Understanding the Active Sites of Ag/Zeolites and Deactivation Mechanism of Ethylene Catalytic Oxidation at Room Temperature

Article abstract of DOI:10.1021/acscatal.7b02410

Microporous Ag/zeolite (ZSM-5, Beta, Y, and Mordenite) catalysts were found to be potentially good catalytic materials for ethylene oxidation at room temperature. These catalysts were evaluated under both dry and humid atmospheres, achieving 100% conversion of 100 ppm of ethylene mineralized into CO₂ at 25 °C. Moreover, the zeolite framework type and relative humidity had a significant effect on catalytic stability. Pyridine Fourier transform infrared spectra (FTIR) and solid-state hydrogen-1 (¹H) magic angle spinning nuclear magnetic resonance (MAS NMR) studies revealed that Br?nsted acid sites were the active sites in Ag/zeolites; the deactivated Ag/zeolites had no available Br?nsted acid sites. When the number of Br?nsted acid sites of Ag/ZSM-5-humid and Ag/Beta-humid were reduced, the catalytic activities of Ag/ZSM-5-humid and Ag/Beta-humid decreased. Temperature-programmed oxidation mass spectrometry (TPO-MS) and water adsorption/desorption results indicated that H₂O adsorbed on Br?nsted acid sites led to the disappearance of available Br?nsted acid sites and thus resulted in Ag/zeolite catalyst deactivation. An understanding of the active sites and deactivation mechanism in ethylene oxidation on zeolite catalysts is helpful in the synthesis of a better ethylene oxidation catalyst and development of an effective technology of eliminating trace ethylene.

Full text of DOI:10.1021/acscatal.7b02410

Subscriber access provided by AUSTRALIAN NATIONAL UNIV
Article
Understanding the active sites of Ag/zeolites and deactivation
mechanism of ethylene catalytic oxidation at room-temperature
Hongling Yang, Chunyan Ma, Xin Zhang, Yang Li, Jie Cheng, and Zhengping Hao
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.7b02410 • Publication Date (Web): 29 Dec 2017
Downloaded from http://pubs.acs.org on December 31, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 44
ACS Catalysis
1
2
3
4
5
6
7
1
2
Understanding the active sites of Ag/zeolites and deactivation
mechanism of ethylene catalytic oxidation at room-temperature
8
9
3
4
Hongling Yang,1,2 Chunyan Ma,*,1,2 Xin Zhang,1,2 Yang Li,1 Jie Cheng,1,2
Zhengping Hao*,1,2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
6
1. Department of Environmental Nanoꢀmaterials and Technologies, Research
Center for EcoꢀEnvironmental Sciences, Beijing 100085, China
7
8
9
2. National Engineering Laboratory for VOCs Pollution Control Material &
Technology, University of Chinese Academy of Sciences, Beijing 100049,
China
10
11
*Eꢀmail: cyma@rcees.ac.cn; zpinghao@rcees.ac.cn
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 44
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
ABSTRACT
Microporous Ag/zeolite (ZSMꢀ5, Beta, Y and Mordenite) catalysts were
found to be potential good catalytic materials for ethylene oxidation at
room temperature. These catalysts were evaluated in both dry and humid
atmosphere, achieving 100% of 100 ppm of ethylene mineralized into
CO2 at 25 oC. Moreover, the zeolite framework type and relative humidity
had a significant effect on catalytic stability. Pyridine fourier transform
infrared spectra (FTIR) and solidꢀstate hydrogenꢀ1 (1H) magic angle
spinning nuclear magnetic resonance (MAS NMR) studies revealed that
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10 Brønsted acid sites were the active sites in Ag/zeolites, the deactivated
11 Ag/zeolites had no available Brønsted acid site. When the number of
12 Brønsted acid sites of Ag/ZSMꢀ5ꢀhumid and Ag/Betaꢀhumid reduced, the
13 catalytic activities of Ag/ZSMꢀ5ꢀhumid and Ag/Betaꢀhumid decreased.
14 Temperatureꢀprogrammed oxidationꢀmass spectrometry (TPOꢀMS) and
15 water adsorption/desorption results indicated that H2O adsorbed on
16 Brønsted acid sites led to the disappearance of available Brønsted acid
17 sites, and thus resulted in Ag/zeolite catalyst deactivation. The
18 understanding of active sites and deactivation mechanism in ethylene
19 oxidation on zeolite catalysts is helpful to synthesize a better ethylene
20 oxidation catalyst and develop an effective technology of eliminating
21 trace ethylene.
22 Keywords: ethylene, lowꢀtemperature oxidation, Ag, zeolite, deactivation
ACS Paragon Plus Environment

Page 3 of 44
ACS Catalysis
1
2
3
4
1
1. INTRODUCTION
5
6
7
8
9
2
3
4
5
6
7
8
9
Ethylene (C2H4), a lowꢀmolecularꢀweight volatile organic compound
(VOC), is a plant hormone, which enhances ripening and decaying of the
fruits and vegetables.1,2 Even low concentrations of ethylene can cause a
faster deterioration of fresh fruits and vegetables, so it is important to
remove ethylene from storage facilities. Lowꢀtemperature catalytic
oxidation is the most promising costꢀeffective method for removing
ethylene.3ꢀ5
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To the best of our knowledge, the complete oxidation of ethylene at
10 low temperatures are achieved mainly on mesoporous Co3O4 supported
11 Au catalysts and Pt/MCMꢀ41 catalyst.6,7 Their catalytic stabilities need to
12 be further improved, moreover, complicated preparation process of
13 mesoporous Co3O4 and low hydrothermal stability of MCMꢀ41 support
14 extremely restricted their application in ethylene oxidation. In contrast
15 with mesoporous support of Co3O4 and MCMꢀ41, the microporous
16 aluminosilicate zeolites are easily obtained, can act as a support to load
17 and disperse metal nanoparticles. Microporous aluminosilicate zeolites as
18 catalysts are used in different reactions, the acidic properties of zeolites
19 determine the catalytic performances in the acidꢀcatalyzed reactions.8ꢀ14
20 Thus, it is desirable that zeolites acting as catalysts or catalyst supports
21 are used for ethylene oxidation at room temperature and obtain good
22 catalytic performances, specially the catalytic stabilities.
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 44
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
Except for acidity of microporous zeolite, framework type of
microporous zeolite is of critical importance for the given catalytic
reaction. Sultana et al. investigated the conversion of levulinic acid to
aromatics over small pore FER, medium pore ZSMꢀ5, and large pore Y,
BEA and MOR, the highest catalytic activity of ZSMꢀ5 was attributed to
its threeꢀdimensional pore geometry but not the acidity.12 Bokhoven et al.
reported that the number of Brønsted acid sites varied with framework
topologies of zeolites. ZSMꢀ5, MOR, Beta and Y showed different
catalytic activities in propane cracking and dehydrogenation.13 Hunger et
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10 al. compared the catalytic activity of ethylbenzene disproportionation
11 over mediumꢀpore HZSMꢀ5 and the largeꢀpore HY, HZSMꢀ5 showed
12 higher reactivity due to its larger numbers of Brønsted acid sites and
13 smaller pores.14 In this case, it is essential to study the catalytic
14 performances of zeolites with different framework types for a given
15 reaction, which will help us to understand the effect of framework type or
16 acidity of zeolites on the chemical reaction.
17
In the study of catalytic stability of mesoporous Pt/MCMꢀ41 in
18 ethylene oxidation, Fukuoka et al. reported the gradual deactivation of
19 Pt/MCMꢀ41 was ascribed to the adsorption of H2O on the catalyst surface
20 for ethylene oxidation,7 but no unequivocal identification of the active
21 sites has been proposed. Moreover, there is very limited understanding of
22 mechanistic pathways over the mesoporous zeolite deactivation by
ACS Paragon Plus Environment

