Ozone catalytic oxidation for ammonia removal from simulated air at room temperature

Article abstract of DOI:10.1039/c4cy01269k

Ozone catalytic oxidation (OZCO) for removing ammonia from simulated air over the AgMn/HZSM-5 (AgMn/HZ) catalyst with high ammonia conversion and high N₂ selectivity at room temperature is reported for the first time. HZ, Ag/HZ, Mn/HZ and AgMn/

Full text of DOI:10.1039/c4cy01269k

View Article Online
View Journal
Catalysis
Sc ience &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: Y. Liu, X. Li, J. Liu,
C. shi, X. Zhu, A. Zhu and B. W.-L. Jang, Catal. Sci. Technol., 2015, DOI: 10.1039/C4CY01269K.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/catalysis

Page 1 of 41
Catalysis Science & Technology
1
View Article Online
DOI: 10.1039/C4CY01269K
1
2
Ozone catalytic oxidation for ammonia removal from simulated air at
room temperature
a
a
a
a
a,
a
3
4
Yang Liu , Xiao-Song Li , Jing-Lin Liu , Chuan Shi , Xiaobing Zhu *, Ai-Min Zhu ,
b,
Ben W.-L. Jang *
a
5
6
7
8
Laboratory of Plasma Physical Chemistry, Dalian University of Technology, 116024 Dalian,
China
b
Department of Chemistry, Texas A&M University-Commerce, PO Box 3011, Commerce, TX
75429, United States
9
0
1
---------------------
1
1
* Corresponding authors. Fax: +86-411-84706094 (X. Zhu), +1-903-468-6020 (B.W.-L. Jang).
E-mail addresses: xzhu@dlut.edu.cn (X. Zhu), Ben.Jang@tamuc.edu (B.W.-L. Jang).
1
1
1
1
1
1
1
1
2
2
2
2
2
2
3
4
5
6
7
8
9
0
1
2
3
4
ABSTRACT
Ozone catalytic oxidation (OZCO) for removing ammonia from simulated air
over AgMn/HZSM-5 (AgMn/HZ) catalyst with high ammonia conversion and high N2
selectivity at room temperature is reported for the first time. HZ, Ag/HZ, Mn/HZ and
AgMn/HZ catalysts were compared in the OZCO reactions of gaseous and adsorbed
NH . In OZCO of gaseous NH , N was the major product and N O was the minor
3
3
2
2
product. NH conversion dropped quickly with time-on-stream (TOS) over HZ and
3
Ag/HZ catalysts while it remained almost constant at a high level over Mn/HZ and
AgMn/HZ catalysts during the entire test. N selectivity of AgMn/HZ catalyst was
2
higher than that of Mn/HZ catalyst. When the initial concentration of NH was 521
3
ppmv and the ratio of initial concentration of O to NH was 1.73, 99% of NH
3
3
3
conversion with 94% of N selectivity was obtained over AgMn/HZ catalyst at room
2

Catalysis Science & Technology
Page 2 of 41
2
View Article Online
.
-1. -1
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
temperature and 150,000 ml g h of gas hourly space velocity (GHSV). Finally, the
pathways for OZCO of NH were proposed for the four catalysts based on the
3
comparative investigation of the gaseous products and surface species during OZCO
of adsorbed and gaseous NH3.
Keywords: Catalytic oxidation; Ozone; Ammonia removal; AgMn/HZSM-5 catalyst.

Page 3 of 41
Catalysis Science & Technology
3
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
1. Introduction
Ammonia, which is a toxic, corrosive and reactive gas with an irritating odor, is
considered as a highly hazardous pollutant and of great importance in PM 2.5
1
formation and acid rain deposition. Due to the increasing environmental concerns,
NH removal from air is becoming a serious issue. Selective catalytic oxidation (SCO)
3
2
-12
of NH to N is an ideal technology for NH abatement.
Several catalysts have
3
2
3
been proposed for SCO of NH , however, they usually require high temperatures and,
3
therefore, are energy-consuming. Typically, Pt or PtO based catalysts, supported on
γ-Al O or HZSM-5 (HZ) zeolites, required temperatures of 200-250 °C to achieve
2
3
‡
1
1
1
1
1
1
1
1
1
1
2
2
2
100% of NH conversion using O (gas hourly space velocity (GHSV ) = 60,000
3 2
.
-1. -1
2
5
ml g h with 1000 ppmv of NH ). Long et al. have prepared several Fe based
3
catalysts, such as Fe O -TiO , Fe O -ZrO , Fe O -Al O and Fe O -SiO , for SCO of
2
3
2
2
3
2
2
3
2
3
2
3
2
NH and found that 95-99% of NH conversion could be obtained at temperatures as
3
3
.
-1. -1
10
high as 500 °C (GHSV = 150,000 ml g h with 1000 ppmv of NH ). Lippits et al.
3
1
2
and Liang et al. reported that, over Cu/γ-Al O catalysts, 100% of NH conversion
2
3
3
.
-1. -1
was achieved at temperatures exceeding 300 °C (GHSV = 12,000 ml g h with
1
0
.
-1. -1
12
20,000 ppmv of NH3 and 30,000 ml g h with 1000 ppmv of NH , respectively).
3
O , which is a strong oxidant, has attracted extensive attention and is now being
3
widely investigated in the catalytic removal of toxic substances from polluted gases
because its performance in such reactions is superior to that of O . On the basis of the
2
1
3-16
results from Mehandjiev and co-authors,
using O instead of O permitted the
3 2
reaction to proceed in a low-temperature range and achieved a high catalytic activity.

Catalysis Science & Technology
Page 4 of 41
4
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
The data obtained using manganese oxide as the catalyst demonstrated that the
activation energy in catalytic benzene oxidation using O was three times lower than
3
1
3
13,17-26
that using O2. O has been frequently used for the removal of benzene,
3
2
7-33
34,35
36,37
14,16,38-40
toluene,
chlorobenzene,
formaldehyde
and carbon monoxide,
etc.
However, to the best of our knowledge, very few studies focused on the ozone
1
catalytic oxidation (OZCO) of NH . One is reported by Kastner and co-authors, who
3
investigated OZCO of NH at room temperature using wood fly ash and biomass char
3
as catalysts. However, the best NH conversion achieved was around 60% at
3
.
-1. -1
relatively low GHSV of 76,200 ml g h and low inlet NH concentration of about 11
3
41
1
1
1
1
1
1
1
1
1
1
2
2
2
ppmv. Another is reported by Ichikawa et al., but it is about aqueous-phase reaction

of NH4 with O3.
1
3,17,25,27,42
Manganese oxide catalysts are highly active for ozone decomposition
6
and the supported Ag catalysts were reported to be active for NH oxidation. HZ
3
zeolites are often used as catalyst supports in NH oxidation processes, since the acid
3
sites in HZ can enhance the adsorption and activation of NH molecules and the pore
3
4,7
structure of HZ favors NH oxidation to N . Herein, the investigation on OZCO of
3
2
NH over the well-defined HZ supported Ag and/or Mn catalysts is reported in this
3
study. To clarify the reaction mechanisms for OZCO of NH , OZCO of adsorbed NH
3
3
is also examined comparatively over the same catalysts.
2. Experimental
2.1. Catalyst preparation

