Methane to Methanol: Structure-Activity Relationships for Cu-CHA

Article abstract of DOI:10.1021/jacs.7b06472

Cu-exchanged zeolites possess active sites that are able to cleave the C-H bond of methane at temperatures ≤200 °C, enabling its selective partial oxidation to methanol. Herein we explore this process over Cu-SSZ-13 materials. We combine activity tests and X-ray absorption spectroscopy (XAS) to thoroughly investigate the influence of reaction parameters and material elemental composition on the productivity and Cu speciation during the key process steps. We find that the Cu^II moieties responsible for the conversion are formed in the presence of O₂ and that high temperature together with prolonged activation time increases the population of such active sites. We evidence a linear correlation between the reducibility of the materials and their methanol productivity. By optimizing the process conditions and material composition, we are able to reach a methanol productivity as high as 0.2 mol CH₃OH/mol Cu (125 μmol/g), the highest value reported to date for Cu-SSZ-13. Our results clearly demonstrate that high populations of 2Al Z₂Cu^II sites in 6r, favored at low values of both Si:Al and Cu:Al ratios, inhibit the material performance by being inactive for the conversion. Z[Cu^IIOH] complexes, although shown to be inactive, are identified as the precursors to the methane-converting active sites. By critical examination of the reported catalytic and spectroscopic evidence, we propose different possible routes for active-site formation.

Full text of DOI:10.1021/jacs.7b06472

Subscriber access provided by University of Birmingham
Article
Methane to methanol: structure-activity relationships for Cu-CHA
Dimitrios K. Pappas, Elisa Borfecchia, Michael Dyballa, Ilia Pankin, Kirill A. Lomachenko,
Andrea Martini, Matteo Signorile, Shewangizaw Teketel, Bjørnar Arstad, Gloria Berlier, Carlo
Lamberti, Silvia Bordiga, Unni Olsbye, Karl Petter Lillerud, Stian Svelle, and Pablo Beato
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b06472 • Publication Date (Web): 25 Sep 2017
Downloaded from http://pubs.acs.org on September 25, 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.
Journal of the American Chemical Society 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 17
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Methane to methanol: structure-activity relationships for Cu-CHA
Dimitrios K. Pappas1, Elisa Borfecchia*2,3, Michael Dyballa1, Ilia A. Pankin3,4, Kirill A. Lomachenko4,5,
Andrea Martini3, Matteo Signorile3, Shewangizaw Teketel2, Bjørnar Arstad6, Gloria Berlier3, Carlo
Lamberti3,4, Silvia Bordiga1,3, Unni Olsbye1, Karl Petter Lillerud1, Stian Svelle*1 and Pablo Beato*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
1Center for Materials Science and Nanotechnology (SMN), Department of Chemistry, University of Oslo, 1033 Blindern,
0315 Oslo (Norway)
2Haldor Topsøe A/S, Haldor Topsøes Allé 1, 2800 Kongens Lyngby (Denmark)
3Department of Chemistry and INSTM Reference Center, University of Turin, via P. Giuria 7, 10125 Turin (Italy)
4IRC “Smart Materials”, Southern Federal University, Zorge Street 5, 344090 RostovꢀonꢀDon, (Russia)
5European Synchrotron Radiation Facility (ESRF), 71 avenue des Martyrs, CS 40220, 38043 Grenoble Cedex 9, (France)
6SINTEF Materials and Chemistry, Forskningsveien 1, 0373 Oslo (Norway)
ABSTRACT: Cuꢀexchanged zeolites possess active sites able to cleave the C−H bond of methane at temperatures ≤200 °C, enaꢀ
bling its selective partial oxidation to methanol. Herein we explore this process over CuꢀSSZꢀ13 materials. We combine activity
tests and XꢀRay Absorption Spectroscopy (XAS) to thoroughly investigate the influence of reaction parameters and material eleꢀ
mental composition on the productivity and Cuꢀspeciation during the key process steps. We find that the CuII moieties responsible
for the conversion are formed in the presence of O2 and that high temperature together with prolonged activation time increases the
population of such active sites. We evidence a linear correlation between the reducibility of the materials and their methanol
productivity. By optimizing the process conditions and material composition, we are able to reach a methanol productivity as high
as 0.2 molCH3OH/molCu (125 µmol/g), the highest value reported to date for CuꢀSSZꢀ13. Our results clearly demonstrate that high
populations of 2Al Z2CuII sites in 6r, favored at low values of both Si:Al and Cu:Al ratios, inhibit the material performance by beꢀ
ing inactive for the conversion. Z[CuIIOH] complexes, although shown to be inactive, are identified as the precursors to the meꢀ
thaneꢀconverting active sites. By critical examination of the reported catalytic and spectroscopic evidences, we propose different
possible routes for active site formation.
