Selective catalytic reduction of NO with CO and C3H6 over Rh/NbOPO4

Article abstract of DOI:10.1016/j.cattod.2018.07.027

The selective catalytic reduction of NO with C₃H₆ and CO was investigated over platinum group metal (PGM; Pd, Rh, or Pt) supported on NbOPO₄. Rh/NbOPO₄ exhibits a lower light-off temperature of NO reduction to N₂ compared to Pd- or Pt-loaded NbOPO₄ catalysts and Rh/Al₂O₃ under stoichiometric conditions. Rh/NbOPO₄ shows superior catalytic activity for NO reduction to Rh/Al₂O₃ under a wide range of lean conditions. Propylene (C₃H₆) was partially oxidized to reactive aldehyde species on Rh/NbOPO₄, whereas less-reactive carboxylate species was generated and adsorbed on Rh/Al₂O₃. A highly dispersed Rh⁰ nanoparticles on NbOPO₄ was slowly oxidized by exposure to excess O₂, but Rh on Al₂O₃ was readily oxidized to less-active Rh₂O₃. The high activity of Rh/NbOPO₄ for C₃H₆ oxidation to reactive aldehyde species and the resistance of Rh⁰ against oxidation results in the enhancement of NO reduction activity in a slightly lean region.

Full text of DOI:10.1016/j.cattod.2018.07.027

Accepted Manuscript
Title: Selective catalytic reduction of NO with CO and C3H6
over Rh/NbOPO4
Authors: Shinsuke Imai, Hiroki Miura, Tetsuya Shishido
PII:
S0920-5861(18)31023-X
DOI:
Reference:
https://doi.org/10.1016/j.cattod.2018.07.027
CATTOD 11572
To appear in:
Catalysis Today
Received date:
Revised date:
Accepted date:
3-5-2018
22-6-2018
9-7-2018
Please cite this article as: Imai S, Miura H, Shishido T, Selective catalytic
reduction of NO with CO and C3H6 over Rh/NbOPO4, Catalysis Today (2018),
https://doi.org/10.1016/j.cattod.2018.07.027
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.

Title
Selective catalytic reduction of NO with CO and C3H6 over Rh/NbOPO4
Authors
Shinsuke Imaia, Hiroki Miuraa,b,d, and Tetsuya Shishido*a,b,c,d
aDepartment of Applied Chemistry for Environment, Graduate School of Urban
Environmental Sciences, Tokyo Metropolitan University, 1-1 Minami-Osawa, Hachioji,
Tokyo 192-0397, Japan
bResearch Center for Hydrogen Energy-Based Society, Tokyo Metropolitan University,
1-1 Minami-Osawa, Hachioji, Tokyo 192-0397, Japan
cResearch Center for Gold Chemistry, Tokyo Metropolitan University, 1-1 Minami-
Osawa, Hachioji, Tokyo 192-0397, Japan
dElements Strategy Initiative for Catalysts and Batteries, Kyoto University, 1-30 Goryo-
Ohara, Nishikyo-ku, Kyoto 615-8245, Japan
*Corresponding author: shishido-tetsuya@tmu.ac.jp
Running title
Selective catalytic reduction of NO over Rh/NbOPO4
1

Graphical abstract
2

Highlights
 Rh/NbOPO4 exhibits higher activity for NO reduction than Pd- or Pt-loaded
catalysts
 Rh/NbOPO4 shows high NO reduction activity in a slightly lean region.
 Rh/NbOPO4 showed high activity for C3H6 oxidation to reactive aldehyde species.
 Rh/NbOPO4 possesses resistance of Rh0 against oxidation.
Abstract
The selective catalytic reduction of NO with C3H6 and CO was investigated over platinum
group metal (PGM; Pd, Rh, or Pt) supported on NbOPO4. Rh/NbOPO4 exhibits a lower
light-off temperature of NO reduction to N2 compared to Pd- or Pt-loaded NbOPO4
catalysts and Rh/Al2O3 under stoichiometric conditions. Rh/NbOPO4 shows superior
catalytic activity for NO reduction to Rh/Al2O3 under a wide range of lean conditions.
Propylene (C3H6) was partially oxidized to reactive aldehyde species on Rh/NbOPO4,
whereas less-reactive carboxylate species was generated and adsorbed on Rh/Al2O3. A
highly dispersed Rh0 nanoparticles on NbOPO4 was slowly oxidized by exposure to
excess O2, but Rh on Al2O3 was readily oxidized to less-active Rh2O3. The high activity
of Rh/NbOPO4 for C3H6 oxidation to reactive aldehyde species and the resistance of Rh0
against oxidation results in the enhancement of NO reduction activity in a slightly lean
region.
3