Page 5 of 41
Catalysis Science & Technology
5
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
HZ-supported catalysts were prepared via incipient wetness impregnation
‡
2. -1
method using HZ (SiO /Al O = 360, BET surface area = 341 m g , Catalyst Plant
2
2
3
.
2
of Nankai University, China) as the support. AgNO and Mn(CH COO) 4H O were
3
3
2
used as Ag and Mn precursors, respectively. The impregnated samples were aged
overnight at ambient temperature in darkness followed by drying at 110 °C for 6 h and
calcination in static air in an oven at 450 °C for 3 h. The obtained powders were
grounded, tableted, crushed and sieved to 20-40 mesh.
2.2. ICP & BET characterization and the chemical states of Ag and Mn species
Inductively coupled plasma-atomic emission spectroscopy (ICP-AES, optima
2000DV, USA) was used to determine Ag and Mn loadings. The Ag loading of Ag/HZ
catalyst was 0.7 wt% and the Mn loading of Mn/HZ catalyst was 2.3 wt%. The Ag
and Mn loadings of AgMn/HZ catalyst were 0.8 wt% and 2.4 wt%, respectively. The
reason why the Ag loading of 0.8 wt% and Mn loading of 2.4 wt% were chosen for
1
1
1
1
1
1
1
1
1
1
2
2
2
2
6
AgMn/HZ catalyst is based on our previous study, which showed such AgMn/HZ
catalyst possessing the best activity and stability in benzene oxidation by O3.
The specific surface areas of the supported catalysts were measured by the N2
adsorption BET method at -196 °C (Autosorb-1, Quantachrome, USA). The BET
2. -1
surface areas were 341, 334 and 336 m g for Ag/HZ, Mn/HZ and AgMn/HZ
catalysts, respectively.
25
26
As revealed by the XPS and UV-vis Diffuse Reflectance Spectra of the
catalysts, Ag was present in metallic state over Ag/HZ and AgMn/HZ catalysts and
3
+
4+
Mn existed as a mixture of Mn and Mn over Mn/HZ and AgMn/HZ catalysts after

Catalysis Science & Technology
Page 6 of 41
6
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
calcination.
2.3. Quantification methods of the concentrations of NH , N O, O , NO, NO and
3
2
3
2
N2
For the quantification of NH , N O and O , the standard curves of “Strongest
3
2
3
FT-IR peak height versus Concentration” were established on a Fourier
transform-infrared (FT-IR) spectrometer (Nicolet-Antaris IGS Analyzer, Thermo,
USA). For the standard curves of NH and N O, their concentrations were given by
3
2
their standard gases. For O standard curve, due to O standard gas being unavailable,
3
3
O concentrations were calibrated by the iodometric titration method. The strongest
3
-
1
-1
-1
1
1
1
1
1
1
1
1
1
1
2
2
2
peak heights at 1000-900 cm , 2300-2150 cm and 2150-2000 cm were used for the
quantification of NH , N O and O , respectively.
3
2
3
-1
-1
However, for NO (2000-1800 cm ) and NO (1700-1500 cm ), their
2
concentrations could not be precisely evaluated by the FT-IR spectrometer alone due
-1
to the superposition of H O (2000-1200 cm ). Therefore, NO and NO were
2
2
quantified with mass spectrometer (HPR 20 QIC, HIDEN Analytical, England) by the
internal standard method. The product of N O was used as the internal standard
2
because its concentration could be determined simultaneously by the FT-IR
spectrometer. The specific quantification methods of NO, NO and N are described in
2
2
Supplementary Data.
2.4. OZCO of gaseous NH and investigation of the surface species after OZCO
3
process
Catalytic reactions were carried out with a fixed-bed flow reactor. A quartz

Page 7 of 41
Catalysis Science & Technology
7
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
U-tube reactor (i.d. 4 mm) with about 0.1 g of catalyst was immersed in water bath at
17 °C. Ozone was generated from pure O (purity ＞ 99.999%) by a homemade
2
ozone generator and monitored online by the FT-IR spectrometer. Before performing
the experiments, all catalysts were pretreated in a dry simulated air stream at 450 °C
for 1 h and then cooled down to room temperature in the same atmosphere. The feed
‡
gas of 250 SCCM (Standard Cubic Centimeters per Minute), containing NH , O , 20
3
3
vol.% of O and balanced by N , was flowed through the catalyst bed, which
2
2
.
-1. -1
corresponded to a GHSV of 150,000 ml g h . Unless otherwise stated, NH and O
3
3
9
0
concentrations were c.a. 530 ppmv and 450 ppmv, respectively. NH conversion
3
Conv. Conv.
1
1
(
XNH3 ), O conversion ( X O3 ), ratio of converted O to NH (
n
/n
), N O
3
3
3
2
O3
NH3
1
2
selectivity (
S
) and N selectivity ( SN2 ) were calculated as follows:
N O
2
2
in
NH3
out
 C
C
NH3
1
1
X NH3 (%) 
100%
(1)
(2)
(3)
in
NH3
C
in
O3
out
C
 CO
3
3
XO3 (%) 
100%
in
O3
C
in
O3
C
 XO3
Conv. Conv.
1
1
4
5
n
/n

O3
NH3
in
NH3
C
 XNH3
C
in
2
N O
2
S
(%)
100%
(4)
(5)
N O
2
C
 XNH3
NH3
1
1
1
1
6
7
8
9
S
(%) 1 S
N2
N O
2
in
NH3
out
where
C
and
C
are the inlet and outlet NH concentrations in the gas stream
3
NH3
in
O3
out
O3
during OZCO of gaseous NH , respectively;
C
and
C
are the inlet and outlet O3
3
concentrations in the gas stream during OZCO of NH , respectively;
C
is the outlet
3
N O
2
2
0
N O concentration in the gas stream during OZCO of NH .
2
3

Catalysis Science & Technology
Page 8 of 41
8
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
To verify the reliability of N selectivity derived from eqn. (5), additional
2
experiments using Ar as the balanced gas for direct measurement of N were carried
2
out: 0.2 g of AgMn/HZ catalyst, 500 SCCM of total flow rate with 1000 ppmv of NH3
and 500 ppmv of O , 20 vol.% of O and Ar balanced. The results are shown in Fig.
3
2
S1. The N selectivity in Fig. S1e was derived from the direct measurement according
2
to eqn. (6):
CN2 2
7
SN2 (%)
100%
(6)
in
C
 XNH3
NH3
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
whereCN2 is the outlet N concentration in the gas stream during OZCO of NH and
2 3
could be obtained from eqn. S(16). And the N selectivities derived from eqns. (5) and
2
1
1
1
1
1
1
1
1
1
1
2
2
2
(6) were 80% and c.a.75%, respectively. This confirms that the N selectivity derived
2
from eqn. (5) is reliable in this work.
After OZCO of NH , the used catalyst pellets were unloaded from the reactor,
3
ground into powders for ATR (attenuated total reflectance) measurements and
temperature programmed desorption (TPD) studies.
The ATR spectra were recorded on a Nicolet 6700 spectrometer (Thermo, USA)
with an MCT detector (Mercury Cadmium Telluride detector, which needs to be
cooled with liquid nitrogen). The catalyst powder was loaded onto the Smart ATR
accessory with diamond as the ATR material. The background spectrum was recorded
with the bare diamond and it was subtracted from the spectra of HZ, Mn/HZ, Ag/HZ
and AgMn/HZ catalysts. The beam path inside the spectrometer was purged with
purified CO -free dry air. The spectra were recorded by averaging 32 scans at a
2
-1
resolution of 4 cm .

