NH3 oxidation mechanism over Cu/sapo-34 catalysts prepared by different methods

Article abstract of DOI:10.1002/cctc.201402048

NH₃ oxidation mechanism and its inhibition effect on the selective catalytic reduction of NO_x by NH₃ (NH ₃-SCR) reaction over two series of Cu/SAPO-34 catalysts were studied. It was found that the different NO conversions were induced by the diverse NH₃ oxidation degree. X-ray diffraction, scanning electron microscopy, H₂ temperature-programmed reduction (H₂-TPR), and CO diffuse reflectance infrared Fourier transform spectroscopy (CO-DRIFTS) analyses were conducted to estimate the Cu species distribution. The TPR results showed that the CuO species was dominant in impregnated Cu/SAPO-34 samples, and the ion-exchanged Cu/SAPO-34 samples contained more isolated Cu²⁺ species. Furthermore, H₂-TPR, electron paramagnetic resonance, and CO-DRIFTS analyses were utilized to quantify the CuO, Cu²⁺, and Cu⁺ species. The results together with the kinetic results revealed the CuO is the active site for NH₃ oxidation, and the NH₃ oxidation over Cu/SAPO-34 sample consists two steps: (a) NO production; (b) further NO consumption by adsorbed NH₃ species on Cu²⁺ sites. Finally, the inhibition of NH₃ oxidation on NH₃-SCR activity was explained and the NO conversion at high temperature was related to the Cu²⁺/CuO ratio in Cu/SAPO-34 sample.

Full text of DOI:10.1002/cctc.201402048

CHEMCATCHEM
FULL PAPERS
DOI: 10.1002/cctc.201402048
NH3 Oxidation Mechanism over Cu/SAPO-34 Catalysts
Prepared by Different Methods
Tie Yu,[a] Jun Wang,[a] Yu Huang,[a] Meiqing Shen,*[a, b] Wei Li,[c] and Jianqiang Wang*[a]
NH3 oxidation mechanism and its inhibition effect on the selec-
tive catalytic reduction of NOx by NH3 (NH3–SCR) reaction over
two series of Cu/SAPO-34 catalysts were studied. It was found
that the different NO conversions were induced by the diverse
NH3 oxidation degree. X-ray diffraction, scanning electron mi-
croscopy, H2 temperature-programmed reduction (H2–TPR),
and CO diffuse reflectance infrared Fourier transform spectros-
copy (CO–DRIFTS) analyses were conducted to estimate the Cu
species distribution. The TPR results showed that the CuO spe-
cies was dominant in impregnated Cu/SAPO-34 samples, and
the ion-exchanged Cu/SAPO-34 samples contained more isolat-
ed Cu2+ species. Furthermore, H2–TPR, electron paramagnetic
resonance, and CO–DRIFTS analyses were utilized to quantify
the CuO, Cu2+, and Cu+ species. The results together with the
kinetic results revealed the CuO is the active site for NH3 oxida-
tion, and the NH3 oxidation over Cu/SAPO-34 sample consists
two steps: (a) NO production; (b) further NO consumption by
adsorbed NH3 species on Cu2+ sites. Finally, the inhibition of
NH3 oxidation on NH3–SCR activity was explained and the NO
conversion at high temperature was related to the Cu2+/CuO
ratio in Cu/SAPO-34 sample.
Introduction
Selective catalytic reduction of NOx by NH3 (NH3–SCR) was
used as an effective technology to eliminate the NOx from the
exhaust of diesel engine.[1–4] In recent years, zeolite SCR cata-
lysts such as Cu/ZSM-5 and Fe/beta,[5–8] etc. have attracted
more attention than oxide and noble-metal catalysts, because
of their excellent activity and N2 selectivity, However, the alu-
minosilicate zeolite catalysts (Cu/ZSM-5 and Fe/beta) display
inferior hydrothermal stability at high temperature as a result
of dealumination causing the skeleton collapse and agglomer-
ation of active sites.[9–13] Nevertheless, SAPO-34, a small-pore-
sized molecular sieve with chabazite structure built from infin-
ite TO4 tetrahedral units (T=Si, Al, and P), maintains its crystal-
linity even after treatment at 1173 K.[14] Naturally, the Cu/SAPO-
34 zeolite has been reported to be a promising SCR catalyst
owing to both the superior NOx conversion and the excellent
hydrothermal stability at high temperature.[15–21]
The NH3–SCR activity of Cu/SAPO-34 catalyst was firstly re-
ported by Bull et al. and Andersen et al. in early US pat-
ents,[15,16] and the Cu/SAPO-34 sample, with 2.41 wt% CuO, ex-
hibits 90% NO conversion at 240–4608C. In addition, Fickel
and others found that Cu/SAPO-34 catalysts perform superior
NH3–SCR activity to that of Cu/zeolites, and their prominent
hydrothermal stability at 7508C was related to their unique mi-
croporous structure.[17,20] Meanwhile, our previous research-
es[18,19] proved the isolated Cu2+ species are the SCR active
sites by kinetic tests. Similarly, the active sites of isolated Cu2+
in Cu/SAPO-34 were also studied by Korhonen[22] by EXAFS
and Gao[23] by kinetic methods. Throughout the above activity
studies on the Cu/SAPO-34 catalyst, there is a common re-
markable phenomenon: the NO conversion decreases at high
temperature, but the NH3 conversion still keeps growing. It is
predicted NH3 oxidation inhibits NO conversion. On the other
hand, in the practical after treatment system of diesel, an NH3
oxidation catalyst is always set after SCR catalyst to eliminate
the NH3 slip and prevent the secondary pollution. Therefore, it
is important to study the NH3 oxidation activity of Cu/SAPO-34
catalyst and understand its determinant. If a screened Cu/
SAPO-34 catalyst simultaneously performs superior SCR and
NH3 oxidation activity, the NH3 oxidation catalyst can be omit-
ted for economy reasons. However, there has been no special
research on the NH3 oxidation mechanism and the active site
over Cu/SAPO-34 catalyst, and its detailed influence on the
NH3–SCR reaction has not been concluded yet. To confirm the
active site of NH3 oxidation, various Cu species in Cu/SAPO-34
catalyst should be firstly studied. It is known that the prepara-
tion methods could affect the Cu distribution in Cu/SAPO-34
catalyst. Previous studies[19,20,24,25] reported that Cu/SAPO-34
samples, prepared by one-pot, precipitation, and ion-exchang-
[a] Dr. T. Yu, Prof. J. Wang, Y. Huang, Prof. M. Shen, Dr. J. Wang
Key Laboratory for Green Chemical Technology of State Education Ministry
School of Chemical Engineering & Technology
Tianjin University
Tianjin 300072 (P.R. China)
Fax: (+86)22-27892301
E-mail: jianqiangwang@tju.edu.cn
[b] Prof. M. Shen
State Key Laboratory of Engines
Tianjin University
Tianjin 300072 (P.R. China)
[c] Prof. W. Li
General Motors Global Research and Development
Chemical Sciences and Materials Systems Lab
3500 Mound Road, Warren, Michigan 48090 (United States)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201402048.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&1
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
ing methods, all contained Cu2+, CuO, and Cu+ species. The
one-pot and ion-exchanged method were found optimal
methods to prepare Cu/SAPO-34 samples with more isolated
Cu2+ contents, performing good NO conversions, whereas the
precipitation method generated more CuO species in Cu/
SAPO-34 catalyst. Consequently, herein, the Cu/SAPO-34 cata-
lysts prepared by different methods are used to study the
active site of NH3 oxidation and its influence on the NH3–SCR
reaction.