Most reaction approaches found in the literature are based
on a multiꢀstep process, with an activation of the catalyst as
1. INTRODUCTION
the first step in order to generate the active sites. For the actiꢀ
vation, H2O2,16 N2O,24, 28 NO,29 and O211ꢀ15, 20ꢀ24, 26, 30 have led to
the formation of Cu centers that were able to convert methane.
Generally, a relatively low activation temperature (500 °C)
combined with a cheap and available oxidant (i.e. air) would
be preferred for forming the methaneꢀconverting centers in the
zeolite. After this activation, methane is reacted over the mateꢀ
rial at different conditions (partial pressure, reaction time, and
temperature) in order to cleave one C−H bond on the active
sites. Finally, the product (methanol) is extracted using a polar
solvent or by passing steam through the reactor. Recently, an
alternative approach to this stepwise process was reported for
CuꢀMOR such that methane is fed at 7 bar and H2O is utilized
to partially oxidize methane and simultaneously regenerate the
active sites while facilitating the product desorption.6
The direct lowꢀtemperature conversion of methane to methꢀ
anol is currently a heavily investigated field, due to its enorꢀ
mous potential for the energy and chemicals sector.1ꢀ7 Howevꢀ
er, the high energy required for cleavage of the C−H bond in
CH4 combined with the facile overꢀoxidation to COx pose
great challenges for the realization of such a process. In naꢀ
ture, monoꢀoxygenase enzymes in methanotrophic bacteria are
able to transform methane to methanol.8ꢀ10 These enzymes
have inspired researchers in the field of heterogeneous catalyꢀ
sis to mimic such a unique reactivity. Metalꢀexchanged zeoꢀ
lites are able to stabilize wellꢀdefined metalꢀoxo sites: differꢀ
ent zeolite topologies have been demonstrated to stabilize
copper,11ꢀ15 iron,16ꢀ17 and other18ꢀ19 metalꢀoxo clusters and all of
them have shown activity for stoichiometric transformation of
methane to methanol.
Apart from the process conditions, the methanol yield will
obviously depend on the amount of copper present as active
sites. However, the actual nature of the active Cu sites formed
in different zeolites is still a heavily debated subject. For Cuꢀ
MOR, monoꢀ(ꢁꢀoxo) dicopper sites have been proposed by
Alayon et al.;13, 31ꢀ32 isolated trimeric Cuꢀoxo centers have been
suggested by Grundner et al.,12 and even larger Cuꢀoxo clusꢀ
ters have been discussed.33ꢀ34 Recently, CuꢀSSZꢀ13 has also
The ability of Cuꢀzeolites to selectively convert methane to
methanol was evidenced for the first time by Groothaert and
coꢀworkers.11 The authors proposed the methane C−H bond
cleavage to occur over bis(µꢀoxo) dicopper sites present in the
pores of MFI (ZSMꢀ5) and MOR frameworks. Following theꢀ
se findings, a wide range of zeolite frameworks have been inꢀ
vestigated with MOR,6, 11ꢀ15, 20ꢀ23 CHA,15, 23ꢀ25 and MFI11, 14, 26ꢀ27
exhibiting the highest yields of methanol.
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 17
been identified as an active material for the direct conversion
been reported to strongly impact the Cu speciation in the mateꢀ
of methane to methanol.15, 23, 25 For CuꢀSSZꢀ13, a zeolite with
CHA topology, both monomeric active species, in the form of
[CuIIOH]+ complexes,35 and multimeric species36 have been
envisaged from theory. Ipek et al.25 proposed the existence of
[Cu2O2]2+ and [Cu2O]2+ species in CuꢀSSZꢀ13 zeolites based
on PXRD, Raman and UVꢀVis data, suggesting their involveꢀ
ment in the direct methane conversion.
rials30, 40, 47, 52 and therefore they are also expected to have an
effect on the methanol productivity. Recently, Kim et al. exꢀ
plored the effect of different oxidants in the activation process
and showed that, in the case of O2, 500 ºC is a limit for Cuꢀ
MOR catalysts,52 while Park et al. studied the effect of activaꢀ
tion temperature over numerous structures finding a volcanoꢀ
like behavior at 450 °C for most.53 Here, we first study the efꢀ
fect of temperature, partial pressure and time during the oxyꢀ
gen activation step on the productivity over 0.5CuꢀSSZꢀ
13(12).
