Positional influence of Ru on Perovskite structured catalysts for efficient H2 production process by heavy-hydrocarbon source

Article abstract of DOI:10.1016/j.apcata.2019.117111

Direct conversion of heavy-hydrocarbon source into H₂–rich gas fuels has been highlighted as an efficient ultraclean system. Dictating the productivity of the system, the most important factor lies with the activity and physical/chemical stabil

Full text of DOI:10.1016/j.apcata.2019.117111

Accepted Manuscript
Title: Positional influence of Ru on Perovskite structured
catalysts for efficient H2 production process by
heavy-hydrocarbon source
Authors: Yukwon Jeon, Ohchan Kwon, Chanmin Lee,
Gicheon Lee, Jae-ha Myung, Sang Sun Park, John T.S. Irvine,
Yong-gun Shul
PII:
DOI:
Article Number:
S0926-860X(19)30265-0
https://doi.org/10.1016/j.apcata.2019.117111
117111
Reference:
APCATA 117111
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
10 March 2019
3 June 2019
12 June 2019
Please cite this article as: Jeon Y, Kwon O, Lee C, Lee G, Myung J-ha, Sun Park S,
Irvine JTS, Shul Y-gun, Positional influence of Ru on Perovskite structured catalysts
for efficient H2 production process by heavy-hydrocarbon source, Applied Catalysis A,
General (2019), https://doi.org/10.1016/j.apcata.2019.117111
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Positional influence of Ru on Perovskite structured catalysts
for efficient H2 production process by heavy-hydrocarbon
source
b,
c
b
d
b
Yukwon Jeona,b,†, Ohchan Kwon , Chanmin Lee , Gicheon Lee , Jae-ha Myung , Sang Sun Park ,
†
John T. S. Irvinea and Yong-gun Shulb,*
a School of chemistry, University of St Andrews, St Andrews, Fife KY16 9ST, United Kingdom.
b Department of Chemical and Biomolecular Engineering, Yonsei University, 50 Yonsei-ro,
Seodaemun-gu, Seoul, 03722, Republic of Korea.
c Research Institute of Sustainable Manufacturing System, Intelligent Sustainable Materials
R&D Group, Korea Institute of Industrial Technology (KITECH), 89 Yangdaegiro-gil, Ipjang-
myeon, Seobuk-gu, Cheonan-si, Chungcheongnam-do, 31056, Republic of Korea.
d Department of Materials Science and Engineering, Incheon National University, 119 Academy
street, Songdo 1-dong, Yeonsu-gu, Incheon, 22012, Republic of Korea.
†These authors contributed equally to this work.
*Corresponding author: Yong-Gun Shul
E-mail: shulyg@yonsei.ac.kr
Tel.: +82-2-2123-2758, Fax: +82-02-2123-3554
1

Table of Contents Graphic
Highlights
 Perovskites for direct conversion of heavy-hydrocarbon source into H2–rich gas fuels
 Lattice and Surface type Ru were designed by changing the position and loading
 Positional and chemical changes of Ru species affected to the reaction mechanism
 Lattice doped Ru shows great activity with less coke formation/sulfur poisoning
2

ABSTRACT
Direct conversion of heavy-hydrocarbon source into H2–rich gas fuels has been
highlighted as an efficient ultraclean system. Dictating the productivity of the system, the most
important factor lies with the activity and physical/chemical stability of the catalytic materials. In
this work, perovskite hollow fiber catalysts by partial substitution of Cr with Ru in the B site for
LaCrO3 (LaCr1-xRuxO3, x=0–0.4) are designed to investigate the effects of the positional and
chemical changes of Ru on the catalytic properties at autothermal reforming of heavy-hydrocarbon.
From a novel synthesis method, a hollow fibrous perovskite nanoparticle network is prepared to
provide high surface area. Depending on the Ru doping levels, the optimal Ru amount exists as
lattice type Ru, while an excess Ru resulted in the formation of surface type Ru on the perovskite.
Especially, a slight structural modification of the orthorhombic structure owing from the doped Ru
results in the increase of the surface oxygen and reducibility due to the high-valanced Cr and Ru
states. After autothermal reforming of heavy-hydrocarbon including sulfur, it is obvious that the
mixed lattice and surface Ru type shows a large deactivation due to the strong binding energy with
carbon and sulfur. In contrast, less coke formation and sulfur poisoning are revealed for the catalyst
with mostly lattice type Ru through the stable Ru species and high amount of active oxygen.
Consequently, the hollow perovskite fiber catalyst of LaCr0.8Ru0.2O3 displays the optimal activity
to extract the highest amount of H2 from the heavy-hydrocarbon source with exceptional stability.
KEYWORDS: Perovskite catalyst • Ru position • Stability • H2 production • coke/sulfur
resistance
3

1. Introduction
Hydrogen (H2) holds great promise as a fuel owing to the increasing demand for alternative
energy sources such as new chemical processes and fuel cell systems.[1] Among many
technologies used for producing H2, reforming of fossil fuels is the most practical method because
of its high energy efficiency. Heavy-hydrocarbon source, like diesel or gasoline, is one of the fuel
candidate to practically establish H2 infrastructure due to its rich H2 content with high gravimetric
and volumetric energy densities, easy storage and transport with accessibility.[2] The technology
development to extract H2 from heavy-hydrocarbons could also aid in the transition to use new
energy sources such as gas oil, bio-oil and biomass etc. Autothermal reforming (ATR), a
combination of steam reforming and partial oxidation reactions, has been widely studied for H2
production from heavy-hydrocarbons due to its high yields of H2 with low coke formation and fast
reaction start-up, etc.[3] However, the main problem is the decrease in catalyst stability caused by
an agglomeration of the metal active sites, sulfur poisoning and carbon coking, especially in such
severe operating conditions.[4] Thus, high-efficiency H2 production is mainly dependent on the
catalyst design including catalytic activity/stability and resistance to the deposited carbon and
sulfur. Precious metals, including Pt, Rh, Ru, and Pd, are typically used as catalysts deposited onto
carefully engineered supports.[5] However, more catalyst studies are still needed to alleviate the
significant deactivations in heavy-hydrocarbon reforming.
In recent years, perovskite catalysts have been investigated as candidates in the H2
production reactions, due to their stability at high temperature, redox, and H2-rich
environments.[6] In addition, perovskites are resistant to coke deposition and have a low chemical
binding energy with sulfur, which make them suitable for heavy-hydrocarbon reforming.
Meanwhile, the compound of LaCrO3 is a promising catalyst owing to its structural stability and
good reducibility.[7] Perovskites have the formula ABO3, where the B cation is believed to be
4

