense LSGM electrolyte
interaction, improving the performance of methanol fuel. Carter et al. support with a thickness of about 200 μm. A barrier layer of
[32] fabricated a 3DOM cathode for zirconia microtubule SOFCs, La0.4Ce0.6O2‑δ (LDC) was applied to both sides of the LSGM pellets to
proving that the design of the cathode microstructure is effective in avoid unnecessary interface reaction. To create an anode slurry, STFN-
enabling SOFCs to function at a lower temperature. Previous beneficial 3DOM/STFN-Solid and binder (V-737, Heraeus) were combined in a
explorations on ordered pore structure Ni/Gd0.1Ce0.9O1.95(GDC) anodes specific ratio and then screen-printed on one side of the electrolyte and
have shown that the application of ordered pore electrode materials in sintered at 1150℃ for 2 h. LSCF-GDC cathode and LSCF collector were
SOFC is feasible [33,34], while the reports of ordered mesoporous oxide prepared by a similar method, but the sintering temperature was 950℃.
anodes are rare. Therefore, we combine the two strategies of A-site The active area of electrodes was all 0.196 cm2. Silver grids are attached
deficient and the construction of 3DOM microstructures. The A-site to the electrodes on both sides, and two groups of silver wires are con­
deficient is utilized to promote B-site exsolution, and the anode with nected as the single cell’s current collectors.
ordered mesoporous microstructure is further constructed to increase
the density of exsolved nanoparticles and to enhance the gas-phase 2.3. Materials characterization
transfer process in the porous anode. We expect that the synergistic ef­
fect between them will help to promote the exsolution of transition The crystal structures of the STFN-Solid and STFN-3DOM powders in
metal elements at B site, and then strengthen the activity of alkane pristine, reduced, and reoxidized conditions were identified via X-ray
reforming and the kinetics of hydroxide reaction. diffraction at a scanning rate of 1◦ min− 1 (XRD, D8 ADVANCE, Bruker
In this work, A-site deficient Ni-doped Sr0.9Ti0.3Fe0.63Ni0.07O3‑δ AXS Corp. German with Cu Kα radiation, λ = 1.54060 Å). Scanning
perovskite anode with ordered macroporous structures (STFN-3DOM) electron microscopy (SEM) image was obtained using a JSM-6301F
was prepared with PMMA as a template, and was tested as the anode (JEOL Ltd., Japan) operating at 1–5 kV. STFN-3DOM powders after
toward H2 and alkane hydrocarbon fuels. During the operation of reduction were characterized by High-resolution transmission electron
SOFCs, Ni-Fe nanoparticles were formed in situ on the surface of the microscopy (HRTEM) analysis (TECNAI G2 F30, operating at 200 kV),
material, along with the transformation from perovskite structure to RP which are also equipped with a CCD camera and respective EDAX de­
structure. To understand the exsolution behavior, the elemental reduc­ tectors. The Digital Micrograph software package (Gatan GmbH) was
tion behavior and valence change of STFN-3DOM under reduction used to analyze the high-resolution TEM images. Brunauer − Emmett −
conditions were studied by quasi-in-situ XPS. The single cell using STFN- Teller (BET, BELSORP-Max) method was used to obtain the nitrogen
3DOM with 3DOM structure as anode showed superior performance, physisorption isotherms and the pore size distribution of STFN-Solid and
with a Pmax of 1.33 W cm− 2 in wet H2 and 0.838 W cm− 2 in wet C3H8 at STFN-3DOM, both samples were degassed at 200℃ for 12 h. Quasi-in
850 ◦ C, respectively. Density functional theory calculation revealed that situ X-ray photoelectron spectroscopy (XPS, ESCALAB 250 Xi + X-ray
the improved propane reforming activity was due to the synergistic ef­ Photoelectron Spectrometer, Thermo Fisher Scientific, US) was
fect between Ni-Fe alloy nanoparticles and the oxygen vacancies in employed to investigate the oxidation states of the powders in pristine
parent perovskite. and reduced conditions using Al Kα radiation. The test procedure is as
follows: the fresh samples are reacted in the reaction chamber connected
2. Experimental to the vacuum chamber at different conditions (10 % H2: 90 % Ar or 20
% C3H8:80 %Ar with a steam-carbon ratio= 1) under high temperature,
2.1. Synthesis of material the pressure of the reaction chamber is 0.5 bar, after the reaction the
chamber will be cooled down in an Ar atmosphere, evacuated. And then
The STFN-3DOM powders were prepared by a colloidal crystal the samples will be transferred to the XPS analyzer chamber for testing.
templating method. Well-arrayed PMMA microspheres templates Raman spectroscopy (InVia Qontor, Renishaw) was used to test the
(~200 nm in diameter) were synthesized by a modified self-assembly carbon resistance of STFN-3DOM and Ni/Al2O3. The elemental analyzer