1
2
3
4
5
6
7
8
Cuꢀexchanged SSZꢀ13 is also a highly efficient deNOx cataꢀ
lyst via NH3ꢀassisted Selective Catalytic Reduction (NH3ꢀ
SCR),37ꢀ39 and has been extensively characterized in the literaꢀ
ture, focusing on the Cuꢀspecies formed inside the zeolite latꢀ
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 1a shows a linear dependency of the methanol
productivity on the activation temperature. Moving from 350
to 500 °C, an increase of 160% in the methanol output was
tice as a function of the temperature and gaseous environꢀ
38ꢀ39
ment,25,
e.g. during thermal activation and into SCRꢀ
relevant conditions.39ꢀ48
observed, reaching
a
productivity equal to 0.041
molCH3OH/molCu (25.3 µmol/g) with 86% selectivity. The
selectivity remained approximately constant over the entire
activation temperature range. Further, the material was actiꢀ
vated at 450 °C for 120 min with the partial pressure of O2
ranging from 0 to 1000 mbar (Figure 1b). Increasing the oxyꢀ
gen partial pressure from 67 to 187 mbar leads to a uniform
increase of the productivity, which then levels to 0.038
molCH3OH/molCu (23.2 µmol/g) at higher partial pressures.
Activation of the material in pure He at 450 °C (first point in
Figure 1b) resulted in very low productivity of 0.014
molCH3OH/molCu. Looking into the effect of activation time
(Figure 1c) it is observed that increasing the activation time
from 60 to 480 min results in an increase of the methanol yield
by almost 100 %. After 480 minutes of activation, a plateau is
observed, corroborating the fact that the activation process reꢀ
quires time to reach an optimum at these conditions (187 mbar
O2 and at 450 °C). The maximum productivity is observed afꢀ
ter 720 minutes of activation, 0.058 molCH3OH/molCu (35.8
µmol/g). These results show that increased activation time,
oxygen partial pressure, and reaction time all promote the
population of Cu species responsible for the cleavage of the
C−H bond of methane. The maximum methanol yield is 0.065
molCH3OH/molCu (40.6 µmol/g) and is reached when the
sample is activated in pure oxygen at 500 °C for 480 min.
Based on the extended knowledge accumulated in the conꢀ
text of NH3ꢀSCR over the past decade, and driven by the iniꢀ
tial experimental insights about methane activation on Cuꢀ
SSZꢀ13,15, 23ꢀ25 we identified this system as highly suited for
establishing structure–activity relationships. Herein, we comꢀ
bine catalytic testing, in situ/operando Xꢀray Absorption Specꢀ
troscopy (XAS)49 and complementary insights from Raman
and FTIR spectroscopy to explore the methane to methanol
conversion over CuꢀSSZꢀ13 for a wide range of materials and
reaction conditions. In doing so, optimization of the process
conditions has been achieved, thereby arriving at the greatest
reaction yield reported so far for this material. Moreover, a
linear correlation between the amount of reducible Cu species
and the reaction yield has been found.
2
RESULTS AND DISCUSSION
The CuꢀSSZꢀ13 materials investigated in this work have
been prepared by conventional aqueous ion exchange, and
subjected to compositional and physicoꢀchemical characterizaꢀ
tion by EDX, N2ꢀphysisorption, SEM and PXRD. Additional
details on the synthesis and laboratory characterization can be
found in the Supporting Information (SI), Sections 1 and 2,
respectively. The studied materials are named using
(‘Cu:Al’)CuꢀSSZꢀ13(‘Si:Al’) code, to identify their composiꢀ
tion.