responsible for the catalytic activity. Also, cation A stabilizes the oxidation state of cation B, which
can evenly disperse the active phase at the B-site throughout the perovskite material.[8] To further
improve the activity and stability, partial substitution of B-sites by small amounts of other cations
can alter the structure and electronic configuration of the perovskites.[5,9] The development of
tailored functional materials that consist the catalytic nanoparticles on external surfaces or on inner
lattices of porous crystals are also a key importance in the catalysis field.[10] In the case of sulfur
poisoning and carbon coking, they usually take place on the active metal species by a strong
chemical bonding with S and C.[11] Therefore, there is an issue on controlling the doped active
metals in the perovskite structure compared to the deposited catalysts. However, not many studies
have been dealt with the location of the substituted metal to the perovskite structure associated
with the better reforming activity/stability and these mechanisms.
Here, as Ru is one of the most active and stable ingredients at the H2 production reactions,
Ru-substituted perovskite hollow fibers with the formula LaCr1-xRuxO3 (x=0-0.4) have been
designed for an autothermal reforming catalyst to produce H2 from model diesel fuel. The
perovskite catalysts were prepared by an ACF (Activated carbon fiber) templating method to
synthesize hollow fiber architectures with multi-mixed perovskite nanoparticles for an efficient
heat/mass transfer including short diffusion and high contact time with the reactants in the
autothermal reforming system.[12] Different catalytic behaviours of the heavy-hydrocarbon ATR
were investigated by differentiating the structural and chemical states in the perovskite, which
were characterized by the means of SEM/TEM, XRD, XPS, H2-TPR, XANES spectra, and
EXAFS spectra. The initial activities of the designed perovskite catalysts were then measured, and
each stability was compared in terms of the structural changes, carbon coking and sulfur poisoning
5

after ATR by heavy-hydrocarbon source. From the results, reaction mechanisms for both lattice
and surface Ru type perovskites are investigated.
2. Results and Discussion
2.1 Ru doped hollow perovskite fibers
In Fig.1A, the hollow perovskite fibers with a composition of a perovskite nanoparticle
network were successfully produced for each LaIIICrIII1-xRuIII /ACF composites, by a mild heat
x
treatment at 1027 K under air flow to combust the carbon content.[12] The chemical compositions
of the synthesized samples (Table 1) were found to agree well with the intended LaCr1-xRuxO3
compositions (x = 0, 0.05, 0.1, 0.2, 0.3 and 0.4), which are denoted as LC and LCR-x. SEM images
clearly portray the obtained perfect micro-tubular features with uniform size and length (See
details in Fig. S1), which is known to provide larger specific surface area than the perovskite grains
prepared by the absence of an ACF template.[12a,b] An interesting finding is shown from the
magnified SEM image of the hollow fiber surfaces (Fig. S1). The sizes of the spherical
nanoparticles connected to the micro-tubular feature were seen to vary by the amount of Ru. This
behaviour correlated well with the powder X-ray diffraction (XRD) results of the calculated
average particle sizes, as summarized in Table 1, which shows a remarkable decrease from the LC
to LCR-x (x = 0-0.2) samples. Pore volumes, BET surface areas, and average pore diameters were
also calculated from N2 adsorption–desorption isotherms (Table 1). As small mean crystallite sizes
should be related to the total pore volume and surface area, it was found that the pore volume and
surface area of the perovskites increased by the incorporation of Ru into the perovskite lattice. On
the other hand, particles size increase was clearly shown with higher Ru ratios of 0.3 and 0.4,
which may be due to the excess Ru agglomeration outside the perovskite surface. Among all, the
6

perovskite hollow fiber catalyst of LCR-0.2 showed an optimized high BET surface area of 14.4
m2/g, which may support the catalytic reaction in the heavy-hydrocarbon ATR.
Homogeneity and the high crystallinity of the samples have been confirmed according to
the XRD patterns in Fig.1B. These all showed the major diffraction peaks at similar angles,
corresponding to the (hkl) indices of (112), (022), (220), (222), (312), and (040). These reflections
correspond to those for the orthorhombic structure of LaCrO3 perovskite with monoclinic La2O3
peaks within the Pbnm space group, reported in powder diffraction file JCPDF 024-1016.[9b,13]
Unsubstituted LaCrO3 has a structure with only one distinct B-site, whereas Ru doping introduces
a monoclinic distortion into the ideally orthorhombic perovskite structure.[8] This reduction in
symmetry creates two distinct B-sites that allow the cation to sit on two sites with slightly different
bond lengths to the neighbouring oxygen atoms. From the magnified image of the (112) reflection
at the right-hand side of Fig.1B, substitution of RuIII into LaCrO3 caused peak shifts to a lower
angle but does not indicate significant changes in the perovskite structure. Also, an increased basal
spacing (d-value) from 2.739 Å to 2.765 Å as well as changes in the unit cell parameter and volume
can be observed (Table 1). The d-values were consistent with the basal spacing observed in the
TEM images of the perovskite surfaces (Fig.S2) also increases in the length of a, b, and c axes in
the unit cell were seen. The a, b, c axis increased from 5.499 to 5.519 Å (△a = 0.020 Å), from
5.485 to 5.511 Å (△b = 0.026 Å) and from 7.762 to 7.791 Å (△c = 0.029 Å), respectively. These
changes can be recognized as a partial B-site substitution due to the different crystal ionic radii of
CrIII (0.755 Å) and RuIII (0.820 Å).[14] Such lattice expansion indicates that the crystal structure
may contain a high percentage of octahedral RuO6 configuration sites at the designed
stoichiometric ratios.
7