2
Y. Jiang et al. Chemical Engineering Journal 491 (2024) 151865

(vario EL cube, CEIEC) was used to detect the content of the C element in 2.5. Electrochemical characterization
SFTN-3DOM before and after the catalytic performance test.
To study the electrochemical properties of the STFN-3DOM anode, a
2.4. Catalytic testing single cell was placed on a ceramic tube and sealed. The anode was
exposed to Ar, while the cathode was exposed to open air. The tem­
The reforming activity of STFN-3DOM for propane was tested in a perature of the single cell was gradually increased to 850℃ at a rate of
quartz tube with an inner diameter of 9 mm at 700-850℃. No additional 5℃ min− 1. Then, it was switched to wet H2, and the temperature was
pressure was applied to the whole pipeline, and the test temperature maintained for 1 h to ensure that the anode was completely reduced.
interval was 50℃. 200 mg of STFN-3DOM powder was fixed in the Subsequently, the electrochemical test was started, and humidified H2
center of the quartz tube reactor with quartz fiber. According to SOFC (3 % H2O/H2, 100 mL min− 1), humidified CH4(3 % H2O/CH4, 100 mL
operating conditions, the reaction steam-carbon ratio (S/C) was set to 1, min− 1), and C3H8/water vapor mixture (S/C = 1, C3H8/Ar, 10 mL
Ar was used as a carrier gas, and the total flow rate was 100 ml/min min− 1/40 mL min− 1) was supplied to the anode respectively. The EIS
(C3H8 20 mL/min, Ar 80 mL/min). After the test, the samples were taken and j-V-P curves under different temperatures and fuel were tested with
out for elemental analysis and Raman test. a Princeton electrochemical workstation (Versa STAT3), EIS tests were
During the test, the reactor temperature was increased to 850 ◦ C conducted in the frequency range from 0.1 Hz to 1 MHz with an
under Ar flow at first and under the pure H2 flow for 1 h to help the Ni-Fe amplitude of 10 mV. The long-term stability and redox stability test of
nanoparticles exsolve. The reactor temperature was then reduced to the cells were tested by electronic load (ITECH IT9000).
700 ◦ C at a rate of 10 ◦ C min− 1 and remained for 30 min to reach thermal
stability. The test procedure for the reforming of C3H8 with commercial 2.6. Density functional theory calculation details
Ni/Al2O3 catalyst was consistent with STFN-3DOM, with a minimum
temperature of 400 ◦ C. The reaction products from low S/C steam The VASP package was employed for all DFT calculations [35]. The
reforming of propane were fed into Agilent GC-systems-7890A, which PBE [36] functional with D3[37] dispersion correction was utilized,
were equipped with TCD and FID detectors for online analysis. The while the PAW method [38] was employed to describe the core treat­
conversions for propane and the yields for hydrogen and methane were ments of the atoms. The wavefunction kinetic energy was fixed at 450
determined according to the following equations: eV. According to the XRD test results, the bulk oxide corresponds to the
RP tetragonal phase oxide (Sr2TiO4, I4/mmm, PDF#39–1471), and the
F[C3 H8 ]in − F[C3 H8 ]out
X C 3 H8 = × 100 (1) Ni, Fe dissolved nanoparticles can correspond to the Ni3Fe alloy (Pm-3
F[C3 H8 ]in
m) phase (PDF#04–004-6754), the approximate structure CIF file was
searched in the structure library, and performed the structure optimi­
F[H2 ]out
Y H2 = × 100 (2) zation based on the downloaded CIF structure, with Ni3Fe (1 1 0) crystal
4 × F[C3 H8 ]in + F[H2 O]in ( )
1 0
face, consistent with the results of XRD. Periodic slabs with
[ ] 1 − 2
ni × F CProduct,i out
YCProduct,i = × 100 (3) supercell was applied to model the (1 1 0) surfaces. The distance between
3 × F[C3 H8 ]in
the top and the bottom of the adjacent slab was set at approximately 18
YCProduct,i Å. The two bottom layers of the slab were kept frozen during optimi­
SCProduct,i = × 100 (4) zation. The rotationally invariant DFT + U was employed to describe the
X C 3 H8
local Coulomb interactions between 3d electrons [39,40]. The effective
F[H2 ]out Hubbard parameters were set at 2.00, 4.00 and 4.00 eV for Ti, Fe and Ni,
S H2 = ∑ (y [ ] ) (5) respectively [41,42]. The wavefunctions were fully relaxed until the
F[H2 ]out + 2
× F CProduct,i out electronic energy change and eigenvalues change were below 10-5 eV.
The structures were fully relaxed until the maximal force on each atom
Where XC3 H8 is the conversion values for C3H8, YH2 and YCProduct,i are the was less than 0.05 eV/Å. The CI-NEB method [43] was used to determine
yield values for H2 and carbon-based product, F[C3 H8 ]in and F[C3 H8 ]out reaction pathway and energy barrier.
are the molar flow rates for C3H8 at the inlet of the reactor expressed in
mol/min, SH2 and SCproduct,i are the selectivities of H2 and carbon- 3. Results and discussion
containing product, respectively. F[H2]out is the molar flow rates for
[ ]
H2 at the reactor outlet, F CProduct,i out are the molar flow rates of carbon- 3.1. Crystalline and morphological structure analysis
containing products at the reactor outlet, expressed in mol/min. And ni
is the carbon number of the carbon-containing component, y is the Fig. 1a showed the X-ray diffraction (XRD) pattern of the freshly
hydrogen number of the carbon-containing component. prepared STFN-3DOM powders, the reduced STFN-3DOM powders and
The carbon balance is calculated as follows: the reoxidized STFN-3DOM powders. The original STFN-3DOM exhibi­
ted a cubic perovskite structure with a Pm-3 m spatial group (SrTiO3-x,
Carbonin = Vin × 3 × CC3 H8 (6) PDF#35–0734). By reducing STFN-3DOM with hydrogen, Ni and Fe
∑ were exsolved to form Ni3Fe allow (Pm-3 m) phase (PDF#04–004-6754)
Carbonout = Vout × Ci × ni (7) and the perovskite phase was transformed into RP tetragonal phase
(Sr2TiO4, I4/mmm, PDF#39–1471). Fig. 1b and 1c exhibited the SEM
Carbonin − Carbonout image of different sample with 0.1-Sr deficient and 0.05-Sr deficient of
Error = × 100 (8)
Carbonin reduced STFN-Solid, the Ni-Fe nanoparticles can be seen at the surface of
both samples (representative particles are marked by yellow circles).
Where Carbonin and Carbonout are total carbon in inlet and outlet gas, Vin
The comparison of Fig. 1 b-c also proved that more A-site deficient help
and Vout are inlet and outlet flow, Ci is the ratio of carbon-containing
to bring more exsolved nanoparticles, and the related DES images are
components, ni is the carbon number of the carbon-containing
displayed in Fig. S1. As shown in Fig. S2, the XRD data showed that the
components.
reduced STFN-Solid sample still retained the ABO3 structure. Consid­
ering that this phenomenon has not been observed in the previous
literature of similar materials [19], the difference in the parent oxide
phases of the two reduced samples may be because more exsolution

3
Y. Jiang et al. Chemical Engineering Journal 491 (2024) 151865

Fig. 1. (a) XRD of STFN-3DOM before and after redox cycle; SEM image of (b) reduced STFN-Solid powders with Sr deficient of 0.1; (c) reduced STFN-Solid powders
with Sr deficient of 0.05; (d) fresh STFN-3DOM and (e) STFN-3DOM finish the calcination treatment at 1150 ℃ for 2 h; (f) HRTEM and EDS mapping images of
reduced STFN-3DOM.

occurs in STFN-3DOM, promoting the transition from a perovskite main peaks representing Fe 2p1/2 and Fe 2p3/2 were located at 724.5 and
structure to an A2BO4 RP phase, which is caused by the combination of 710.9 eV respectively. And the fitting peaks at 713.1, 727.4, and 734.0
A-site deficient and 3DOM. From the SEM images at lower magnification eV were related to Fe3+, while the fitting peaks at 711.1 eV, 716.6, and
(Fig. 1d), it can be seen intuitively that the STFN-3DOM presented a 730.8 eV were related to Fe2+ [44], indicating that Fe element had
three-dimensionally ordered macroporous structure, and the average significant mixed valence. After reduction, the Ni2+ cation was reduced
pore size is 130 nm. The SEM images of the PMMA template and the to Ni0, which can be proved by the obvious reduction of the Ni2+ peak
histogram of the pore size of STFN-3DOM are demonstrated in Fig. S3 area and the increase of the Ni0 peak area at 853.3 eV. For the Fe 2p
and Fig. S4, respectively. Fig. 1e showed the SEM image of STFN-3DOM spectrum of the reduced sample, Fe 2p1/2 and Fe 2p3/2corresponded to
finished the calcination treatment at 1150 ℃ for 2 h, which proved that two main peaks at 724.4 eV and 710.9 eV, respectively, among which
STFN-3DOM still maintained an intact macroporous structure. For the the three peaks at 710.6 eV and 724.4 eV belong to Fe2+, 714.0 eV and
reduced STFN-3DOM, the powder was characterized by TEM (Fig. 1f), 727.5 eV corresponded to Fe3+ and 707.6 eV corresponded to Fe0 [45].
and it was clearly observed that the exsolved nanoparticles with a It can be seen that Fe3+ cation was partially reduced to Fe2+ cation,
diameter of 10–50 nm were firmly bonded to the surface of the bulk which corresponded to the increase of Fe2+ peak area and the decrease
phase. Combined with TEM EDS spectrum analysis, elements such as Sr, of Fe3+ peak area, and the satellite characteristics of Fe 2p region also
Ti, Fe and O were uniformly distributed on the skeleton of the reduced showed that the existence of Fe2+ in perovskite in reducing atmosphere
STFN-3DOM, and Ni and Fe can be observed to coexist at the dissolved increases. Following reduction, the average valence states of the tran­
nanoparticles, indicating that Ni-Fe nanoparticles are formed, which sition metal elements Ni and Fe, situated at the B-site, decreased
was consistent with XRD spectrum results. The crystal plane spacing of (Table S1). As shown in Table S3, it was difficult to observe the
dissolved nanoparticles was 0.205 nm, which corresponded to the (1 1 1) enrichment of Ni0 or Fe0 on the surface of the sample reduced at 700 ◦ C
crystal plane of Ni-Fe alloy [18]. for 1 h, but there was a significant increase in the amount of Fe2+. This
indicated that during the reduction process, especially for the Fe
element in the B-site, it also contains the change from Fe3+ to Fe2+. The
3.2. Oxidation states of elements spectra of reduced Ni 2p + Fe 2 s and Fe 2p shifted to lower Eb (Fig. 2a
and b), which means that the density of local electron clouds increases
To understand the surface chemistry and oxidation state changes of [46]. In addition, combined with the results of the XPS spectra under
exsolved Ni-Fe under SOFC operating conditions, the reduction behavior 700 and 750 ℃, it can be found that the reduction of Ni was easier than
of STFN-3DOM samples was analyzed by quasi-in situ XPS. Fig. 2 that of Fe during the heating process, caused by the lower segregation
showed the spectra of Ni 2p + Fe 2 s, Fe 2p, O 1 s, and Sr 3d induced by energy of Ni [18]. In the event of a reducing atmosphere, as illustrated in
STFN-3DOM after the reduction at 750℃ and 0.5 bar. Considering the Fig. 2 c-d, O 1 s and Sr 3d showed the binding energy transition of ΔEb =
high vacuum and reducing conditions, all spectra were calibrated with 0.44 and 0.27 eV, respectively, which corresponded to the change of
the Ti 2p3/2 peak of 458.5 V. In the pristine STFN-3DOM powders, the Fermi energy level of perovskite caused by ΔpO2 between oxidation and
characteristic peak of the Ni element was located at 854.8 eV, corre­ reducing atmosphere, and other STFN perovskite also showed similar
sponding to Ni2+ 2p3/2 (Fig. 2a). It can be seen from Fig. 2b, the two

