Methane Conversion to Ethylene and Aromatics on PtSn Catalysts

Article abstract of DOI:10.1021/acscatal.6b02724

Pt and PtSn catalysts supported on SiO₂ and H-ZSM-5 were studied for methane conversion under nonoxidative conditions. Addition of Sn to Pt/SiO₂ increased the turnover frequency for production of ethylene by a factor of 3, and pretreatment of the catalyst at 1123 K reduced the extent of coke formation. Pt and PtSn catalysts supported on H-ZSM-5 zeolite were prepared to improve the activity and selectivity to non-coke products. Ethylene formation rates were 20 times faster over a PtSn(1:3)/H-ZSM-5 catalyst with SiO₂:Al₂O₃ = 280 in comparison to those over PtSn(3:1)/SiO₂. H-ZSM-5-supported catalysts were also active for the formation of aromatics, and the rates of benzene and naphthalene formation were increased by using more acidic H-ZSM-5 supports. These catalysts operate through a bifunctional mechanism, in which ethylene is first produced on highly dispersed PtSn nanoparticles and then is subsequently converted to benzene and naphthalene on Br?nsted acid sites within the zeolite support. The most active and stable PtSn catalyst forms carbon products at a rate, 2.5 mmol of C/((mol of Pt) s), which is comparable to that of state-of-the-art Mo/H-ZSM-5 catalysts with same metal loading operated under similar conditions (1.8 mmol of C/((mol of Mo) s)). Scanning transmission electron microscopy measurements suggest the presence of smaller Pt nanoparticles on H-ZSM-5-supported catalysts, in comparison to SiO₂-supported catalysts, as a possible source of their high activity. A microkinetic model of methane conversion on Pt and PtSn surfaces, built using results from density functional theory calculations, predicts higher coupling rates on bimetallic and stepped surfaces, supporting the experimental observations that relate the high catalytic activity to small PtSn particles. (Figure Presented).

Full text of DOI:10.1021/acscatal.6b02724

Subscriber access provided by Fudan University
Article
Methane Conversion to Ethylene and Aromatics on PtSn Catalysts
Duygu Gerceker, Ali Hussain Motagamwala, Keishla R. Rivera-Dones, James
B. Miller, George W. Huber, Manos Mavrikakis, and James A. Dumesic
ACS Catal., Just Accepted Manuscript • Publication Date (Web): 03 Feb 2017
Downloaded from http://pubs.acs.org on February 3, 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 50
ACS Catalysis
1
2
3
4
5
6
7
8
9
Methane Conversion to Ethylene and Aromatics on
PtSn Catalysts
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
a,b
a
a
Duygu Gerceker , Ali Hussain Motagamwala , Keishla R. Rivera-Dones , James B. Miller ,
a
a
a
George W. Huber , Manos Mavrikakis , James A. Dumesic *
a
Department of Chemical and Biological Engineering, University of Wisconsin-Madison,
Madison, Wisconsin 53706, USA
b
Great Lakes Bioenergy Research Center, University of Wisconsin-Madison, 1552 University
Ave., Madison, WI 53726, USA
Abstract: Pt and PtSn catalysts supported on SiO and H-ZSM-5 were studied for methane
2
conversion under non-oxidative conditions. Addition of Sn to Pt/SiO increased the turnover
2
frequency for production of ethylene by a factor of 3, and pretreatment of the catalyst at 1123 K
reduced the extent of coke formation. Pt and Pt-Sn catalysts supported on H-ZSM-5 zeolite were
prepared to improve the activity and selectivity to non-coke products. Ethylene formation rates
were 20 times faster over a Pt-Sn(1:3)/H-ZSM-5 catalyst with SiO :Al O = 280, than over Pt-
2
2
3
Sn(3:1)/SiO . H-ZSM-5 supported catalysts were also active for the formation of aromatics, and
2
the rates of benzene and naphthalene formation were increased by using more acidic H-ZSM-5
supports. These catalysts operate through a bifunctional mechanism, in which ethylene is first
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 50
1
2
3
4
5
6
7
8
9
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
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
produced on highly dispersed PtSn nanoparticles, and then is subsequently converted to benzene
and naphthalene on Brønsted acid sites within the zeolite support. The most active and stable
PtSn catalyst forms carbon products at a rate, 2.5 mmol-C/mol-Pt.s, that is comparable to that of
state-of-the-art Mo/H-ZSM-5 catalysts with same metal loading and operated under similar
conditions (1.8 mmol-C/mol-Mo.s). Scanning transmission electron microscopy measurements
suggest the presence of smaller Pt nanoparticles on H-ZSM-5 supported catalysts, compared to
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
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
SiO -supported catalysts, as a possible source of their high activity. A microkinetic model of
2
methane conversion on Pt and PtSn surfaces, built using results from density functional theory
calculations, predicts higher coupling rates on bimetallic and stepped surfaces, supporting the
experimental observations that relate the high catalytic activity to small PtSn particles.
Keywords: methane conversion, platinum, tin, H-ZSM-5, ethylene, aromatics, microkinetic
model, particle size
1. Introduction
1
Methane is a stable molecule that is difficult to activate without significant energy input. The
goal of activating methane and converting it into more valuable chemicals has long been
2
recognized as an important challenge. Many studies have addressed oxidative coupling of
methane, usually over catalysts that are composed of or modified by alkali and alkali earth metal
1
compounds. As an alternative approach, dehydroaromatization (DHA), a non-oxidative coupling
of methane to ethane, ethylene and aromatics, has received significant attention. However,
thermodynamic considerations limit methane conversion of dehydroaromatization to less than
1
0% for C products, even at high temperatures (970 K), while formation of carbon is highly
2
2
favored. Accumulation of carbon on the catalyst surface results in deactivation and loss of
2
ACS Paragon Plus Environment