2.2 Lattice and Surface type Ru
Meanwhile, in any of the XRD patterns (Fig.1B) of the prepared catalysts with the Ru
fraction between 0.05–0.2 exhibited no additional phases. This result could be indicative of the
existence of the Ru in the perovskite lattice structure. However, when the fraction of Ru was 0.3–
0.4, monoclinic diffraction with individual peaks were observed with a decrease in the intensity of
the perovskite structure. The individual metal oxide structures were mainly the formation of
tetrahedral RuO2 (JCPDS 083-1344),[15] which may be explained by the excess Ru deposited on
the perovskite surface as separated Ru species. Based on the XRD analysis, a change in Ru species
at the perovskites is assumable, shown in the schematic diagram of Fig.2A. Incorporation of Ru in
a single-phase perovskite structure is expected until the molar ratio of Ru reaches 0.2, which could
be denoted as lattice type Ru. This is shown in the trend of the d-value and cell volume in Fig.2B
shows a gradual increase up to the Ru fraction of 0.2, due to the larger atomic sizes of Ru.
Furthermore, these values do not change at higher Ru fractions, which may indicate a maximum
amount of Ru for doping in LaCrO3. Under such circumstances, the excess Ru may aggregate to
the surface mostly as RuO2, labelled as surface type Ru.
To compare the dispersion and composition of the Ru phase in the perovskite structure,
SEM-EDX and TEM-EDX analysis of LCR-0.2 and LCR-0.4 perovskite hollow fibers were
compared in Fig.3. From Fig.3A, it was revealed that all of the constituting elements are
homogeneously distributed through out the LCR-0.2 sample on the low magnification scale
employed. On the other hand, Ru species in the LCR-0.4 hollow fibers were found to aggregate in
some regions, while the other elements were still homogeneously distributed. This could be a clue
that excess Ru exists as a different Ru phase on the catalyst surface. The synthesized catalysts were
further examined by TEM and EDX to investigate lattice d-space and metal composition. Fig.3B(a)
8

shows TEM micrographs derived from LCR-0.2 with an interplanar distance of 0.277 nm at the
regular lattice fringes (112) of the perovskite structure. Identified by their acute angle, the observed
data were in accordance with the phase found in XRD. Furthermore, many stacking faults and
micro-twinnings of typical perovskite structures were observed. In case of the LCR-0.4 in
Fig.3B(b), spherical Ru based nanoparticles with diameters of 20–30 nm were found in a separated
LaCr/RuO2 phase with an interplanar distance of 2.753 Å at (112). The nanoparticles on the surface
showed an interplanar distance of 0.240 nm for the crystal planes of RuO2 lattice fringes (111)
direction with their characteristic acute angle.
2.2 Chemical structure analysis
Changes in the nature of the chemical bonds are evident analysed by the XPS analysis. In
Figs.S3 and 4(A-B), XPS peaks corresponding to La 3d, Cr 2p, and Ru 3p3/2 were displayed
respectively in detail. The binding energies and atomic ratios were determined by the peak fittings
and summarized in Table S1. As expected, Ru is basically doped at the B site of the perovskite
structure, no significant differences in the peak shape were observed for the La 3d5/2 doublets
(La2O3 and La(OH)3) for all LaCr1-xRuxO3 hollow fibers as shown in Fig.S3.[6a,8a] On the other
hand, the Cr and Ru species were changed as the ratio changed, which may suggest a structural
distortion of the B site. As from Fig.4A, the Cr 2p3/2 peaks were shifted from the low-valence (Cr3+)
toward the high-valence states (Cr(3+δ)+), but no significant change in peak shape was
observed.[9a,b] Therefore, it is apparent that the portion of the Cr3+ decreased with an elevated
proportion of Cr(3+δ)+ as the amount of Ru was increased, due to the sharing of the electron density
with the Ru and less electronegative Cr. However, for LCR-0.4, the Cr 2p3/2 peaks were further
9

shifted toward the higher valence states of Cr(3+δ)+ with a change in the peak shape, indicating
partial distortion of the typical perovskite structure. The Ru 3p3/2 peak in Fig.4B can be also used
to determine the proportion of the two different Ru species of Ru4+ and Ru(4+δ)+.[8c,15a-c] From
the increase in the peak intensity, the increase in the total Ru amount as designed were confirmed.
About the proportion of the Ru species in Table S1, the perovskite with a Ru incorporation
exhibited higher binding energy for Ru(4+δ)+, generally mixed-valence state, located in the BO3
lattice rather than on the surface. The proportion of Ru(4+δ)+ rises incredibly for the LCR-0.4,
whereas Ru4+ valence state relatively decreased. This partial oxidation may have occurred due to
the formation of surface type Ru by the aggregation of unsubstituted Ru and exist as Ru oxides, as
assumed in Fig.3.[15c]
To investigate more about the chemical bonding character of Ru by the electronic and
geometric structures of absorbing ions, LCR-0.2 and LCR-0.4 hollow fibers were examined using
Ru K-edge XANES and filtered/transformed EXAFS spectra. In Fig.4C, the main peaks of the
XANES spectra showed similar shapes for both compounds, though small changes in the intensity
and shifts in the energies were observed. No pre-edge peaks observed at the positions of the edge
jump which suggests a strongly stabilized local symmetry for Ru, likely by the nearly regular
octahedral symmetry at the orthorhombic lattice. In the main edge regions I and II, intense and
sharp resonance peaks of the Ru species were essentially identical to those of the RuO2 standard.
In the main edge region II, the peak position for LCR-0.2 shifted to a higher energy field and
further increased at the Ru portion of 0.4. The position of the peaks in the main edge regions are
generally assigned to the dipole allowed 1s→5p transitions and provide indirect information on the
Ru(4+δ)+ species.[16a-c] Because the higher effective charge of Ru mainly contributes to the
increase in the transition energy, the shift to a higher energy region indicates an increase in the
10

proportion of Ru(4+δ)+ at the changed perovskite structure. This result is in good agreement with
the XPS curve fitting results (Table S1).
In Fig.4D, Fourier-transformed Ru K-edge EXAFS spectra of the perovskites and
references were displayed at the range of 1–6 Å. For all measured compounds, an intense and
highly isotropic first peak in the R space of 1–2 Å were clearly observed, which is attributed to the
ordered O symmetry of the RuO octahedron. In the higher R range of 2–4 Å, it is obvious that the
shape of the Fourier-transformed spectra is different for the perovskites with different amounts of
Ru. The peaks in this range are principally due to the second nearest neighbor (A-site cation) and
third nearest neighbor (B-site cation) at the ABO3 perovskite structure, and the corresponding
collinear multiple scattering contribution.[16a,b] The collinear multiple scattering contributions
via the linear Ru–O–B chain, particularly, provides a contribution for 2.5–6 Å which can be
correlated with the Ru species scattering in the perovskite structure. It has been reported that the
decrease in the intensity of this peak indicates a decrease in the size of the Ru species, meaning an
increase in the metal dispersion in the perovskite.[16b,d] Based on such observations, the peaks of
the second shell for LCR-0.2 with a lower intensity indicates that the Ru species were highly
dispersed because of lattice type Ru that has strong interactions between the Ru–O–B chains. The
fitting results generated from the Fourier transformation corresponding to the Ru–O bonding pair
are shown in Fig.S4 and Table S2 to obtain more information about the chemical bonding
characteristics. The fitted plots exhibited low R-factors which imply fair fitted results. It was clear
that the bonding structure for LCR-0.4 is almost the same as the RuO2 reference, as we assumed a
surface type Ru formation. On the other hand, most of the bonding features for LCR-0.2 are
different, escpecially the main Ru-O(1) bonding length of 2.01 Å (RuO2=1.98 Å), which is due to
the distorted RuO6 octahedral bonding configuration in the perovskite structure.[16e-f]
11