4
Y. Jiang et al. Chemical Engineering Journal 491 (2024) 151865

Fig. 2. Quasi-in-suit XPS spectra of (a)Ni 2p + Fe 2 s, (b)Fe 2p, (c)Sr 3d and (d)O 1 s of STFN-3DOM samples obtained under wet H2 at 700℃ and 750℃,
respectively. The three conditions of pristine, reduced, and redox compared in each panel.

changes in binding energy [47,48]. The O 1 s spectrum was subjected to voltage measured was approximately 1.1 V, which closely approached
deconvolution, resulting in the identification of three distinct peaks, the the ideal theoretical value obtained through the Nernst equation,
fitting peak at 529.4 eV corresponded to lattice oxygen (OLatt), the fitting signifying the presence of a dense electrolyte. According to Fig. 3a-b, the
peak at 530.3 eV corresponds to metal–oxygen bond (OM-O) came from SOFC that used the STFN-3DOM anode produced a Pmax of 1.33 W cm− 2
OH– or O2– and the fitting peak at 532.2 eV corresponds to adsorbed at 850℃ under wet H2, which was about 34 % higher than the Pmax of
oxygen (OAds) [49–51]. In the spin–orbit splitting Sr 3d3/2 and 3d5/2 the STFN-Solid anode (0.99 W cm− 2) under the same condition, indi­
spectra, the peaks at 133.01 eV and 134.75 eV can correspond to Sr in cating that the application of STFN-3DOM as the anode material can
the perovskite lattice (SrLatt). In comparison, two peaks at 133.64 eV and greatly improve the performance of SOFCs. Fig. 3c showed Arrhenius
135.44 eV can come from SrO or Sr(OH)2 species (SrO/OH) [46]. As plots for area-specific resistance (ASR) of STFN-3DOM and STFN-Solid
shown in Table S1, in the reduced STFN-3DOM sample, the accounts of in H2, it can be found that in the whole temperature range of 700-
OM-O and OAds were 7.38 % and 22.98 %, respectively, which was lower 850℃, STFN-3DOM showed a better performance than those of STFN-
than that of the pristine sample (23.62 % and 54.42 %). Compared with Solid. Fig. 3d exhibited the EIS curve of the single cell measured at
OLatt, OM-O and OAds tended to exist on the surface of STFN-3DOM. A 850℃ under open-circuit conditions in humidified H2 with STFN-3DOM
discernible reduction in the surface oxygen to lattice oxygen ratio was and STFN-Solid anode, respectively. At 850 ◦ C, the single cell with
observed following the reduction process. This reduction was indicative STFN-3DOM anode exhibited an ASR of 0.086 Ω cm2, which was nearly
of the loss of OAds and the formation of oxygen vacancies under reduc­ 1.7 times lower than that of the anode made of STFN-Solid (0.147 Ω
tion conditions [46]. Notably, this generation of additional oxygen va­ cm2). This proved that the three-dimensional ordered mesoporous
cancies and electrochemical active sites enhanced the surface oxidation structure can increase the activity of anodic hydrogen oxidation (HOR),
reaction of the anode. because the 3D structure provided a higher density of catalytic active
sites per unit volume. Different SEM images of the reduced STFN-3DOM
are exhibited in Fig. 3h. And a large number of pore structures and
3.3. Electrochemical performance exsolved nanoparticles attached to them can be clearly observed. And
the specific surface areas of the STFN-Solid anode and STFN-3DOM
LSGM’s high ionic conductivity and negligible electronic conduc­ anode are 2.358 and 33.513 m2 g− 1, respectively, which further leads
tivity make it suitable as the electrolyte for the electrolyte-supporting to the difference in particle density on their surfaces. Fig. 3e also dis­
system. Fig. S5 provided a fractured view of a single cell, and the played the Bode plot, based on the relaxation time distribution (DRT)
interface between the electrode and the dense LSGM electrolyte with a analysis, the specific subprocess that affects the HOR reaction was
thickness of about 200 μm can be clearly observed. The electrodes, identified. There were four peaks observed in both cells. The peak
characterized by a porous structure, played a crucial role in promoting ranging from10-1-101 and 102-103 Hz, was observed in the gas diffusion
the diffusion of gaseous reactants and products. and charge transform region of the anode, respectively [52]. Both re­
Fig. 3a illustrated the j-V-P curve of a single cell with STFN-3DOM gions exhibited a decrease in the STFN-3DOM anode, indicating
anode under wet H2 conditions at 700-850◦℃. The open circuit

5
Y. Jiang et al. Chemical Engineering Journal 491 (2024) 151865

Fig. 3. (a) j-V-P curves of single cells with STFN-3DOM anode fed with wet H2 at different temperatures; (b) Comparison of peak power density of single cells with
STFN-3DOM and STFN-Solid as anodes in wet H2; (c) Arrhenius plots for ASR of STFN-3DOM and STFN-Solid in wet H2; (d) Nyquist plot and Bode plot of LSTFN-
3DOM and STFN-Solid in wet H2 under 850 ℃; (e)DRT results of STFN-3DOM and STFN-Solid cells at 850 ◦ C under wet H2; (f) Redox stability test for the single cell
with STFN-3DOM anode; (g) Long-term stability test for the single cell with STFN-3DOM anode; (h) Different SEM image of reduced STFN-3DOM powders.