Key Words
SCR, NO, Rh, NbOPO4
1. Introduction
The automotive three-way catalysts (TWCs) work efficiently to remove NO, carbon
monoxide (CO) and residual hydrocarbons (HCs), which are the major pollutants in
exhaust emissions, simultaneously near the stoichiometric air-to-fuel (A/F) ratio [1]. Both
CO and HCs are oxidized to CO2 and H2O, but NO must be completely reduced to N2
despite the presence of oxygen. Although engine control under fuel-lean conditions (in
the presence of excess O2) is desirable in terms of fuel efficiency and less CO2 emissions,
NOx reduction under fuel-lean conditions is a challenging issue. Rh plays an important
role in TWCs, especially for NOx reduction (NOx into N2). However, NOx reduction is
strongly inhibited in excess O2, which leads to the combustion of reducing gases (CO and
HCs) and the oxidation of metallic Rh (Rh0) to the less-active rhodium oxides such as
Rh2O3 [2-8]. On the other hand, the redox behavior of metal nanoparticles (NPs) often
depends on the interaction at the metal-support interface. Therefore, both the activity and
stability of supported metal catalysts is dependent on the combination of metal and
support material.
Machida and his co-workers have reported that various metal phosphates such as
AlPO4, ZrP2O7, and LaPO4 work as thermally stable supports for Rh catalysts [9-18].
They proposed that the interaction between Rh species and phosphate supports (AlPO4,
4

YPO4, ZrP2O7, and LaPO4) via Rh−O−P bonding affects the electronic states and redox
behaviors of Rh. For instance, Rh2O3 on AlPO4 and ZrP2O7 is more reducible and the
formed metallic Rh possesses a higher tolerance against reoxidation. Moreover, this
unusual redox behavior of the Rh catalysts is expected to cause a promotion effect on
their performance for NO reduction under lean conditions.
In NOx reduction under fuel-lean conditions, the reductant (HCs and CO) must react
effectively with O2 and/or NO to reduce NO to N2. Many researchers have proposed the
possible roles of HCs: (i) the reductants remove adsorbed oxygen to enable NO
dissociation to form N and O and subsequent desorption of N2; (ii) the reductants are
partially oxidized to activated intermediate species that react with adsorbed NO [19-23].
These roles suggest that the controlling of oxidizability of catalyst and adsorption
property are also important.
The present study examined the catalytic activity for the selective catalytic reduction
of NO with CO and C3H6 over supported Rh catalysts under a variety of oxygen
concentrations. The effect of support (niobium compounds (NbOPO4 and Nb2O5) and
Al2O3) on both the redox behavior of Rh and catalytic performance was investigated. It
is well known that niobium compounds are widely used as the components of oxidation
catalysts and thermally stable. Supported Rh catalysts were characterized by XRD, FT-
IR, Rh K-edge XANES, and H2-TPR to reveal the reason for improved catalytic
performance of Rh/NbOPO4, which exhibited the highest activity for NO reduction under
lean conditions among all the supported Rh catalysts tested.
2. Experimental
2.1 Catalyst preparation
5

γ-Al2O3 (JRC-ALO-8) was provided by the Catalysis Society of Japan. Niobium
phosphate (NbOPO4) was kindly supplied by CBMM. A series of supported PGM (Rh,
Pd, and Pt) catalysts (metal loading amount: 0.5 wt% as metal) were prepared by the
impregnation method. The following PGM precursors were used as source materials:
RhCl3 3H2O, PdCl2, and H2PtCl6 (Wako Pure Chemical Industries, Ltd.). The support
was added to an aqueous solution of PGM precursor. After stirring at 353 K for 2 h, the
resulting powder was dried overnight at 353 K, and then calcined at 773 K for 5 h in dry
air. The loading amount of PGM was fixed at 0.5 wt%.
2.2 Catalytic activity test
Catalytic experiments were carried out in a fixed-bed flow reactor at atmospheric
pressure. A 200 mg sample of the catalyst (25/50 mesh) was placed in a tubular reactor
(i.d. 8 mm). The total flow rate was 100 mL min−1 (W/F = 0.12 g s mL−1, gas hourly space
velocity, GHSV ≈ 15000 h−1). Before each experiment, the catalyst was heated to 773 K
under He flow (30 mL min-1) and held for 1 h. The effluent gases from the reactor were
analyzed by gas chromatography (Shimadzu GC-8A, Porapak Q and MS-5A columns)
equipped with a thermal conductivity detector (TCD). The temperature-programmed
reaction was carried out by increasing the temperature from 373 K in a stepwise manner
(50 K intervals) under stoichiometric conditions (λ=1): 1000 ppm NO, 1000 ppm CO,
250 ppm C3H6, 1125 ppm O2, and He as the balance.
The λ value was used as an indicator of oxygen concentration and was calculated
according to the following equation:
λ = ([CO]a + [NO]a + [O2]a × 2)/([CO]s + [NO]s + [O2]s × 2)
6