Page 3 of 50
ACS Catalysis
1
2
3
4
5
6
7
8
9
selectivity. A promising class of catalysts for methane dehydroaromatization is comprised of
shape selective silica-alumina zeolites, such as ZSM-5, ZSM-8, ZSM-11 and MCM-22 modified
3
, 4, 5
with transition metal species, such as Mo.
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
Early work by Wang, et al. reported that benzene could be produced on Mo or Zn modified
6
ZSM-5 catalysts at 973 K under non-oxidizing conditions. A bifunctional mechanism was
proposed, with activation of methane on molybdenum containing sites, dimerization of CH3
radicals to form ethane and the primary product ethylene, followed by aromatization of ethylene
7
,8
to benzene on Br
ø
nsted acid sites. The presence of Mo C was found to be necessary for the
2
8,9,10
formation of hydrocarbons.
Furthermore, partial oxidation of Mo C resulted in higher
2
hydrocarbon formation rates, suggesting that Mo C-MoO with an oxygen deficiency provided
2
2
1
1
the active catalytic sites for methane activation. Pretreatment of Mo/H-ZSM-5 with an n-
butane/hydrogen mixture formed β-Mo C and α-MoC species that were responsible for the
2
1-x
1
2
higher production rate of aromatic compounds and slower deactivation of the catalyst.
A
reaction mechanism involving the formation of acetylene as an important reaction intermediate
1
3
was suggested.
Various modifications in the reaction conditions and catalyst formulation for Mo/H-ZSM-5
have been studied to improve catalytic activity and minimize deactivation due to coke formation.
Co-feeding with H shifted the equilibrium back toward methane, resulting in both decreased
2
14
coke deposition and lower hydrocarbon formation rates. Modification of Mo/H-ZSM-5 with Rh
provided a catalyst that was stable when operated at 1023 K with a 6% H /CH feed. Rh was
2
4
proposed to suppress the coke formation on Brønsted acid sites, thereby enhancing catalyst
1
5
stability. Co-feeding of up to 5% CO was found to result in lower coke formation rates on
2
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 50
1
2
3
4
5
6
7
8
9
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
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
16, 17
Mo/H-ZSM-5 and Fe or Co modified Mo/H-ZSM-5.
High pressures of CO were observed to
2
1
8
oxidize the Mo C formed during the induction period, resulting in lower catalytic activity.
2
Under cyclic operation between CH and H flows, addition of Fe to Mo/H-ZSM-5 improved
4
2
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
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
catalyst stability through formation of carbon nanotubes with H release under CH flow. The
2
4
carbon nanotubes were subsequently removed by flowing H , regenerating the active sites. The
2
1
9
cyclic evolution of H provided decreased coke formation. In a recent study of the effect of Co
2
addition in the range of 0-1.6%, a Mo/H-ZSM-5 catalyst containing 0.8 wt% Co showed optimal
2
0
performance for benzene formation and favored formation of heavier products. The effects of
21
22
23
Ga , Zr , W and Ru modification of Mo/H-ZSM-5 system were also investigated to achieve
2
4,25
more stable and active catalysts with decreased coke formation.
The structures and locations of the carbonaceous deposits that lead to deactivation of Mo/H-
ZSM-5 have been examined to gain insight about their interactions with methane and formation
of intermediates during reaction. XPS characterization of spent Mo/H-ZSM-5 samples revealed
three types of carbon species on the spent catalyst: the first type is a graphitic-like carbon present
in the zeolite channels; the second type is carbidic-like carbon in Mo C mainly located at the
2
outer surface of the zeolite; and the third type is hydrogen-poor pre-graphitic type, the quantity
of which increased with time on stream. Located on the outer portions of the zeolite, the third
2
6
carbon type was thought to be responsible for deactivation of the catalyst. In another study, it
was shown that some of the carbon species deposited on the surface were reactive and could be
27
hydrogenated to yield ethylene and benzene at 839 K and 913 K, respectively. In a recent
study, three stages of Mo-H-ZSM-5 deactivation were identified. The first stage was rapid
deactivation due to carburization of MoO to Mo C accompanied by carbon deposition on the
3
2
external surface of the catalyst. The second stage exhibited slower deactivation due to coke being
4
ACS Paragon Plus Environment