To explore whether the CuO species is the active site for NH3
oxidation in Cu/SAPO-34 catalyst, four samples with various
CuO contents (1, 2, 4, and 8 wt%) were prepared by the im-
pregnation method to ensure the dominant CuO state, and
they were further applied to study the NH3 oxidation mecha-
nism. In addition, three Cu/SAPO-34 samples, with Cu loadings
of 0.7, 1, and 1.5 wt% by the ion-exchange method were also
prepared to confirm the active site obtained in impregnated
samples. Firstly, the Cu location and distribution over two
series of Cu/SAPO-34 samples were identified by XRD, SEM,
and hydrogen temperature-programmed reduction (H2–TPR).
Then, electron paramagnetic resonance (EPR) spectroscopy
and CO diffuse reflectance infrared Fourier transform spectros-
copy (CO–DRIFTS) were used in combination with H2–TPR to
quantify the Cu2+, CuO, and Cu+ species. Furthermore, kinetic
NH3 oxidation analysis was performed to estimate the appar-
ent activation energy (Ea) and the active site of NH3 oxidation.
Finally, the NH3–SCR and the NH3 oxidation activities over im-
pregnated and ion-exchanged Cu/SAPO-34 samples were uti-
lized to explain the competition between them at high tem-
perature, and the reaction network of the two reactions during
NH3–SCR process was concluded.
Figure 1. XRD profiles of a) impregnated Cu/SAPO-34 and b) ion-exchanged
Cu/SAPO-34 catalysts.
with the decrease in Cu loading from Figure 2b to 2d. To iden-
tify the chemical composition of the scrap, energy dispersive
spectrometry area scanning was performed on the IM-Cu8
sample, because this has the highest Cu content. The results
presented in Figure 2g and 2h show that the scrap contains
23.66 wt% Cu, which is much higher than the Cu loading in
IM-Cu8 sample. Additionally, this scrap is ascribed to the CuO
species on the external of Cu/SAPO-34 samples. This result also
is also consistent with the XRD results that the CuO species in-
creases with the increment of Cu loading. Nevertheless, for the
EX-Cu1.5 with the highest Cu loading in ion-exchanged sam-
ples, there are almost no CuO particles detected in Figure 2e
compared with the SAPO-34 support in Figure 2 f.
Results and Discussion
Results of XRD and SEM
The XRD profiles of the synthesized molecular sieves in
Figure 1 present the typical chabazite structure, illustrating the
formation of SAPO-34 supports. For the four impregnated sam-
ples in Figure 1a, the loading of Cu does not affect the crystal-
linity of the molecular sieves. However, two CuO peaks appear
at 35.5 and 38.78, and their intensities rise with the increment
of Cu loading. The particle sizes of CuO, calculated by the
38.78 peaks in the XRD profiles according to the Scherrer equa-
tion, are 0.42, 0.46, 0.49, and 0.52 nm for IM-Cu1, IM-Cu2, IM-
Cu4, and IM-Cu8, respectively. In addition, the ion-exchanged
Cu/SAPO-34 samples present the same XRD spectra with the
SAPO-34 support, and no CuO peaks are detected, even for
EX-Cu1.5. It is concluded that the impregnation method tends
to generate large CuO particles and the ion-exchange method
generates smaller ones.
The specific surface areas (SSA) of Cu/SAPO-34 samples pre-
pared by the impregnation method (Table 1) decrease relative
to that of the SAPO-34 support, which is caused by pore block-
ing by the CuO particles. Nevertheless, the ion-exchange Cu/
SAPO-34 samples present SSA results equivalent to those of
SAPO-34 support, implying less CuO formation or smaller CuO
particles in them.
In Figure 2, the SEM images of partial Cu/SAPO-34 samples
are shown. Firstly, the preparation process does not have
a clear influence on the particles structure, based on the same
morphology in all images. For the impregnated samples, the
IM-Cu8 presents much scrap on the surface of Cu/SAPO-34
particles as seen in Figure 2a, and this scrap strongly declines
The above bulk characterization presents the impregnated
Cu/SAPO-34 samples all contain big particle sized CuO species,
and their sizes increase with the Cu loading. The Cu species in
ion-exchanged Cu/SAPO-34 display well dispersion or the
formed CuO species cannot be detected.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
2
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Table 1. Composition of all Cu/SAPO-34 catalysts.