improved gas diffusion and charge transform under 3DOM structure. To (0.092 Ω cm2) under wet CH4 conditions (Fig. 4c). This may be due to
observe the redox stability of the STFN-3DOM anode, the single cell with the presence of more exsolved Ni-Fe nanoparticles, which makes CH4
STFN-3DOM anode was driven at a constant voltage of 0.8 V at 800◦℃ more reactive on STFN-3DOM. To evaluate the methane tolerance of the
under the condition of irregular gas switching between H2 and air at the single cell with STFN-3DOM anode, the stability test was carried out
anode side. In Fig. 3f, it can be seen that during the redox operation, under a constant voltage of 0.8 V at 800℃. As shown in Fig. 4d, single-
when STFN-3DOM was oxidized in air condition, the current density of a cell performance degraded less than 6 % after more than 80 h of testing.
single cell was reduced to about 0 A cm− 2. However, under the condition Additionally, a comprehensive assessment of the electrochemical effi­
of H2 reduction, the current density returned to the initial value and ciency of the STFN-3DOM single cell in C3H8 with a low S/C ratio at
remained stable, indicating the great redox stability. Fig. 3g provided 700–850 ◦ C was conducted. The j-V-P curve and EIS curve of the single
the long-term stability test of STFN-3DOM. The results showed that the cell were presented in Fig. 4e-f. It was shown that the single cell with
STFN-3DOM Ni-Fe nano-particle exsolution anode had good redox STFN-3DOM anode could exhibit excellent electrochemical catalytic
performance and long-term stability. Good redox stability and long-term activity in the C3H8 atmosphere even in the case of low steam supply (S/
stability were largely due to the interaction between Ni-Fe nanoparticles C = 1), the Pmax reached 0.838 Wcm− 2 at 850℃. Considering the
and RP-STFN skeleton, which has been proved by many experimental complexity of the SOFC anode reaction, the good internal reforming
results [7,17,25,26]. activity of the STFN-3DOM anode may also come from the rich oxygen
The electrochemical catalytic activity of STFN-3DOM anode in wet vacancies in the electrode.
CH4 and wet C3H8 was further studied. Fig. 4a-b presented the j-V-P
curves of single cells with STFN-3DOM and STFN-Solid anode under wet
CH4 conditions, respectively. The single cell with STFN-3DOM anode 3.4. Evaluation of catalytic performance
achieved a Pmax of 0.909 W cm− 2 at 850 ◦ C in wet CH4, with an OCV of
1.02 V, while the Pmax of that with STFN-Solid anode was 0.716 W cm− 2. Fig. 5 illustrated the catalytic performance of STFN-3DOM during
It is worth noting that the STFN-3DOM anode also showed a small ASR C3H8 steam reforming. The catalytic activity of STFN-3DOM and its
relationship with temperature were investigated, and the results were

6
Y. Jiang et al. Chemical Engineering Journal 491 (2024) 151865

Fig. 4. (a)j-V–P curves of single cells with STFN-3DOM anode fed with wet CH4 at different temperatures; (b) Comparison of peak power density of single cells with
STFN-3DOM and STFN-Solid as anodes in wet CH4; (c) EIS plot of the single cell with STFN-3DOM and STFN-Solid anode in wet CH4 at 850 ◦ C; (d) Long-term stability
test for the single cell with STFN-3DOM anode under wet CH4; (e) j-V–P curves of single cells with STFN-3DOM anode fed with C3H8 (S/C = 1) at different tem­
peratures; (f) EIS plot of STFN-3DOM cells fed with C3H8 (S/C = 1) at 850 ◦ C and 800 ◦ C.

Fig. 5. (a) Relationship between C3H8 conversion and H2 yield with temperature; (b) The relationship between CO and C2H4 yield with temperature; (c)Diagram of
product selectivity with temperature; (d)C 1 s quasi-in situ XPS spectra of STFN-3DOM after the addition of C3H8 (S/C = 1); (e) O 1 s quasi-in situ XPS spectra of
STFN-3DOM after the addition of C3H8 (S/C = 1) (f) Schematic diagram of C3H8 steam reforming principle.

presented in Fig. 5a. It was observed that the catalytic activity of STFN- balance calculation is shown in Table S4. The elemental analysis and
3DOM was positively correlated with temperature, and at 850 ◦ C, the Raman results of the tested samples are shown in Table S2 and Fig. S7.
STFN-3DOM exhibited a C3H8 conversion rate of 96.86 %. The yield of Under the same propane test conditions, the samples of STFN-3DOM
H2 and CH4 also increased with the increasing temperature (Fig. 5a, c). were tested by quasi-in-situ XPS to observe the reaction of C3H8 on
Fig. 5b-c showed that the yield and selectivity of CO increased with the active site. Fig. 5d and 5e presented quasi-in-situ XPS spectra of C 1 s
increasing temperature. In contrast, the selectivity of C2H4 decreased and O 1 s, respectively, obtained from the STFN-3DOM surface in the
significantly at 850 ◦ C. At the same time, the selectivity of C3H6 presence of 0.5 bar in the C3H8 condition at 750 ◦ C with the low S/C. The
decreased with increasing temperature, indicating that the cracking presence of a peak at 286.5 eV in the C 1 s region was evidence of the
reaction was inhibited with the increase in temperature. The carbon formation of CH3O– species on the STFN-3DOM surface during propane

7
Y. Jiang et al. Chemical Engineering Journal 491 (2024) 151865

reforming [53]. And the peak at 285.2 eV can be attributed to methyl were characterized by Raman spectroscopy. The results established that
carbon (–CH3) in the decomposition product, which was directly STFN-3DOM demonstrates superior anti-carbon deposition capabilities
adsorbed on the surface of STFN-3DOM [54–56]. And no graphite car­ (Fig. S6). Therefore, it was obvious that the great performance of the
bon peak at 284.6 eV was observed, suggesting there was not much STFN-3DOM sample at high temperatures and low S/C ratio was more
carbon deposition during this process. When propane existed, the broad suitable for the harsh operating conditions of SOFC. The decline in
peak at 532 eV may come from the methoxy group and hydroxyl group, catalytic activity of commercial Ni/Al2O3 catalysts can be attributed to
which may be produced when propane was adsorbed and dissociated at the inadequate interaction between the metal and substrate. This sub­
the Ni-Fe site and transferred to the surface of reduced STFN-3DOM to optimal interaction led to the deposition of coke and the accumulation of
combine with lattice oxygen. The weak peak band at 536–537 eV cor­ nanoparticles, which impeded the efficient functioning of the catalyst
responded to the gaseous products (CO and CO2) formed when propane [18].
was oxidized on the surface [57,58]. Because of the low S/C ratio in this
experimental condition, the production of the two products was not 3.5. Theoretical analysis
obvious, which was also consistent with the composition of gas phase
products of propane reforming at 750℃. From the above results, it is Two possible paths of adsorption and dissociation of propane were
speculated that the main source of its catalytic activity came from the simulated by DFT calculation. The kinetics of C-H bond activation and
various active metal nanoparticles exsolved on its porous structure dissociation of C3H8 and the subsequent H transform process were
surface and the large concentration of deficient (Fig. 5f). studied when the primary carbon and secondary carbon of C3H8 were
Considering the excellent conversion efficiency of STFN-3DOM for adsorbed on the (1 1 0) surface of reduced STFN-3DOM, respectively.
propane steam reforming, the commercial Ni/Al2O3 catalyst was The motivation for this decision was due to a recognized cognition that
selected to compare the anti-carbon deposition performance. The TEM in the reforming reaction, the decisive step of the initial propane
image of Ni/Al2O3 catalyst was shown in Fig. S8. Unfortunately, the Ni/ dehydrogenation process to produce C1 intermediate was the dissocia­
Al2O3 catalyst developed severe carbon deposition at 400 ◦ C, making tion of C3H8* (adsorbed state), and oxygen addition was more likely to
further experiments impossible. This may be due to the difference in the occur on CH/C intermediates[41,59,60]. The first dehydrogenation ac­
performance of the two samples in anti-carbon deposition at high tem­ tivity of propane through primary and secondary carbon was calculated
peratures and low S/C ratios. Therefore, both samples were tested for to understand the influence of the surface chemistry of STFN-3DOM on
their resistance to carbon deposition in pure C3H8. STFN-3DOM and Ni/ the initial activation of propane. Fig. 6a-b illustrated the dehydrogena­
Al2O3 were placed in a dry C3H8 atmosphere at 750℃, and the samples tion paths of propane adsorbed on the (1 1 0) surface of reduced STFN-