Looking into the methane loading step, initially we introꢀ
duced 100 mbar of methane at different temperatures, for 60
min (Figure 1d). Loading methane at temperatures lower than
135 °C results in a very low productivity; a 35 °C increase in
the temperature yields a roughly doubled output. A temperaꢀ
ture of 200 °C is optimal for activation of methane; the correꢀ
sponding productivity is equal to 0.034 molCH3OH/molCu
(21.3 µmol/g) with a selectivity of 86%. Increasing the temꢀ
perature up to 300 °C does not affect the total productivity, as
the sum of the products remains constant. However, a signifiꢀ
cant decrease in selectivity is observed, since more COx is deꢀ
tected, leading to 63% selectivity at 300 °C. As mentioned beꢀ
fore, the quantification of the products is obtained during the
H2Oꢀassisted extraction step, so that the resulting COx might
originate from different intermediates, such as formate, which
evolve during the methane loading step. It appears that such
species leading to full oxidation are favored at higher temperaꢀ
ture. Moreover, a further rise of the temperature to 300ꢀ400 °C
results in a decreased methanol and total productivity, indicatꢀ
ing that high temperature hinders the stabilization of intermeꢀ
diates on the active centers. Figure 1e depicts the effect of CH4
partial pressure on the productivity. The productivity doubled
moving from 100 mbar (0.034 molCH3OH/molCu) to 1000
mbar (0.061 molCH3OH/molCu). The higher CH4 pressure
We will first describe the effects of reaction conditions in
the different steps of the process. Then, we will investigate the
effect of characteristic pretreatments on the speciation via
XAS and performance and follow the entire process with opꢀ
erando XAS. XAS measurements were carried out at the
BM2350 and BM26A51 beamlines of the European Synchrotron
Radiation Facility (ESRF) (for experimental details see SI,
Section 1.4). Testing experiments were performed in our home
laboratory, at comparable conditions in a standard fixed bed
setup (for experimental details see SI, Section 1.3). Finally,
the effects of material composition on productivity are exꢀ
plored. Complementary spectroscopic techniques, such as
Raman and Fourier Transformed InfraꢀRed (FTIR) are utilized
to assist the formulation of a complete scheme critically conꢀ
sidering different active species as well as their formation
mechanisms based on our observations and findings from the
literature.
2.1 Impact of reaction parameters on the performance
High temperature activation of Cuꢀexchanged zeolites in
oxygenꢀcontaining atmosphere is the initial step for the conꢀ
version. Temperature, oxygen partial pressure and time have
2
ACS Paragon Plus Environment

Page 3 of 17
Journal of the American Chemical Society
leads to a greater coverage of activated species from methane. results, in Figure 1f, show an increase in productivity from 30
1
2
3
4
A similar trend has been observed over CuꢀMOR zeolites for
highꢀpressure methane loading, where an increased CH4 presꢀ
sure is suggested to transform spectator species into species
that participate in the bond cleavage.7 Finally the effect of conꢀ
tact time for 100 mbar CH4 flow at 200 °C was explored. The
min to 120 min, reaching 0.047 molCH3OH/molCu (38.0
µmol/g). Further increasing the time leads to a moderate rise
in productivity, and a plateau is observed after ca. 300 min, at
0.058 molCH3OH/molCu.
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. Normalized methanol productivity (molCH3OH/molCu ꢀ circles) and selectivity (% ꢀ triangles) during different reaction condiꢀ
tions and multiple cycles. Figures aꢀg follow the reference reaction conditions (i.e. O2 activation: 450 °C for 120 min in 187 mbar O2, CH4
loading: 60 min in 100 mbar CH4 at 200 °C and H2O assisted extraction: online with a 10% H2O stream at 135 °C) while for each a single
parameter is studied: (a) O2 activation temperature; (b) O2 partial pressure; (c) O2 activation time, (d) CH4, loading temperature; (e) CH4
partial pressure; (g) CH4 contact time; (g) H2O assisted extraction temperature. Figure h shows the effect of reaction conditions for achievꢀ
°
ing the highest productivity (i.e. O2 activation: 500 °C for 480 min in 1000 mbar O2, CH4 loading: 360 min in 1000 mbar CH4 at 200 C
°
and H2O assisted extraction: online with a 10% H2O stream at 200 C) and the effect of multiple cycles of the process. Solid lines in the
figure are utilized as a guide to the eye.