Page 9 of 41
Catalysis Science & Technology
9
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
For the TPD experiments of the used catalysts, a helium stream was flowed
through the catalyst bed of 0.03 g at a flow rate of 100 SCCM for 1 h until the
baseline of the mass spectrometer became stable. The temperature was raised linearly
.
-1
at 10 °C min to 450 °C and held for 30 min afterwards.
2.5. OZCO of adsorbed NH3
2.5.1. Investigation of surface adsorption sites on the catalysts towards NH by
3
NH -TPD
3
NH -TPD tests were carried out with 0.1 g of catalyst under an Ar stream of 100
3
SCCM. Before the TPD measurements, the samples were pretreated in 100 SCCM of
o
1
1
1
1
1
1
1
1
1
1
2
2
2
Ar flow at 450 C for 1 h and then cooled down to room temperature in the same
atmosphere. The samples were then treated with 540 ppmv of NH for 25 min. After
3
that, the NH stream was replaced with 100 SCCM of Ar flow for 1 h to remove the
3
physisorbed NH . Subsequently, the temperature was increased in an Ar stream of 100
3
o
.
-1
SCCM to 450 C with a ramp of 10 °C min .
2.5.2. OZCO of adsorbed NH and investigation of the surface species after
3
OZCO
After pretreated in dry simulated air at 450 °C for 1h and cooled down to room
temperature, the samples were treated with 540 ppmv of NH for 25 min at room
3
temperature. Then the NH stream was replaced with 250 SCCM of He flow
3
containing 20 vol.% of O for 40 min to remove the physisorbed NH . After that, a
2
3
feed gas of 250 SCCM, containing 450 ppmv of O , 20 vol.% of O and balanced by
3
2
He, was flowed through the catalyst bed for 30 min. In order to present how much

Catalysis Science & Technology
Page 10 of 41
1
0
View Article Online
DOI: 10.1039/C4CY01269K
1
2
N O was produced during OZCO of adsorbed NH , N O / (N  N O) was defined as
2
3
2
2
2
follows:
t
F  C dt
2
N O

0
3
N O / (N  N O )（ %）
100%
(7)
2
2
2
t
t





F  C dt  C dt

N
2
N O

2

0
0
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
where F and t are the total flow rate and oxidation time during OZCO of adsorbed
NH , respectively. And the integral is the area under the curve for N or N O versus
3
2
2
time during OZCO of adsorbed NH3.
Thereafter, ATR and TPD experiments were performed over the used catalysts
under the conditions described in Section 2.4.
1
1
1
1
1
1
1
1
1
1
2
2
3. Results and discussion
3.1. OZCO of gaseous NH3
3.1.1. Comparison of OZCO of gaseous NH over HZ, Mn/HZ, Ag/HZ and
3
AgMn/HZ catalysts
4
3
Based on the results of Olszyna et al., mixtures of NH and O could react at
3
3
30 °C in gaseous phase. Therefore, a control consisting of a reactor system without
catalyst was operated at room temperature, 250 SCCM of dry simulated air with 527
ppmv of NH and 446 ppmv of O . The result shows that almost no NH conversion
3
3
3
was observed, which demonstrates that the gaseous reaction of O and NH could be
3
3
neglected in this study.
The catalytic performance of HZ and HZ-supported catalysts was evaluated in
the OZCO of gaseous NH and the results are presented in Fig. 1. Fig. 1a shows the
3

Page 11 of 41
Catalysis Science & Technology
11
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
variation of NH conversion with time-on-stream (TOS) during NH oxidation using
3 3
O . As shown in Fig. 1a, Ag/HZ and Mn/HZ catalysts possessed much higher NH
3
3
conversion than the bare HZ support did, which indicates that the addition of Ag or
MnO greatly enhanced the NH oxidation. Compared with the Mn/HZ catalyst, the
x
3
increase of NH conversion on AgMn/HZ catalyst was not significant, since Mn/HZ
3
catalyst had already possessed very high NH conversion before the addition of Ag.
3
In terms of catalyst stability, NH conversion dropped quickly with TOS over HZ
3
and Ag/HZ catalysts. However, the conversion was almost constant around 50% over
Mn/HZ and AgMn/HZ catalysts during the entire reaction. The results demonstrate
1
1
1
1
1
1
1
1
1
1
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
that the addition of MnO greatly suppressed catalyst deactivation and enhanced the
x
catalyst stability.
Additionally, as shown in Fig. 1b, O conversion remained relatively constant at
3
60% and 80% over HZ and Ag/HZ catalysts, respectively. The relatively constant
conversion of O with decreasing conversion of NH suggests that the usage of O for
3
3
3
NH oxidation decreases with time over HZ and Ag/HZ. However, for Mn/HZ and
3
AgMn/HZ catalysts, O conversion was at 100% during the entire test. This should be
3
ascribed to the very high activity of manganese oxides on O decomposition and the
3
1
3,17,25,27,42
consistent usage of O for NH oxidation throughout the test.
3
3
In all four cases, N was the main gaseous product with small amount of N O.
2
2
NO or NO was not detected during OZCO of gaseous NH . As shown in Fig. 1c, N O
2
3
2
and N selectivities of the catalysts were compared among all four catalysts. In
2
general, the N selectivities were high on all samples, exceeding 80%. However,
2

Catalysis Science & Technology
Page 12 of 41
1
2
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
0
Ag/HZ catalyst possessed the highest N selectivity of 95% while Mn/HZ catalyst was
2
the least selective to N among all catalysts. The oxidation of NH can be described as
2
3
4
4
the abstraction of hydrogen process, in which the H atoms in the NH molecules
3
were abstracted one by one to form adsorbed NH species and N atoms. Then the N
x
atoms could either be further oxidized into NO, which can react with NH to form N ,
x
2
44
or further react with NO formed and release as N O. The reason why that the N
2
2
selectivities over HZ and Ag/HZ catalysts were slightly higher than those of Mn/HZ
and AgMn/HZ catalysts (Fig. 1c) is not clear at this point. Since the N selectivity of
2
AgMn/HZ catalyst (about 87%) was higher than that of Mn/HZ catalyst (c.a. 80%)
1
(Fig. 1c), it indicates that Ag is beneficial to promote the selective oxidation of NH to
3
1
1
1
2
N during OZCO of gaseous NH .
2
3
Conv. Conv.
The
n
/n
ratio during NH oxidation is summarized in Fig. 1d. As shown
3
O3
NH3
Conv. Conv.
1
3
in Fig. 1d, the
n
/n
ratio was relatively constant around 1.7 over Mn/HZ and
O3
NH3
1
1
4
5
AgMn/HZ catalysts during the entire reaction. However, for HZ and Ag/HZ catalysts,
Conv. Conv.
their
n
/n
ratios were low at the beginning of OZCO reaction and then
O3
NH3
Conv. Conv.
1
1
6
7
increased fast with TOS, especially for HZ, the
n
/n
ratio of which arrived at
O3
NH3
about 2 at TOS = 60 min. Again, these results suggest that Mn/HZ and AgMn/HZ are
more efficient and stable in O utilization for NH oxidation.
1
1
8
9
3
3
in
O3
in
NH3
3.1.2. Effects of
C
/ C ratio over AgMn/HZ catalyst
in
O3
in
NH3
2
0
Fig. 2 shows the effects of
C
/ C ratio on the NH oxidation by O over
3
3
2
2
1
2
AgMn/HZ catalyst at TOS = 60 min. Based on the proposed O decomposition
3
4
5
mechanism by Oyama and co-workers, it is assumed that only one oxygen atom