Fig. 6. Optimized geometry structures for the C3H8 reforming reaction for (a) primary carbon path and (b) secondary carbon path; Electron density difference maps
of (c) the primary carbon of C3H8 adsorption process and (d) the secondary carbon of C3H8 adsorption process; (e) Potential energy profile for the C3H8 reforming
reaction on reduced STFN-3DOM in two paths; (f) Projected density of states of the reduced STFN-3DOM.

8
Y. Jiang et al. Chemical Engineering Journal 491 (2024) 151865

3DOM with primary carbon and secondary carbon, respectively. In the

primary carbon path, the primary carbon atom of C3H8 was adsorbed on
the Ni-Fe nanocluster, and the H atom was combined with the Fe-Ni
atom on the Ni-Fe nanocluster, and then the C-H bond cracked. The
reaction energy was 0.26 eV, and the kinetic barrier was 1.22 eV. In the
secondary carbon path, the kinetic barrier of C-H bond breakage was
1.19 eV and the reaction energy was 0.24 eV, and both paths show
satisfactory reactivity (Fig. 6e). It can be seen from the DFT calculation
that the process of hydrogen transform tended to be spontaneous in
thermodynamics. C3H8 was firstly adsorbed on the Ni-Fe active nano­
particles and then dissociated. The dissociated H+ proton will be further
adsorbed on the oxygen of the RP-STFN skeleton to form an H-O bond so
that the subsequent HOR reaction can take place.
To understand the origin of the interaction between reduced STFN-
3DOM and C3H8, the electronic structure was analyzed by calculating
the charge density difference of selected atoms (Fe and O atoms, C atoms
and H atoms on the surface of STFN-3DOM) and projected density of
states (PDOS) of selected atoms (Fe atoms, O atoms, Sr atoms and Ni
atoms on the surface of reduced STFN-3DOM) in the transition state of
propane dehydrogenation and dissociation. In Fig. 6c-d, the electron
density difference diagram showed that the density around C3H8-Ni-Fe
nanoparticles-O was high, which can confirm the interaction between Fig. 7. Schematic diagram of predicted propane reforming reaction path.
C3H8 and Ni-Fe-O interface. As illustrated in Fig. 6f, the PDOS bias of Fe,
Ni, and O in proximity to the Fermi level in the transition state of pro­ 4. Conclusions
pane dehydrogenation and dissociation was significant, thus promoting
the adsorption and transformation of propane. In this work, the innovative A-site deficient Ni-doped
According to the results of the catalyst test, STFN-3DOM had high Sr0.9Ti0.3Fe0.63Ni0.07O3‑δ perovskite anode with a three-dimensionally
performance to catalyze the steam reforming of propane. It was worth ordered macroporous structure has been developed. The large specific
noting that the A-site deficient of perovskite materials and the close surface area and A-site deficient of STFN-3DOM provide more exsolved
combination of bimetallic Ni-Fe nanoparticles and support are of great Ni-Fe active sites, which remarkably improved the performance of SOFC
significance to the reforming reaction. Under the SOFC operation, RP- fueled with hydrocarbon. XRD and quasi-in-suit XPS are used to reveal
STFN substrate creates many RP-STFN-Ni-Fe interfaces in SOFCs due the change of crystal structure and chemical state of elements during the
to in-situ exsolved Ni-Fe nanoparticles and the consequent excess of exsolution process. At 850 ◦ C, the single cell with an STFN-3DOM anode
oxygen vacancies. The resulting interfaces were of particular signifi­ reached a Pmax of 1.33 W cm− 2 in wet H2; while the fuel was C3H8 with a
cance in enhancing the electrochemical performance of the single cell, as low S/C, the single cell also reached a Pmax of 0.838 W cm− 2. In the
they were one of the main places where C3H8 reforming, H2 dissociation, meantime, the C3H8 conversion achieved 98.75 % at 850 ◦ C, and the
and transform reactions occurred. These processes greatly improve the yields of H2 and CH4 increased with the increasing temperature without
adsorption, dissociation, and electrochemical oxidation kinetics of H2 obvious carbon deposition. The single cell with STFN-3DOM anode
and hydrocarbon fuel while effectively avoiding carbon deposition. The exhibited a stable current output of more than 110 h and 80 h in wet H2
primary reaction process we predict was shown in Fig. 7. Propane can be at 750 ◦ C and wet CH2 at 800 ◦ C, and the redox stability was evaluated
easily adsorbed to the Ni-Fe active site on the surface of STFN-3DOM for 80 h under a constant voltage of 0.8 V at 800 ◦ C. on the development
through a C-M bond (M = Ni, Fe), and dissociated on it, forming an of direct hydrocarbon fuel anodes of SOFCs for applications in stationary
intermediate species C3H8*. C-H cleavage occurred at the edge or middle combined heat and power cogeneration and mobile APUs.
methyl group and produced C3H7* and H*. H* was transferred from the
Ni site to lattice oxygen of RP-STFN to form hydroxyl (OH*), in which CRediT authorship contribution statement
the dissociation of the C-H bond was carried out through a transition
state. With the deep oxidation of propane, several C-H cleavage steps Yuying Jiang: Writing – original draft, Visualization, Methodology,
occurred before one C-C cleavage step; that was, the reaction will follow Formal analysis. Jiamei Liu: Writing – original draft. Bo Cheng:
the path of C3H6 *+H *-C3H5 *+H *, and finally, the product of CH3C* Methodology. Xinyue Dang: Investigation. Huaiqiang Su: Writing –
will be generated. At the same time, OH* on the surface of RP-STFN review & editing. Yani Hua: Writing – review & editing. Zhan Gao:
gained electrons from transition metal ions, forming H2O molecules Writing – review & editing.
and leaving oxygen vacancies on the surface. Under the working con­
dition of SOFC, The O2– was continuously transported from the cathode
to the anode to regenerate the lattice oxygen consumed and facilitate the Declaration of competing interest
catalytic cycle. It is important to note that the reaction efficiency is
dependent largely on the temperature at a constant S/C ratio. It can be The authors declare that they have no known competing financial
observed that during the C3H8 reforming process, although the yields of interests or personal relationships that could have appeared to influence
H2 and CH4 increased as the temperature rose, hydrogen generation was the work reported in this paper.
easier. This is because lower temperatures are more conducive to the
production of CH4, as the methanation reaction between CO or CO2 and Data availability
H2 is a highly exothermic process. Conversely, higher temperatures
leaded to higher hydrogen production, as propane reforming was an Data will be made available on request.
endothermic process. In addition, the product consistently maintained a
low level of CO content, which may be due to the low S/C ratio. Acknowledgments