It is also noteworthy that the oxygen species also transform into a highly active phase for
the LCR-0.2. Generally, the surface adsorbed oxygen is more involved in catalytic reactions than
the lattice oxygen, which can be an important factor for heavy-hydrocarbon reforming.[8b] These
different oxygen species can be identified from the O XPS spectra displayed in Fig.5A. Typically,
the O spectra peak is split into two peaks of lower binding energy (529.3 eV) for lattice oxygen
(O2-) and higher binding energy (531.3 eV) for surface adsorbed oxygen (O-).[7b,9b] A higher
proportion of adsorbed oxygen was observed on the surface as Ru substitution was increased until
0.2, and the spectra were used to calculate the ratio of Osurface/Olattice for each sample. The values
of 0.71, 0.84, 0.92, 1.24, and 1.01 were found for LC, LCR-0.05, LCR-0.1, LCR-0.2, and LCR-
0.3, respectively, while LCR-0.4 can be considered due to the dual phase from different peak shape
by the moderated perovskite structure with surface type Ru, separately. The same tendency of the
oxygen species was observed by the XANES analysis of the O K-edge, as shown in Fig.5B. O K-
edges can be divided into two regions, corresponding to Oα and Oβ, at the range of ~533 eV and
~536 eV, respectively. Oα is assigned as oxygen atoms located at the surface/near-surface region
from a transition of an O 1s electron into unoccupied anti-bonding O 2p, and Oβ expressed the
oxygen atoms in the structural oxide phase from the convolution of O 2p states in the perovskite
lattice.[17a,b] With the consistent XPS results, it was found that the surface oxygen increases as
the Ru amount increases maybe due to the structural changes after Ru doping. On the other hand,
sudden decreases of LCR-0.3 and LCR-0.4 were observed by the separation of the Ru species.
Compare to all, LCR-0.2 possess the highest amount of active oxygen, which may provide a higher
catalytic activity for the ATR reactions.
Meanwhile, the the perovskite and Ru species changes can also modify the reduction
behavior of the perovskites. Therefore, H2-TPR was carried out in Fig.5C. In general, the reduction
12

peak position is affected by the crystallite size, metal oxidation state or location, bonding strength,
oxygen defects, and doping level of the samples.[6a,9b,18] The H2-TPR profiles of the perovskites
showed peaks corresponding to a two-step consecutive reduction. The first peak contains the
reductions of Ru and Cr ions on the surface from their oxidization to the lower state. The second
reduction peak at higher temperature is attributed to the reduction of Ru and Cr ions in the
perovskite structure.[6a] By partial incorporation of Ru to Cr in the B-site, the first reduction peak
shifted to a lower temperature. When the fraction of Ru is >0.3, a bigger peak with a larger area is
observed for the first reduction below 200 °C;[18] because Ru may be located outside the
perovskite lattice rather than the inner lattice. Compared to others, the second reduction peak for
LCR-0.2 was seen at a higher temperature with significantly larger peak area. This result is
attributed to the presence of an optimal Ru quantity incorporated into the lattice, along with the
high portion of CrHigh-Valence species that are strongly bonded to O.[9b] As a result, this fine profile
for LCR-0.2 indicates enhanced reducibility with higher H2 consumption and better structural
stability compared with the other Ru-substituted samples. Therefore, it is plausible that relatively
high percentages of Ru(4+δ)+, Cr(3+δ)+, and O- species in LCR-0.2 is correlated to the formation of
the lattice type Ru, which enhance the mobility of oxygen in the lattice and are related to the
oxygen surface storage capacity. These compositions may demonstrate the prevalent role of Ru
and Cr redox cycles on catalytic activity. Usually, heavy-hydrocarbon species are adsorbed and
interact with the O atoms during ATR. Here, the electron transfer to the reactants from the active
metals neighboring the vacancy at the B-site facilitates the C–H bond cracking for H2 release. This
feature is also crucial for the stability and resistance of the coke formation during the catalytic
hydrocarbon reforming processes. Consequently, LCR-0.2 is expected to show the greatest activity
13

and stability as a redox catalyst in the reforming of heavy-hydrocarbons fuels to produce H2 gas
under harsh reaction conditions.
2.3 Heavy-hydrocarbon autothermal reforming
To evaluate the catalytic activity and stability of the prepared catalysts, ATR tests using
hexadecane (C16H34) as a fuel with 100 ppm of sulfur were carried out at the operating conditions
of 800 °C, 4000/h, and carbon/H2O/O2 = 1/1.25/0.4. The product mole percentage toward the H2
and by-products versus time in stream over the catalysts of LCR-x (x = 0-0.4) hollow fibers are
shown in Fig.6. The product production time curves displayed by the usual scattered data for
reforming reaction which is due to the non-homogeneous mixture of the reactants. All hollow
fibrous catalysts exhibited good reforming efficiencies for ATR using the sulfur-containing heavy-
hydrocarbon. The high performance of the catalyst is not only attributed to the Ru composition
which provides a great catalytic efficiency but also, due to the improved mass/heat transfer rates
from the high effective surface area of the micro-tubular structures.[12] Comparing to LC, the
initial activities for H2 production are higher for the Ru-substituted catalysts, proving that the
substitution of Cr by Ru favorably affects the H2 extraction capacity from the hydrocarbon
molecules. However, it was obvious that the initial amount of H2 produced up to 10 h varied
depending on the amount of Ru in the perovskite structure. Values of 38.8, 45.5, 48.6, 62.7, 63.4,
and 69.7 mol% were obtained for LCR-x hollow fibers (x = 0, 0.05, 0.1, 0.2, 0.3, and 0.4),
respectively, with the order of x=0 << 0.05 ≤ 0.1 << 0.2 = 0.3 < 0.4, which are comparable with
other supported noble metals for this type of reaction.[5a-c] In addition, the higher reforming
capacity was accompanied by the lower proportion of non-reformed hydrocarbons (CH4, C2H4,
14