Page 5 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
adsorbing H2O in ethylene oxidation.
Herein, we report the catalytic performances of ethylene oxidation
over microporous zeolites (ZSMꢀ5, Beta, Y and Mordenite) and the
corresponding Ag/zeolites in dry and humid atmosphere. To the best of
our knowledge, Ag/zeolites have not been studied so far for ethylene
oxidation reaction. The catalytic active sites and the deactivation reason
of Ag/zeolites in ethylene oxidation reaction are explored through
characterizing the fresh and deactivated catalysts. The reaction
mechanism of ethylene oxidation over Ag/zeolites is proposed to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10 understand Ag/zeolite catalysts deactivation in ethylene oxidation. These
11 are very important to develop the meaningful environmental catalytic
12 materials and technology for elimination of ethylene.
13 2. EXPERIMENTAL SECTION
14
2.1. Catalyst preparation. Commercial ZSMꢀ5, Beta and
15 Mordenite zeolites with SiO2/Al2O3 ratio of 25, and Y zeolite with
16 SiO2/Al2O3 ratio of 5.2 were supplied by the Nankai Catalyst Plant. All
17 the Ag/zeolite catalysts were prepared by using the vacuumꢀassisted
18 impregnation method. The zeolite support material (1.0 g) was
19 impregnated with 3.2 mL of 0.15 mol·Lꢀ1 AgNO3 in a vacuum
20 environment (P = –0.1 MPa gauge pressure) for 1 h, and impregnated at
21 atmospheric pressure for 12 h. The resulting solid was dried at 100 °C in
22 air overnight, followed by calcination under air flow at 600 °C for 2 h.
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 44
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
2.2. Catalyst characterization. Xꢀray photoelectron spectroscopy
(XPS) measurement was carried to determine the elemental composition
and chemical states of the elements on the surface of catalysts. This data
was accumulated on an Axis Ultra ESCA system with a monochromatic
Al Kα standard radiation source. The binding energies were calibrated by
referencing the C1s at 284.8 eV.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Solidꢀstate aluminumꢀ27 (27Al) magic angle spinning nuclear
1
magnetic resonance (MAS NMR) and H MAS NMR spectra were
collected in a Bruker Advance 400 spectrometer. Prior to the MAS NMR
10 experiment, 0.2 g of sample was packed into a glass tube, which was later
11 connected to a vacuum line. The temperature was gradually increased at a
12 rate of 1.5 °C·minꢀ1, and was held at 300 °C at a pressure below 1.3 × 10ꢀ6
13 kPa for 12 h to dehydrate the sample. After the sample was cooled to
14 room temperature, the glass tube was sealed. Next, the sealed sample was
27
15 transferred into a 4 mm NMR rotor. The Al MAS NMR spectra were
3+
16 obtained at 104.23 MHz, and the chemical shifts referred to Al(H2O)6 .
17 The chemical shifts of 1H MAS NMR were referenced to TMS at 0 ppm.
18
Fourier TransformꢀInfrared Spectra (FTꢀIR) were collected using a
19 Bruker Tensor 27 FTIR spectrometer. Selfꢀsupporting thin wafers, used
20 in pyridine adsorption studies, were placed in a quartz holder with an
21 attached heating filament. This was mounted in an in situ cell connected
22 to a vacuum system. Prior to the measurements, each sample was
ACS Paragon Plus Environment