[CO]a, [NO]a, and [O2]a represent the actual gas concentrations, and [CO]s, [NO]s, and
[O2]s are the concentrations under stoichiometric condition. The chemical reaction
equation under stoichiometric condition is:
CO + C3H6 + 4 NO + 9/2 O2 → 2 N2 + 7CO2 + 3H2O
Therefore, the oxygen concentration corresponding to λ = 1, [O2]s, is 1125 ppm, and A
λ value of 1 is considered to correspond to A/F = 14.6. The catalytic activity for various
oxygen concentrations was evaluated at 773 K under the following reaction conditions:
1000 ppm NO, 1000 ppm CO, 250 ppm C3H6, 0–2812.5 ppm O2, and He as balance gas,
and the λ value in the catalytic reaction ranged from 1.79 to 0.47. The outlet gases were
analyzed after the reaction at each oxygen concentration had proceeded for 30 min. The
catalytic reaction was commenced under lean conditions (2812.5 ppm O2) and the oxygen
concentration was changed to 0 ppm (rich conditions) in a stepwise manner while
maintaining the NO and CO concentrations.
2.3 Characterization
X-ray diffraction (XRD) patterns were obtained using a powder X-ray diffractometer
(SmartLab, Rigaku, Tokyo Japan) with Cu Kα radiation (=1.5405 Å).
The Brunauer–Emmett–Teller (BET) specific surface areas of catalysts were estimated
from N2 isotherms at 77 K. N2 adsorption isotherms were measured using an adsorption
analyzer (BELSORP-mini II-HSP, MicrotracBEL, Osaka, Japan).
Rh K-edge XAFS measurements were performed at the BL01B1 beam line at SPring-
8 operated at 8 GeV using a Si(111) double-crystal monochromator. XAFS spectra were
obtained at room temperature in fluorescence mode.
7

Diffuse reflectance infrared Fourier transform (DRIFT) spectra were acquired on an
FT/IR-4700 spectrometer (JASCO, Tokyo, Japan), equipped with a mercury-cadmium-
tellurium detector (MCT) cooled by liquid N2. The catalyst was pre-treated with He (100
mL min−1) at 773 K for 1 h. The reaction gas corresponding to a lean condition (total flow
rate, 100 mL min−1: NO, 1000 ppm; CO, 1000 ppm; C3H6, 250 ppm; O2, 1687.5 ppm; He
balance) was introduced into the cell. DRIFT spectra were obtained after the reaction had
proceeded for 30 min at 773 K, followed by cooled to 573 K in a He flow. Then, the
sample was exposure to 1% NO/He. The spectrum was recorded in He stream at the same
temperature.
H2-TPR profiles recorded using an Ohkura BP-2 instrument (Ohkura Riken, Saitama,
Japan) was used for determination of H2 consumption for reduction of Rh species in
supported Rh catalysts. The sample was treated with Ar at 473 K for 1 h prior to
determination of the TPR profile. The catalysts were mounted in quartz tubes and heated
in 5 vol. % H2 diluted with argon without any pretreatment at a heating rate of 10 K min-
1. The change in the H2 concentration was monitored by a thermal conductivity detector
(TCD). Since water is formed during the reduction process, the effluent gas from the
reactor was dried over 5A molecular sieves before entering the TCD. The amount of
consumed H2 was quantified by a calibration curve method.
3. Results and discussion
The XRD patterns of PGM/NbOPO4 and PGM/Al2O3 were similar to that of NbOPO4
and γ-Al2O3, respectively (Fig. S1). No peak due to the PGM species were observed in
the XRD patterns of PGM/NbOPO4 and PGM/Al2O3, indicating that a highly dispersed
PGM species was stabilized on NbOPO4 and γ-Al2O3 in PGM-loaded catalysts.
8