This work is supported by the Youth Talents Program of China, the

9
Y. Jiang et al. Chemical Engineering Journal 491 (2024) 151865

National Science Foundation of China-NSAF Joint Fund (CN) (Grant No. [18] A.I. Tsiotsias, B. Ehrhardt, B. Rudolph, L. Nodari, S. Kim, W. Jung, et al., Bimetallic
Exsolved Heterostructures of Controlled Composition with Tunable Catalytic
U2230113), the Key Research and Development Projects of Shaanxi
Properties, ACS Nano 16 (2022) 8904–8916, https://doi.org/10.1021/
Province (2021GXLH-Z-072), Natural Science Basic Research Program acsnano.1c11111.
of Sichuan (NO. 23NSFSC2969), Natural Science Basic Research Pro­ [19] T.L. Zhu, H.E. Troiani, L.V. Mogni, M.F. Han, S.A. Barnett, Ni-Substituted Sr(Ti, Fe)
gram of Shaanxi (NO. 2023-JC-QN-0127), the Higher Education Insti­ O3 SOFC Anodes: Achieving High Performance via Metal Alloy Nanoparticle
Exsolution, Joule. 2 (2018) 478–496, https://doi.org/10.1016/j.
tution Academic Discipline Innovation and Talent Introduction Plan joule.2018.02.006.
(“111 Plan”) (No.B23025), Project of Shaanxi Innovative Talent Pro­ [20] Z.H. Du, H.L. Zhao, S. Yi, Q. Xia, Y. Gong, Y. Zhang, et al., High-Performance
motion Plan-Technology Innovation Team (No. 2024RS-CXTD-35). The Anode Material Sr2FeMo0.65Ni0.35O6-delta with In Situ Exsolved Nanoparticle
Catalyst, ACS Nano 10 (2016) 8660–8669, https://doi.org/10.1021/
Research project of the State Key Laboratory of Electrical Insulation and acsnano.6b03979.
Power Equipment. Additionally, we appreciate the teachers at the In­ [21] O. Kwon, K. Kim, S. Joo, H.Y. Jeong, J. Shin, J.W. Han, et al., Self-assembled alloy
strument Analysis Center of Xi’an Jiaotong University for their help with nanoparticles in a layered double perovskite as a fuel oxidation catalyst for solid
oxide fuel cells, J. Mater. Chem. A 6 (2018) 15947–15953, https://doi.org/
material characterizations and analyses. 10.1039/c8ta05105d.
[22] O. Kwon, S. Sengodan, K. Kim, G. Kim, H.Y. Jeong, J. Shin, et al., Exsolution trends
Appendix A. Supplementary data and co-segregation aspects of self-grown catalyst nanoparticles in perovskites, Nat.
Commun. 8 (2017), https://doi.org/10.1038/ncomms15967.
[23] C.Y. Dong, Y.L. Li, D.Y. Cheng, M.T. Zhang, J.J. Liu, Y.G. Wang, et al., Supported
Supplementary data to this article can be found online at https://doi. Metal Clusters: Fabrication and Application in Heterogeneous Catalysis, ACS Catal.
org/10.1016/j.cej.2024.151865. 10 (2020) 11011–11045, https://doi.org/10.1021/acscatal.0c02818.
[24] I. Ro, J. Resasco, P. Christopher, Approaches for Understanding and Controlling
Interfacial Effects in Oxide-Supported Metal Catalysts, ACS Catal. 8 (2018)
References 7368–7387, https://doi.org/10.1021/acscatal.8b02071.
[25] W. Kobsiriphat, B.D. Madsen, Y. Wang, L.D. Marks, S.A. Barnett, La0.8Sr0.2Cr1-
[1] T.M. Gur, Comprehensive review of methane conversion in solid oxide fuel cells: xRuxO3-delta-Gd0.1Ce0.9O1.95 solid oxide fuel cell anodes: Ru precipitation and
Prospects for efficient electricity generation from natural gas, Prog. Energy electrochemical performance, Solid State Ion. 180 (2009) 257–264, https://doi.
Combust. Sci. 54 (2016) 1–64, https://doi.org/10.1016/j.pecs.2015.10.004. org/10.1016/j.ssi.2008.12.022.
[2] J.G. Lee, J.H. Park, Y.G. Shul, Tailoring gadolinium-doped ceria-based solid oxide [26] T. Tan, M.X. Qin, K. Li, M.Y. Zhou, T.K. Liu, C.H. Yang, et al., In-situ exsolved NiFe
fuel cells to achieve 2Wcm-2 at 550 degrees C. Nature, Communications 5 (2014), alloy nanoparticles on Pr0.8Sr1.2(NiFe)O4-delta for direct hydrocarbon fuel solid
https://doi.org/10.1038/ncomms5045. oxide fuel cells, Int. J. Hydrogen Energy 45 (2020) 29407–29416, https://doi.org/
[3] Park S, Kim H, McIntosh S, Worrell W, Gorte RJ, Vohs JM. Anodes for direct 10.1016/j.ijhydene.2020.07.250.
oxidation of hydrocarbons in solid oxide fuel cells. 7th International Symposium on [27] T. Gan, X.Q. Fan, Y. Liu, C.Y. Wang, H.R. Mei, L.J. Fan, et al., A highly active Ni/
Solid Oxide Fuel Cells. Tsukuba, Japan2001. p. 712-8. https://doi.org/10.1002/ Ce0.8Sm0.2O1.9 anode catalyst with a three-dimensionally ordered macroporous
1521-4095(200010)12:19<1465::AID-ADMA1465>3.0.CO;2-9. structure for solid oxide fuel cells, J. Mater. Chem. A 8 (2020) 7792–7800, https://
[4] Z.P. Shao, S.M. Haile, A high-performance cathode for the next generation of solid- doi.org/10.1039/c9ta14034d.
oxide fuel cells, Nature 431 (2004) 170–173, https://doi.org/10.1038/ [28] D.H. Kim, J.H. Jeong, H.C. Woo, M.H. Kim, Synthesis of highly porous polymer
nature02863. microspheres with interconnected open pores for catalytic microreactors, Chem.
[5] S.W. Tao, J.T.S. Irvine, A redox-stable efficient anode for solid-oxide fuel cells, Nat. Eng. J. 420 (2021), https://doi.org/10.1016/j.cej.2020.127628.
Mater. 2 (2003) 320–323, https://doi.org/10.1142/9789814317665_0037. [29] R. Li, P.Q. Wang, S.B. Ma, F.L. Yuan, Z.B. Li, Y.J. Zhu, Excellent selective catalytic
[6] B. Hua, M. Li, Y.F. Sun, J.H. Li, J.L. Luo, Enhancing Perovskite Electrocatalysis of reduction of NOx by NH3 over Cu/SAPO-34 with hierarchical pore structure, Chem.
Solid Oxide Cells Through Controlled Exsolution of Nanoparticles, ChemSusChem Eng. J. 379 (2020), https://doi.org/10.1016/j.cej.2019.122376.
10 (2017) 3333–3341, https://doi.org/10.1002/cssc.201700936. [30] N. Patel, R. Fernandes, S. Gupta, R. Edla, D.C. Kothari, A. Miotello, Co-B catalyst
[7] H.S. Kim, Y. Jeon, J.H. Kim, G.Y. Jang, S.P. Yoon, J.W. Yun, Characteristics of Sr1- supported over mesoporous silica for hydrogen production by catalytic hydrolysis