Page 13 of 41
Catalysis Science & Technology
1
3
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
derived from O molecule is responsible for NH oxidation. The production of N and
3 3 2
N O could be written as eqns. (8) and (9), respectively:
2
1.5O  NH 0.5 N 1.5O 1.5H O
(8)
(9)
3
3
2
2
2
2O  NH 0.5 N O 2O 1.5H O
3
3
2
2
2
5
6
It can be seen from eqn. (8) that the stoichiometric ratio of O to NH is 1.5 for OZCO
3 3
in
in
NH3
of NH to N . When the
C
/ C
ratio was 0.85, which is about half of the
3
2
O3
7
8
stoichiometry, close to half of the NH feed was excessive with respect to O .
3
3
in
in
NH3
Therefore, the NH and O conversions are 51% and 100% at
C
/ C = 0.85,
3
3
O3
in
O3
in
NH3
9
respectively, as shown in Fig. 2. As the
C
/ C
ratio increased to 1.32, which is
1
0
slightly lower than the stoichiometric ratio of 1.5, the NH conversion rose to 84%
3
1
1
1
2
and nearly all the O was consumed (O conversion was 95%, as shown in Fig. 2).
3
3
in
in
NH3
Moreover, almost 100% of NH conversion could be achieved at
C
/ C = 1.73,
3
O3
1
3
which is slightly higher than the stoichiometric ratio of 1.5. In this case, O was
3
1
1
4
5
slightly excessive and the O conversion dropped to 84% (Fig. 2). In terms of N
3
2
in
in
NH3
selectivity, it was high in all three cases and increased from 87% (
C
/ C = 0.85) to
O3
in
O3
in
NH3
in
O3
in
NH3
1
1
6
7
94% (
C
/ C = 1.73), which indicates that higher
C
/ C ratio promotes selective
in
O3
in
oxidation of NH to N during OZCO process. When the
C
/ C ratio was 1.73, the
3
2
NH3
1
1
2
2
2
8
9
0
1
2
NH conversion and N selectivity were as high as 99% and 94%, respectively, over
3 2
.
-1. -1
AgMn/HZ catalyst at room temperature and GHSV of 150,000 ml g h .
3.1.3. Studies on the surface species after OZCO of gaseous NH3
Fig. 3 shows the ATR spectra of the fresh samples and the used catalysts after
OZCO of gaseous NH for 90 min. Compared with the fresh samples (Fig. 3a), new
3

Catalysis Science & Technology
Page 14 of 41
1
4
View Article Online
-
1
-1
-1
-1
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
bands at 1328 cm , 1432 cm , 3080 cm and 3254 cm were observed on all used
samples (Fig. 3b), which indicates the formation of new species after NH oxidation.
3
Nitrates are one of the frequently recognized intermediates or by-products that would
4
6,47
be formed in NH oxidation process.
But the assignment of these new bands is not
3
straightforward, since the characteristic vibrations are different in literature due to
different catalyst preparation and reaction conditions; especially for the
-1
high-frequency regions (above 3000 cm ), the assignments are rare. In order to
facilitate the assignment of the new bands on the used catalysts, the spectrum of the
NH NO -HZ, prepared by impregnation of HZ with NH NO , was also given in Fig.
4
3
4
3
1
1
1
1
1
1
1
1
1
1
2
2
2
3b for reference. The same bands as those on the used samples were also observed on
-1
-1
the NH NO -HZ sample. Hence, bands at 1328 cm and 1432 cm could be assigned
4
3
to the symmetric and asymmetric stretching vibration of N=O in monodentate
4
8
-1
-1
nitrates, respectively; bands at 3080 cm and 3254 cm are attributed to the
4
9
symmetric and asymmetric stretching of N-H, respectively. Therefore, it is
reasonable to deduce that monodentate nitrates and adsorbed ammonia ([NH ] ) were
3
a
deposited on the used catalyst surfaces after OZCO of gaseous NH3.
Moreover, TPD studies were also performed over the used catalysts to further
confirm the nature of the surface species after OZCO of gaseous NH . Fig. 4 shows
3
the TPD results of the used HZ, Ag/HZ, Mn/HZ and AgMn/HZ catalysts after 90-min
OZCO of gaseous NH . The desorption of NH was detected on all samples (Fig. S2)
3
3
and no NO was detected by the FT-IR spectrometer. The MS data are shown in Fig.
S3. As shown in Figs. S3a and S3b, no N was formed over HZ and Ag/HZ catalysts.
2

Page 15 of 41
Catalysis Science & Technology
1
5
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
Figs. 4a and 4b clearly indicate N O as the main product over HZ and Ag/HZ
2
catalysts (at 203 °C and 202 °C for HZ and Ag/HZ catalysts, respectively), together
5
0
with negligible amount of NO . Based on the results of Colombo et al. and the ATR
2
results of the used catalysts (Fig. 3b), it is tentatively deduced that N O came from the
2
5
0
decomposition of ammonia-nitrates complex. H O was not produced on HZ and
2
Ag/HZ catalysts, as shown in Fig. 4e. However, besides the evolution of N O, N was
2
2
detected by the mass spectrometer (Figs. S3c and S3d) in a considerable amount at
208 °C and 213 °C over Mn/HZ and AgMn/HZ catalysts, respectively (Figs. 4c and
50
4d). Colombo et al. also proposed that surface nitrate species ([-NO ] ) could be
3
s
1
1
1
1
1
1
1
1
1
1
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
reduced by [NH ] to produce N . Hence, it is deduced that N came from the
3 a 2 2
reduction of monodentate nitrates by [NH ] over Mn/HZ and AgMn/HZ catalysts.
3
a
However, during TPD process, N was not formed on the used HZ and Ag/HZ
2
catalysts, which indicates that MnO is necessary for the selective catalytic reduction
x
of [-NO ] by [NH ] to N . Additionally, considerable amount of H O was also
3
s
3 a
2
2
detected over Mn/HZ and AgMn/HZ catalysts at 218 °C, as shown in Fig. 4e, which is
almost at the same temperature as the evolution of N (at 208 °C and 213 °C for
2
Mn/HZ and AgMn/HZ catalysts, respectively, as shown in Figs. 4c and 4d) and is
attributed as the co-product of the reduction of [-NO ] by [NH ] to N .
3
s
3 a
2
3.2. OZCO of adsorbed NH3
In order to explore the reaction mechanisms for OZCO of gaseous NH , OZCO
3
of adsorbed NH was then comparatively performed over HZ, Ag/HZ, Mn/HZ and
3
AgMn/HZ catalysts at room temperature. The surface adsorption sites on the catalysts

Catalysis Science & Technology
Page 16 of 41
1
6
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
towards NH were first investigated by NH -TPD technique.
3
3
3.2.1 NH -TPD results
3
Fig. 5 shows the NH profiles of HZ, Ag/HZ, Mn/HZ and AgMn/HZ catalysts
3
during NH -TPD measurements. Over the HZ support, three NH desorption peaks
3
3
o
could be resolved, centered at 116, 161 and 345 C, respectively, as displayed in Fig.
5a. It is commonly recognized that there are three types of NH adsorption sites over
3
H-form zeolite. Brønsted acid (B-acid) sites are formed by the so-called “bridging
hydroxyl”, which connects the aluminum and silicon atoms in the framework of
51
zeolite; Lewis acid (L-acid) sites are usually composed of aluminum species with
5
1
1
1
1
1
1
1
1
1
1
1
2
2
2
low coordination; hydrogen-bonded-physisorption (H-bond) sites are formed by the
5
2
preceding adsorption of NH on the B-acid sites and this type of sites is non-acidic.
3
5
3
o
o
Ma et al. have assigned the peak at 150 C to the H-bonded NH , the peak at 223 C
3
o
to NH adsorbed on acid sites (L- and B- acid sites) and the peak at 428 C to NH
3
3
strongly adsorbed on B-acid sites. Similar assignments were also reported by Wang et
54
al., in which it is demonstrated that the NH desorption peaks at 150, 250 and 385
3
o
C are due to weakly adsorbed NH , NH adsorbed on L-acid sites and B-acid sites,
3
3
respectively. Additionally, two NH desorption peaks were observed by Al-Dughaither
3
5
5
and co-authors and it is indicated that the nature of the weak site was a mixture of
L-acid and H-bond sites while the strong site is of predominantly of B-acid sites.
o
According to these results, in this study, the desorption peak at 116 C is assigned to
o
NH on H-bond sites and the peaks at 161 °C and 345 C corresponded to the
3
desorption of NH on L-acid and B-acid sites, respectively. Compared with the bare
3