Figure 1 shows the results of NO-selective reduction at stoichiometric oxygen
concentration over PGM catalysts supported on NbOPO4. The oxidation activity of
NbOPO4 was quite low (< 7 %) even at 773 K, and NO reduction was also a low (ca.
14 % NO conversion) even at 773 K (Fig. S2). These catalytic activities were drastically
enhanced by PGM loading. Rh/NbOPO4 showed higher catalytic activity than
Pd/NbOPO4 and Pt/NbOPO4; CO and C3H6 oxidation commenced at 473 K and NO
reduction proceeded from 523 K. γ-Al2O3 and Nb2O5 are inactive for NO reduction below
773 K. Al2O3-supported Pd and Pt catalysts exhibited lower light-off temperature than
those of NbOPO4-supported catalysts (Fig. S3). On the contrary, Rh/NbOPO4 exhibited
lower light-off temperature than that of Rh/Al2O3. Because Rh/NbOPO4 catalyst showed
the highest activity for NO reduction among PGM/NbOPO4 catalysts tested, the following
studies were focused on this catalyst compared with Rh/Al2O3 and Rh/Nb2O5, which were
used as reference catalysts.
Figure 2 shows the results of the reactions of NO-CO-C3H6 at 773 K over supported
Rh catalysts on γ-Al2O3, Nb2O5, and NbOPO4 under various oxygen concentration
conditions. Supported Rh catalysts exhibited the activity for NO reduction and NO
conversion to N2 decreased monotonically with increasing oxygen concentration (λ
values). However, the negative slope under lean conditions (1.1<λ<1.79) was more
moderate for Rh/NbOPO4 catalyst than that for Rh/Al2O3 catalyst, and Rh/NbOPO4
catalyst shows superior catalytic activity for NO reduction over a wide range of oxygen
concentrations to Rh/Al2O3. Under stoichiometric (λ=1) to slightly rich (λ>0.8) oxygen
conditions, both Rh/NbOPO4 and Rh/Al2O3 effectively reduced NO to N2 regardless of
the support. On the other hand, in the absence of oxygen, the activity of Rh/NbOPO4
catalyst was lower than that of Rh/Al2O3 catalyst.
9

To elucidate the reasons for the high efficiency for NO reduction of Rh/NbOPO4
catalyst under lean conditions, the adsorbed species was characterized by DRIFT spectra
during the NO-CO-C3H6-O2 reactions under high oxygen concentration condition (lean
condition: λ=1.26). Figure 3 shows DRIFT spectra of adsorbed species over Rh/Al2O3
and Rh/NbOPO4 catalysts. The assignment of absorption bands described below is
consistent with those previously reported [24-30]. Rh/Al2O3 showed the bands assignable
to νas(O-C-O) of bidentate acetate at 1580 cm−1, νs(O-C-O) of bidentate acetate at 1460
cm−1, νas(O-C-O) of carboxylate at 1550 cm−1, and δ(C-H) of formate at 1380 cm−1. Weak
bands due to NO adsorbed on Rh (1910 cm-1), CO adsorbed on Rh (2096 cm-1 and 2018
cm-1), and NCO species bound for Al ions (Al-NCO) at 2232 cm-1 were also observed
(Fig. 3 (a)). Rh/NbOPO4 yielded less intense bands in this region. These bands were
assignable to allylaldehyde− CHO (1678 cm-1) and CO adsorbed on Rh (2122 cm-1 and
2047 cm-1) (Fig. 3(c)). Generally, CO adsorbs on Rh and an electron-deficient Rh species
may be generated by electron drawing effect of phosphate supports. As a result, reducing
back-donation from the Rh d-orbitals to the antibonding π* orbitals of CO leads to a shift
of the bands due to adsorbed CO to higher wavenumbers [14].
It has been reported that the oxidation of C3H6 proceeds via π-allyl complexes, which
may undergo an abstraction of α-hydrogen followed by the insertion of oxygen to an
allylic C−H bond [31, 32]. In general, C3H6 adsorbs on the Lewis acid sites and the
abstraction of α-hydrogen takes place on the Brønsted aicd sites [33]. The surface of
NbOPO4 are terminated by P−OH and Nb−OH groups [14], which possibly act as
Brønsted acid sites. Indeed, pyridine adsorbed FT-IR spectra demonstrated that both
Brønsted and Lewis acid sites coexisted on AlPO4 and ZrP2O7 [14, 16]. On the other hand,
it is well known that γ-Al2O3 had only Lewis acid sites. Therefore, because acid property
10