xYxTi1-yRuyO3+/ - delta and Ru-impregnated Sr1-xYxTiO3+/-delta perovskite of Ammonia Borane: A study on influence of pore structure, Applied Catalysis B-
catalysts as SOFC anode for methane dry reforming, Appl. Surf. Sci. 510 (2020), Environmental. 140 (2013) 125–132, https://doi.org/10.1016/j.
https://doi.org/10.1016/j.apsusc.2020.145450. apcatb.2013.03.046.
[8] M.X. Qin, T. Tan, K. Li, Z.M. Wang, H.Y. Yang, Z.J. Liu, et al., In-situ exsolved FeRu [31] L. Almar, A. Morata, M. Torrell, M. Gong, T. Andreu, M. Liu, et al., A Durable
alloy nanoparticles on Ruddlesden-Popper oxides for direct hydrocarbon fuel solid Electrode for Solid Oxide Cells: Mesoporous Ce0.8Sm0.2O1.9 Scaffolds Infiltrated
oxide fuel cells, Int. J. Hydrogen Energy 45 (2020) 21464–21472, https://doi.org/ with a Sm0.5Sr0.5CoO3-delta Catalyst, Electrochim. Acta 235 (2017) 646–653,
10.1016/j.ijhydene.2020.05.242. https://doi.org/10.1016/j.electacta.2017.03.149.
[9] J.W. Zhang, M.R. Gao, J.L. Luo, In Situ Exsolved Metal Nanoparticles: A Smart [32] B. Liang, T. Suzuki, K. Hamamoto, T. Yamaguchi, H. Sumi, Y. Fujishiro, et al.,
Approach for Optimization of Catalysts, Chem. Mater. 32 (2020) 5424–5441, A reduced temperature solid oxide fuel cell with three-dimensionally ordered
https://doi.org/10.1021/acs.chemmater.0c00721. macroporous cathode, J. Power Sources 212 (2012) 86–92, https://doi.org/
[10] M. Li, T. Hong, J. Cheng, Z. Chen, C. Xu, Enhanced proton conductivity of 10.1016/j.jpowsour.2012.03.107.
La0.95Ca0.05NixNb1-xO4 (0.01 <= x <= 0.3) through in-situ exsolution of [33] L. Almar, B. Colldeforns, L. Yedra, S. Estrade, F. Peiro, A. Morata, et al., High-
metallic nanocatalysts, Mater. Chem. Phys. 208 (2018) 226–236, https://doi.org/ temperature long-term stable ordered mesoporous Ni-CGO as an anode for solid
10.1016/j.matchemphys.2018.01.054. oxide fuel cells, J. Mater. Chem. A 1 (2013) 4531–4538, https://doi.org/10.1039/
[11] Y. Wang, X. Lei, Y. Zhang, F. Chen, T. Liu, In-situ growth of metallic nanoparticles c3ta10439g.
on perovskite parent as a hydrogen electrode for solid oxide cells, J. Power Sources [34] G. Mueller, R.-N. Vannier, A. Ringuede, C. Laberty-Robert, C. Sanchez,
405 (2018) 114–123, https://doi.org/10.1016/j.jpowsour.2018.10.023. Nanocrystalline, mesoporous NiO/Ce0.9Gd0.1O2-delta thin films with tuned
[12] J. Li, J. Hou, X. Xi, Y. Lu, M. Li, Y. Fan, et al., Cogeneration of ethylene and microstructures and electrical properties: in situ characterization of electrical
electricity in symmetrical protonic solid oxide fuel cells based on a responses during the reduction of NiO, J. Mater. Chem. A 1 (2013) 10753–10761,
La0.6Sr0.4Fe0.8Nb0.1Cu0.1O3-δ electrode, J. Mater. Chem. A 8 (2020) 25978–25985, https://doi.org/10.1039/c3ta11175j.
https://doi.org/10.1039/d0ta08974e. [35] G. Kresse, J. Furthmuller, Efficiency of ab-initio total energy calculations for metals
[13] M. Cimenti, M.A. Buccheri, J.M. Hill, Direct Utilization of Methanol and Ethanol on and semiconductors using a plane-wave basis set, Comput. Mater. Sci 6 (1996)
La0.75Sr0.25Cr0.5Mn0.5O3-δ Anodes for Solid Oxide Fuel Cells, Electrocatalysis 3 15–50, https://doi.org/10.1016/0927-0256(96)00008-0.
(2012) 59–67, https://doi.org/10.1007/s12678-011-0076-8. [36] J.P. Perdew, K. Burke, M. Ernzerhof, Generalized gradient approximation made
[14] H. Cai, L. Zhang, J. Xu, J. Huang, X. Wei, L. Wang, et al., Cobalt-free simple (vol 77, pg 3865, Phys. Rev. Lett. 1997;78:1396- (1996), https://doi.org/
La0.5Sr0.5Fe0.9Mo0.1O3-δ electrode for symmetrical SOFC running on H2 and CO 10.1103/PhysRevLett.77.3865.
fuels, Electrochim. Acta 320 (2019), https://doi.org/10.1016/j. [37] S. Grimme, J. Antony, S. Ehrlich, H. Krieg, A consistent and accurate ab initio
electacta.2019.134642. parametrization of density functional dispersion correction (DFT-D) for the 94
[15] A. Marcucci, F. Zurlo, I.N. Sora, E. Placidi, S. Casciardi, S. Licoccia, et al., A redox elements H-Pu, J. Chem. Phys. 132 (2010), https://doi.org/10.1063/1.3382344.
stable Pd-doped perovskite for SOFC applications, J. Mater. Chem. A 7 (2019) [38] G. Kresse, D. Joubert, From ultrasoft pseudopotentials to the projector augmented-
5344–5352, https://doi.org/10.1039/c8ta10645b. wave method, Phys. Rev. B 59 (1999) 1758–1775, https://doi.org/10.1103/
[16] A. Donazzi, T.A. Schmauss, S.A. Barnett, Catalytic and electrocatalytic PhysRevB.59.1758.
performance of Sr(Ti0.3Fe0.7Ru0.07)O3-δ for applications in solid oxide fuel cells [39] S.L. Dudarev, G.A. Botton, S.Y. Savrasov, C.J. Humphreys, A.P. Sutton, Electron-
supplied with ethanol steam reforming mixtures, J. Power Sources 551 (2022), energy-loss spectra and the structural stability of nickel oxide: An LSDA+U study,
https://doi.org/10.1016/j.jpowsour.2022.232215. Phys. Rev. B 57 (1998) 1505–1509, https://doi.org/10.1103/PhysRevB.57.1505.
[17] K.J. Kim, C. Lim, K.T. Bae, J.J. Lee, M.Y. Oh, H.J. Kim, et al., Concurrent promotion [40] K.M. Ho, C.L. Fu, B.N. Harmon, W. Weber, D.R. Hamann, Vibrational Frequencies
of phase transition and bimetallic nanocatalyst exsolution in perovskite oxides And Structural-Properties Of Transition-Metals Via Total-Energy Calculations,
driven by Pd doping to achieve highly active bifunctional fuel electrodes for Phys. Rev. Lett. 49 (1982) 673–676, https://doi.org/10.1103/PhysRevLett.49.673.
reversible solid oxide electrochemical cells, Applied Catalysis B-Environmental. [41] T. Tan, Z. Wang, K. Huang, C. Yang, High-Performance Co-production of Electricity
314 (2022), https://doi.org/10.1016/j.apcatb.2022.121517. and Light Olefins Enabled by Exsolved NiFe Alloy Nanoparticles from a Double-