Page 17 of 41
Catalysis Science & Technology
1
7
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
HZ support, the loading of Ag or (and) MnO created new adsorption sites for NH .
x 3
Besides NH desorption from the HZ support (desorption peaks at 116, 181 and 355
3
o
o
C, respectively), a new peak at 445 C was observed on Ag/HZ catalyst, as shown in
Fig. 5b, which is due to NH adsorbed on Ag sites. For the Mn/HZ catalyst (Fig. 5c),
3
o
the NH peak at 260 C is attributed to the desorption of NH adsorbed on MnO .
3
3
x
Based on the results of Ag/HZ and Mn/HZ catalysts, it can be seen that the addition of
Ag caused an even stronger interaction with NH than MnO did. As revealed in Fig.
3
x
5d, the NH adsorbed on Ag and MnO sites over AgMn/HZ catalyst desorbed at 447
3
x
o
o
C and 215 C, respectively.
1
1
1
1
1
1
1
0
1
2
3
4
5
6
3.2.2 OZCO of adsorbed NH and studies on the surface species after OZCO
3
Fig. 6a exhibits the variations of N O and N concentrations with TOS during
2
2
OZCO of adsorbed NH on HZ, Ag/HZ, Mn/HZ and AgMn/HZ catalysts. It should be
3
noted that, except for N O and N , no other gaseous products were detected over the
2
2
four catalysts. From Fig. 6a, it can be seen that N O was detected in a noticeable
2
amount with respect to N in gaseous products over HZ and Ag/HZ catalysts and the
2
N O / (N  N O) ratios are 43% and 38% for HZ and Ag/HZ catalysts, respectively. It
2
2
2
1
1
1
2
2
2
7
8
9
0
1
2
can also be seen from Fig. 6a that more N and N O were detected over Ag/HZ than
2 2
HZ, which indicates that the addition of Ag enhanced the OZCO reaction of adsorbed
NH . However, only small amount of N O was detected and no N was formed over
3
2
2
Mn/HZ and AgMn/HZ catalysts during OZCO of adsorbed NH3.
Time course for O conversion during OZCO of adsorbed NH is shown in Fig.
3
3
6b. On HZ and Ag/HZ catalysts, O conversion decreased quickly and finally
3

Catalysis Science & Technology
Page 18 of 41
1
8
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
maintained around 10% and 30%, respectively. On Mn/HZ and AgMn/HZ catalysts,
O conversion remained high, but gradual decrease was observed after 20-min and
3
13-min TOS, respectively. Again, MnO is necessary to maintain high and consistent
x
conversion of O in OZCO processes.
3
In addition, ATR and TPD techniques were also adopted to characterize the
surface species after OZCO of adsorbed NH3.
In order to clearly reflect the changes taking place on the catalyst surface after
OZCO of adsorbed NH , ATR spectra of fresh samples after adsorption of NH are
3
3
-1
shown in Fig. 7a. A new band at 1459 cm was observed with respect to the fresh

1
1
1
1
1
1
1
1
1
1
2
2
2
samples (Fig. 3a), which is attributed to the bending vibration of NH4 indicating
7
,53,54
adsorbed NH on B-acid sites.
After OZCO of adsorbed NH , the band attributed
3
3
to NH adsorbed on B-acid sites disappeared, which implies the oxidation of adsorbed
3
-1
-1
-1
NH by O , and new bands at 1443 cm , 3100 cm and 3219 cm appeared (Fig. 7b).
3
3
-1
-1
According to the assignments in Section 3.1.3, bands at 3100 cm and 3219 cm are
4
9
due to the symmetric and asymmetric stretching of N-H, respectively. Moreover,
4
8
Hadjiivanov reported that the stretching vibration of N=O in nitrates also occurred
-1
-1
at 1380 cm . Therefore, the band at 1443 cm is assigned to the stretching vibration
of N=O of [-NO ] . Therefore, [-NO ] and [NH ] were deposited on the used catalyst
3
s
3 s
3 a
surfaces after OZCO of adsorbed NH3.
During TPD measurements, the evolution of NH was detected only on HZ and
3
Ag/HZ catalysts, but not on Mn/HZ and AgMn/HZ as shown in Fig. S4. N O was
2
detected as main product (at 210 °C and 246 °C for HZ and Ag/HZ catalysts,

Page 19 of 41
Catalysis Science & Technology
1
9
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
respectively) with negligible amount of NO over HZ and Ag/HZ catalysts during
2
TPD measurements, as shown in Figs. 8a and 8b, respectively, and N was not
2
detected on either of these two samples, as shown in Figs. S5a and S5b. Similar to the
TPD measurements after OZCO of gaseous NH , N O could also come from the
3
2
5
0
decomposition of ammonia-nitrates complex. H O was not produced on HZ and
2
Ag/HZ catalysts, as shown in Fig. 8e, which might be due to its low concentration.
However, in the cases of Mn/HZ and AgMn/HZ catalysts, N was detected by the
2
mass spectrometer (Figs. S5c and S5d) at 201 °C and 211 °C, respectively, and in a
considerable amount (Figs. 8c and 8d) during TPD measurements, which came from
5
0
1
1
1
1
1
1
1
1
1
1
2
2
0
1
2
3
4
5
6
7
8
9
0
1
the reduction of [-NO ] by [NH ] . N O was also formed on Mn/HZ and AgMn/HZ
3 s 3 a 2
catalysts, as shown in Figs. 8c and 8d, which was derived from the decomposition of
ammonia-nitrates complex. Large amount of H O was also detected over Mn/HZ and
2
AgMn/HZ catalysts at 250 °C, as shown in Fig. 8e and is attributed to the
decomposition of ammonia-nitrates complex and the reduction of [-NO ] by [NH ] .
3
s
3 a
No NH was evolved over Mn/HZ and AgMn/HZ catalysts (Fig. S4) during TPD
3
measurements since all the residual NH after OZCO of adsorbed NH reacted with
3
3
nitrates in TPD processes.
3.3. Proposed reaction pathways for OZCO of gaseous NH3
In the cases of HZ and Ag/HZ catalysts, the same gaseous products (N O and N )
2
2
as those of OZCO of gaseous NH (Fig. 1c) were also detected during OZCO of
3
adsorbed NH , as shown in Fig. 6a. Although the N O / (N2  N O) ratios were 43%
3
2
2
2
2
and 38% for HZ and Ag/HZ catalysts, respectively, which are higher than those of