differs from each other as mentioned above, the different partially oxidized species
derived from C3H6 adsorbed on Rh/NbOPO4 and Rh/Al2O3.
The reactivity of the partially oxidized C3H6 species adsorbed on Rh/NbOPO4 and
Rh/Al2O3 with NO was investigated (Fig. 3 (b) and (d)). After exposure to 1% NO/He at
573 K, the intensity of band due to aldehyde (1678 cm−1) on Rh/NbOPO4 was weakened
and a band corresponding to NO adsorbed on Rh (1926 cm−1) appeared. Considering that
aldehyde species formed on NbOPO4 even in the presence of excess O2, it can be thought
that aldehyde species may play an important role as reducing agents in NO reduction
under the lean conditions. In the case of Rh/Al2O3, weak bands due to CO adsorbed on
Rh (2096 cm-1), and NCO species bound for Al ions (Al-NCO) (2232 cm-1) were
weakened. On the other hand, a negligible change in the bands due to carboxylate was
observed, indicating lower reactivity of carboxylate species formed on Rh/Al2O3
compared to the aldehyde species on Rh/NbOPO4.
Because Rh oxide (Rh2O3) is thermodynamically stable in O2 up to approximately 1173
K, excess oxygen affects both the state of Rh and the catalytic activity in the lean region.
The oxidation state of Rh under lean conditions was studied by Rh K-edge XANES.
Figure 4 shows the Rh K-edge XANES spectra of Rh/NbOPO4 and Rh/Al2O3. The
fractions of metallic Rh (Rh0) and oxidized Rh (Rh3+) were estimated by liner
combination analysis (LCA) using two references (a Rh foil and Rh2O3). When Rh/Al2O3
was treated in gas mixtures containing 1000 ppm NO, 1000 ppm CO, 250 ppm C3H6,
1350 ppm O2 balanced with He (corresponding to lean conditions, λ=1.1) at 773 K, the
spectrum was quite similar to that of Rh2O3 (< 5% Rh0), suggesting that the metallic Rh
species on Al2O3 is easily oxidized and/or the reduction of Rh2O3 on Al2O3 is slow even
under such slightly lean conditions (λ=1.1). In contrast, the height of white line of Rh K-
11

edge XANES spectrum of Rh/NbOPO4 was much lower than that of Rh2O3 and the
fraction of metallic Rh was estimated to be ca. 60 % after the treatment under lean
conditions (λ=1.1). This result clearly indicates that the metallic Rh species on NbOPO4
is substantially stabilized against oxidation.
To characterize different reduction/reoxidation behaviors of Rh/NbOPO4 and
Rh/Al2O3 catalysts, H2-TPR experiments were performed. Figure 5 showed H2-TPR
profiles of Rh catalysts treated under slightly lean conditions (λ=1.1 and 1.26). H2
consumption due to the reduction of Rh oxide to metallic Rh yielded a peak at 380 K for
Rh/Al2O3. The total amount of H2 consumed during the process was approximately 25 %
of the stoichiometric reaction (Rh2O3 + 3H2 = 2Rh + 3H2O). On the other hand, in the
case of Rh/NbPO4, H2 consumption was almost exactly zero, suggesting a very slow
oxidation rate of metallic Rh caused by the stabilization of NbOPO4. Combining the
results of the Rh K-edge XANES and H2-TPR, the NbOPO4 support enables a higher
reducibility of Rh oxide to metallic Rh, which is highly stable against reoxidation.
A similar unusual redox behavior of Rh on AlPO4, which is associated with and
Rh−O−P interfacial linkage via surface phosphate units, was reported [11, 12, 14, 16].
The formation of Rh−O−P bonding in Rh supported on AlPO4, ZrP2O7 and LaPO4 was
confirmed by Rh K-edge EXAFS. The Rh oxide anchored by this interfacial linkage is
easily reduced to active metallic Rh, whereas Rh−O−Al bonding formed in Rh/Al2O3 is
very difficult to reduce. On the other hand, an electron-drawing effect from the phosphate
group decreases the electron density of Rh metal and thus stabilizes against reoxidation
to Rh oxide, compared with Rh metal on Al2O3. In the case of Rh/NbOPO4, it is likely
that Rh−O−P interfacial linkage via surface phosphate units of NbOPO4 is present as well
12