Sample
Cu loading
[wt%]
SSA
[m2 gꢀ1
Cu2+/CuO
[molmolꢀ1
]
]
SAPO-34
IM-Cu1
IM-Cu2
IM-Cu4
IM-Cu8
EX-Cu0.7
EX-Cu1
EX-Cu1.5
–
1
2
4
8
0.7
1
564
482
480
467
450
540
554
539
–
0.31
0.12
0.28
0.14
0.5
0.88
1
1.5
NH3–SCR and NH3 oxidation activities
It is known that the isolated Cu2+ species are the active sites
for NH3–SCR over Cu/SAPO-34 catalyst.[18,19,22] As seen in Fig-
ure 3a, the four impregnated Cu/SAPO-34 samples present
narrow NH3–SCR activity windows. At low temperature, the IM-
Cu4 performs the highest NO conversion, whereas the IM-Cu1
exhibits the best SCR activity at high temperature. The
switched NO conversions are related to the diverse Cu distribu-
tion in them, which will be discussed later. In addition, it is
clearly seen that the NO conversions are less than 0 above
4008C (Figure 3a), whereas the NH3 conversions do not decline
(Figure 3b), which is a result of the strong NH3 oxidation. The
NO2 generation in Figure 3c increases sharply at high tempera-
ture. Compared with the impregnated Cu/SAPO-34 samples,
the ion-exchanged Cu/SAPO-34 samples reveal a wide NH3–
SCR activity window. The NO conversions improve with the in-
crement of Cu loading (Figure 3d), and the three samples still
perform superior NH3–SCR activities at high temperature. The
NH3 conversions (Figure 3e) present the same sequence with
Figure 2. SEM images of Cu/SAPO-34 catalysts. a) IM-Cu8; b) IM-Cu4; c) IM-
Cu2; d) IM-Cu1; e) EX-Cu1.5; f) SAPO-34 support; g and f) the EDS images of
IM-Cu8.
Figure 3. NH3–SCR activities results. a) NO conversion, b) NH3 conversion, and c) NO2 concentrations over impregnated Cu/SAPO-34 catalysts; d) NO conver-
sion, e) NH3 conversion, and f) NO2 concentrations over ion-exchanged Cu/SAPO-34 catalysts. The feed contains 500 ppm NH3, 500 ppm NO, 5% O2, N2 bal-
ance. The flow rate and the volume hourly space velocity in all experiment were controlled at 1000 mLminꢀ1 and 600000 hꢀ1
.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
3
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
NO conversions and increase with the temperature. The NO2
generation in Figure 3 f is much less than impregnated Cu/
SAPO-34 samples, illustrating the weakening of NH3 oxidation
over ion-exchanged samples. However, the NO conversions
also start to decrease above 4008C.
Considering the different influences of NH3 oxidation on the
SCR results as seen in Figure 3, the NH3 oxidation was per-
formed to estimate the diverse competition degree of NH3
consumption over two series of Cu/SAPO-34 samples at high
temperature. As shown in Figure 4a, the four impregnated Cu/
SAPO-34 samples present superior NH3 conversion above
3008C, which is the turning point in NH3–SCR. In addition, the
NH3 conversions improve with the increment of Cu loading,
and the main byproduct is NO as seen in Figure 4b. Therefore,
it is likely to understand that both the competitive consump-
tion of reactant NH3 and the NO production at high tempera-
ture can directly decrease the NH3–SCR activities shown Fig-
ure 3a. The NH3 oxidation over ion-exchanged samples is
showed in Figure 4c. The NH3 conversions also rise with the in-
crement of Cu loading, and the NH3 oxidation occurs above
3008C. The NO productions in Figure 4d are all less than
20 ppm, and this wide difference from those of the impregnat-
ed Cu/SAPO-34 samples in Figure 4b is related to their diverse
Cu distribution. Combining the above activities results, it is
concluded that the decline of NH3–SCR activity at high temper-
ature is mainly induced by the NH3 oxidation. In addition, it is
very important to study the active site of NH3 oxidation to ex-
plain the difference over the two series of Cu/SAPO-34 cata-
lysts.
Cu distribution in Cu/SAPO-34 catalysts
It is known that the isolated Cu2+, CuO, and Cu+ species
always coexist in Cu/SAPO-34 catalysts regardless of the Cu
loading.[18,21,27] To determine the distribution and contents of
various Cu species in two series of Cu/SAPO-34 samples, three
different methods are utilized to quantify the Cu species, H2–
TPR, EPR spectrometry, and CO-Drifts.
In Figure 5 the H2–TPR profiles of Cu/SAPO-34 samples are
shown, and the TPR curve of HSAPO-34 is subtracted as the
baseline of the Cu/SAPO-34 sample. The H2 consumption peak
of impregnated samples in Figure 5a from 200 to 3508C was
found to contain two peaks after deconvolution and curve sto-
chastic fitting procedures. The peak at 2508C represents the
reduction of isolated Cu2+ to Cu+, and the peak at higher tem-
perature represents the reduction of the copper oxide in zeo-
lites from CuO to Cu0.[18,21] The Cu2+ species and CuO contents
are listed in Figure 6a. It is seen that the CuO contents in-
crease with the increment of Cu loading, but the IM-Cu4 con-
tains the highest isolated Cu2+ species among them. The Cu+
species are not shown because of its few contents. Notably,
the CuO reduction peaks in Figure 5a shift to lower tempera-
ture with the rise of Cu loading, which may be caused by the
increase of the particle sizes. The bigger the CuO particles in
the Cu/SAPO-34 sample are, the weaker the interaction is be-
tween the SAPO-34 supports and CuO species. This result is
consistent with the sequence of particle sizes calculated from
Figure 4. NH3 oxidation results. a) NH3 conversion and b) NO concentration
over impregnated Cu/SAPO-34 catalysts; c) NH3 conversion and d) NO con-
centration over ion-exchanged Cu/SAPO-34 catalysts. The inlet gases con-
tained 500 ppm NH3, 5% O2 with N2 as the balance. The flow rate and the
volume hourly space velocity in all experiment were controlled at
1000 mLminꢀ1 and 600000 hꢀ1
.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
4
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
5008C contain three peaks as shown Figure 5b. The peak at
low temperature represents the reduction of isolated Cu2+ to
Cu+. The one at moderate temperature represents the reduc-
tion of the copper oxide in zeolites from CuO to Cu0, and the
third peak at the higher temperature comes from the reduc-
tion from Cu+ to Cu0.[18,21] In addition, another H2 consumption
peak above 6008C in EX-Cu1.5 can be attributed to the reduc-
tion of highly stable Cu+ to Cu0, which is named as “H (High)-
Cu+”.[28–31] For the ion-exchanged Cu/SAPO-34 samples, the
quantification of the content of various Cu species (Figure 6b)
reveals that the CuO and isolated Cu2+ species both improve
with the increment of Cu loading. Furthermore, the Cu2+/CuO
(mole ratio), listed in Table 1, is very important to explain the
NO conversion at high temperature, which will be discussed
later.