Page 5 of 50
ACS Catalysis
1
2
3
4
5
6
7
8
9
deposited in the micropores of the support during benzene and naphthalene formation. A third
deactivation period was characterized by a sharp decrease in aromatics formation, attributed to
2
8
the blocking of the channels with carbon deposits near the pore mouths.
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
In addition to studies that investigated the activity of various transition metals such as Fe, Re,
V and Cr supported on SiO and H-ZSM-5, methane conversion has also been studied on Pt-
2
2
9,30,31,32
containing catalysts.
In an early work that addressed low temperature activation of
methane, a 6 wt% Pt/SiO catalyst was first exposed to methane at 523 K and then flushed with
2
H , resulting in the formation of hydrocarbon products ranging from C to C Pt/HX and Pt/HY
2
2
7.
catalysts with the same Pt loading exhibited higher methane conversion and higher formation
rates of C and C products. The catalysts were not effective at higher temperatures because of
3
5
3
3,34
coke deposition on Pt.
At 973 K, a 2 wt% Pt/H-ZSM-5 catalyst exhibited low methane
conversions. However, adding Pt to Mo/H-ZSM-5 improved the catalytic activity of Mo/H-
3
5,36
ZSM-5 by decreasing the rate of coke formation.
The active sites of the catalyst were
3
7
hypothesized to be associated with a Pt-MoO -H-ZSM-5 structure on the catalyst. At 973 K,
3
addition of 0.1 wt% Sn to the Pt-Mo/H-ZSM-5 catalyst yielded slightly lower methane
conversion while decreasing the extent of coke deposition and increasing the selectivity to
aromatics. Catalysts prepared by sequential impregnation of Pt followed by Sn exhibited higher
benzene selectivity and better coke mitigation compared to catalysts prepared by co-
3
8, 39
impregnation.
In this work, we present a study of non-oxidative methane conversion on Pt-Sn bimetallic
catalysts that exhibit high activity, selectivity, and stability for production of ethylene. We also
show that the addition of Brønsted acid sites to Pt-Sn/ZSM-5 leads to catalysts that are active for
the conversion of methane to benzene and naphthalene, through a bifunctional mechanism,
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 50
1
2
3
4
5
6
7
8
9
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
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
whereby ethylene is produced on highly dispersed PtSn nanoparticles, followed by its conversion
to benzene and naphthalene on Brønsted acid sites within the zeolite support. A microkinetic
model built using results from density functional theory calculations (DFT) for terrace and step
sites of Pt and PtSn surfaces predicts higher rates of ethylene formation on bimetallic step sites.
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
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2. Experimental
Catalyst preparation: Catalysts consisting of Pt and Pt-Sn supported on SiO and H-ZSM-5
2
were prepared by impregnation. For SiO -supported catalysts, Davisil (Grade 646, 35-60 mesh,
2
Sigma Aldrich) was impregnated with an aqueous solution of tetraammine platinum (II) nitrate
precursor (Sigma Aldrich, 99%) to the target loading of 6 wt% Pt. After drying overnight at 373
3
K, the catalyst was calcined under air flow (50 cm (STP)/min) at 573 K for 2 h, followed by
3
reduction in flowing hydrogen (50 cm (STP)/min) at 773 K for 2 h. Prior to contacting with air,
the catalyst was cooled to room temperature under Ar flow and passivated with flowing 1%
O /Ar for 30 min. Addition of Sn to Pt/SiO was performed through evaporative impregnation.
2
2
To achieve an overall atomic ratio of Pt/Sn equal to 3:1, the required amount of the Sn precursor,
tri-n-butyl tin (Strem Chemicals Inc, min 94%), was dissolved in 80 mL n-pentane (Sigma
Aldrich), stirred for 5 min and then Pt/SiO was added. The mixture was stirred until all the
2
pentane had evaporated, and the catalyst was then dried for 30 min in an oven at 373 K.
Calcination, reduction and passivation steps were repeated for the resulting catalyst. Catalysts
prepared via this procedure were labeled as Pt/SiO and PtSn/SiO . For zeolite-supported
2
2
catalysts, four different ZSM-5 supports with nominal SiO /Al O ratios of 23, 50, 80 and 280
2
2
3
+
were obtained in NH form (Zeolyst). Prior to impregnation, supports were calcined in air at
4
873 K for 18 h to remove ammonia and obtain the H-ZSM-5 form. Based on the targeted Pt-Sn
atomic ratio for a 0.58 wt% Pt catalyst, the required amount of the Sn precursor was dissolved in
6
ACS Paragon Plus Environment

Page 7 of 50
ACS Catalysis
1
2
3
4
5
6
7
8
9
methanol and impregnated onto the zeolite support. After drying overnight at 373 K, the catalyst
3
was calcined under air flow (50 cm (STP)/min) at 573 K for 2 h. Subsequently, Sn/H-ZSM-5
was impregnated with Pt using an aqueous solution of tetraammine platinum (II) nitrate
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
precursor. The resulting catalyst was dried overnight at 373 K, treated in Ar (50 cm (STP)/min)
at 573 K for 1 h, and reduced at 773 K for 6 h. Prior to contacting with air, the catalyst was
cooled to room temperature under Ar flow and passivated with flowing 1% O /Ar for 30 min.
2
Catalysts prepared by this procedure were labeled as Pt/Z-280, PtSn(1:1)/Z-280, PtSn(1:2)/Z-
280, PtSn(1:2.5)/Z-280, PtSn(1:3)/Z-280, PtSn(1:2)/Z-80, PtSn(1:2)/Z-50 and PtSn(1:2)/Z-23
based on their composition and SiO /Al O ratio. For zeolite supported samples, higher Sn
2
2
3
4
0
content was targeted to overcome the possible interaction of Sn with alumina sites.
Catalyst characterization: Platinum surface site densities of the fresh catalysts were
determined by CO chemisorption at 308 K (Micromeritics, ASAP, 2020C Analyzer). Turnover
frequencies (TOF) for product formation were normalized by the Pt site density value of the
catalyst.
Textural properties of the blank and impregnated H-ZSM-5 catalysts were determined from N2
adsorption-desorption isotherms obtained at 77 K using the same instrument. Prior to the
adsorption measurements, samples were degassed at 423 K and evacuated for 6 h. Specific
surface areas of the catalysts were calculated by the BET method.
The acid site densities of fresh and spent catalysts were measured by NH –temperature
3
programmed desorption (NH -TPD). 100 mg of each catalyst sample was heated to 473 K under
3
helium flow and kept at constant temperature for 1 h. After cooling to 423 K, NH was adsorbed
3
on the sample from a 1% NH /He flow for 45 min. Following a purge with helium at 423 K for
3
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 50
1
2
3
4
5
6
7
8
9
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
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
9
0 min to remove weakly adsorbed NH , the sample was heated to 973 K under helium flow with
3
a 10 K/min ramp and held at 973 K for 2 h.
Thermogravimetric analysis (TGA) was performed on spent catalysts to quantify the amount of
deposited coke (TGA 500). In this analysis, 25-30 mg of the spent catalyst sample was heated
from room temperature to 973 K with a ramp of 20 K/min under oxygen flow. The change in the
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
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
sample mass due to combustion of coke and release of CO was used in the calculation of the
2
amount of coke deposition.
Scanning transmission electron microscopy (STEM) imaging was performed using a FEI Titan
STEM with Cs aberration correction operated at 200 kV in high-angle annular dark field
(HAADF) mode. Catalyst samples were suspended in ethanol then deposited on a holey carbon
Cu TEM grid. Samples were plasma cleaned before being loaded into the microscope. ImageJ
software was used to analyze the micrographs and to calculate particle size distributions.
Methods: Reaction kinetics experiments were conducted in a ½ - inch stainless steel down
flow reactor. The required mass of catalyst (2 g for SiO supported catalysts, 1 g for zeolite
2
supported catalysts) was loaded in the reactor in three sections separated with quartz wool to
minimize channeling. Following the reduction with H for 1 h at 773 K, the catalyst was heated
2
3
to 973 K under flowing He, and then 42 cm (STP)/min CH flow was introduced as feed. For the
4
experiments comparing the effect of support acidity, 0.25 g catalyst was packed with 2-2.5 g of
3
silica chips for each experiment, and the CH flowrate was decreased to 10.5 cm (STP)/min. For
4
certain runs with the SiO -supported catalysts, the catalysts were further reduced at 1123 K for 2
2
h prior to the reaction. For experiments with H co-feeding, H gas corresponding to 5% of the
2
2
CH flow was introduced from a 10% H /He tank (Airgas). Analysis of the products, except
4
2
naphthalene, was carried out by an online gas chromatograph (Shimadzu Corp., GC-2014)
8
ACS Paragon Plus Environment