Catalysis Science & Technology
Page 20 of 41
2
0
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
OZCO of gaseous NH (The N O selectivities were less than 10% over HZ and
3 2
Ag/HZ catalysts, as shown in Fig. 1c), N was still the main product during OZCO of
2
adsorbed NH . The reasonable explanation to the higher ratios of N O / (N2  N O)
3
2
2
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
during OZCO of adsorbed NH is as follows: during OZCO of adsorbed NH , O was
3 3 3
continuously fed into catalyst bed after NH adsorption. Therefore, O species
3
,
*
including O (g) and adsorbed atomic oxygen [O ] , were excessive with respect to the
3
a
decreasing amount of adsorbed NH . As a result, the adsorbed NH was more likely to
3
3
be further oxidized by O species to produce N O in contrast to OZCO of gaseous
2
NH3.
1
1
1
1
1
1
1
1
1
1
2
2
2
In addition, as shown in Fig. 6b, O conversion during OZCO of adsorbed NH3
3
also exhibited similar stability as that of OZCO of gaseous NH (Fig. 1b). O
3
3
conversions of HZ and Ag/HZ catalysts were comparable to those of Mn/HZ and
AgMn/HZ catalysts in the initial several minutes due to O consumption by the
3
pre-adsorbed NH and thereafter stabilized at around 10% and 30% (Fig. 6b),
3
respectively. These values of the O conversion were much lower than those in OZCO
3
of gaseous NH (60% and 80% for HZ and Ag/HZ, respectively, as shown in Fig. 1b),
3
which can be ascribed to that HZ and Ag/HZ are intrinsically less active for O3
decomposition in the absence of NH3.
Moreover, the surface species on the used HZ and Ag/HZ catalysts after OZCO
of gaseous and adsorbed NH were all composed of [-NO ] and [NH ] , as indicated
3
3 s
3 a
by the ATR characterization in Figs. 3b and 7b. TPD results of the used catalysts after
OZCO of adsorbed NH were also very similar to those after OZCO of gaseous NH :
3
3

Page 21 of 41
Catalysis Science & Technology
2
1
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
only NH and N O were detected, besides the negligible amount of NO , and no N
3 2 2 2
was observed. Based on the similarities of ATR and TPD results of the used catalysts
between OZCO of gaseous and adsorbed NH and the low activity of HZ and Ag/HZ
3
catalysts in O decomposition, it is reasonable to deduce that OZCO of gaseous NH
3
3
over HZ and Ag/HZ catalysts proceeds through Pathway (A) in Fig. 9, which means
that NH (g) is first adsorbed on the catalyst surface to form [NH ] and then the
3
3 a
[NH ] is oxidized by O species to final product N . Besides [NH ] oxidation in
3
a
2
3 a
Pathway (A), O could also be consumed via catalytic decomposition (Pathway (C) in
3
Fig. 9) over HZ and Ag/HZ catalysts.
1
1
1
1
1
1
1
1
1
1
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
On the other hand, for Mn/HZ and AgMn/HZ catalysts, gaseous products were
barely detected during OZCO of adsorbed NH (Fig. 6a), which is totally different
3
from the result of OZCO of gaseous NH (N was detected as the major gaseous
3
2
product during OZCO of gaseous NH , as shown in Fig. 1c). In addition, O
3
3
conversion also showed different stability in OZCO of gaseous (Fig. 1b) and adsorbed
(Fig. 6b) NH . It could be attributed to the fast accumulation of nitrates on the surface
3
of Mn/HZ and AgMn/HZ catalysts during OZCO of adsorbed NH , which would
3
occupy the active sites for O decomposition. Moreover, the surface species
3
accumulated on the catalyst surface after OZCO of adsorbed NH were different from
3
those after OZCO of gaseous NH , which can be deduced from the huge difference in
3
the ratios of N to N O formed in TPD measurements after OZCO processes (The
2
2
ratios of N to N O were 1 for the former case (Figs. 8c and 8d) and 8 for the latter
2
2
case (Figs. 4c and 4d)). Based on the differences between OZCO of gaseous and

Catalysis Science & Technology
Page 22 of 41
2
2
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
adsorbed NH and the high activity of Mn/HZ and AgMn/HZ catalysts in O
3 3
decomposition, it is reasonable to deduce that OZCO of gaseous NH over Mn/HZ
3
and AgMn/HZ catalysts proceeds through Pathway (B) in Fig. 9, in which O (g) is
3
*
first catalytically decomposed to form adsorbed atomic oxygen [O ] on the catalyst
a
*
surface and then the NH species are oxidized by [O ] to final product N .
x
a
2
Based on the proposed reaction mechanism, the results of OZCO of gaseous NH3
in Fig. 1 could be explained as follows. For HZ and Ag/HZ catalysts, the NH3
adsorption rate is much higher than the [NH ] oxidation rate by O species in Pathway
3
a
(A) , so the latter is the rate-controlling step in OZCO of gaseous NH over HZ and
3
1
1
1
1
1
1
1
1
Ag/HZ catalysts. However, as the reaction proceeded, more active sites on HZ and
Ag/HZ catalysts would be occupied by [NH ] resulting in the decrease in NH
3
3
a
conversion with TOS, as shown in Fig. 1a. In addition, O could be consumed through
3
both Pathways (A) and (C) over HZ and Ag/HZ catalysts. At the initial stage of the
OZCO reaction, O was consumed mainly by Pathway (A) due to the large amount of
3
active sites on the catalyst surface. However, as the reaction went on, Pathway (A)
was gradually inhibited by the adsorbed NH molecules, so Pathway (C) became the
3
main path responsible for O consumption. Therefore, the O conversion remained
3
3
1
1
8
9
almost constant in general with TOS over HZ and Ag/HZ catalysts during OZCO of
Conv. Conv.
gaseous NH , as shown in Fig. 1b. This is also supported by the
n
/n
ratios of
3
O3
NH3
Conv. Conv.
2
0
the reactions. As shown in Fig 1d, the initial
n
/n
ratios of reactions over HZ
O3
NH3
2
2
1
2
and Ag/HZ were around 1.4 and 1.3, respectively, which are lower than the
stoichiometric ratio of 1.5 in eqn. (8). This is likely due to the fact that more than one

Page 23 of 41
Catalysis Science & Technology
2
3
View Article Online
DOI: 10.1039/C4CY01269K
1
2
oxygen atoms per O molecule are involved in NH oxidation to N in Pathway (A).
3
3
2
Conv. Conv.
As the OZCO reaction went on, the
n
/n
ratios of HZ and Ag/HZ catalysts
O3
NH3
3
4
5
6
7
8
9
increased and were more than the stoichiometric ratio of 1.5 (Fig. 1d) in eqn. (8),
which indicates that Pathway (C) had become the main route for O consumption over
3
HZ and Ag/HZ catalysts.
However, for Mn/HZ and AgMn/HZ catalysts, since the O decomposition and
3
*
NH (g) oxidation by [O ] are both at high reaction rate, NH and O conversions
3
a
3
3
were almost constant during OZCO of gaseous NH over Mn/HZ and AgMn/HZ, as
3
shown in Figs. 1a and 1b, respectively. Based on the two parallel reactions in eqns. (8)
Conv. Conv.
1
1
0
1
and (9), it is possible to calculate the theoretical
n
/n
ratio of the reaction
O3
NH3
according to N O and N selectivities. Since the selectivities of N O & N were 20%
2
2
2
2
1
1
2
3
& 80% and 13% & 87% for Mn/HZ and AgMn/HZ catalysts (Fig. 1c), respectively,
Conv. Conv.
the theoretical
n
/n
ratios can be calculated (1.5SN2  2 SN2O ) to be about
O3
NH3
1
1
1
1
1
1
2
2
2
4
5
6
7
8
9
0
1
2
1.60 and 1.57 for Mn/HZ and AgMn/HZ, respectively, which are consistent with the
experimental results, ~1.7 for Mn/HZ and ~1.65 for AgMn/HZ.
There also exists another possible mechanism where the reaction pathway might
be the same over the four catalysts but the reaction rates of Mn/HZ and AgMn/HZ
catalysts are much faster than those of HZ and Ag/HZ catalysts, which results from
the high ability of Mn/HZ and AgMn/HZ catalysts to activate ozone. The mechanism
will be further studied in the future work.
Conclusions