Page 7 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
2
degassed in vacuum (< 10ꢀ4 Pa) by heating to 500 °C at a rate of
10 °C·minꢀ1 and held at this temperature for 4 h to desorb any possible
physisorbed species. After the wafer was cooled down to room
temperature, the adsorption of pyridine was carried out for 1 h.
Desorption was performed in vacuum environment (< 10ꢀ4 Pa) at elevated
temperatures (150, 250, 350 and 450 °C). Relative determination of
Brønsted and Lewis acidity was derived from the areas of the IR bands at
ca. 1540 and 1450 cm−1, respectively, using the extinction coefficients
given by Emeis.15
3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
5
6
7
8
9
10
Xꢀray diffraction (XRD) patterns were recorded on an X’Pert PRO
11 (PANalytical) diffractometer equipped with a Cu Kα radiation source (λ
12 = 0.15418 nm). A continuous mode was used for collecting data in the 2θ
13 range from 5° to 90° at a scanning speed of 5 °·minꢀ1.
14
Static adsorption equilibrium measurements of H2O on Ag/zeolite
15 catalysts were investigated using an intelligent gravimetric analyzer
16 (Model IGAꢀ002; Hiden Isochema Instrument, Warrington, UK) with a
17 sensitivity of 0.1 µg. The apparatus has an ultrahigh vacuum system with
18 a fully computerized microbalance, allowing adsorption isotherms to be
19 determined by setting pressure steps. Before the measurement, each
20 sample was degassed at 300 °C (< 10ꢀ9 kPa) for 4 h.
21
The H2Oꢀtemperature programmed desorption (H2OꢀTPD) was
22 performed in the aforementioned IGA apparatus, each sample (60 mg)
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 44
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
was degassed at 300 °C (< 10ꢀ9 kPa) for 4 h. Before H2OꢀTPD
measurement, the adsorption process was performed by allowing the
sample to adsorb saturated water vapor at 25 °C (P = 2.86 kPa). Then, the
sample was heated to 405 °C at a rate of 0.1 °C·minꢀ1 at 2.86 kPa. The
desorbed water was measured by using the weight loss of the sample. The
weight loss curves were recorded, which were called the TPD curves of
H2O.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Inꢀsitu infrared spectra of the samples were recorded on a Bruker
Tensor27 using the diffuse reflectance infrared fourier transform (DRIFT)
10 spectroscopy scanned from 4000 to 850 cmꢀ1 at a resolution of 4 cmꢀ1. For
11 Ag/ZSMꢀ5 in dry atmosphere, before the infrared spectra were recorded,
12 the sample was swept with He gas at 500 °C for 1 h, and then the catalyst
13 bed temperature was lowered to 25 °C. The mixed gas (100ppm C2H4/21%
14 O2/He balance) with a total gas flow of 25 mL·minꢀ1 passed through 5A
15 molecule sieve to be dried, and then passed through the sample cell, after
16 30 min one spectrum was recorded. Next 21% O2/He balance gas passed
17 through 5A molecule sieve, and then passed through the sample cell, the
18 spectra were recorded after 5, 15, 25 and 35 min, respectively. For
19 Ag/ZSMꢀ5 in humid atmosphere, the mixed gas (100ppm C2H4/21%
20 O2/He balance) with a total gas flow of 25 mL·minꢀ1 passed through the
21 sample cell, after 30 min one spectrum was recorded. Then 21% O2/He
22 balance gas passed through the sample cell, the spectra were recorded
ACS Paragon Plus Environment

Page 9 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
after 5, 15, 25 and 35 min, respectively.
2.3. Ethylene oxidation activity measurements. Ethylene
oxidation experiments were performed in a Uꢀshape quartz tubular
reactor loading with 0.2 g of catalyst (40ꢀ60 mesh), placed in a water bath
at 25 °C. The catalytic performance was evaluated under dry and humid
atmosphere. The reaction gas of 100 ppm C2H4/21% O2/79% He,
followed by drying or wetting, was flowed through the catalyst bed at a
flow rate of 25 mL·minꢀ1, which corresponded to a space velocity (SV) of
7500 mL·hꢀ1·gꢀ1. The drying reaction gas is that 100 ppm C2H4/21%
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10 O2/79% He passed through 5A molecule sieve to be dried. The relative
11 humidity of 50% (water vapor in air) was adjusted by varying the 100
12 ppm C2H4/21% O2/79% He gas flow rate through a bubbler placed in a
13 water bath at 25 °C. The water vapor concentration was measured online
14 using a hygrometer (Apresys, USA), which was expressed as relative
15 humidity (RH) at 25 °C.
16
The reactants and products were analyzed by using an Agilent
17 7890B gas chromatograph equipped with a hydrogen flame ionization
18 detector (Agilent HPꢀPLOT U column), and a thermal conductivity
19 detector (TCD) (PorapakꢀQ column and 5A molecular sieve column). The
20 conversion of ethylene into CO2 was calculated based on the effluent
21 content of CO2, which was based on the formula [CCO2out/(2CC2H4inlet)] ×
22 100%, where CCO2out and CC2H4inlet were the concentrations of the
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 44
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
production of CO2 and C2H4 introduced into the reactor, respectively. The
carbonaceous and hydrous species deposited on the catalyst during
ethylene
oxidation
reaction
were
characterized
by
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
temperatureꢀprogrammed oxidationꢀmass spectrometry (TPOꢀMS). In a
typical TPOꢀMS experiment, 20 mg of deactivated Ag/zeolite catalyst in
dry atmosphere (21%O2/He) was loaded in a continuous flow quartz
reactor that was then coupled to a mass spectrometer (Omnistar TM).
8
9
3. RESULTS AND DISCUSSION
3.1. Catalytic performances for ethylene oxidation. The catalytic
10 performances of zeolite and Ag/zeolite catalysts for ethylene oxidation
11 were investigated at 25 °C under 100 ppm C2H4/21% O2/He in dry
12 atmosphere. We defined the value after the first 5 minute reaction as the
13 “initial conversion”. As shown in Figure 1a, in dry atmosphere, the initial
14 conversion of ethylene over ZSMꢀ5, Beta, Y and Mordenite was 38%,
15 35%, 25% and 13%, respectively. Thus, it is obvious that there are active
16 sites in microporous aluminosilicate zeolites for ethylene oxidation. The
17 initial conversion of ethylene over all the Ag/zeolite catalysts was nearly
18 100% in dry atmosphere (Figure 1b). The variation in the ethylene
19 conversion with time on stream strongly depended on the zeolite type.
20 Ag/ZSMꢀ5 and Ag/Beta catalysts provided the highest stability, showing
21 no considerable change in the first 6 h, although the conversion decreased
22 slowly to zero in the following 18 h. While Ag/Y and Ag/Mordenite
ACS Paragon Plus Environment