Page 9 of 50
ACS Catalysis
1
2
3
4
5
6
7
8
9
equipped with a Rt®-Alumina BOND/MAPD capillary column and a flame ionization detector.
To detect naphthalene, the same gas chromatograph equipped with an Agilent J&W GS-GasPro
column was used. All the lines between reactor outlet and GC sampling loop inlet were heat
traced to 493 K to prevent product condensation.
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. Results and Discussion
3.1. Catalyst Characterization:
CO chemisorption: The platinum surface site densities of the catalysts determined by CO
chemisorption are shown in Table 1. Because all the catalysts were initially reduced at 773 K
during catalyst preparation, the CO uptakes of the fresh catalysts were measured following an in
situ reduction at 773 K. On both supports, compared to monometallic Pt catalysts, the presence
of Sn decreased the number of Pt sites accessible for CO chemisorption, suggesting that
formation of bimetallic particles took place. To identify the effect of treatment at high
temperatures, samples of Pt/SiO and PtSn/SiO catalysts were reduced at 1123 K, and samples
2
2
of Pt/Z-280 and PtSn(1:3)/Z-280 were reduced at 973 K, prior to CO chemisorption
measurements. The number of surface Pt sites on the Pt/SiO catalyst decreased by 75% after
2
treatment at 1123 K, and surface site density on PtSn/SiO decreased only by 55%, suggesting
2
that the presence of Sn decreased the extent of sintering. The number of surface Pt sites on Pt/Z-
2
80 and PtSn(1:3)/Z-280 decreased by approximately 50% after treatment at 973 K.
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 50
1
2
3
4
5
6
7
8
9
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
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 1. Platinum surface site densities of catalysts determined by CO chemisorption
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
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Pretreatment
Temperature
Pretreatment
Temperature
(K)
Site density
(ꢀmol Pt/g)
Site density
(ꢀmol Pt/g)
Catalyst
Catalyst
(K)
#
#
Pt/SiO2
773
773
773
773
773
773
773
76
30
Pt/SiO2
1123
1123
973
773
773
773
973
19
13
#,a
#,a
PtSn/SiO2
PtSn/SiO2
Pt/Z-280
7.5
6.6
6.3
8.9
3.7
Pt/Z-280
3.4
5.6
4.8
4.9
1.8
PtSn(1:1)/Z-280
PtSn(1:2)/Z-80
PtSn(1:2)/Z-23
PtSn(1:2)/Z-280
PtSn(1:2)/Z-50
PtSn(1:2.5)/Z-280
PtSn(1:3)/Z-280
PtSn(1:3)/Z-280
#
SiO supported catalysts had Pt loading of 6 wt% and zeolite supported catalysts had Pt
2
loading of 0.58 wt%.
a
PtSn/SiO catalysts have an overall atomic ratio of Pt/Sn equal to 3:1
2
Surface area characterization: N physisorption was employed to obtain surface areas of the
2
blank supports and metal impregnated supports with varying acidity. BET surface areas of the
6
,41
blank H-ZSM-5 supports were in agreement with values reported in the literature.
As shown
in Table 2, catalysts impregnated with only-Sn and with Sn followed by Pt, had lower BET
surface areas than the blank supports. This decrease suggests that the impregnated metals entered
the pores of the support. The BET surface area of a spent PtSn(1:3)/Z-280 catalyst sample
yielded an even lower BET surface area, probably due blockage of the pores by carbonaceous
deposits formed during methane conversion reactions.
10
ACS Paragon Plus Environment

Page 11 of 50
ACS Catalysis
1
2
3
4
5
6
7
8
9
Table 2. BET surface areas of supports and catalysts measured by N physisorption*
2
2
2
Support
BET surface area (m g)
Catalyst
BET surface area (m g)
/
/
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
Z-23
Z-50
Z-80
Z-280
355
397
351
378
Pt/Z-280
Sn/Z-280
384
342
333
306
PtSn(1:3)/Z-280
PtSn(1:3)/Z-280-spent
2
(
* +/- 5 m /g )
NH -Temperature Programmed Desorption: The acid site densities of the blank supports,
3
the impregnated catalysts, and the spent samples are shown in Table 3. As expected, the
calculated acid site density from the desorption of ammonia decreased with increasing
SiO :Al O ratio of the H-ZSM-5 support (from 850 ꢀmol/g to 62 ꢀmol/g). The same trend was
2
2
3
observed with the samples of fresh catalysts on which Sn and Pt were impregnated (from 710
mol/g to 41 ꢀmol/g). Similarly, on the spent catalysts collected at the end of the reaction, the
acid site density decreased with increasing SiO :Al O ratio; however, the range of the values
ꢀ
2
2
3
was narrower (from 140 ꢀmol/g to 90 ꢀmol/g). This difference may be related to the
accumulation of coke deposits on the acid sites of the support.
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 50
1
2
3
4
5
6
7
8
9
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
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 3. Acid site density of supports and catalysts determined by NH -TPD
3
Acid site
density
Acid site
density
Acid site
density
(ꢀmol/g)
Support
Catalyst (fresh)
Catalyst (spent)
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
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
ꢀmol/g)
(ꢀmol/g)
Z-23
Z-50
Z-80
Z-280
850
PtSn(1:2)/Z-23
PtSn(1:2)/Z-50
PtSn(1:2)/Z-80
PtSn(1:2)/Z-280
710
280
250
41
PtSn(1:2)/Z-23
PtSn(1:2)/Z-50
PtSn(1:2)/Z-80
PtSn(1:2)/Z-280
140
120
120
90
270
250
62
3
.2. Effect of pretreatment temperature on SiO supported catalysts:
2
Pt/SiO and bimetallic PtSn/SiO catalysts, pretreated at 773 and 1123 K, were studied for coke
2
2
formation trends under reaction conditions both with and without 5% H co-feeding. It was
2
observed that catalysts pretreated at 773 K had carbon balances as low as 65% during the initial
period of reaction (e.g., 0.5 h) due to coke deposition. Pretreatment at 1123 K was performed to
decrease the surface concentration of silanol groups on the silica surface, in an attempt to
minimize the extent of coke formation and to improve the carbon balance. Spent catalysts were
collected at the end of 6 h time on stream for each reaction, and thermogravimetric analyses
(TGA) were carried out. In parallel, the difference between the inlet and outlet moles of carbon
flow of each catalyst integrated over the same time on stream period of 6 h was used to obtain an
estimate for the extent of carbon deposition on the catalyst. The calculated amounts of deposited
carbon (from TGA analysis after reaction) and carbon loss (from comparison of carbon flow
rates into and out of the reactor versus time on stream) for SiO -supported catalysts are shown in
2
Table 4.
12
ACS Paragon Plus Environment