C2H6, and C3H6) in the products from each figure in Fig.6, which indicates that the increase in H2
production was supplemented by the decrease in COx formation.
From Fig.6, the stability of the prepared perovskite catalysts can be also examined
against long periods on stream of heavy-hydrocarbon ATR in the presence of sulfur. Interestingly,
the catalysts derived from the LCR-0.05, LCR-0.1, and LCR-0.2 hollow fibers with lattice type
Ru were not suffered from the decrease in H2 production over the same time period. As expected,
the LCR-0.2 hollow fiber catalyst especially exhibited high activity and great stability with a lower
degree of deactivation, which definitely shows a different behavior to the Ru supported LaCrO3
catalyst with a dramatic decrease in performance at the same loading (Fig.S5). It is strongly
associated with the reactivity owing to the relatively higher amount of lattice type Ru in the
perovskite. Furthermore, the stable reactivity was achieved together with a high tolerance against
carbon deposition and sulfur poisoning on the feeds of heavy-hydrocarbon. This result was even
comparable with the reforming results using Pt-GDC (Fig.S6) where a decrease in catalytic activity
was accompanied under harsh reaction conditions (fuel with C and S species), since it is usually
known that a high deactivation occurs by S poisoning at the Pt metal surface.[11] The catalysts
derived from LCR-0.3 and LCR-0.4 hollow fibers, having surface type Ru, show also a noticeable
decrease in H2 production of around 6% and 34%, respectively over time, even though the initial
performances were relatively high. The decrease in H2 production was also accompanied by a
substantial increase in CO2 production for the LCR-0.3 in a minor way, and especially for the
LCR-0.4 with surface type Ru. This result can be explained by the increased occurrence of the
Boudouard reaction (2CO →CO2 + C), maybe owing to a greater formation of coke. Coking usually
blocks the active sites and sinters the catalyst, which results in the loss of surface area and a
reduction in the number of active surface sites.[4] Therefore, after each heavy-hydrocarbon
15

reforming runs, the presence of deposited carbon and sulfur on the catalytic surface may provide
a clue on the stability depending on the position of the active Ru at the perovskite catalysts.
2.4 Material characterization after the catalyst tests
To investigate the deactivation behavior during heavy-hydrocarbon ATR in the presence
of sulfur for 50 hr, all catalysts were post-characterized. After each ATR, different quantities of
carbonaceous residue deposits on the used catalysts. The amount of the deposited carbon was
calculated by TGA analysis (Fig.S7) and the results are summarized in Fig.7A. The total coke
quantities for LaCr1-xRuxO3 (x = 0, 0.05, 0.1, and 0.2, respectively) were quite low and even
decreased close to the order of the higher Ru amount, whereas a slight increase was observed for
x=0.3 and a significant increase was observed for x=0.4. Remarkably, these results are consistent
with the stability results (Fig.6), depending on the amount of Ru and its location in the active
region. From the SEM results in Fig.7B, the micrographs exhibited a retained hollow fibrous
architecture composed of the perovskite nanoparticle networks. No carbon deposition on LCR-0.2
surface was seen whereas an obvious covering with many carbon filaments (confirmed also from
the large coke quantity in Fig.7A) was displayed on the LCR-0.4 fiber surface. From the TEM
analysis of LCR-0.2 after ATR in Fig.S8A, there is also almost no carbon deposition on the surface
seen without any structural change of the original and surface/bulk structures. Moreover, it is
interesting that a base carbon growth was displayed due to the position of Ru inside the perovskite
structure with strong interaction, which shows usually slower coke formation.[19] The presence
of the deposited carbons on the perovskite catalysts can be further proved by the XRD analysis in
Fig.S8B. XRD spectra of LCR-0.2 exhibited low-intensity diffraction peaks of carbon, indicating
limiting effect on carbon formation. On the other hand, the XRD result for LCR-0.4 highlights a
large amount of carbonaceous from a huge peak of graphitic carbons (JCPDS 75-1621)[4b,c]. The
16

structural change of the used LCR-0.4 catalysts was also indicative through the XRD with a change
in the d-value and unit cell parameters after the long reaction time (Table S3). Moreover, the Ru
species of the LCR-0.4 outside the catalyst as a tetragonal RuO2 (Fig.1B) partially changes to
metallic Ru at the reductive reaction environment, which may be a factor in the increased coke
formation and sulfur poisoning through the strong bonding with Ru metal.[11,19]
The change in the Ru chemical structure after the heavy-hydrocarbon reforming was easily
detectable by the Ru K-edge XANES and filtered/transformed EXAFS spectra, as shown in
Fig.8A-B. From the XANES spectra in Fig.8A, no changes of the peak shape and proportion of
Ru4+/Ru(4+δ)+ species in the LCR-0.2 after ATR were visible. Furthermore, there was no reflection
on the pre-edge, which means a stable local symmetry in the orthorhombic structure. Compared to
the LCR-0.4 after ATR, a weak pre-edge jump in region I was clearly observed at around 22150
eV. Generally, pre-edge peaks are due to the dipole forbidden 1s→4d transition, which can explain
a broken O symmetry of the LCR-0.4 during the harsh ATR reaction.[16a,b] When the pre-edge
peaks become intensified because of d–p mixing or the quadruple allowed transition, Oh symmetry
is broken as the distortion of the Ru species from O symmetry around the absorber increases. The
peak at the main edge region II was also different and reflected a change in the original perovskite
structure to the separated RuO2 species. To obtain more information about the change in the
chemical bonding characteristics, the fitting results generated from the Fourier transformation, are
shown in Fig.8B corresponding to the Ru–O bonding pair. The fitted Ru K-edge EXAFS spectra
for LCR-0.2 do not change at all after the ATR test. However, in the fitted Ru K-edge EXAFS
spectra for LCR-0.4, the highly isotropic peaks became smooth and collinear multiple scattering
in the whole R space disappeared, which may reveal the broken ordered O symmetry of the RuO6
octahedron and linear Ru–O–B chain, respectively.
17