Page 11 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
showed more obvious deactivation, the initial conversion of ethylene over
Ag/Y was only kept for 3 h, and then the conversion decreased to zero in
the following 5 h. For Ag/Mordenite, the initial conversion was kept for
0.6 h, and then the conversion decreased to zero in the following 4.5 h.
An initial gas of 100 ppm C2H4/21% O2/He with relative humidity of
50% was also used to investigate catalytic performances over Ag/zeolite
catalysts, as also shown in Figure 1b. Compared with a dry atmosphere
stream, water vapor had almost no effect on the initial activity of ethylene
oxidation for all these Ag/zeolite catalysts. The initial conversion of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10 ethylene over all the Ag/zeolite catalysts under humid atmosphere was
11 still nearly 100%. However, in the case of humid atmosphere, the initial
12 conversion of ethylene over Ag/ZSMꢀ5 was only kept for the first 1.5 h,
13 and then subsequently decreased slightly from 100% to 65% in the
14 following 1 h. There was no extra decrease of ethylene conversion during
15 the following test and the conversion was remained at 65% for Ag/ZSMꢀ5
16 in the following 21.5 h. In contrast, the initial conversion over Ag/Beta
17 was sustained for the first 2.5 h and decreased rapidly to 30% in the next
18 0.5 h, and then the ethylene conversion was not changed during the
19 subsequent test of 21 h under humid atmosphere. Meanwhile, the initial
20 conversion of ethylene over Ag/Y and Ag/Mordenite catalysts was only
21 kept for 1.5 and 0.5 h, respectively, and then lost their activities
22 completely in the following 5.5 h.
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 44
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
These results clearly showed that the initial ethylene conversion of
100% could be achieved on all Ag/zeolite catalysts in both dry and humid
atmosphere. In humid atmosphere, Ag/ZSMꢀ5 and Ag/Beta catalysts kept
ethylene conversions of 65% and 30% after reacting for 3 h, respectively.
Until reacting for 24 h, the ethylene conversion over Ag/ZSMꢀ5 and
Ag/Beta was still kept 65% and 30%, respectively. In dry atmosphere, the
Ag/ZSMꢀ5 and Ag/Beta lost their activities completely and there was no
ethylene conversion after reacting for 24 h. However, the initial ethylene
conversion of 100% over Ag/ZSMꢀ5 and Ag/Beta was kept for much
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10 longer time in dry atmosphere than in humid atmosphere. It is suggested
11 that, at the beginning of the reaction, H2O has a negative effect on the
12 catalytic activities of ethylene oxidation over Ag/ZSMꢀ5 and Ag/Beta,
13 which make Ag/ZSMꢀ5 and Ag/Beta deactivate more quickly. However,
14 as the reaction goes on, H2O has a positive effect on the catalytic activity
15 of ethylene oxidation over Ag/ZSMꢀ5 and Ag/Beta, which make
16 Ag/ZSMꢀ5 and Ag/Beta show a certain activity of ethylene oxidation and
17 not completely deactivate. Furthermore, H2O only has a negative effect
18 on catalytic activity and stability of ethylene oxidation over Ag/Y and
19 Ag/Mordenite. Compared with the reaction in dry atmosphere, poor
20 catalytic activities and stabilities were shown in humid atmosphere for
21 both Ag/Y and Ag/Mordenite.
22
3.2. Physical parameters of Ag/zeolites. The textural properties of
ACS Paragon Plus Environment

Page 13 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
zeolites and the corresponding Ag/zeolite catalysts are presented in Table
S1. It could be seen that the surface areas and pore volumes of
Ag/zeolites were nearly identical to those of the zeolites, indicating the
structure of zeolites was largely maintained after the incorporation of
silver. All Ag/zeolites exhibited type I isotherms (Figure S1), implying
zeolites were dominated by a microporous structure. The present
hysteresis loops was associated with pore condensation into mesopores.11
Although the framework type and ring number for all the Ag/zeolites
were different, the micropore diameters of all Ag/zeolites were mainly
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10 distributed around 0.6~0.9 nm, besides, there were a ~2.0 nm pore
11 distribution in Ag/ZSMꢀ5 and a 1.1 nm pore distribution in Ag/Mordenite,
12 respectively (Figure S1a inset). The average diameter of Ag nanoparticles
13 (NPs) on Ag/zeolite catalysts before and after reaction (in dry atmosphere)
14 obtained from TEM images was about 2.3 nm, and Ag NPs dispersed
15 uniformly on all Ag/zeolite catalysts (Figure 2, Figure S2), indicating Ag
16 nanoparticles did not aggregate during the reaction. And it is difficult for
17 Ag particles to enter into the micropores of the zeolite, and the
18 micropores may not be blocked by the Ag particles. Considering the
19 molecule size of ethylene is 0.4 nm,16 the diffusion of ethylene into the
20 pores of these Ag/zeolite catalysts is relatively easy.
21
XPS analyses of Ag/zeolite catalysts before and after reaction (in dry
22 atmosphere) were performed to confirm the valence state of Ag in
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 44
1
2
3
4
5
6
7
8
9
1
2
catalysts, the results being shown in Figure 3. The typical binding
energies Ag 3d5/2 at 368.6 and 367.8 eV on the Ag/zeolite catalysts were
assigned to Ag0 and Ag+, respectively.17 All these Ag/zeolites suggested
that there are two valence states corresponding to metallic Ag0 and ionic
Ag+. It can be seen from Table S2 that the relative concentration of
Ag0/Ag+ of Ag/zeolite catalysts after being used in dry atmosphere
decreases 77%–93% compared with that of Ag/zeolite before reaction. It
is indicated that Ag0 species were oxidized to Ag+ in ethylene oxidation
reaction.
3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
5
6
7
8
9
10
3.3. Surface acidity of zeolite and Ag/zeolite catalysts. The acidic
11 properties of fresh zeolites and Ag/zeolites were examined by pyridine
12 FTꢀIR spectra in the region of 1580ꢀ1400 cmꢀ1 with the desorption
13 temperature of 150, 250, 350, 450 °C, as shown in Figure 4. The typical
14 IR bands for pyridine adsorbed on Brønsted and Lewis acid sites, which
15 were observed at 1540 cmꢀ1 and 1450 cmꢀ1, respectively.18ꢀ20 Both the
16 Brønsted and Lewis acid sites existed in all the fresh zeolites and
17 Ag/zeolites. Due to sorption of pyridine allowed differentiating between
18 Brønsted acid sites and Lewis acid sites, the concentrations of the
19 Brønsted and Lewis acid sites are summarized in Table 1. There are less
20 Brønsted acid sites and Lewis acid sites in zeolites than in the
21 corresponding Ag/zeolites. The number of acid sites of zeolites can be
22 modulated by Ag interaction with framework Al and nonꢀframework Al in
ACS Paragon Plus Environment