Page 13 of 50
ACS Catalysis
1
2
3
4
5
6
7
8
9
Table 4. Carbon deposition and calculated carbon loss for SiO -supported catalysts
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
Pretreatment at 773 K
Carbon deposition Carbon loss
Pretreatment at 1123 K
Carbon deposition Carbon loss
Catalyst
Pt/SiO2
(TGA) (mmol/gcat) (mmol/gcat) (TGA) (mmol/gcat) (mmol/gcat)
14
13
12
4.1
15
14
21
7.7
6.7
3.6
5.1
1.8
6.0
11
Pt/SiO – 5%H
2
2
#
PtSn/SiO2
6.1
4.5
#
2
PtSn/SiO – 5%H
2
#
PtSn/SiO catalysts had overall atomic ratio of Pt/Sn equal to 3:1
2
As seen from carbon deposition measurements and the calculated carbon losses in each
experiment, catalysts pretreated at 1123 K had lower coke deposition as measured by TGA and
had higher closure of the carbon balance. As the calculated carbon losses were found to be
higher in amount than the carbon detected on the spent catalyst samples in TGA analyses, it is
likely that some undetected carbon was deposited on the reactor walls or elsewhere in the reactor
system. Hydrogen co-feeding decreased the extent of coke formation, particularly on the
PtSn/SiO catalyst. When the catalyst was pretreated at 773 K, addition of 5% H resulted in
2
2
6
0% decrease in coke deposition.
3.3. Effect of Sn addition on catalytic activity of SiO supported catalysts: Figure 1a
2
compares the TOF values for ethylene formation as a function of time on stream for the Pt/SiO2
and bimetallic PtSn(3:1)/SiO catalysts pretreated at 1123 K. Addition of Sn to Pt/SiO2
2
increased the ethylene TOF, and after five hours on stream, when ethylene TOF achieved a
steady value, the PtSn/SiO catalyst had approximately 3 times higher ethylene TOF than
2
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 50
1
2
3
4
5
6
7
8
9
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
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Pt/SiO Figure 1b shows the same comparison for the benzene TOF. The PtSn/SiO catalyst
2
.
2
-3
-1
pretreated at 1123 K yielded a benzene TOF of 8.6 x 10 ks at the end of five hours on stream,
which was approximately 2% of its ethylene TOF. Addition of Sn to Pt/SiO increased the initial
2
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
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
benzene activity, and the steady state benzene TOF over PtSn/SiO was approximately 2 times
2
higher than Pt/SiO . Figure 1c and Figure 1d display the methane conversion to detectable
2
products (left axis) and the selectivities of the catalyst towards ethylene, ethane and benzene
(
right axis) over total hydrocarbon products excluding coke for Pt/SiO and PtSn/SiO ,
2 2
respectively. The PtSn/SiO catalyst yielded higher methane conversion to products and higher
2
selectivity towards ethylene compared to Pt/SiO . Also, both catalysts showed low selectivity
2
towards ethane and benzene.
14
ACS Paragon Plus Environment

Page 15 of 50
ACS Catalysis
1
2
3
4
5
6
7
8
9
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
Figure 1. Effect of Sn addition on Pt/SiO pretreated at 1123 K. (a) Ethylene TOF and (b)
2
benzene TOF over Pt/SiO and PtSn/SiO , methane conversion to detectable products and
2
2
product selectivities excluding coke for (c) Pt/SiO and (d) PtSn/SiO (overall atomic ratio of
2
2.
3
Pt/Sn equal to 3:1 ) Conditions: CH flowrate = 42 cm (STP)/min, T = 973 K, catalyst mass = 2
4
g.
3
.4. Effect of H co-feeding on catalytic activity of SiO supported catalysts:
2
2
Co-feeding hydrogen with methane has been used in the literature to decrease coke formation,
although it has resulted in lower product formation rates. Accordingly, 5% H was co-fed with
2
methane on Pt and PtSn/SiO catalysts pretreated at 1123 K. Figure 2a and Figure 2b display
2
ethylene TOF values for Pt and PtSn/SiO catalysts under no co-feed and 5% H co-feeding
2
2
conditions. On the Pt/SiO catalyst with 5% H co-feeding, following a short term increase, the
2
2
activity decreased monotonically, and at the end of 6 h, it was 60% of the activity of the catalyst
with only methane feed. On the PtSn/SiO catalyst, the ethylene TOF decreased monotonically,
2
and the value at steady state with 5% H co-feeding was three times lower than for the catalyst
2
with no co-feeding. As shown in Figure 2c and Figure 2d, the benzene TOF values of the
catalysts showed a similar trend; however, the decrease in the TOF due to H co-feeding was
2
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 50
1
2
3
4
5
6
7
8
9
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
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
more pronounced for the initial period of the time on stream. Methane conversion to detectable
products (left axis) and the selectivities of the catalyst towards ethylene, ethane and benzene
(
right axis) over total hydrocarbon products excluding coke for Pt/SiO and PtSn/SiO with 5%
2 2
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
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
H co-feeding are plotted in Figure 2e and Figure 2f. As indicated by the decreased product TOF
2
values, both catalysts had lower methane conversion when H is co-fed with methane. This trend
2
is in accordance with results from previous studies.
16
ACS Paragon Plus Environment