EPR analysis is an effective way to characterize the isolated
Cu2+ species because only Cu2+ species are active, but CuO,
binuclear species, or Cu+ ions are inactive. In Figure 7 the EPR
spectra of two series of Cu/SAPO-34 samples are presented to
further evaluate the Cu2+ state. It is seen that all the Cu/SAPO-
34 catalysts contain the same kind of isolated Cu2+ species
with g// =2.335, illustrating no influence of preparation meth-
ods on the Cu2+ location. Furthermore, the quantified isolated
Cu2+ contents are shown in Figure 8 by deconvolution of the
EPR spectra. Relative to the Cu2+ contents from TPR results,
the two characterization methods display the same sequence:
Figure 5. H2–TPR results of a) impregnated and b) ion-exchanged Cu/SAPO-
34 catalysts.
Figure 6. Isolated Cu2+ and CuO contents in a) impregnated and b) ion-ex-
changed Cu/SAPO-34 catalysts from H2–TPR results in Figure 4.
Figure 7. EPR spectra of a) impregnated and b) ion-exchanged Cu/SAPO-34
catalysts. Prior to the tests, the Cu/SAPO-34 catalysts were dehydrated at
500 8C for 1 h.
XRD results in Figure 1a. For the three ion-exchanged Cu/
SAPO-34 samples, the H2 consumption profiles from 200 to
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
5
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Figure 8. Correlation of isolated Cu2+ in a) impregnated and b) ion-ex-
changed Cu/SAPO-34 catalysts by EPR and H2–TPR results.
(a) IM-Cu4 sample has the highest Cu2+ content within the im-
pregnated Cu/SAPO-34 samples; (b) the Cu2+ contents rise
with the increment of Cu loading in ion-exchanged samples.
Both Cu2+ and CuO species cannot interact with CO to gen-
erate stable carbonyls, but they can be adsorbed by Cu+ as
carbonyls at ambient temperature.[32,33] Therefore, CO-DRIFTS is
considered to be an effective method to quantify the Cu+ spe-
cies in Cu/SAPO-34 samples. As the Cu+ species could not be
completely identified by the H2–TPR results, CO-DRIFTS at am-
bient temperature is used to probe the Cu+ species in the two
series of Cu/SAPO-34 samples. As seen in Figure 9, the Cu+
Figure 9. CO adsorption over a) impregnated and b) ion-exchanged Cu/
SAPO-34 catalysts.
ment.[19,21] To confirm the same state but various sizes of CuO
species by different preparation methods, the IM-Cu4 sample
was chosen herein to be hydrothermally treated at 7508C for
12 h. The XRD results shown in the Supporting Information,
Figure S1, reveal that the CuO peaks at 35.5 and 38.78 disap-
pear in aged IM-Cu4 sample, and the SEM results in Figure S2
clearly demonstrate diminishing of the CuO scrap on the aged
sample. These results imply that the CuO species can also exist
in the internal of Cu/SAPO-34 sample, and this migration and
its influence on NH3 oxidation will be discussed later. Shortly,
the CuO species in impregnated Cu/SAPO-34 samples is locat-
ed on the external surface of the samples, whereas the CuO
species of ion-exchanged Cu/SAPO-34 samples is situated in
the internal.
species in all Cu/SAPO-34 samples exhibits
a peak at
2133 cmꢀ1, assigned to CO adsorbed onto Cu+ species with
the coexistence of H2O as Cu+(H2O)(CO).[32,33] These peak inten-
sities are also displayed in Figure 9 to compare the Cu+ con-
tents. It is seen that the Cu+ contents in impregnated Cu/
SAPO-34 samples follow the sequence: IM-Cu4>IM-Cu2>IM-
Cu8>IM-Cu1, whereas the Cu+ contents in ion-exchanged Cu/
SAPO-34 samples increase with the Cu loading.
Actives sites of NH3 oxidation
By combining the above quantification results on the Cu
species in Cu/SAPO-34 samples, it is concluded that the prepa-
ration methods mainly affect the distribution of Cu species.
The impregnation method generates more CuO species, which
are mainly located on the external of Cu/SAPO-34 samples,
whereas the ion-exchange method generates much less CuO
than isolated Cu2+ species. It is clear from XRD results that the
CuO species in ion-exchanged samples disperses well, and it
locates in the frameworks of samples. It is known from our pre-
vious studies that the CuO cluster can migrate into the frame-
work of Cu/SAPO-34 samples through hydrothermal treat-
The NH3–SCR and NH3 oxidation results in Figure 3 and 4
reveal that the NH3 oxidation is the main reason to decrease
the NO conversion at high temperature. It is already proven
that the isolated Cu2+ species are the active sites of NH3–SCR
reaction over Cu/SAPO-34 catalyst.[18,19,22] To finally explain
their competition relation, firstly the active site of NH3 oxida-
tion has to be found.