Page 15 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
zeolites.21,22
Both the Brønsted acid site and Lewis acid site concentrations of
Ag/zeolites determined by desorbing pyridine at 150 °C decreased as the
following order:
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ag/ZSMꢀ5 > Ag/Beta > Ag/Y > Ag/Mordenite.
Thus, it can be seen that the total acidity decreased as the above
order. Furthermore, the distribution of very weak, weak, medium and
strong Brønsted acid sites were different. All the Ag/zeolites displayed
mainly in very weak and weak Brønsted acid sites. Ag/Y showed an
10 increase in the number of strong Brønsted acid sites compared with other
11 Ag/zeolite catalysts. Moreover, it can be observed that C(B)/C(L) of
12 Ag/ZSMꢀ5 and Ag/Beta were greater than 1, indicating that Brønsted acid
13 sites in Ag/ZSMꢀ5 and Ag/Beta accounted for more than half of the total
14 acidity (Table 1). The C(B)/C(L) of Ag/Y and Ag/Mordenite were 0.57
15 and 0.11, respectively. The results showed that the number of Brønsted
16 acid sites was much less than Lewis acid sites, especially for
17 Ag/Mordenite. As it was presumed that the majority of the Brønsted acid
18 sites and a minor portion of Lewis acid sites were located in the zeolite
19 channels, the decrease of the C(B)/C(L) ratio could be explained by the
20 less tetraꢀcoordinated aluminum or decreasing accessibility of pyridine
21 molecules inside the zeolite channels.23 These results showed that there
22 were more Brønsted acid sites occurred than Lewis acid sites in both
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 44
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
Ag/ZSMꢀ5 and Ag/Beta, which showed better catalytic stabilities than
Ag/Y and Ag/Mordenite, in which there were more Lewis acid sites than
Brønsted acid sites.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.4. Surface acidity changes of Ag/zeolites during reaction. The
27Al MAS NMR spectra of fresh Ag/zeolites and used Ag/zeolites in dry
or humid atmosphere for 24 h are shown in Figure 5a. We named the
Ag/zeolites used in dry atmosphere for 24 h as Ag/zeolitesꢀdry; and
named the Ag/zeolites used in humid atmosphere for 24 h as
Ag/zeolitesꢀhumid. It can be seen that Ag/zeolitesꢀdry, Ag/Yꢀhumid and
10 Ag/Mordeniteꢀhumid deactivated completely, but Ag/ZSMꢀ5ꢀhumid and
11 Ag/Betaꢀhumid showed 65% and 30% activity, respectively (Figure 1b).
12 In general, the chemical shifts at 55 ppm and 0 ppm are attributed to
13 tetraꢀcoordinated and hexaꢀcoordinated aluminum, respectively.24 Table 2
14 summarizes the relative integrated areas of various aluminum species in
15 fresh Ag/zeolites, Ag/zeolitesꢀdry, and Ag/zeolitesꢀhumid. It was found
16 that all the fresh Ag/zeolites had tetraꢀcoordinated aluminum and
17 hexaꢀcoordinated aluminum. The relative amount of tetraꢀcoordinated
18 aluminum for fresh Ag/zeolites ranked as follows:
19
20
Ag/ZSMꢀ5 > Ag/Beta > Ag/Y > Ag/Mordenite.
The tetraꢀcoordinated aluminum results were in accord with pyridine
21 FTꢀIR findings. Compared with fresh Ag/zeolites, the integrated area of
22 tetraꢀcoordinated aluminum decreased for the Ag/zeoliteꢀdry. However,
ACS Paragon Plus Environment

Page 17 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
the integrated area of hexaꢀcoordinated aluminum increased for the
Ag/ZSMꢀ5ꢀdry, Ag/Betaꢀdry and Ag/Yꢀdry. The integrated area of
hexaꢀcoordinated aluminum decreased only for Ag/Mordeniteꢀdry. There
were less tetraꢀcoordinated aluminum in Ag/ZSMꢀ5ꢀhumid and
Ag/Betaꢀhumid than the corresponding fresh Ag/ZSMꢀ5 and Ag/Beta, but
more than the corresponding Ag/ZSMꢀ5ꢀdry and Ag/Betaꢀdry. For
Ag/Yꢀhumid and Ag/Mordeniteꢀhumid, the integrated area of
tetraꢀcoordinated aluminum was similar to that of Ag/Yꢀdry and
Ag/Mordeniteꢀdry. Thus, it is evident that more tetraꢀcoordinated
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10 aluminum in the Ag/zeolites contributes to a better catalytic activity.
The 1H MAS NMR spectra of fresh Ag/zeolites, Ag/zeolitesꢀdry, and
11
12 Ag/zeolitesꢀhumid are exhibited in Figure 5b and are used to probe acid
13 sites. Two groups of resonances were seen for all the fresh Ag/zeolites at
14 4.2 and 1.7 ppm, which were attributed to Brønsted acid sites and silanol
15 groups, respectively.24ꢀ26 The resonance at 2.9 ppm was assigned to
16 extraꢀframework AlOH species. The relative peak areas of the fresh
17 Ag/zeolites, Ag/zeolitesꢀdry, and Ag/zeolitesꢀhumid were listed in Table 3.
18 It could be observed that the percentage of Brønsted acid sites for
19 Ag/ZSMꢀ5 was the highest among all the fresh Ag/zeolites. The
20 percentage of Brønsted acid sites decreased in the following order:
21
22
Ag/ZSMꢀ5 > Ag/Beta > Ag/Y > Ag/Mordenite.
The absence of the 4.2 ppm feature in the spectra of Ag/zeolitesꢀdry,
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 44
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
Ag/Yꢀhumid, and Ag/Mordeniteꢀhumid confirmed that Brønsted acid site
disappeared as Ag/zeolites deactivated. However, although the number of
Brønsted acid sites of Ag/ZSMꢀ5ꢀhumid and Ag/Betaꢀhumid was less
than the corresponding fresh Ag/ZSMꢀ5 and Ag/Beta, there were a certain
number of Brønsted acid sites in Ag/ZSMꢀ5ꢀhumid and Ag/Betaꢀhumid. A
new broad peak at 5–9 ppm appeared for all the deactivated Ag/zeolites.
Fuchs et al. assigned this 5–9 ppm peak to residual water associated with
a Brønsted acid site, and this residual “water” was either due to water
molecules Hꢀbonds to Brønsted acid sites, or to H3O+ ions, or both.27
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10 White et al. also attributed the broad 5–9 ppm peak to residual water
11 feature.28 Thus, it revealed that no available Brønsted acid site for
12 ethylene oxidation led to the deactivation of Ag/zeolite catalysts. And the
13 dominant interaction that Brønsted acid sites were covered by residual
14 water which generated from ethylene oxidation resulted in the Ag/zeolite
15 deactivation.
16
3.5. TPO-MS analysis and adsorption-desorption properties.
17 TPOꢀMS is one of the most commonly used techniques to characterize
18 species deposition. Representative TPOꢀMS data of deactivated
19 Ag/zeolites are shown in Figure 6, which illustrate the evolution of CO2
20 and H2O, as well as O2 consumption. There was no O2 consumption and
21 no formation of CO2, indicating no deposited carbonaceous species was
22 removed from the deactivated Ag/zeolites. The H2O production profile
ACS Paragon Plus Environment