XANES spectra for each carbon and sulfur in Fig.8C-D can serve as evidences to prove
the carbon/sulfur deposition mechanism of the lattice and surface type Ru. Fig.8C shows the C K-
edge XANES spectra overlapping with the Ru M5,4-edge. Firstly, the existence of Ru in both LCR
perovskites is clear by the shoulder at 283.5 eV assigned to Ru M5-edge.[20a,b] In the case of C
K-edge, two distinct peaks, assigned as graphite π* and σ* transitions at ~285.5 eV and ~291.9 eV
respectively, are observed even for the small amount of carbon coking occurred at LCR-0.2. Yet,
the intensity for LCR-0.4 was much higher than the LCR-0.2, due to the higher carbon deposit.
The most interesting feature is the bigger peak intensity of a new peak at ~289.6 eV for LCR-0.4,
indicating a stronger oxidized C environment than that for –COOH in the deposited carbon.[20b]
Therefore, it could be interpreted to a strong interaction with the surface type Ru through chemical
bonding with the produced carbon species. Similar results were seen at the S L-edge XANES
spectra in Fig.8D. The peak intensity for LCR-0.4 was essentially bigger than LCR-0.2,
demonstrating a larger amount of poisoned surface by sulfur. Moreover, the spectra for LCR-0.4
shows three different peaks in the 161-168 eV region.[20c,d] Each arising peaks by several
electronic transitions exactly matched with the used precursor of DPT in the fuel, specifying sulfur
poisoning through a strong binding with the Ru surface which could turn also to a carbon growth
site.[4] Remarkably, a positive shift of the main peaks was observed for LCR-0.2 with also an
intensity decrease. The positive shift reveals oxidation of the sulfur species to a gas phase like SOx,
which might be promoted by the oxygen activation from the high amount of surface oxygen.[7b,9b]
This mechanism for LCR-0.2 hollow fiber catalyst provides another understanding of the slow
deactivation during the heavy-hydrocarbon reforming with sulfur species.
2.5 Expected reaction mechanism.
18

Summing up all the results, we could finally come to the reaction mechanism of the
surface/lattice type Ru and lattice type Ru. The comparisons are illustrated in Fig. 8E for the
autothermal reforming by heavy-hydrocarbon fuel in the presence of sulfur. It was found that the
deactivation of the surface/lattice type Ru is mainly caused by carbon coking and sulfur poisoning,
which can be also the reason for the low durability of the Ru supported perovskite catalyst.
Especially, carbon coking usually comes from the cracking of the carbon chain in the fuel, which
is more likely when heavy-hydrocarbon fuel used. Ideally, the hydrocarbon species are adsorbed
on the surface of the active metal site and converts to H2 and COx by reaction with O2 and H2O
molecule. However, from the strong interaction of the metallic surface Ru species, the stabilization
step of the active precursor was carried out to form filamentous coke (C*+C*).[11b,c, 19] The
large sulfur poisoning was also due to the strong binding energy of Ru-S, which were accelerated
by the change in the oxide symmetric and Ru species. On the contrary, the lattice type Ru fully
doped in perovskites shows less carbon coking and sulfur poisoning. As it is very important to
transform coke precursors (C*) into gases for example by the Boudouard reaction
(C*+CO2↔2CO), gasification (C*+H2O→CO+H2) and methanation (C*+2H2↔CH4), the
catalyst with lattice type Ru shows sufficient activity.[11] Moreover, the fully doped perovskite
catalysts possess much more active oxygen by oxygen defects, which enhances the formation of
C–O active bonds to minimize C–C bond formation. In the case of sulfur poisoning, the lattice
type Ru has also lower binding energy with sulfur species and the active oxygen provides a
reactivity to oxidize the S species to gaseous SOx.[4b,11a] This mechanism suggests that the nature
of Ru species inside the perovskite lattice has enough activity to crack the heavy-hydrocarbons
that possess unsaturated C=C bonds by having high coordination ability. Furthermore, this protects
the catalyst from carbon coking and even sulfur poisoning for great catalytic stability.
19

3. Conclusions
In this work, hollow fibers networked with perovskite nanoparticles of LaCr1-xRuxO3 (x =
0–0.4) were perpared, which have a favorable effect on the capacity to yield H2 directly from
heavy-hydrocarbon fuel reforming. The most important findings arrived at: (i) At the Ru amount
of 0.05–0.2, lattice type Ru was dominant whereas surface type Ru was formed at the higher Ru
amount of 0.3-0.4. (ii) Partial substitution of Cr by Ru in the perovskite lattice led to the strucutral
modification which had a positive influence on the enhanced catalytic property through more
surface oxygen and better reducibility. (iii) From the characterization after ATR reactions, the
stability of the perovskite structure was affected by the difference in the Ru chemical nature. The
formation of surface nanometric Ru particles was the factor in the accumulation of carbon and
sulfur on the catalyst surface due to the strong metal–C/S chemical bonding, which is related to
the decrease in the H2 production and increase of the CO2 by-product. (iv) The optimal composition
of LCR-0.2 hollow fiber, having only lattice type Ru, showed comparable activity and was less
prone to deactivation by coke and sulfur deposits, through the inner location of the active sites and
structural distortion with more active oxygens at the surface. These highly active and durable
catalyst materials could possibly be applied to the actual reforming process that is being
constructed for the infrastructure in the upcoming H2 society.
4. Experimental details
4.1 Synthesis of Hollow Perovskite Fiber Catalysts
Synthesis procedure of the hollow LaCr1-xRuxO3 fiber was shown in the Fig.1 [12]. The
perovskite hollow fiber (LaCr1-xRuxO3, x=0, 0.05, 0.1, 0.2, 0.3, and 0.4) was prepared using
aqueous (solution) impregnation synthesis which is a rapid and convenient method to provide high
20

surface area oxides. The acid-treated ACF template (1 g) with H2SO4 (6 M) and HNO3 (6 M) was
immersed in a water based solution of La(NO3)3·9H2O, Cr(NO3)3·9H2O, and HN4O10Ru with a
certain stoichiometric LaIII, CrIII, and RuIII ions, resulting in ionic binding of the metal cations to
the negatively charged surface of the ACF. After washing with deionized water for 24 hours, the
hollow fiber composed of a perovskite nanoparticle network could then be successfully formed by
mild heat treatment of the ACF impregnated with LaIII/CrIII1-x/RuIIIx (x=0, 0.05, 0.1, 0.2, 0.3, and
0.5) ions at 1027 K for 6 h in sufficient air flow to remove the C content via combustion with a
formation of crystalline structure. For comparison, Ru supported LaCrO3 (Same Ru loading with
LCR-0.2) and 5wt.% Pt–GDC (gadolinium-doped CeO2) catalysts were prepared by incipient
wetness impregnation in a similar manner [2b].
4.2 Synthesis of Catalyst characterization
The chemical composition of each sample was analyzed by Inductively Coupled Plasma-
Atomic Emission Spectroscopy (ICP-AES, OPTIMA 4300DV, Perkin-Elmer, USA). For crystal
structure analysis, X-ray diffraction (XRD) was performed by a Rigaku Miniflex AD11605 with
X-ray diffractometer using Cu Kαradiation (λ= 1.5405 nm) at 30 kV and 30 mA. The morphology
of the as-synthesized samples was monitored by using scanning electron microscope (SEM, JEOL
JSM-6701F) operated at 15 kV, equipped with a an energy-dispersive X-ray (EDX) spectrometer.
High-resolution transmission electron microscopy (HR-TEM) imaging as well as selected area
electron diffraction (SAED) was performed at 200 kV in a JEOL JEM-2100F with an energy-
dispersive X-ray (EDX) spectrometer. The BET (Brunauer-Emmett-Teller) surface area, pore
volume and pore diameter were determined by a distribution graph of N2 adsorption-desorption at
77 K (Micromeritics ASAP 2010), using the BET equation. To probe the chemical states of the
elements and their proportions on the surface of the perovskites, the X-ray photoelectron
21