According to the above results on Cu species, there are
three kinds of Cu species existing in Cu/SAPO-34 samples: the
isolated Cu2+, CuO, and Cu+ species, and the preparation
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
6
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
methods affect their distribution. In the four impregnated Cu/
SAPO-34 samples, the CuO species takes up the biggest por-
tion, and it increases with the Cu loading. The increase of Cu
loading also causes the growth of CuO size on the external of
samples. In addition, the content of the Cu2+ and Cu+ species
is far less than that of CuO in impregnated Cu/SAPO-34 ac-
cording to the TPR results in Figure 5. Both species present no
clear sequence in abundance among these samples, and IM-
Cu4 contains the most of Cu2+ and Cu+ species. For the three
ion-exchanged Cu/SAPO-34 samples, the content of the isolat-
the quantification results of Cu species (see above), only the
CuO species is fit for this requirement. Furthermore, calculation
of NH3 oxidation turnover frequency (TOF) is the best way to
prove the active site of NH3 oxidation, based on the CuO con-
tents shown in Figure 6. However, the CuO species in different
impregnated Cu/SAPO-34 samples present diverse particle
sizes, which indicates the active CuO site is not equal to the
actual CuO content shown in Figure 6. The TOFs, calculated
from the CuO content in impregnated Cu/SAPO-34 as shown
in Figure S3a, decrease with the increment of CuO content as
a result of its increasing particle size. Therefore, only the CuO
species is considered to be the active site of NH3 oxidation
over impregnated Cu/SAPO-34 samples. For the ion-exchanged
Cu/SAPO-34 samples in Figure 10b, there are also three paral-
lel lines of NH3 conversion rates, and the calculated Ea values
are 93.35 kJmolꢀ1. The SEM results present there are no CuO
species on the external of ion-exchanged Cu/SAPO-34 samples,
and the CuO species in them are on the nanoscale owing to
the smaller limiting pore apertures of SAPO-34 molecular
sieves. The TOF results in Figure S3b prove the CuO species is
the active site for NH3 oxidation for ion-exchanged Cu/SAPO-
34 samples because the CuO species is in the same state with
nanoscale sizes in the framework. In addition, it is noteworthy
that the hydrothermally treated IM-Cu4 sample, compared
with the fresh one, proves the migration of CuO species, and
their H2–TPR results in Figure S4 present the same state of
CuO species at 3008C in both samples. The same Ea values also
demonstrate the same NH3 oxidation mechanism in Figure S5.
Consequently, the active site for NH3 oxidation is the CuO spe-
cies over Cu/SAPO-34 samples, and the different preparation
methods could not affect the NH3 oxidation mechanism.
ed Cu2+ species is much higher, and the contents of the Cu2+
CuO, and Cu+ species all rise with the Cu loading.
,
The kinetic NH3 oxidation was performed between 380 and
4608C (Figure 10), at which temperatures the NH3 conversion
Influence of NH3 oxidation on NH3–SCR at high temperature
It is clear from Figure 4 that the main difference between the
impregnated and ion-exchanged Cu/SAPO-34 samples is the
NO production. The impregnated Cu/SAPO-34 samples pro-
duce much more NO than the ion-exchanged ones. However,
the main composition difference over the two series of Cu/
SAPO-34 samples is the Cu2+/CuO ratio as listed in Table 1.
With the decrease of Cu2+/CuO ratios in samples, the more NO
is generated in Figure 4. Therefore, it is predicted that the NH3
oxidation proceeds in two steps: (a) the reactant NH3 is oxi-
dized to NO over CuO sites; (b) the produced NO is consumed
over Cu2+ sites as standard NH3–SCR reaction. In Figure 3, the
impregnated Cu/SAPO-34 samples exhibit the following se-
quence of NO conversion: IM-Cu1>IM-Cu4>IM-Cu8>IM-Cu2,
presenting the same sequence in their Cu2+/CuO ratios in
Table 1. Two problems should be pointed out: (a) the NO
yields are less than the NH3 conversion in Figure 4a and 4b,
and this is caused by the coexistence of Cu2+ species in im-
pregnated Cu/SAPO-34 samples. In previous studies we found
that it was difficult to prepare the Cu/SAPO-34 catalysts only
with one kind of Cu species, which is consistent with the re-
sults by others.[18,21,23,27] The Cu2+ species are the SCR active
sites and reduce the NO by NH3 to N2 as an NH3–SCR reaction.
(b) In our previous studies, the SCR reaction rates were higher
Figure 10. Kinetic results of NH3 oxidation over a) impregnated and b) ion-
exchanged Cu/SAPO-34 catalysts. The inlets consisted of 500 ppm NH3, 5%
O2 with N2 as the balance. The kinetic steady-state measurements were ob-
tained at 380, 400, 420, and 4358C for impregnated samples and 420, 440,
and 4608C for ion-exchanged samples.
is less than 20%. According to Figure 10a, the NH3 conversion
rates rise with the increment of Cu loading, explaining the NH3
conversion sequence shown in Figure 4. Moreover, the four
lines have an equivalent slope, indicating the same NH3 oxida-
tion mechanism. The apparent activation energy (Ea) for the
NH3 oxidation, calculated from the slope in Figure 10a, is
93.35 kJmolꢀ1. In addition, the active-site contents for NH3 oxi-
dation should also increase with the increment of Cu loading
for the impregnated Cu/SAPO-34 samples, consistent with the
sequence of NH3 conversion rates in Figure 10a. Combining
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
7
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
than the NH3 oxidation reaction rates herein, and the Ea value
for the former one (37.49 kJmolꢀ1) is far less than the value ob-
tained herein (93.35 kJmolꢀ1).[26] Consequently, though the NH3
oxidation consists of two steps, the kinetic results and the Ea
value from Figure 10 just reflect the mechanism of step (a).