Catalysis Science & Technology
Page 24 of 41
2
4
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
OZCO for removing ammonia from simulated air at room temperature was
studied in this paper. Compared with HZ, Ag/HZ and Mn/HZ catalysts, AgMn/HZ
catalyst exhibited high activity, good stability and high N selectivity, which could be
2
attributed to the outstanding ability of MnO to decompose O and the promoting
x
3
effect of Ag. The NH conversion and N selectivity were as high as 99% and 94%,
3
2
6
7
respectively, over AgMn/HZ catalyst at room temperature, GHSV of 150,000
.
-1. -1
in
in
ml g h and
C
/C ratio of 1.73 that is slightly higher than the stoichiometric ratio
O3
NH3
8
9
0
1
2
3
4
5
6
7
8
9
of OZCO of NH to N . The possible pathway for OZCO of NH over AgMn/HZ
3 2 3
*
catalyst was proposed as follows: O (g) is first adsorbed and decomposed into [O ]
3
a
*
1
1
1
1
1
1
1
1
1
1
and then NH species are oxidized by [O ] to form N . This was supported by the
x
a
2
results of comparison between OZCO of adsorbed and gaseous NH3.
Acknowledgements
This work is supported by National Natural Science Foundation of China
(11079013, U1201231) and by the Welch Foundation in US (#T-0014).
‡
Notes
1 GHSV (Gas Hourly Space Velocity) represents the amount of gas mixture flowing
over catalysts per unit gram of catalyst per unit time and can be calculated by the
following equation:
2
2
0
1
F
GHSV 
mcat
2
2
where F and mcat represent the total volume flow rate of reaction gas and mass of

Page 25 of 41
Catalysis Science & Technology
2
5
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
catalyst, respectively.
2 BET surface area is calculated using the theory of adsorption of polymolecular layer
defined by Brunauer, Emmett and Teller.
3 SCCM is an abbreviation of Standard Cubic Centimeters per Minute indicating
.
-1
o
ml min at a standard temperature of zero C and pressure of 101.325 kPa.
References
8
9
1 J.R. Kastner, J. Miller, P. Kolar and K.C. Das, Chemosphere, 2009, 75, 739.
2 Y. Li and J.N. Armor, Appl. Catal. B: Environ., 1997, 13, 131.
3 L. Lietti, G. Ramis, G. Busca, F. Bregani and P. Forzatti, Catal. Today, 2000, 61, 187.
4 R.Q. Long and R.T. Yang, J. Catal., 2001, 201, 145.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
5 R.Q. Long and R.T. Yang, J. Catal., 2002, 207, 158.
6 L. Gang, B.G. Anderson, J. van Grondelle and R.A. van Santen, Appl. Catal. B: Environ., 2003, 40,
101.
7 G. Qi, J.E. Gatt and R.T. Yang, J. Catal., 2004, 226, 120.
8 L. Chmielarz, P. Kuśtrowski, A. Rafalska-Łasocha and R. Dziembaj, Appl. Catal. B: Environ., 2005,
58, 235.
9 A.C. Akah, G. Nkeng and A.A. Garforth, Appl. Catal. B: Environ., 2007, 74, 34.
10 M.J. Lippits, A.C. Gluhoi and B.E. Nieuwenhuys, Catal. Today, 2008, 137, 446.
11 M.S. Kim, D.W. Lee, S.H. Chung, Y.K. Hong, S.H. Lee, S.H. Oh, I.H. Cho and K.Y. Lee, J. Hazard.
Mater., 2012, 237-238, 153.
12 C. Liang, X. Li, Z. Qu, M. Tade and S. Liu, Appl. Surf. Sci., 2012, 258, 3738.
13 A. Naydenov and D. Mehandjiev, Appl. Catal. A: Gen., 1993, 97, 17.
14 D. Mehandjiev, K. Cheshkova, A. Naydenov and V. Georgesku, Reac. Kinet. Catal. Lett., 2002, 76,
287.
15 S. Dimitrova, G. Ivanov and D. Mehandjiev, Appl. Catal. A: Gen., 2004, 266, 81.
16 P. Konova, M. Stoyanova, A. Naydenov, St. Christoskova and D. Mehandjiev, Appl. Catal. A: Gen.,
2006, 298, 109.
17 H. Einaga and S. Futamura, J. Catal., 2004, 227, 304.
18 H. Einaga and A. Ogata, J. Hazard. Mater., 2009, 164, 1236.
19 H. Einaga and A. Ogata, Environ. Sci. Technol., 2010, 44, 2612.
20 H. Einaga, N. Maeda and Y. Teraoka, Appl. Catal. B: Environ., 2013, 142-143, 406.
21 H. Einaga, Y. Teraoka and A. Ogata, J. Catal., 2013, 305, 227.
22 C.R. Lee, J. Jurng, G.N. Bae, J.K. Jeon, S.C. Kim, J.M. Kim, M. Jin and Y.K. Park, J. Nanosci.
Nanotechno., 2011, 11, 7303.
23 J.H. Park, J.M. Kim, M. Jin, J.K. Jeon, S.S. Kim, S.H. Park, S.C. Kim and Y.K. Park, Nanoscale Res.
Lett., 2012, 7, 14.