Page 17 of 50
ACS Catalysis
1
2
3
4
5
6
7
8
9
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
Figure 2. Effect of H co-feeding with CH on SiO -supported catalysts pretreated at 1123 K.
2
4
2
Ethylene TOF over (a) Pt/SiO and (b) PtSn/SiO benzene TOF over (c) Pt/SiO and (d)
2
2,
2
PtSn/SiO , methane conversion to detectable products and selectivities excluding coke for (e)
2
Pt/SiO with 5% H co-feeding and (f) PtSn/SiO 5% H co-feeding. (overall atomic ratio of
2
2
2
2
3
Pt/Sn equal to 3:1) Conditions: CH flowrate = 42 cm (STP)/min, T=973 K, catalyst mass = 2 g.
4
3
H flowrate = 2 cm (STP)/min for the co-feeding runs.
2
3.5. Effect of support and Pt:Sn ratio on the activity of H-ZSM-5 supported catalysts:
H-ZSM-5 zeolite with SiO /Al O = 280 was investigated for its performance as a support for
2
2
3
Pt and PtSn catalysts. Figure 3a shows the ethylene TOF as a function of time on stream.
PtSn(1:1)/Z-280 and PtSn(1:2.5)/Z-280 catalysts were also studied for CH conversion, and they
4
performed similar to PtSn(1:2)/Z-280. These data are omitted here for the sake of clarity. As the
Sn content increased, the ethylene TOF of the zeolite-supported catalysts increased. At the end of
6
h, the ethylene TOF on the Pt-Sn(1:3)/Z-280 catalyst was 4 times higher than the ethylene TOF
on the Pt/Z-280 catalyst and 15 times faster than that of the PtSn/SiO catalyst. In evaluating the
2
activity of zeolite-supported catalysts for benzene formation, the formation rate of the product
per mass of catalyst was used instead of TOF values normalized per Pt site density of the catalyst
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 50
1
2
3
4
5
6
7
8
9
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
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
surface. The rationale behind this choice was the bifunctional nature of these catalysts leading to
the formation of aromatic compounds on acid sites of the zeolite. Figure 3b displays the benzene
formation rate of Pt and PtSn catalysts supported on H-ZSM-5 per mass of catalyst. The rate of
benzene formation was promoted by the zeolite support, as evidenced by the 4 times increase in
the benzene formation rate with the use of a bimetallic catalyst Pt-Sn on the zeolite support. A
significant difference between the formation of ethylene and benzene was that while the benzene
formation rate deactivated within the first two hours, the rate of ethylene formation kept
increasing up to 10 h time on stream, particularly on the bimetallic catalysts. Figure 3c and
Figure 3d display the methane conversion to detectable products (left axis) and the selectivities
of the catalyst towards ethylene, ethane and benzene (right axis) over total hydrocarbon products
excluding coke for Pt/Z-280 and Pt-Sn(1:3)/Z-280 catalysts. In parallel with the variation of
ethylene TOF profiles of the catalysts, Pt/Z-280 showed a flat conversion profile around 0.05%,
while the conversion over Pt-Sn(1:3)/Z-280 exceeded 0.10% at the end of 8 h. Ethylene was the
product with highest selectivity on both catalysts, while Pt-Sn(1:3)/Z-280 also yielded a
significant benzene selectivity around 10%. The zeolite-supported catalysts performed better
than their silica supported counterparts, and bimetallic PtSn catalysts were more active than Pt/Z-
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
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
280 for formation of both ethylene and benzene.
18
ACS Paragon Plus Environment

Page 19 of 50
ACS Catalysis
1
2
3
4
5
6
7
8
9
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
Figure 3. Effect of Sn addition to Pt/Z-280. (a) Ethylene TOF (b) benzene formation rate per
mass of catalyst over Pt/Z-280 and PtSn/Z-280, methane conversion to detectable products and
selectivities excluding coke for (c) Pt/Z-280 and (d) PtSn(1:3)/Z-280. Ethylene TOF and benzene
formation rate over PtSn/SiO (overall atomic ratio of Pt/Sn equal to 3:1) are given for
2
3
comparison. Conditions: CH flowrate = 42 cm (STP)/min, T = 973 K, catalyst mass = 1 g.
4
As explained in more detail below, a microkinetic model based on results from DFT
calculations suggests that smaller Pt-Sn particles should be more active for methane conversion.
19
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 50
1
2
3
4
5
6
7
8
9
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
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Thus, it can be hypothesized that PtSn/H-ZSM-5 catalysts have smaller particle sizes compared
to SiO -supported catalysts. STEM images were collected to obtain information about the
2
particle size distributions of the studied catalysts. Figure 4 displays STEM images of Pt/SiO and
2
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
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Pt/H-ZSM-5 samples. The histograms of the particle size distributions of the two catalysts are
shown in Figure 5. The particle size distributions for these catalysts showed that the surface-
averaged particle sizes of the Pt/SiO and Pt/Z-280 catalysts were 2.5 nm and 1.8 nm,
2
respectively. The origin of the smaller metal particle size on the H-ZSM-5 catalyst can be
2
/
2
attributed to its higher surface area compared to SiO (425 m g for H-ZSM-5 versus 300 m /g
2
for SiO ). As another way to investigate the effect of particle size of the catalyst, the variation of
2
the fraction of total metal surface area was calculated as a function of the metal particle diameter.
This variation of the fraction of total metal surface area versus particle size is shown in the inset
of Figure 5. Sixty percent of the surface area for the Pt/Z-280 catalyst was in the particle
diameter range from 1-2 nm. In contrast, only 15% of the surface area of Pt/SiO was in this
2
same region. To investigate the effect of Sn on the particle size, STEM images were collected for
the PtSn/SiO catalyst with overall atomic ratio of Pt/Sn equal to 3:1 and for the PtSn(1:3)/Z-280
2
catalyst. The average particle size of the PtSn/SiO catalyst was 2.5 nm, while the average
2
particle size of the PtSn(1:3)/Z-280 catalyst was 2.3 nm. EDS analysis of the catalysts showed
that the bimetallic particles on the PtSn/SiO catalyst were composed of 4.5 at% Sn, whereas the
2
target value was 25%. The same analysis yielded a composition of 23 at% Sn on the Pt-
Sn(1:3)/Z-280 catalysts that had the target atomic fraction of 75 at% Sn. Although both catalysts
had lower than expected amounts of Sn interacting with Pt, formation of bimetallic particles was
more successful on the zeolite support, as shown by slightly increased particle size and EDS
analysis of the PtSn(1:3)/Z-280 catalyst.
20
ACS Paragon Plus Environment