Compared with the NO consumption in step (a), the NH3 oxida-
tion to NO is the rate-determining step. Moreover, in our previ-
ous study[26] we found the adsorbed NH3 species on Cu2+ sites
performed superior SCR activity, whereas the CuO almost pre-
sented no acidity and could not adsorb NH3 species. And the
DRIFTS results confirmed the much faster SCR rate than NH3
oxidation over Cu/SAPO-34 sample. For the ion-exchanged Cu/
SAPO-34 samples, they perform superior SCR activity and less
NO production at high temperature, which is mainly related to
its higher Cu2+/CuO ratio (see Table 1). Thus, the NO genera-
tion decrease with the rise of Cu loading, and enough Cu2+
active sites ensure the consumption of generated NO from
step (a). From this, the NH3–SCR and NH3 oxidation results can
be explained by their active sites and competition relation as
follows. At first, the impregnated Cu/SAPO-34 samples produce
more CuO species and less Cu2+ species, whereas the ion-ex-
changed Cu/SAPO-34 samples contain more Cu2+ species than
CuO species. At low temperatures, the NH3 oxidation displays
no activity, and the NH3–SCR activity is related to the contents
of Cu2+ species. The NO conversions in Figure 3a and 3c
reveal the same sequence of Cu2+ contents as in Figure 8. At
high temperatures, the NH3 oxidation has an increasing activity
with the increment of temperature, and the CuO species is the
active site. The NH3 is firstly oxidized to NO as shown in
Scheme 1. For the impregnated samples, the CuO contents are
cies on Cu2+ sites. And the appropriate acidity of SAPO-34 and
the Cu2+/CuO ratios in ion-exchanged Cu/SAPO-34 samples
ensure the excellent SCR activity and N2 selectivity at the
whole temperature range. However, the NO conversion still
begin to fall from 4008C in Figure 3c, proving the influence of
NH3 oxidation.
Conclusions
Two kinds of Cu/SAPO-34 catalysts, prepared by the impregna-
tion method and ion-exchanging method, respectively, were
applied to study their NH3 oxidation mechanism and its influ-
ence on the selective catalytic reduction of NOx by NH3 (NH3–
SCR). It was found that impregnated Cu/SAPO-34 samples con-
tain more CuO species than Cu2+ and Cu+ species. In addition,
the CuO is mainly located on the external of the molecular
sieves. The ion-exchanged Cu/SAPO-34 samples contain more
Cu2+ species, and their content increases with the Cu loading
until 1.5 wt%.
CuO species is the NH3 oxidation active site on Cu/SAPO-34
samples. The apparent activation energy (Ea) for NH3 oxidation
is 93.35 kJmolꢀ1, and the different preparation methods do not
affect the Ea value and the mechanism of NH3 oxidation. The
NH3 oxidation consists of two steps over Cu/SAPO-34 catalyst:
(a) the reactant NH3 is oxidized by O2 to produce NO on the
CuO site; (b) the generated NO is further reduced by the ad-
sorbed NH3 on Cu2+ sites as the SCR reaction. Furthermore,
the NH3 oxidation mainly decreases the NH3–SCR activity at
high temperature range for NH3 consumption competition. At
low temperatures, the NO conversion is related to the Cu2+
contents, and the Cu2+/CuO ratio is responsible for the NO
conversion at high temperatures.
Experimental Section
Preparation of Cu/SAPO-34 Catalysts
The SAPO-34 was synthesized with the mole composition of 0.2
morpholine (MA): 0.1Al2O3: 0.1P2O5: 0.1SiO2: 6.5H2O by the hydro-
thermal method, and the MA was used as a structure-directing
agent. The sources of Si, P, and Al were silica sol, 85% phosphoric
acid, and pseudoboehmite, respectively. During the synthesis pro-
cess, the phosphoric acid and the pseudoboehmite were firstly
mixed with H2O, and fiercely stirred for 1 h. Then the silica sol and
MA were added and blended until smooth. The homogeneous
mixture was sealed in an autoclave and heated at 2008C for 24 h
for crystallization, and the as-synthesized sample was obtained
through centrifuging, washing, and drying at 1008C for 6 h. Finally,
the sample was treated at 6508C for 6 h in air to remove the tem-
plate. The compositions of SAPO-34 were tested by X-ray fluores-
cence spectrometry (BRUKER, S4Pioneer) as Si0.124P0.407Al0.469O2.
Scheme 1. NH3 oxidation mechanism over Cu/SAPO-34 catalysts.
higher than that of Cu2+ species. The excessively produced NO
could not be eliminated timely through NH3–SCR on the lower
Cu2+ species, and the NO is the main byproduct during the
NH3 oxidation process (Figure 4a and 4b). During the NH3–SCR
process (Figure 3a and 3b), NH3 oxidation is dominant at high
temperature, and the competition between NH3 consumption
and NO production make the impregnated samples have infe-
rior SCR activity. For the ion-exchanged samples, the Cu2+ spe-
cies is present at higher contents than the CuO species. If the
NO is produced on CuO sites as in step (a), it can be complete-
ly consumed by the adsorbed NH3 species on the Cu2+ sites in
the presence of O2. The mechanism in Scheme 1 illustrates the
lower NO production in Figure 4d over ion-exchanged sam-
ples. During the NH3–SCR process (Figure 3c and 3d), the
NH3–SCR reaction is dominant at high temperature, and the
produced NO can be mostly reacted by the adsorbed NH3 spe-
Two series of Cu/SAPO-34 catalysts were prepared by the ion-ex-
changing method and impregnation method. For the ion-ex-
changed samples, the NH4-SAPO-34 was firstly obtained by ex-
changing H/SAPO-34 in ammonium nitrate solution at 808C. Then,
the NH4-SAPO-34 was stirred with copper acetate solution to
obtain Cu/SAPO-34. After each ion-exchange processes, the slurry
was filtered, washed and the solid was dried at 908C for 16 h. For
the impregnated samples, copper acetate solution was gradually
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
8
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
instilled and evenly stirred with NH4-SAPO-34 support, and the ob-
tained mixture was dried at 908C overnight. All the Cu/SAPO-34
samples were finally calcined at 5508C for 4 h in muffle furnace.