as AlPO4, ZrP2O7 and LaPO4. As a result, unusual redox behavior of Rh was observed on
NbOPO4.
The present study demonstrates that Rh/NbOPO4 exhibit a higher activity for NO
reduction in slightly lean regions compared with Rh/Al2O3. Many researchers have
proposed the possible roles of HCs such as C3H6 as follows: (i) the reductants remove
adsorbed oxygen to enable NO dissociation to form N and O and subsequent desorption
of N2; (ii) the reductants are partially oxidized to activated intermediate species that react
with adsorbed NO [19-23]. A reactive aldehyde species was formed on Rh/NbOPO4
whereas less-reactive carboxylate species was formed on Rh/Al2O3. The aldehyde species
on Rh/NbOPO4 shows high reactivity toward NO, and exists on the surface of
Rh/NbOPO4 even under lean conditions as shown in Fig. 3, strongly suggesting that
aldehyde species plays a key role as reducing agents in the NO reduction under lean
conditions. Due to low reactivity of carboxylate species, the carboxylate species are stable,
resulting in the inhibition of adsorption of NO and O2. Furthermore, unusual redox
behavior of Rh on NbOPO4 has also an important role in NO reduction in slightly lean
regions. The reduction of Rh oxide is much easier than that of the Rh oxide formed on
Al2O3 and metallic Rh is further stabilized against oxidation on NbOPO4. A higher
fraction of metallic Rh in Rh/NbOPO4 favors adsorption and activation of NO, and the
formation of reactive aldehyde species on the surface of NbOPO4 favors the reaction
between surface oxygen and/or NO to generate N2. Consequently, 1) the high activity of
Rh/NbOPO4 for the formation of aldehyde species via partial oxidation of C3H6, and 2)
metallic Rh having tolerance against reoxidation may result in the enhanced deNOx
performance of Rh/NbOPO4 in a slightly lean region.
4. Conclusions
13

The selective catalytic reduction of NO with C3H6 and CO was investigated over
platinum group metal (PGM; Pd, Rh, or Pt) catalysts supported on NbOPO4. Rh/NbOPO4
exhibited higher activity than those of Pd- or Pt-loaded NbOPO4 catalysts and Rh/Al2O3
at lower temperatures under stoichiometric conditions. Rh/NbOPO4 could preserve a
higher the activity for NO reduction than Rh/Al2O3 and Rh/Nb2O5 within a wide range of
lean region. Partially oxidized C3H6 was adsorbed on Rh/NbOPO4 in the form of reactive
aldehyde species, in contrast to the less-reactive carboxylate species adsorbed on
Rh/Al2O3. Metallic Rh formed on NbOPO4 was only slowly oxidized on exposure to
excess O2, whereas Rh on Al2O3 was readily oxidized to less-active Rh2O3. The high
activity of Rh/NbOPO4 toward C3H6 oxidation via aldehyde species as well as the
resistance of metallic Rh against reoxidation may result in the enhanced deNOx
performance of Rh/NbOPO4 in a slightly lean region.
Acknowledgement
This study was supported in part by the Program for Element Strategy Initiative for
Catalysts & Batteries (ESICB) commissioned by the Ministry of Education, Culture,
Sports, Science and Technology (MEXT) of Japan. The XAFS experiment at SPring–8
was carried out under the approval (proposal No. 2016B1224 and 2017B1245) of Japan
Synchrotron Radiation Research Institute (JASRI).
14