Page 21 of 50
ACS Catalysis
1
2
3
4
5
6
7
8
9
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
Figure 4. STEM images of (a) Pt/SiO (b) Pt/H-ZSM-5 catalysts.
2,
21
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 50
1
2
3
4
5
6
7
8
9
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
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 5. Particle size distribution of Pt/SiO and Pt/Z-280. Inset displays the fraction of the
2
total catalytic surface area as a function of particle size.
3.6 Effect of support acidity of the zeolite supported catalysts on catalytic activity: To
study the effect of surface acidity on catalyst performance and the product distribution of zeolite-
supported catalysts, we prepared PtSn(1:2) catalysts using more acidic H-ZSM-5 supports: Z-80,
Z-50 and Z-23. Figure 6a shows the ethylene TOF as a function of time on stream over four
PtSn(1:2) catalysts supported on H-ZSM-5 supports with different acidity. The highest rate of
ethylene production was observed on the PtSn(1:2)/Z-50 catalyst. The catalyst showed an
increasing ethylene TOF profile up to more than 20 h, after which it began to decrease. The less
acidic catalyst PtSn(1:2)/Z-80 reached an earlier peak around 10 h with almost half of the
ethylene TOF value of PtSn(1:2)/Z-50, and the least acidic PtSn(1:2)/Z-280 catalyst yielded
almost a flat profile compared to the other three more acidic catalysts. To compare the activity of
the catalysts towards formation of aromatics, the formation rates of benzene and naphthalene per
mass of catalyst are plotted in Figure 6b and Figure 6c, respectively. The catalysts with the two
most acidic supports, PtSn(1:2)/Z-23 and PtSn(1:2)/Z-50, showed the highest activity both for
benzene and naphthalene production. The two less acidic catalysts, PtSn(1:2)/Z-80 and
PtSn(1:2)/Z-280, were significantly less active for production of aromatics and deactivated
during time on stream. On the PtSn(1:2)/Z-23 catalyst, the maximum formation rate was
achieved at 15 h, and on the PtSn(1:2)/Z-50 it was achieved around 20 h. For both catalysts,
these time points were earlier than the time for maximum production of ethylene. The initially
increasing and then decreasing volcano-shaped activity profile for the formation of all the
products suggests the formation of a reactive “hydrocarbon pool” on the acidic surface during the
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
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
earlier period of reaction, similar to the hydrocarbon pool concept extensively studied and
22
ACS Paragon Plus Environment

Page 23 of 50
ACS Catalysis
1
2
3
4
5
6
7
8
9
described in methanol-to-olefin (MTO) literature. It can be hypothesized that oligomerization of
ethylene on the acid sites takes place together with interaction with the higher hydrocarbon
4
2,43
structures of polyaromatics deposited in the pores.
After the maximum rate for ethylene
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
formation is achieved, methane conversion is suppressed by deactivation.
Figure 6. Effect of support acidity on (a) ethylene TOF (b) benzene formation rate and (c)
naphthalene formation rate of zeolite-supported catalysts. Conditions: CH flowrate = 10.5
4
3
cm (STP)/min, T = 973 K, catalyst mass = 0.25 g.
23
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 50
1
2
3
4
5
6
7
8
9
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
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The methane conversion to detectable products (left axis) and the selectivities of the
catalysts towards ethylene, ethane, benzene and naphthalene (right axis) over total hydrocarbon
products excluding coke are plotted for each catalyst in Figure 7 a-d. The PtSn(1:2)/Z-50 catalyst
showed the highest conversion, reaching approximately 0.30%, followed by the PtSn(1:2)/Z-23
catalyst. Conversion profiles of both catalysts were similar to the profiles for aromatics
formation. The least acidic PtSn(1:2)/Z-280 catalyst had low conversion, as indicated by its low
ethylene TOF and aromatic formation rates. On all the catalysts, regardless of their activity,
ethylene was the product with highest selectivity, and benzene formation was significantly lower
during the entire time on stream. Over the two most active catalysts, there were visibly opposite
trends between ethylene and naphthalene selectivities. During the initial periods of the reaction
while ethylene formation was still increasing, its selectivity decreased in the favor of naphthalene
formation, which subsequently decreased when the ethylene selectivity began to increase again.
The benzene selectivities of the catalysts were lower than naphthalene, supporting the argument
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
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
4
that naphthalene was formed from benzene . The delay in the maximum rate of ethylene
formation compared to aromatics can be explained by faster oligomerization of ethylene to
benzene on the more acidic catalysts, compared to desorption of ethylene from the catalyst. After
the maximum rate is reached, the catalyst begins to deactivate due to coke deposition.
24
ACS Paragon Plus Environment