Page 19 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
exhibited a shape centered at around 150°C. From the H2O production
profiles, it was seen that the features of the desorption peaks, in terms of
peak area, indicated the amount as well as the nature of the H2O
deposition produced during ethylene oxidation reaction. It can be
observed that the amount of H2O production changed for different
Ag/zeolites and decreased as the following order:
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ag/Y > Ag/Beta > Ag/ZSMꢀ5 > Ag/Mordenite.
It was indicated that much more H2O deposited on the Ag/Y than
other Ag/zeolite catalysts. The H2O production profile comprised only
10 one distinct feature, a sharp peak from 50ꢀ250 °C, centered at 150 °C.
11 This indicated that the H2O deposition was of simple structure desorption
12 at only one temperature range.
13
Figure 7a exhibits the water adsorption isotherms on Ag/zeolites at
14 25 °C. The adsorption of water vapor on Ag/zeolites mainly occurred at
15 low relative pressures, which was related to the number of hydrophilic
16 sites in the Ag/zeolites.27,29 Due to the different SiO2/Al2O3 ratio of Y or
17 much more the strong Brønsted acid sites, Ag/Y had a dramatically larger
18 capacity of adsorbing water compared with other Ag/zeolites. The other
19 three Ag/zeolites with the same SiO2/Al2O3 ratio also had different
20 amounts of water adsorption. The decreased order was as follows:
21 Ag/Beta > Ag/ZSMꢀ5 > Ag/Mordenite, which was in line with the H2O
22 deposition amounts calculated from TPOꢀMS results. Figure 7b shows
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 44
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
H2OꢀTPD results of Ag/zeolites. It can be seen that the total desorbed
amount of H2O was equal to the adsorbed amount of H2O on the
Ag/zeolite catalysts. Thus, the adsorbed H2O could be desorbed
completely from the catalyst, indicating no chemical change of the
adsorbed H2O.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
C2H4ꢀtemperature programmed desorption (C2H4ꢀTPD) of Ag/zeolite
and H2OꢀAg/zeolite was performed to investigate the effect of water
preꢀadsorption on ethylene adsorption. H2OꢀAg/zeolite refers Ag/zeolite
catalyst which adsorbs water to reach saturation. C2H4ꢀTPD results gave
10 the C2H4 desorption amounts from fresh Ag/zeolite and H2OꢀAg/zeolite,
11 as shown in Figure S3. There are two kinds of desorption peaks in the
12 C2H4ꢀTPD curves of fresh Ag/zeolite catalysts. The peak at low
13 temperature may be ascribed to the presence of physical adsorption C2H4
14 species; and the other at high temperature is attributed to stronger
15 chemical adsorption C2H4 species.30,31 However, C2H4 desorption peak at
16 low temperature of H2OꢀAg/zeolite can be negligible, indicating that the
17 weak adsorption sites were firstly occupied by H2O. In addition, the
18 intensity of C2H4 desorption peak at high temperature of H2OꢀAg/zeolites
19 is much weaker than that of the fresh Ag/zeolites. It is indicated that the
20 total adsorption capacity of C2H4 on the H2OꢀAg/zeolite is considerably
21 decreased by the first adsorption of H2O compared with fresh Ag/zeolite.
22 Therefore, H2O functioned as a competitive adsorption species to inhibit
ACS Paragon Plus Environment

Page 21 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
the adsorption of C2H4 on the surface of Ag/zeolites.
CO2ꢀTPD profiles from Ag/zeolite catalysts were shown in Figure
S4. CO2 is easily desorbed from Ag/ZSMꢀ5 surface because the primary
desorption peak is located at relatively low temperature of 48 °C. CO2, as
the products of ethylene oxidation, its desorption capability may be
correlated to the catalytic performance of the Ag/zeolites.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
H2O is the product of a CꢀH oxidation reaction and H2O vapor
cannot be always removed from the reactants. From the TPOꢀMS and
H2O adsorptionꢀdesorption data, we can conclude that H2O adsorbing on
10 Ag/zeolites existed in the form of H2O molecule. The desorption amount
11 of H2O from the deactivated Ag/zeolites was in accord with the H2O
12 adsorption capacity of the fresh Ag/zeolites. That is to say, the adsorbed
13 H2O saturated the Ag/zeolite catalysts, and resulted in the deactivation of
14 the catalysts. Thus, the Ag/zeolite catalysts could be regenerated after
15 being treated in Helium flow for 2 h at 200 °C, which was attributed to
16 the adsorbed H2O being removed from the Ag/zeolite catalysts (Figure
17 S5).
18
3.6. Crystalline states of Ag/zeolites during deactivation. Figure 8
19 shows the XRD patterns of zeolites, fresh Ag/zeolites, the deactivated
20 Ag/zeolites in dry atmosphere, and the standard XRD patterns of zeolites.
21 As seen from the XRD patterns, all the zeolites presented their typical
22 diffraction patterns, which are in good agreement with the standard cards.
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 44
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
After Ag loading, neither the crystal phase nor the particle size of zeolites
changed. No obvious characteristic diffraction peak of Ag was detected,
presumably due to the small crystalline size and good dispersion of Ag,
which was also confirmed by TEM images.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compared with fresh catalysts, the intensity of all the peaks
increased for the deactivated Ag/ZSMꢀ5 and Ag/Beta, indicating that the
crystallinity improved for the deactivated Ag/ZSMꢀ5 and Ag/Beta. Acid
sites strength in zeolites will increase with the improved crystallinity,
because the strong interaction AlꢀO results in a weak bond OꢀH and
10 increases the strength of the proton in a well crystalline zeolite. The
11 increase of acid sites strength will deduce the weak protonꢀdenoting
12 ability, thus results in a poor catalytic performance in an acidic catalysis
13 reaction. The difference in the relative intensity of (111) and (533) planes
14 between the fresh and the deactivated Ag/Y might be related to
15 preferential crystallographic orientation. The same phenomenon
16 happened to Ag/Mordenite, the relative intensity of (200) and (202)
17 planes for the deactivated Ag/Mordenite changed when comparing with
18 fresh Ag/Mordenite. It is indicated that the crystalline structure of Ag/Y
19 and Ag/Mordenite was not so stable and there was a change in the
20 orientation along a preferred direction for the deactivated Ag/Y and
21 Ag/Mordenite.
22
3.7. Reaction and deactivation mechanism. The results of reaction
ACS Paragon Plus Environment