The Cu loading of Cu/SAPO-34 catalysts was tested by inductively
coupled plasma analysis with the results shown in Table 1. The im-
pregnated and ion-exchanged Cu/SAPO-34 catalysts were named
as “IM-Cu loading” and “EX-Cu loading” for short.
NH3 oxidation was performed in the same reactor as the NH3–SCR
tests, as well as the packing and the pretreatment. The inlet gases
contained: 500 ppm NH3 and 5% O2. The gas flow and the volume
hourly space velocity in all experiments were controlled at
1000 mLminꢀ1 and 600000 hꢀ1, respectively. The NH3 conversion
was calculated from Equation (2):
NH3 inlet ꢀ NH3 outlet
ð2Þ
NH3 conversion ½%ꢁ ¼
ꢂ 100½%ꢁ
NH3 inlet
Characterization
The XRD patterns were performed on an X’Pert Pro diffractometer
operating at 40 kV and 40 mA with nickel-filtered CuKa radiation
(l=1.5418 ꢁ) in the range 58<2q<508 with a step size of 0.028.
The BET surface area (m2 gꢀ1) was determined from the line portion
of the BET plot by measuring the N2 isotherm of the samples at
77 K using F-Sorb 3400 volumetric adsorption/desorption appara-
tus. Prior to the measurement, the samples were degassed at
1508C under vacuum for 3 h. SEM images of the samples were ob-
tained with a HITACHI S4800 field emission microscope. The sam-
ples were pasted on a sample holder using a carbon tape and
then covered with Pt film to become conductive. SEM images
were taken at magnifications of 3000ꢂ with the 3 kV electron
beam.
Kinetic tests of NH3 oxidation
The NH3 oxidation kinetic experiments were performed in a thin
quartz tube by using the 20 mg portions of the catalyst and quartz
sand (100 mg). A relatively small particles size (80–100 mesh) and
the volume hourly space velocity (3600000 hꢀ1) as in the previous
work were chosen to perform kinetic tests over Cu/SAPO-34 sam-
ples.[18,19,26] Prior to the experiments, the samples were pretreated
in 5% O2/N2 at 5008C for 30 min. The inlets consisted of 500 ppm
NH3 and 5% O2 with N2 as the balance. The kinetic steady-state
measurements were obtained from 380 to 4358C, during which
the NH3 conversions were less than 20%. The NH3 oxidation rates
can be calculated from the NH3 conversion:
H2–TPR experiments were performed in a U-shaped tubular quartz
reactor. The samples (100 mg) were firstly pretreated with O2 at
5008C for 30 min. Then, the temperature was elevated to 9008C at
a ramping rate of 58Cminꢀ1 under an H2 flow of 15 mLminꢀ1. The
consumption of hydrogen was monitored by a thermal conductivi-
ty detector. EPR spectra were recorded on a Bruker ESP320 spec-
trometer at RT and atmospheric pressure. Prior to the EPR experi-
ments, the Cu/SAPO-34 catalysts were dehydrated for 1 h at
5008C. The CO-DRIFTS measurements were conducted on an FTIR
spectrometer (Nicolet 6700) by using a heatable environmental re-
action cell with ZnSe windows, which was connected to a gas-
dosing system. Prior to the experiments, the samples were pre-
treated at 5008C for 1 h by O2. The CO adsorption was performed
at 208C and a N2 purge followed to eliminate the physically ad-
sorbed CO. The spectra were recorded with a resolution of 2 cmꢀ1
until the intensity was stable.
XNO ½%ꢁ ꢂ FNO ½LðNO Þ ꢃ minꢀ1
ꢁ
x
x
x
ð3Þ
rate ¼
mcata ꢂ 60½s ꢃ minꢀ1ꢁ ꢂ 22:4½L ꢃ molꢀ1
ꢁ
with the units of rate as [mol NOx·gcata·sꢀ1] and with XNH =NH3 con-
3
version (in %) and FNH =flow rate of NH3 (in LNH minꢀ1
)
3
3
Acknowledgements
The authors are grateful to the financial support from the GM
Global Research & Development (RD-07-312-NV487). This work
was supported by the program of the National High Technology
Research and Development Program of China (863 Program,
2011AA03A405).
Keywords: copper · kinetics · nitrogen oxides · reduction ·
zeolites
NH3–SCR and NH3 oxidation activity
NH3–SCR activities were measured in a quartz reactor by using
0.1 g portions of the sample (60–80 mesh) with quartz (0.9 g, 60–
80 mesh) at atmospheric pressure. The sample was sealed in the
tube with quartz wool. The temperature was controlled by
a type K thermocouple inserted into the center of the mixed
sample. The FTIR spectrometer (MKS-2030) equipped with a 5.11 m
gas cell was used to measure the concentrations of NO, NO2, N2O,
and NH3. The gas flow and the volume hourly space velocity in all
experiments were controlled at 1000 mLminꢀ1 and 600000 hꢀ1, re-
spectively. Prior to the experiment, the sample was pretreated by
O2 at 5008C for 30 min. The inlets gases contained 500 ppm NO,
500 ppm NH3 and 5% O2 without H2O and CO2, because in our pre-
vious studies we found the presence of H2O and CO2 in the inlets
could not decrease the NO conversions.[21,26] The test temperature
was elevated from 120 to 6008C. The NOx conversion was calculat-
ed from Equation (1):
[1] M. V. Twigg, Appl. Catal. B 2007, 70, 2–15.
[2] P. Forzatti, Catal. Today 2000, 62, 51–65.
[3] R. Q. Long, R. T. Yang, J. Catal. 1999, 186, 254–268.