References
[1] J. Kaspar, P. Fornasiero, M. Graziani, Catalysis Today, 50 (1999) 285-298.
[2] R.M. Heck, R.J. Farrauto, S.T. Gulati, Catalytic Air Pollution Control:Commercial
Technology, Wiley, 1995
[3] K.C. Taylor, R.L. Klimisch, J. Catal. 30 (1973) 478
[4] M. Shelef, Catal. Rev. Sci. Eng. 11 (1975) 1.
[5] M. Shelef, G.W. Graham, Catal. Rev. Sci. Eng. 36 (1994) 433.
[6] R.J. Farrauto, R.M. Heck, Catal. Today 51 (1999) 351.
[7] M. Shelef, R.W. McCabe, Catal. Today 62 (2000) 35.
[8] M. Haneda, O. Houshito, H. Takagi, K. Shinoda, Y. Nakahara, K. Hiroe, T.Fujitani,
H. Hamada, Top. Catal. 52 (2009) 1868.
[9] M. Machida, K. Murakami, S. Hinokuma, K. Uemura, K. Ikeue, M. Matsuda, M. Chai,
Y. Nakahara, T. Sato, Chemistry of Materials, 21 (2009) 1796-1798.
[10] K. Ikeue, K. Murakami, S. Hinokuma, K. Uemura, D. Zhang, M. Machida, Bull.
Chem. Soc. Japan, 83 (2010) 291-297.
[11] M. Machida, S. Minami, K. Ikeue, S. Hinokuma, Y. Nagao, T. Sato, Y. Nakahara,
Chem. Mater., 26 (2014) 5799-5805.
[12] H.P. Buwono, S. Minami, K. Uemura, M. Machida, Ind. & Eng. Chem. Res., 54
(2015) 7233-7240.
[13] M. Machida, T. Eidome, S. Minami, H.P. Buwono, S. Hinokuma, Y. Nagao, Y.
Nakahara, J. Phys. Chem. C, 119 (2015) 11653-11661.
[14] M. Machida, S. Minami, S. Hinokuma, H. Yoshida, Y. Nagao, T. Sato, Y. Nakahara,
J. Phys. Chem. C, 119 (2015) 373-380.
[15] M. Matsui, M. Machida, S. Sakaki, J. Phys. Chem. C, 119 (2015) 19752-19762.
[16] Y. Nagao, Y. Nakahara, T. Sato, H. Iwakura, S. Takeshita, S. Minami, H. Yoshida,
M. Machida, ACS Catalysis, 5 (2015) 1986-1994.
[17] H.P. Buwono, S. Takeshita, S. Hinokuma, H. Yoshida, Y. Nagao, Y. Nakahara, M.
Machida, Catal. Today, 281 (2017) 583-589.
[18] H.P. Buwono, M. Yamamoto, R. Kakei, S. Hinokuma, H. Yoshida, M. Machida, J.
Phys. Chem. C, 121 (2017) 17982-17989.
[19] R. Burch, J. P. Breen, F. C. Meunier, Appl. Catal. B: Environ. 39 (2002) 283−303.
[20] A. Fritz, V. Pitchon, Appl. Catal. B: Environ. 13 (1997) 1−25.
[21] M. Iwamoto, H. Hamada, Catal. Today 10 (1991) 57−71.
[22] A. Obuchi, A. Ohi, M. Nakamura, A. Ogata, K. Mizuno, H. Ohuchi, Appl. Catal. B:
Environ. 2 (1993) 71−80.
15

[23] H. Hamada, Y. Kintaichi, M. Sasaki, T. Ito, M. Tabata, Appl. Catal. 75 (1991) L1−L8.
[24] T. I. Halkides, D. I. Kondarides, X. E. Verykios, Catal. Today 73 (2002) 213−221.
[25] C. Chauvin, J. Saussey, J.-C. Lavalley, H. Idriss, J.-P. Hindermann, A. Kiennemann,
P. Chaumette, P. Courty, J. Catal. 121 (1990) 56−69.
[26] E. Finocchio, M. Daturi, C. Binet, J. C. Lavalley, G. Blanchard, Catal. Today 52
(1999) 53−63.
[27] M. Haneda, K. Shinoda, A. Nagane, O. Houshito, H. Takagi, Y. Nakahara, L. Hiroe,
T. Fujitani, H. Hamada, J. Catal 259 (2008) 223–231.
[28] C. Martin, I. Martin, V.J. Rives, Mol. Catal. 73 (1992) 51–63.
[29] E. Finocchio, G. Busca, V. Lorenzelli, J. Chem, Soc, J. Chem. Soc. Faraday Trans.
I, 90 (1994) 3347–3356.
[30] C.C.R. Sutton, G.D. Silva, G.V. Franks, Chem. Eur. J. 21 (2015) 6801–6805.
[31] A. Bielansky, J. Haber, Oxygen in Catalysis; Marcel Dekker: New York, 1991.
[32] R. Grasselli, G.; Centi, F. Trifiro, Appl. Catal. 57 (1990) 149−166.
[33] B. Zhou, K. T. Chuang, X. J. Guo, J. Chem. Soc., Faraday Trans. 87 (1991)
3695−3702.
16