Catalysis Science & Technology
Page 26 of 41
2
6
View Article Online
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
24 M. Jin, J.W. Kim, J.M. Kim, J. Jurng, G.N. Bae, J.K. Jeon and Y.K. Park, Powder Technol., 2011,
DOI: 10.1039/C4CY01269K
214, 458.
25 Y. Liu, X.S. Li, C. Shi, J.L. Liu, A.M. Zhu and Ben W.-L. Jang, Catal. Sci. Technol., 2014, 4, 2589.
26 Y. Liu, X.S. Li, J.L. Liu, C. Shi and A.M. Zhu, Chin. J. Catal., 2014, 35, 1465.
27 E. Rezaei and J. Soltan, Chem. Eng. J., 2012, 198-199, 482.
28 E. Rezaei, J. Soltan, N. Chen and J. Lin, Chem. Eng. J., 2013, 214, 219.
29 E. Rezaei, J. Soltan and N. Chen, Appl. Catal. B: Environ., 2013, 136-137, 239.
30 C.Y.H. Chao, C.W. Kwong and K.S. Hui, J. Hazard. Mater., 2007, 143, 118.
31 M. Li, K.N. Hui, K.S. Hui, S.K. Lee, Y.R. Cho, H. Lee, W. Zhou, S. Cho, C.Y.H. Chao and Y. Li,
Appl. Catal. B: Environ., 2011, 107, 245.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
32 E. Park, S. Chin, J. Kim, G.N. Bae and J. Jurng, Powder Technol., 2011, 208, 740.
33 H.H. Kim, M. Sugasawa, H. Hirata, Y. Teramoto, K. Kosuge, N. Negishi and A. Ogata, Plasma
Chem. Plasma Process, 2013, 33, 1083.
34 H.C. Wang, H.S. Liang and M.B. Chang, J. Hazard. Mater., 2011, 186, 1781.
35 H.S. Liang, H.C. Wang and M.B. Chang, Ind. Eng. Chem. Res., 2011, 50, 13322.
36 D.Z. Zhao, T.Y. Ding, X.S. Li, J.L. Liu, C. Shi and A.M. Zhu, Chin. J. Catal., 2012, 33, 396.
37 D.Z. Zhao, C. Shi, X.S. Li, A.M. Zhu and Ben W.-L. Jang, J. Hazard. Mater., 2012, 239-240, 362.
38 M. Stoyanova, P. Konova, P. Nikolov, A. Naydenov, St. Christoskova and D. Mehandjiev, Chem.
Eng. J., 2006, 122, 41.
39 Z. Hao, D. Cheng, Y. Guo and Y. Liang, Appl. Catal. B: Environ., 2001, 33, 217.
40 M. Petersson, D. Jonsson, H. Persson, N. Cruise and B. Andersson, J. Catal., 2006, 238, 321.
41 S. Ichikawa, L. Mahardiani and Y. Kamiya, Catal. Today, 2014, 232, 192.
42 S.T. Oyama, Catal. Rev.: Sci. Eng., 2000, 42, 279.
43 K.J. Olszyna and J. Heicklen, in Photochemical Smog and Ozone Reactions, ed. R.F. Gould,
American Chemical Society, Washington, DC, 1972, vol. 113, ch. 6, pp. 191-210.
44 E.M. Slavinskaya, S.A. Veniaminov, P. Notté, A.S. Ivanova, A.I. Boronin, Yu.A. Chesalov, I.A.
Polukhina and A.S. Noskov, J. Catal., 2004, 222, 129.
45 W. Li, G.V. Gibbs and S.T. Oyama, J. Am. Chem. Soc., 1998, 120, 9041.
46 A. Savara, W.M.H. Sachtler and E. Weitz, Appl. Catal. B: Environ., 2009, 90, 120.
47 S. Tamm, N. Vallim, M. Skoglundh and L. Olsson, J. Catal., 2013, 307, 153.
48 K.I. Hadjiivanov, Catal. Rev.: Sci. Eng., 2000, 42, 71.
49 The Sadtler Handbooks of Infrared Spectra, Bio-Rad Laboratories, Inc., Informatics Division.
50 M. Colombo, I. Nova and E. Tronconi, Catal. Today, 2012, 197, 243.
51 L. Rodríguez-González, F. Hermes, M. Bertmer, E. Rodríguez-Castellón, A. Jiménez-López and U.
Simon, Appl. Catal. A: Gen., 2007, 328, 174.
52 N. Katada and M. Niwa, Cataly. Surv. Asia, 2004, 8, 161.
53 L. Ma, Y. Cheng, G. Cavataio, R.W. McCabe, L. Fu and J. Li, Appl. Catal. B: Environ., 2014,
156-157, 428.
54 D. Wang, L. Zhang, J. Li, K. Kamasamudram and W.S. Epling, Catal. Today, 2014, 231, 64.
55 A.S. Al-Dughaither and H. de Lasa, Ind. Eng. Chem. Res., 2014, 53, 15303.
4
2

Page 27 of 41
Catalysis Science & Technology
2
7
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
Figure Captions
Fig. 1. Time courses for (a) NH conversion, (b) O conversion, (c) N O and N
3
3
2
2
Conv. Conv.
selectivities and (d)
n
/n
ratio during OZCO of gaseous NH over HZ,
3
NH3
O3
4
5
Mn/HZ, Ag/HZ and AgMn/HZ catalysts. Conditions: 0.1 g of catalysts, feed
gas of 250 SCCM containing 530 ppmv of NH , 450 ppmv of O , 20 vol.% of
3
3
6
7
O and balanced by N .
2
2
in
in
NH3
Fig. 2. Effects of
C
/ C ratio on OZCO of gaseous NH over AgMn/HZ catalyst at
3
O3
8
9
TOS = 60 min. Conditions: 0.1 g of AgMn/HZ catalyst, 250 SCCM of flow
in
in
NH3
in
O3
in
NH3
rate with 530 ppmv (
C
/ C
 0.85), 517 ppmv (
C
/ C
1.32) and 521
O3
in
O3
in
NH3
1
0
ppmv (
C
/ C
1.73) of NH , respectively, in 20 vol.% O / 80 vol.% N .
3
2
2
1
1
1
1
1
1
1
1
1
2
2
2
1
2
3
4
5
6
7
8
9
0
1
2
Fig. 3. ATR spectra of the catalysts: (a) fresh samples and (b) used catalysts after
OZCO of gaseous NH for 90 min. OZCO conditions are the same as those in
3
Fig. 1. NH NO -impregnated HZ (NH NO -HZ) was prepared for reference.
4
3
4
3
Preparation of the NH NO -HZ sample involves impregnation of HZ with
4
3
NH NO solution and then drying at 50 °C for 12 h.
4
3
Fig. 4. TPD profiles of the used catalysts after 90-min OZCO of gaseous NH : the
3
concentrations of N O and NO on (a) HZ and (b) Ag/HZ catalysts; the
2
2
concentrations of N , N O and NO on (c) Mn/HZ and (d) AgMn/HZ catalysts;
2
2
2
-1
(e) the strongest FT-IR peak height of water in the range of 3500 to 4000 cm .
OZCO conditions are the same as those in Fig. 1. TPD conditions: 0.03 g of
.
-1
the used catalysts, 100 SCCM of He, 10 °C min .
Fig. 5. NH profiles of (a) HZ, (b) Ag/HZ, (c) Mn/HZ and (d) AgMn/HZ catalysts
3

Catalysis Science & Technology
Page 28 of 41
2
8
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
during NH -TPD measurements. NH adsorption conditions: 0.1 g of catalysts,
3 3
540 ppmv of NH for 25-min adsorption; TPD conditions: 100 SCCM of Ar,
3
.
-1
10 °C min .
Fig. 6. Time courses for (a) N O and N concentrations and (b) O conversion during
2
2
3
OZCO of adsorbed NH over HZ, Mn/HZ, Ag/HZ and AgMn/HZ catalysts.
3
NH adsorption conditions: 0.1 g of catalysts, 540 ppmv of NH for 25-min
3
3
adsorption; OZCO conditions: a feed gas of 250 SCCM containing 450 ppmv
of O , 20 vol.% of O and balanced by He.
3
2
Fig. 7. ATR spectra of the catalysts (a) after their exposure to certain amount of NH3
1
1
1
1
1
1
1
1
1
1
2
and (b) after 30-min OZCO of adsorbed NH . NH adsorption and OZCO
3
3
conditions are the same as those in Fig. 6.
Fig. 8. TPD profiles of the used catalysts after 30-min OZCO of adsorbed NH : the
3
concentrations of N O and NO on (a) HZ and (b) Ag/HZ catalysts; the
2
2
concentrations of N , N O and NO on (c) Mn/HZ and (d) AgMn/HZ catalysts;
2
2
2
-1
(e) the strongest FT-IR peak height of water in the range of 3500 to 4000 cm .
OZCO conditions are the same as those in Fig. 6. TPD conditions: 0.03 g of
.
-1
the used catalysts, 100 SCCM of He, 10 °C min .
Fig. 9. Proposed reaction pathways for OZCO of gaseous NH over HZ, Ag/HZ,
3
Mn/HZ and AgMn/HZ catalysts.

Page 29 of 41
Catalysis Science & Technology
2
9
View Article Online
DOI: 10.1039/C4CY01269K
1
2
3
Fig. 1
6
5
4
3
2
1
0
0
0
0
0
0
0
(a)
HZ
Ag/HZ
Mn/HZ
AgMn/HZ
-10
0
10 20 30 40 50 60 70 80 90 100
TOS (min)
4
5
6
1
00 (b)
8
6
4
2
0
0
0
0
0
HZ
Ag/HZ
Mn/HZ
AgMn/HZ
-10
0
10 20 30 40 50 60 70 80 90 100
TOS (min)
7
8