[4] P. Forzatti, Appl. Catal. A 2001, 222, 221–236.
[5] S. Sato, Y. Yu-u, H. Yahiro, N. Mizuno, M. Iwamoto, Appl. Catal. 1991, 70,
L1L5.
[6] J. Pꢃrez-Ramꢄrez, J. C. Groen, A. Brꢅckner, M. S. Kumar, U. Bentrup, M. N.
Debbagh, L. A. Villaescusa, J. Catal. 2005, 232, 318–334.
[7] R. Q. Long, R. T. Yang, J. Catal. 1999, 188, 332–339.
[8] X. Feng, W. K. Hall, J. Catal. 1997, 166, 368–376.
[9] A. Subbiah, B. K. Cho, R. J. Blint, A. Gujar, G. L. Price, J. E. Yie, Appl. Catal.
B 2003, 42, 155–178.
[10] J. Y. Yan, G. D. Lei, W. M. H. Sachtler, H. H. Kung, J. Catal. 1996, 161, 43–
54.
[11] J. Y. Yan, W. M. H. Sachtler, H. H. Kung, Catal. Today 1997, 33, 279–290.
[12] M. J. Rokosz, A. V. Kucherov, H. W. Jen, M. Shelef, Catal. Today 1997, 35,
65–73.
[13] C. He, Y. Wang, Y. Cheng, C. K. Lambert, R. T. Yang, Appl. Catal. A 2009,
368, 121–126.
NOx inlet ꢀ NOx outlet
[14] M. Briend, M. J. Peltre, P. Massiani, P. P. Man, R. Vomscheid, M. Derewin-
ski, D. Barthomeuf, Stud. Surf. Sci. Catal. 1994, 84, 613–620.
ð1Þ
NOx conversion ½%ꢁ ¼
ꢂ 100½%ꢁ
NOx inlet
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
9
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
[15] I. Bull, W.-M. Xue, P. Burk, R. S. Boorse, W. M. Jaglowski, G. S. Koermer, A.
Moini, J. A. Patchett, J. C. Dettling, M. T. Caudle, US Patent 7,610,662,
2009.
[25] U. Deka, I. Lezcano-Gonzalez, S. J. Warrender, A. Lorena Picone, P. A.
Wright, B. M. Weckhuysen, A. M. Beale, Microporous Mesoporous Mater.
2013, 166, 144–152.
[26] T. Yu, J. Wang, M. Shen, W. Li, Cata. Sci. Technol. 2013, 3, 3234–3241.
[27] L. Wang, J. R. Gaudet, W. Li, D. Weng, J. Catal. 2013, 306, 68–77.
[28] M. Richter, M. J. G. Fait, R. Eckelt, E. Schreier, M. Schneider, M. M. Pohl, R.
Fricke, Appl. Catal. B 2007, 73, 269–281.
[16] P. J. Andersen, J. E. Bailie, J. L. Casci, H.-Y. Chen, J. M. Fedeyko, R. K. S.
Foo, R. R. Rajaram, US Patent 2010/0290963, 2010.
[17] D. W. Fickel, E. D’Addio, J. A. Lauterbach, R. F. Lobo, Appl. Catal. B 2011,
102, 441–448.
[29] C. Torre-Abreu, M. F. Ribeiro, C. Henriques, G. Delahay, Appl. Catal. B
1997, 12, 249–262.
[18] J. Xue, X. Wang, G. Qi, J. Wang, M. Shen, W. Li, J. Catal. 2013, 297, 56–
64.
[30] C. Torre-Abreu, M. F. Ribeiro, C. Henriques, G. Delahay, Appl. Catal. B
1997, 14, 261–272.
[19] S. Fan, J. Xue, T. Yu, D. Fan, T. Hao, M. Shen, W. Li, Catal. Sci. Technol.
2013, 3, 2357–2364.
[31] M. Richter, M. J. G. Fait, R. Eckelt, M. Schneider, J. Radnik, D. Heidemann,
R. Fricke, J. Catal. 2007, 245, 11–24.
[32] K. Hadjiivanov, L. Dimitrov, Microporous Mesoporous Mater. 1999, 27,
49–56.
[33] K. I. Hadjiivanov, M. M. Kantcheva, D. G. Klissurski, J. Chem. Soc. Faraday
Trans. 1996, 92, 4595–4600.
[20] L. Wang, W. Li, G. Qi, D. Weng, J. Catal. 2012, 289, 21–29.
[21] J. Wang, T. Yu, X. Wang, G. Qi, J. Xue, M. Shen, W. Li, Appl. Catal. B 2012,
127, 137–147.
[22] S. T. Korhonen, D. W. Fickel, R. F. Lobo, B. M. Weckhuysen, A. M. Beale,
Chem. Commun. 2011, 47, 800–802.
[23] F. Gao, E. D. Walter, N. M. Washton, J. Szanyi, C. H. F. Peden, ACS Catal.
2013, 3, 2083–2093.
[24] R. Martꢄnez-Franco, M. Moliner, C. Franch, A. Kustov, A. Corma, Appl.
Catal. B 2012, 127, 273–280.
Received: February 15, 2014
Published online on && &&, 0000
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&10
&
ÞÞ
These are not the final page numbers!

FULL PAPERS
T. Yu, J. Wang, Y. Huang, M. Shen,* W. Li,
J. Wang*
&& – &&
NH3 Oxidation Mechanism over Cu/
SAPO-34 Catalysts Prepared by
Different Methods
The copper makes the difference: The
mechanism of NH3 oxidation on the se-
lective catalytic reduction of NOx by NH3
over Cu/SAPO-34 catalysts prepared by
different methods is studied. CuO is
identified as the active site for the reac-
tion, and the different NO conversions
at high temperatures are related to the
different Cu2+/CuO ratios in the two
Cu/SAPO-34 catalyst series.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&11
&
ÞÞ
These are not the final page numbers!