Figure captions
Figure 1. NO-selective reduction at stoichiometric oxygen concentration over PGM
catalysts supported on NbOPO4.
Conditions: Total flow rate 100 mL min-1, 1000 ppm NO, 1000 ppm CO, 250
ppm C3H6, 1125 ppm O2, and He as the balance (λ=1).
Figure 2. The reactions of NO-CO-C3H6 at 773 K over supported Rh catalysts on γ-Al2O3
and NbOPO4 under various oxygen concentration conditions.
Condition: Total flow rate 100 mL min-1, 1000 ppm NO, 1000 ppm CO, 250 ppm
C3H6, 0–2812.5 ppm O2, and He as the balance (λ=0.47-1.79).
Figure 3. DRIFT spectra of adsorbed species over Rh/Al2O3 and Rh/NbOPO4 catalysts.
(a) and (c) in gas mixture containing NO, 1000 ppm; CO, 1000 ppm; C3H6, 250
ppm; O2, 1687.5 ppm; He balance (λ=1.26) at 773K and (b) and (d) subsequent
admission of 1% NO/He at 573 K.
Figure 4. Rh K-edge XANES spectra of Rh/NbOPO4 and Rh/Al2O3 treated in gas mixture
containing NO, 1000 ppm; CO, 1000 ppm; C3H6, 250 ppm; O2, 1350 ppm; He
balance (λ=1.1) at 773 K.
Figure 5. H2-TPR profiles of Rh/NbOPO4 and Rh/Al2O3.
(a) treated in gas mixture containing NO, 1000 ppm; CO, 1000 ppm; C3H6, 250
ppm; O2, 1350 ppm; He balance (λ=1.1) and (b) treated in gas mixture containing
NO, 1000 ppm; CO, 1000 ppm; C3H6, 250 ppm; O2, 1687.5 ppm; He balance
(λ=1.26).
17

100
80
60
40
20
0
100
80
60
40
20
0
Rh
Pd
Pt
Rh
Pd
Pt
373
473
573
673
773
373
473
573
673
773
Temperature / K
Temperature / K
Fig. 1 NO-selective reduction with CO and C3H6 at stoichiometric oxygen concentration
over PGM catalysts supported on NbOPO4.
Conditions: Total flow rate 100 mL min-1, 1000 ppm NO, 1000 ppm CO, 250 ppm C3H6,
1125 ppm O2, and He as the balance (λ=1).
18

l
l
0.6 0.8 1.0 1.2 1.4 1.6
0.6 0.8 1.0 1.2 1.4 1.6
100
80
60
40
20
0
100
80
60
40
20
0
Rh/NbOPO4
Rh/Al2O3
0
500 1000 1500 2000 2500
O2 concentration (ppm)
0
500 1000 1500 2000 2500
O2 concentration (ppm)
Fig. 2. The reactions of NO-CO-C3H6 at 773 K over supported Rh catalysts on γ-Al2O3
and NbOPO4 under various oxygen concentration conditions.
Condition: Total flow rate 100 mL min-1, 1000 ppm NO, 1000 ppm CO, 250 ppm C3H6,
0–2812.5 ppm O2, and He as the balance (λ=0.47-1.79).
19

0.2
Rh/NbOPO4
(d)
x 4
x 4
(c)
Rh/Al2O3
(b)
(a)
2200 2000 1800 1600 1400
Wavenumber / cm-1
Fig. 3. DRIFT spectra of adsorbed species over Rh/Al2O3 and Rh/NbOPO4 catalysts.
(a) in gas mixture containing NO, 1000 ppm; CO, 1000 ppm; C3H6, 250 ppm; O2, 1687.5
ppm; He balance (λ=1.26) at 773 K and (b) subsequent admission of 1% NO/He at 573
K.
20

1.2
1.0
0.8
0.6
0.4
0.2
0.0
Rh metal
Rh2O3
Rh/Al2O3
Rh/NbOPO4
23200
23250
23300
Photon energy / eV
Fig. 4. Rh K-edge XANES spectra of Rh/NbOPO4 and Rh/Al2O3 treated in gas mixture
containing NO, 1000 ppm; CO, 1000 ppm; C3H6, 250 ppm; O2, 1350 ppm; He balance
(λ=1.1) at 773 K.
21

0.01
Rh/NbOPO4
(b)
(a)
Rh/Al2O3
(b)
(a)
323
373
423
473
Temperature / K
Figure 5. H2-TPR profiles of Rh/NbOPO4 and Rh/Al2O3.
(a) treated in gas mixture containing NO, 1000 ppm; CO, 1000 ppm; C3H6, 250 ppm; O2,
1350 ppm; He balance (λ=1.1) and (b) treated in gas mixture containing NO, 1000 ppm;
CO, 1000 ppm; C3H6, 250 ppm; O2, 1687.5 ppm; He balance (λ=1.26).
22