Page 25 of 50
ACS Catalysis
1
2
3
4
5
6
7
8
9
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
Figure 7. Effect of support acidity on methane conversion to detectable products and
selectivities of (a) PtSn(1:2)/Z-23, (b) PtSn(1:2)/Z-50 (c) PtSn(1:2)/Z-80 and (d) PtSn(1:2)/Z-
3
2
80. Conditions: CH flowrate = 10.5 cm (STP)/min, T = 973 K, catalyst mass = 0.25 g.
4
As seen in Table 5, quantification of the amount of coke deposited on spent zeolite-supported
catalysts showed that the catalysts supported on more acidic supports had higher extents of coke
deposition, while the PtSn catalysts supported on the Z-280 support and the relatively inactive
Pt/Z-280 catalyst had lower amounts of coke.
25
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 50
1
2
3
4
5
6
7
8
9
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
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 5. Coke deposition on spent zeolite-supported catalysts
Catalyst
Carbon deposition by TGA (mmol/gcat)
Pt/Z-280
2.1
3.6
9.6
8.9
10
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
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
PtSn(1:3)/Z-280
PtSn(1:2)/Z-23
PtSn(1:2)/Z-50
PtSn(1:2)/Z-80
PtSn(1:2)/Z-280
6.5
In previously reported product formation rates, benzene was the main aromatic product, with
3,12,18,19,28
selectivity higher than 60%.
However, for our catalysts with the two most acidic
supports (Z-23 and Z-50), naphthalene was the major aromatic product, with formation rates up
to 7 times higher than that of benzene. Over these two catalysts, when naphthalene formation
reached its maximum rate, the selectivity to ethylene was 60%, naphthalene was 30%, and
benzene was less than 5%. Furthermore, the total rate of carbon product formation (the sum of
ethylene, benzene and naphthalene) on the PtSn(1:2)/Z-50 catalyst was approximately 2.5 mmol-
4
5
9
,
17
,
20
19
21
C/mol-Pt.s, which is comparable to the calculated values of 0.1 0.2 0.3 0.6 , 0.7 , 2.1 ,
,
2
4
16
9
.2 and 15 mmol-C/(mol-Mo.s) from reported product formation rates for various Mo/H-
ZSM-5 catalysts operated at 973 K and similar space velocities. In addition, we prepared Mo/H-
ZSM-5 catalysts with 0.58 wt% and 4 wt% Mo, and we studied these catalysts for methane
3
conversion at 973 K with 0.25 g of catalyst under 10.5 cm (STP)/min methane flow. The total
rates of carbon product formation for these catalysts were 1.8 and 0.3 mmol-C/mol-Mo.s,
respectively. To better illustrate this comparison, Figures 8 a-c display the rate of carbon product
formation from ethylene, benzene and naphthalene on the 0.58 wt% Pt-Sn(1:2)/Z-50, 0.58 wt%
26
ACS Paragon Plus Environment

Page 27 of 50
ACS Catalysis
1
2
3
4
5
6
7
8
9
Mo/Z-50 and 4 wt% Mo/Z-50 catalysts. The total rate of carbon product formation on each
catalyst is shown in Figure 8d. The rates in Figure 8 are normalized per Pt atom or per Mo atom
in the catalyst.
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
Figure 8. Rate of carbon product formation from (a) ethylene (b) benzene (c) naphthalene on the
0.58 wt% Pt-Sn(1:2)/Z-50, 0.58 wt% Mo/Z-50 and 4 wt% Mo/Z-50 catalysts. (d) Total rate of
3
carbon product formation on each catalyst. Conditions: CH flowrate = 10.5 cm (STP)/min, T =
4
973 K, catalyst mass = 0.25 g.
27
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 50
1
2
3
4
5
6
7
8
9
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
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
.7. Microkinetic Model
A microkinetic model of methane conversion was developed to gain insight into catalyst
reactivity by using DFT-derived parameters for terrace and stepped surfaces of Pt and Pt Sn. The
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
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Pt(433) terrace, Pt(433) step, Pt Sn(111) terrace and Pt Sn(211) step surfaces studied are shown
3
3
in Figure 9. Details and key results of the model for the studied surfaces are given in the
Supplementary Information. Without any adjustments in the DFT-derived parameters, the model
was used to predict formation rates of the products and variations in surface coverages of the
reaction intermediates under reaction conditions. The effect of Sn addition, H co-feeding and
2
particle size on the catalyst activity were investigated by comparing model predictions with
experimental results.
Figure 9. Model surfaces used in DFT calculations (a) Pt(433) terrace (b) Pt(433) step (c)
Pt Sn(111) terrace and (d) Pt Sn(211) step. The unit cells of the terrace sites are highlighted.
3
3
The variation of the surface coverages of intermediates and ethylene TOF values of the
studied surfaces are shown in Figure 10. The reversibilities (i.e., the ratio of the reverse rate to
the forward rate) of the elementary steps carrying the reaction flux for the various surfaces
predicted by microkinetic model are shown in Table 6.
28
ACS Paragon Plus Environment

Page 29 of 50
ACS Catalysis
1
2
3
4
5
6
7
8
9
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
Figure 10: Surface coverages and ethylene TOF values of the studied surfaces predicted by the
microkinetic model (a) no co-feeding conditions (b) 5% H co-feeding with methane.
2
(Experimental conditions same as Figure 1 and Figure 2) Left axis shows surface coverage of the
indicated species, right axis shows the ethylene TOF values.
The model predicted that terrace and steps of Pt were almost fully covered with adsorbed CH*
and C* fragments, with a slightly lower total coverage on Pt step sites, corresponding to a 2.4
times higher ethylene TOF. The elementary C-C coupling step with the highest rate was found as
CH* + CH* → CHCH* which carried the 34% of the reaction flux on Pt terrace and 46% of the
reaction flux on Pt step sites. On Pt terrace sites, the elementary step with the highest degree of
rate control was the activation of adsorbed methane and desorption of the hydrogen, whereas on
Pt steps, desorption of ethylene from the surface also had significant degrees of rate control.
Coverages of the surface by adsorbed CH* and C* were significantly lower on the Pt Sn(111)
3
terrace than on the Pt(111) terrace. On the Pt Sn(111) terrace, the coverage of adsorbed CH*
3
decreased to 0.28 from 0.92 and 0.77 on terrace and steps of Pt, and the coverage of adsorbed C*
decreased to 0.07 from 0.09 and 0.2 on terrace and steps of Pt. Furthermore, 50% of the sites on
29
ACS Paragon Plus Environment