Page 23 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
investigation and characterizations indicated that the Brønsted acid sites
but not the Ag sites were the active sites. First, zeolites without loading
Ag nanoparticles did show the catalytic activities for ethylene oxidation.
Second, the Ag/zeolites with more Brønsted acid sites showed a better
catalytic stability. Most of the important, the deactivated Ag/zeolites had
no available Brønsted acid site. When the number of Brønsted acid sites
of Ag/ZSMꢀ5ꢀhumid and Ag/Betaꢀhumid reduced, the catalytic activities
of Ag/ZSMꢀ5ꢀhumid and Ag/Betaꢀhumid decreased. O2ꢀTPD
measurements were carried out on zeolite and Ag/zeolite catalysts to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10 study their oxygen species desorption behaviors, as shown in Figure S6.
11 Generally speaking, active oxygen species such as Oꢀ and O2 desorbed
ꢀ
12 below 300 °C, and lattice oxygen will desorb above 300 °C.32 The
13 desorption peaks of Ag/zeolites centered at the range 200ꢀ300 °C shifted
14 to lower temperature compared with the same peak of the corresponding
15 zeolites. Moreover, there is a new peak centered below 150 °C for
16 Ag/zeolites. It can be concluded that the desorption peak of surface active
17 oxygen species generated much easier on Ag/zeolites than zeolites, Thus,
18 Ag can promote the generation of surface active oxygen species on
19 Ag/zeolites, which may offer higher catalytic activities and stabilities of
20 Ag/zeolites compared with zeolites. Furthermore, the lower temperature
21 the corresponding desorption peak centered below 150 °C indicated that
22 it is easier to generate surface active oxygen species. The desorption
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 44
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
temperature of surface active oxygen species obeys the following order:
Ag/ZSMꢀ5 > Ag/Beta > Ag/Y > Ag/Mordenite, which is consistent with
the stability sequence for ethylene oxidation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
DRIFT spectra of Ag/ZSMꢀ5 and Ag/Mordenite in dry and humid
atmosphere are presented in Figure 9. The bands appearing at 2363 and
2339 cmꢀ1 are attributed to the CO2 asymmetric stretch and 1630 cmꢀ1 is
attributed to the H2O stretch.33 Both in dry and humid atmosphere, after
the 100 ppm C2H4 + 21% O2 + He mixed gas flowing through Ag/zeolites
for 30 min, formaldehyde species and carbonic acid species (1730 cmꢀ1
10 for ν(CO) ) were formed, and CO2 and H2O was produced. CO2, H2O and
11 formaldehyde species were dominant on Ag/zeolites after the catalysts
12 were exposed to a flow of O2 + He for 35 min. In dry atmosphere, a
13 narrow peak at 3645 cmꢀ1 was attributed to free OH groups. In humid
14 atmosphere, a broad band at 3200ꢀ3600 cmꢀ1 was attributed to
15 hydrogenꢀbonded OH groups and a peak at 3645 cmꢀ1 overlapped with
16 hydrogenꢀbonded OH groups was also attributed to free OH groups.34
17 Based on these results, we conclude that ethylene oxidation over
18 Ag/zeolites follows a pathway, that is C2H4 → HCHO* → CO2 + H2O.
19 The reaction mechanism of ethylene oxidation in dry atmosphere is
20 different from the mechanism in humid atmosphere. In dry atmosphere,
21 H2O adsorbed on Brønsted acid sites generates free OH groups and the
22 disappearance of Brønsted acid sites deduces the deactivation of
ACS Paragon Plus Environment

Page 25 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
Ag/zeolites. However, in humid atmosphere, except for a part of free OH
groups, hydrogenꢀbonded OH groups generate, which can keep a certain
of Brønsted acid sites.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A generalized mechanism was proposed to describe the conversion
of ethylene to CO2 and H2O and the deactivation process. The first step
involves the adsorption and activation of ethylene on a Brønsted acid site
of Ag/zeolite catalyst, forming adsorbed ethylene species, which is in
accord with reported ethylene activation on a Brønsted acid site.35ꢀ38 The
second step is the attack of active oxygen species on the adsorbed
10 ethylene species that are formed on the Brønsted acid sites of zeolite
11 catalyst. Because Ag can promote the generation of surface active oxygen
12 species from O2ꢀTPD results, an ethylene oxidation catalyst with
13 improved catalytic performance requires the joint participation of
14 Brønsted acid sites and metallic ions sites. The third step is that the
15 carbonꢀcarbon bond of adsorbed ethylene species is broken, and forms
16 formaldehyde species. Fukuoka et al. also reported HCHO adsorbed on Pt
17 is the intermediate species for ethylene oxidation over Pt/MCMꢀ41.7 The
18 fourth step, formaldehyde species are oxidized into carbonic acid species.
19 The last step, carbonic acid species decompose into CO2 and H2O. In dry
20 atmosphere, because the generated H2O adsorbed onto the Brønsted acid
21 site, the Brønsted acid site was consumed and resulted in catalyst
1
22 deactivation, which has been confirmed by H MAS NMR of fresh and
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 44
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
deactivated Ag/zeolite catalysts. However, when Ag/ZSMꢀ5 and Ag/Beta
were used in humid atmosphere, more H2O molecules existed in the
reaction gases, some Brønsted acid sites were covered by H2O and lost,
the other Brønsted acid sites around by H2O molecules might form
hydrogen bonds with H2O molecules. That is the reason for not
deactivation of Ag/ZSMꢀ5 and Ag/Beta when reacting in humid
atmosphere for 24 h. The detailed reaction and deactivation mechanism
of ethylene oxidation over Ag/zeolites are shown in Figure 10.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
4. CONCLUSIONS
10 In summary, microporous Ag/zeolite (ZSMꢀ5, Beta, Y and Mordenite)
11 catalysts have been found to exhibit high catalytic performances for
12 ethylene oxidation at room temperature. The initial activity of ethylene
13 oxidation over all the Ag/zeolite catalysts is 100% at 25 °C both in dry
14 and humid atmosphere, although the maintenance time of the initial
15 activity over the Ag/zeolites is different. Based on characterization results
16 of fresh and deactivated Ag/zeolites, we can conclude that Brønsted acid
17 site is the catalytic active site for ethylene oxidation. Water vapor in the
18 reactant atmosphere has a significant effect on the catalytic performances
19 of the Ag/zeolites. The main reason for catalyst deactivation is that H2O
20 molecules adsorbed onto Brønsted acid sites result in the disappearance
21 of available Brønsted acid sites in Ag/zeolites. It is desirable that zeolite
ACS Paragon Plus Environment