spectroscopy (XPS) were recorded on a K-alpha (Thermo VG, U.K.) equipped with Al
monochromated X-ray sources (Al Kα, 1486.6 eV), using a hemispherical energy analyzer. The
samples were fixed on a carbon tape and all binding energies were calibrated using C1s (284.8
eV). The catalytic sites of the perovskites were examined by temperature programmed reduction
(TPR; Belcat-M) equipped with a thermal conductivity detector (TCD). The catalyst samples (15
mg) were heated from room temperature to 900 oC at a rate of 10 oC/min under the gas feed rate
of 40 cc/min (10% H2/Ar). The X-ray absorption spectroscopy (XAS) results were recorded at the
Pohang Accelerator Laboratory (PAL, Pohang, Korea), beamline 10C (Multipole-wiggler, Ge 13-
element detector) and 10D (Bending magnet, Phoibos 150). Measurements were made at room
temperature with solid samples. The X-ray absorption near edge structure (XANES) data of O K-
edge were collected at beamline 10D XAFS. The XANES and extended X-ray absorption fine
structure (EXAFS) data at Ru K-edges before and after the samples were collected using thin layers
of powder samples deposited on transparent adhesive tape. The measurements were performed in
transmission mode at beamline 10C Wide XAFS using gas-ionization detectors. All the spectra
were calibrated by first measuring the spectrum of RuO2, RuCl3 and Ru metal references.
4.3 Catalytic Testing
The autothermal reforming of heavy hydrocarbon was carried out at atmospheric pressure
in a fixed-bed stainless steel flow reactor with 8 mm internal diameter. The temperature of the
reactor was monitored and controlled by thermocouples that were coaxially inserted into the center
of the catalyst bed. The fuel in this study was hexadecane (Aldrich) with a high carbon number of
16, including DPT (Dibenzothiophene, Sigma-Aldrich) of 100ppm for the sulfur species. The fuels
were injected by liquid pump and ultrasonic injector for homogeneous fuel supply. For ATR
experiments of the hollow fiber catalysts, each feed, with a change of the molar ratios of H2O/C =
22

1−3/1 and O2/C = 0.3−0.7/1, was introduced into the reactor by a mass flow controller (GHSV =
o
2000−5000 h-1), where the temperatures was varied to 700−850 C at the atmospheric pressure.
With an optimum operation conditions, the reforming reaction of each hollow fiber catalysts was
carried out for 50 h. All reformates (H2, CO2, CO, CH4, C2,3Hn) were periodically measured using
on-line GC (Youglin-6000) equipped with a TC detector and an FI detector which were
programmed to operate under high-sensitivity conditions, and the product distributions were
averaged from the steady state. The conversion and H2 mole percentage of products by hexadecane
are defined as follows:
- Hexadecane conversions at the liquid base were over 99%.
- Product distribution: (%) (i: H2, CO, CH4, CO2, C2,3Hn (C2H4+C2H6+C3H6)):
(mole
i)
out
100
mole
i
out
ACKNOWLEDGMENT
This research was supported by the Industrial Technology Innovation Program funded by the
Ministry of Trade, Industry and Energy (MOTIE), Republic of Korea (grant number 10052076).
REFERENCES
[1] (a) A.G. Bhavani, W.Y. Kim, J.S. Lee, ACS Catal., 3(7) (2013) 1537–1544. (b) D. Baudouin,
K. C. Szeto, P. Laurent, A. Mallmann, B. Fenet, L. Veyre, U. Rodemerck, C. Copéret, C.
Thieuleux, J. Am. Chem. Soc. 134 (2012) 20624–20627.
23

[2] (a) R. M. Navarro, M. C. Álvarez-Galván, M. Mota, J. A. Villoria de la Mano, S. M. Al-
Zahrani, J. L. G. Fierro, ChemCatChem 3 (2011) 440–457. (b) I. Kang, J. M. Bae, G. J. Bae,
J. Power Sources 163 (2006) 538–546.
[3] (a) G. Nahar, V. Dupont, Recent Patents on Chemical Engineering, 6 (2013) 8–42. (b) M.
A. Reese, S. Q. Turn, H. Cui, J. Power Sources 195 (2010) 553–558.
[4] (a) Y. Chen, C. Xie, Y. Li, C. Song, T. B. Bolin, Phys. Chem. Chem. Phys., 12 (2010) 5707–
5711. (b) C. Xie, Y. Chen, M. H. Engelhard, C. Song, ACS Catal., 2 (2012) 1127–1137. (c)
E. Nikolla, A. Holewinski, J. Schwank, S. Linic, J. Am. Chem. Soc. 128 (2006) 11354–
11355.
[5] (a) R. K. Kaila, A. Gutierrez, Q. I. Krause, Appl. Catal. B. Environ., 84 (2008) 324–331. (b)
M. C. Alvarez-Galvan, R. M. Navarro, F. Rosa, Y. Briceno, F. Gordillo Alvarez, J. L. G.
Fierro, Int. J. Hydrogen Energy 33 (2008) 652–663. (c) X. Karatzas, K. Jansson, A.
González, J. Dawody, A. Svensson, L. Pettersson, Appl. Catal. B-Environ. 101 (2011) 226–
238.
[6] (a) N. Mota, M. C. Alvarez-Galván, R. M. Navarro, S. M. Al-Zahrani, A. Goguet, H. Daly,
W. Zhang, A. Trunschke, R. Schlögl, J. L. G. Fierro, Appl. Catal. B-Environ. 113 (2012)
271–282. (b) J. R. Mawdsley, J. T. Vaughey, T. R. Krause, Chem. Mater. 21 (2009) 4830–
4838. (c) A. Qia, S. Wang, G. Fu, C. Ni, D. Wu, Appl. Catal. A-Gen. 281 (2005) 233–246.
[7] (a) J. Sfeir, P. A. Buffat, P. Mockli, N. Xanthopoulos, R. Vasquez, H. J. Mathieu, J. V. herle,
K. R. Thampi, J. Catal. 202 (2001) 229–244. (b) K. Rida, A. Benabbas, F. Bouremmad, M.
A. Peña, E. Sastre, A. Martínez-Arias, Appl. Catal. A-Gen. 327 (2007) 173–179.
24