10
Y. Jiang et al. Chemical Engineering Journal 491 (2024) 151865

Perovskite Oxide Anode in Solid Oxide-Ion-Conducting Fuel Cells, ACS Nano 17 photoelectron spectra, Results Phys. 9 (2018) 866–870, https://doi.org/10.1016/j.
(2023) 13985–13996, https://doi.org/10.1021/acsnano.3c03956. rinp.2018.03.045.
[42] H. Xu, D. Cheng, D. Cao, X.C. Zeng, A universal principle for a rational design of [52] Y. Zhang, H. Zhao, Z. Du, K. Swierczek, Y. Li, High-Performance SmBaMn2O5+δ
single-atom electrocatalysts, Nat. Catal. 1 (2018) 339–348, https://doi.org/ Electrode for Symmetrical Solid Oxide Fuel Cell, Chem. Mater. 31 (2019)
10.1038/s41929-018-0127-0. 3784–3793, https://doi.org/10.1021/acs.chemmater.9b01012.
[43] G. Henkelman, B.P. Uberuaga, H. Jónsson, A climbing image nudged elastic band [53] N. Aas, T.J. Pringle, M. Bowker, Adsorption And Decomposition Of Methanol On
method for finding saddle points and minimum energy paths, J. Chem. Phys. 113 TiO2, SrTiO3 And SrO, Journal of the Chemical Society-Faraday Transactions. 90
(2000) 9901–9904, https://doi.org/10.1063/1.1329672. (1994) 1015–1022, https://doi.org/10.1039/ft9949001015.
[44] C.S. Kuivila, J.B. Butt, P.C. Stair, Characterization Of Surface Species On Iron [54] Z. Liu, T. Duchon, H. Wang, D.C. Grinter, I. Waluyo, J. Zhou, et al., Ambient
Synthesis Catalysts By X-Ray Photoelectron-Spectroscopy, Appl. Surf. Sci. 32 pressure XPS and IRRAS investigation of ethanol steam reforming on Ni-CeO2(111)
(1988) 99–121, https://doi.org/10.1016/0169-4332(88)90076-1. catalysts: an in situ study of C-C and O-H bond scission, PCCP 18 (2016)
[45] J. Sun, P. Zan, X. Yang, L. Ye, L. Zhao, Room-temperature synthesis of Fe3O4/Fe- 16621–16628, https://doi.org/10.1039/c6cp01212d.
carbon nanocomposites with Fe-carbon double conductive network as [55] C.T. Au, W. Hirsch, W. Hirschwald, Adsorption And Interaction Of Methanol With
supercapacitor, Electrochim. Acta 215 (2016) 483–491, https://doi.org/10.1016/j. Zinc-Oxide - Single-Crystal Faces And Zinc-Oxide Copper Catalyst Surfaces Studied
electacta.2016.08.139. By Photoelectron-Spectroscopy (XPS and UPS), Surf. Sci. 221 (1989) 113–130,
[46] M. Santaya, C.E. Jimenez, H. Esteban Troiani, E.A. Carbonio, M. Damian Arce, https://doi.org/10.1016/0039-6028(89)90570-0.
L. Maria Toscani, et al., Tracking the nanoparticle exsolution/reoxidation processes [56] Z. Liu, T. Duchon, H. Wang, E.W. Peterson, Y. Zhou, S. Luo, et al., Mechanistic
of Ni-doped SrTi0.3Fe0.7O3-δ electrodes for intermediate temperature symmetric Insights of Ethanol Steam Reforming over Ni-CeOx(111): The Importance of
solid oxide fuel cells, J. Mater. Chem. A 10 (2022) 15554–15568, https://doi.org/ Hydroxyl Groups for Suppressing Coke Formation, J. Phys. Chem. C 119 (2015)
10.1039/d2ta02959f. 18248–18256, https://doi.org/10.1021/acs.jpcc.5b04310.
[47] A. Nenning, A.K. Opitz, C. Rameshan, R. Rameshan, R. Blume, M. Haevecker, et al., [57] D.R. Mullins, P.M. Albrecht, Acetaldehyde Adsorption and Reaction on CeO2(100)
Ambient Pressure XPS Study of Mixed Conducting Perovskite-Type SOFC Cathode Thin Films, J. Phys. Chem. C 117 (2013) 14692–14700, https://doi.org/10.1021/
and Anode Materials under Well-Defined Electrochemical Polarization, J. Phys. jp404752m.
Chem. C 120 (2016) 1461–1471, https://doi.org/10.1021/acs.jpcc.5b08596. [58] D.R. Mullins, P.M. Albrecht, T.-L. Chen, F.C. Calaza, M.D. Biegalski, H.M. Christen,
[48] A. Rothschild, W. Menesklou, H.L. Tuller, E. Ivers-Tiffee, Electronic structure, et al., Water Dissociation on CeO2(100) and CeO2(111) Thin Films, J. Phys. Chem.
defect chemistry, and transport properties of SrTi1-xFexO3-y solid solutions, Chem. C 116 (2012) 19419–19428, https://doi.org/10.1021/jp306444h.
Mater. 18 (2006) 3651–3659, https://doi.org/10.1021/cm052803x. [59] K. Lee, E. Lee, C. Song, M.J. Janik, Density functional theory study of propane
[49] P.A.W. van der Heide, Systematic x-ray photoelectron spectroscopic study of La1- steam reforming on Rh-Ni bimetallic surface: Sulfur tolerance and scaling/
xSrx-based perovskite-type oxides, Surf. Interface Anal. 33 (2002) 414–425, Bronsted-Evans-Polanyi relations, J. Catal. 309 (2014) 248–259, https://doi.org/
https://doi.org/10.1002/sia.1227. 10.1016/j.jcat.2013.10.006.
[50] J.C. Dupin, D. Gonbeau, P. Vinatier, A. Levasseur, Systematic XPS studies of metal [60] Y. Song, S. Laursen, Effect of bulk and surface composition of Ni plus Ga
oxides, hydroxides and peroxides, PCCP 2 (2000) 1319–1324, https://doi.org/ intermetallic compound catalysts in propane steam/wet reforming: Origins of
10.1039/a908800h. nearly ideal experimental product selectivity, J. Catal. 417 (2023) 260–273,
[51] L.L. Ding, L.Q. Wu, X.S. Ge, Y.N. Du, J.J. Qian, G.D. Tang, et al., Study of average https://doi.org/10.1016/j.jcat.2022.12.009.
valence and valence electron distribution of several oxides using X-ray

11