Page 27 of 44
ACS Catalysis
1
2
3
4
5
6
1
2
catalysts may have potential application in ethylene oxidation in the
solution of Brønsted acid sites stability under water vapor.
7
8
9
3
Acknowledgement
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
5
6
7
This work is financially supported by the National Key R&D Program of
China (2016YFC0204203) and National Natural Science Foundation of
China (21337003, 21477148 21477149 and 21777175), and Key
Research Program of Frontier Sciences, CAS (QYZDBꢀSSWꢀDQC018).
8
9
Supporting information
The Supporting Information is available free of charge on the ACS
10 Publications website.
11
BET surface area, pore volume; relative concentration change of
12 Ag0/Ag+1 before and after reaction; HRTEM images and particle diameter
13 distributions; C2H4ꢀTPD; CO2ꢀTPD; regeneration test; O2ꢀTPD and gas
14 chromatography results.
15 References
16 (1) Saltveit, M. E. Postharvest Boil. Tec. 1999, 15, 279–292.
17 (2) Keller, N. M.; Ducamp, N.; Robert, D.; Keller, V. Chem. Rev. 2013,
18 113, 5029–5070.
19 (3) Kumara, D.; Hana, Y. F.; Goodmana, D. W. Top. Catal. 2007, 46,
20 169–174.
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 44
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
(4) Li, W. C.; Zhang, Z. X.; Wang, J. T.; Qiao, W. M.; Long, D. H.; Ling,
L. C. Chem. Eng. J. 2016, 293, 243–251.
(5) Ahn, H. G.; Choi, B. M.; Lee, D. J. J. Nanosci. Nanotechno. 2006, 6,
3599–3603.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(6) Ma, C. Y.; Mu, Z.; Li, J. J.; Jin, Y. G.; Cheng, J.; Lu, G. Q.; Hao, Z. P.;
Qiao, S. Z. J. Am. Chem. Soc. 2010, 132, 2608–2613.
(7) Jiang, C. X.; Hara, K.; Fukuoka, A. Angew. Chem. Int. Ed. 2013, 52,
6265–6268.
(8) Coma, A. Chem. Rev. 1995, 95, 559–614.
10 (9) Lin, B. M.; Zhang, Q. H.; Wang, Y. Ind. Eng. Chem. Res. 2009, 48,
11 10788–10795.
12 (10) Kubo, K.; Iida, H.; Namba, S.; Igarashi, A. Appl. Catal. A 2015, 489,
13 272–279.
14 (11) Chen, H.; Wang, Q. F.; Zhang, X. W.; Wang, L. Ind. Eng. Chem. Res.
15 2014, 53, 19916–19924.
16 (12) Sultana, A.; Fujitani, T. Catal. Commun. 2017, 88, 26–29.
17 (13) Xu, B.; Sievers, C.; Hong, S. B.; Prins, R.; Bokhoven, J. A. J. Catal.
18 2006, 244, 163–168.
19 (14) Huang, J.; Jiang, Y. J.; Marthala, V. R. R.; Bressel, A.; Frey, J.;
20 Hunger, M. J. Catal. 2009, 263, 277–283.
21 (15) Emeis, C. A. J. Catal. 1993, 141, 347–354.
22 (16) Li, J. R.; Kuppler, R. J.; Zhou, H. C. Chem. Soc. Rev. 2009, 38,
ACS Paragon Plus Environment

Page 29 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
1477–1504.
(17) Hoflund, G. B.; Hazos, Z. F. Phys. Rev. B 2000, 62, 11126−11133.
(18) Farneth, W. E. Chem. Rev. 1995, 95, 615–635.
(19) Zhao, X. B.; Guo, X. W.; Wang, X. S. Energ. Fuel. 2006, 20,
1388–1391.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(20) Zhang, J. G.; Qian, W. Z.; Kong, C. Y.; Wei, F. ACS Catal. 2015, 5,
2982–2988.
(21) Wang, S. R.; Guo, W. W.; Zhu, L. J.; Wang, H. X.; Qiu, K. Z.; Cen,
K. J. Phys. Chem. C, 2015, 119, 524–533.
10 (22) Blasco, T.; Boronat, M.; Concepción, P.; Corma, A.; Law, D.;
11 Alejandro, J.; Moya, V. Angew. Chem. Int. Ed. 2007, 46, 3938–3941.
12 (23) Isernia, L. F. Mater. Res. 2013, 16, 792–802.
13 (24) Yu, Z. W.; Li, S. H.; Wang, Q.; Zheng, A. M.; Jun, X.; Chen, L.;
14 Deng, F. J. Phys. Chem. C 2011, 115, 22320–22327.
15 (25) Wang, X.; Dai, W. L.; Wu, G. J.; Li, L. D.; Guan, N. J.; Hunger, M.
16 Micropor. Mesopor. Mat. 2012, 151, 99–106.
17 (26) Dai, W. L.; Sun, X. M.; Tang, B.; Wu, G. J.; Li, L. D.; Guan, N. J.;
18 Hunger, M. J. Catal. 2014, 314, 10–20.
19 (27) Bellat, J. P.; Paulin, C.; Jeffroy, M.; Boutin, A.; Paillaud, J. L.;
20 Patarin, J.; Lella, A. D.; Fuchs, A. J. Phys. Chem. C 2009, 113,
21 8287–8295.
22 (28) Chen, K.; Damron, Z. J.; Pearson, C.; Resasco, D.; Zhang, L.; White,
ACS Paragon Plus Environment