[8] (a) M. A. Pena, J. L. G. Fierro, Chem. Rev. 101 (2001) 1981–2017. (b) R. J. H. Voorhoeve,
D. W. Johnson, J. P. Remeika, P. K. Gallagher, Science 4 (1977) 827–833. (c) J. Suntivich,
H. A. Gasteiger, N. Yabuuchi, H. Nakanishi, J. B. Goodenough, Y. Shao-Horn, Nat. Chem.
3 (2011) 546–550. (d) S. Ida, C. Ogata, M. Eguchi, W. J. Youngblood, T. E. Mallouk, Y.
Matsumoto, J. Am. Chem. Soc. 130 (2008) 7052–7059.
[9] (a) N. Mota, M. C. A lvarez-Galvan, S. M. Al-Zahrani, R. M. Navarro, J. L. G. Fierro, Int.
J. Hydrogen Energy 37 (2012) 7056–7064. (b) D. Fino, N. Russo, E. Cauda, G. Saracco, V.
Specchia, Catal. Today 114 (2006) 31–39. (c) S. Gaur, D. Pakhare, H. Wu, D. J. Haynes, J.
J. Spivey, Energ. Fuels 26 (2012) 1989–1998.
[10] (a) P. Munnik, P. E. de Jongh, K. P. de Jong Chem. Rev. 115 (2015) 6687−6718. (b) D. Li,
X. Li, J. Gong, Chem. Rev. 116 (2016) 11529−11653.
[11] (a) C. Gillan, M. Fowles, S. French, S. D. Jackson, Ind. Eng. Chem. Res. 52 (2013)
13350−13356. (b) S. L. Lakhapatri, M. A. Abraham, Catal. Sci. Technol. 3 (2013) 2755–
2760. (c) S. De, J. Zhang, R. Luque, N. Yan, Energy Environ. Sci. 9 (2016) 3314–3347.
[12] (a) Y. Jeon, D. H. Park, J. I. Park, S. H. Yoon, I. Mochida, J. H. Choy, Y. G. Shul, Sci. Rep.
3 (2013) 2902. (b) Y. Jeon, C. Lee, J. Rhee, G. Lee, J. H. Myung, M. Park, J. I. Park, H.
Einaga, Y. G. Shul, Fuel 187 (2017) 446–456. (c) Y. Jeon, H. Kim, C. Lee, S. Lee, S. Song,
Y. G. Shul, Fuel 207 (2017) 493–502.
[13] J. Sunarso, A. A. J. Torriero, W. Zhou, P. C. Howlett, M. Forsyth, J. Phys. Chem. C 116
(2012) 5827–5834.
[14] R. D. Shannon, Acta Cryst. A32 (1976) 751–767.
25

[15] (a) X. Ma, R. Lin, C. Beuerle, J. E. Jackson, S. O. Obare, R. Ofoli, Nanotechnology 25 (2014)
045701. (b) S. Petrovic, V. Rakic D.M. Jovanovic, A.T. Baricevic, Appl. Catal. B. 66 (2006)
249–257. (c) M. Lü, M. Deng, J.C. Waerenborgh, X. Wu, J. Meng, Dalton Trans. 41 (2012)
11507–11518.
[16] (a) K. Getty, M. U. Delgado-Jaime, Kennepohl, P. Inorg. Chim. Acta, 361 (2008) 1059–
1065. (b) Y. Masuda, S. Hosokawa, M. Inoue, J. Ceram. Soc. Jpn. 119 (2011) 850. (c) I.
Arcon, A. Bencan, A. Kodre, M. Kosec, X-Ray Spectrom. 36 (2007) 301–304. (d) J. Y. Kim,
S. H. Hwang, S. J. Kim, G. Demazeau, J. H. Choy, H. Shimada, J. Synchrotron Rad. 8 (2001)
722–724. (e) M. Retuerto, L. Pascual, F. Calle-Vallejo, P. Ferrer, D. Gianolio, A. G. Pereira,
Á. García, J. Torrero, M. T. Fernández-Díaz, P. Bencok, M. A. Peña, J. L. G. Fierro, Sergio
Rojas, Nat. comm. 10 (2019) 2041. (f) H. T. Jeng, S. H. Lin, C. S. Hsue, Phys. Rev. Lett. 97
(2006) 067002–067001.
[17] (a) Z. Wang, L. Wu, J. Zhou, Z. Jiang, B. Shen, Nanoscale 6 (2014) 12298–12302. (b) W.
T. Hong, K. A. Stoerzinger, B. Moritz, T. P. Devereaux, W. Yang, Y. Shao-Horn, J. Phys.
Chem. C 119 (2015) 2063−2072.
[18] D. L. Hoang, A. Dittmar, J. Radnik, K. W. Brzezinka, K. Witke, Appl. Catal., A. 239 (2003)
95−110.
[19] (a) A. Sacco, J. C. Caulmare, ACS Symposium Series, 202 (1983) 177–192. (b) Y.H. Lai,
Y.L. Chen, Y. Chi, S. Liu, A.J. J. Carty, S.M. Peng, G.H. Lee, J. Mater. Chem. 13 (2003)
1999−2006.
[20] (a) J. Zhou, J. Wang, H. Fang, C. Wu, J. N. Cutlera, T. Kong Sham, Chem. Commun. 46
(2010) 2778–2780. (b) J. Yang, J. Wang, Y. Tang, D. Wang, X. Li, Y. Hu, R. Li, G. Liang,
26

T. Shamb, X. Sun, Energy Environ. Sci. 6 (2013) 1521–1528. (c) F. Jalilehvand, Chem. Soc.
Rev., 2006, 35, 1256. (d) Z. Wang, J. Wang, T. Sham, S. Yang, Nanoscale 6 (2014) 9783–
9790.
Scheme, Figures and Tables
Figure 1. A. synthesis of the hollow perovskite fiber catalysts and B. XRD patterns and evolution
of phases of the LaCr1-xRuxO3 hollow fibers: (a) x=0, (b) x=0.05, (c) x=0.1, (d) x=0.2, (e) x=0.3
and (f) x=0.4. (￭: monoclinic La2O3, ★: tetragonal RuO2).
27

Figure 2. A. Schematic figure of the expected surface and lattice type Ru features, and the
correlation of the d-value and cell volume depending on the Ru amount in the LaCr1-xRuxO3
material system.
28

Figure 3. A. SEM-EDX mapping images and B. TEM images with a TEM-EDX of the LCR-0.2
and LCR-0.4 hollow fibers.
29