The final step of the process is the H2Oꢀassisted extraction
of methanol. Passing 10% steam (in He) through the fixed bed
reactor, the selectively oxidized products (i.e. methanol and
dimethyl ether, accounted as two methanol molecules) are exꢀ
tracted along with overꢀoxidized COx species. Prior to extracꢀ
tion, the sample was exposed to reference reaction conditions
(see SI, Section 3). Figure 1g shows productivity and selectiviꢀ
ty obtained by performing the extraction at temperatures rangꢀ
ing from 110 to 200 °C. A temperature above 135 °C is reꢀ
quired to fully desorb methanol within the fixed time. When
raising the temperature, an increase in the total productivity is
observed, indicating that a higher extraction temperature faꢀ
vors the desorption of adsorbed Cꢀcontaining species in the
form of overꢀoxidation products, as the methanol yield is conꢀ
stant. In particular, comparing extraction at 135 and 200 °C, a
constant methanol yield of 0.037 molCH3OH/molCu (23.1
µmol/g) is observed, whereas the selectivity decreases from 86
to 70%. From a processꢀdesign perspective, the constant yield
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 17
over this broad temperature range is an important feature, the lack of other important scattering contributions in this Rꢀ
1
2
3
4
5
6
7
8
since it potentially allows methane loading and methanol exꢀ
traction to be operated isothermally at 200 °C.
space range, we associate the third maximum in the FTꢀ
EXAFS spectra with neighboring Cu sites in CuxOy moieties,
bestꢀfitted to average metalꢀmetal separation of (3.41 ± 0.06)
Å. The FTꢀEXAFS spectra in Figure 2b are also reported in
Figure S13 in an extended Rꢀspace range up to ca. 6 Å. Here,
the weak peak at 3.4 Å can be clearly distinguished from the
noise, dominating the spectra for R > 3.7 Å. Noteworthy, the
experimental conditions employed here (activation up to 500
°C in pure O2) yielded a more pronounced third maximum in
comparison with our previous study mentioned before40 conꢀ
ducted on a CuꢀSSZꢀ13 zeolite with very similar composition
(Si:Al ≈ 13; Cu:Al ≈ 0.44) but by activating the material only
up to 400 °C in 50% O2/He.
The evaluation of the individual reaction parameters was
translated into optimized conditions and applied in multiple
process cycles. Activation at 500 °C for 480 min in a 1000
mbar O2 feed, followed by methane loading at 200 °C for 360
min in a 1000 mbar CH4 feed, and a H2Oꢀassisted extraction of
methanol at 200 °C (see also SI, Section 3) yield 0.172
molCH3/molCu (107 µmol/g), the highest value reported so far
for CuꢀCHA zeolites. After completion of the first process cyꢀ
cle, the material was purged in He, at 200 °C and subsequently
subjected to in total four complete cycles (Figure 1h). Interestꢀ
ingly, productivity and selectivity gradually increase after each
process cycle. After four cycles, the methanol productivity
reaches 0.202 molCH3OH/molCu (125 µmol/g), a 17 % inꢀ
crease compared to the first cycle. An increase in productivity
during successive process cycles has also been observed for
Cuꢀexchanged MOR zeolites.20, 30
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
When CuꢀSSZꢀ13 is pretreated in He at 500 °C (Figure 2,
black lines), the overall XANES spectral shape is representaꢀ
tive of a dominant CuI state. The characteristic CuI 1s → 4p
peak at 8982.5 eV is evident, while the preꢀedge peak is sigꢀ
nificantly reduced. Indeed, highꢀtemperature treatment in inert
gas is known to trigger, from ~ 230 °C upward, the so called
‘selfꢀreduction’ from CuII to CuI.31, 40, 47, 54, 62ꢀ63 In the correꢀ
sponding FTꢀEXAFS spectrum, we observe a lower intensity
in the whole range, supporting the observations from XANES.
Indeed, this state mostly consists of ‘bare’ ZCuI ions (where Z
indicates coordination to Ofw next to one isolated Al in a
framework Tꢀsite, i.e. 1Al site) in both 6r and 8r of the CHA
framework.40, 64ꢀ65 Both are twofoldꢀcoordinated sites, but with
different local environments in the second shell. Such heteroꢀ
geneity, together with the higher mobility of CuI with respect
to CuII,66 causes a broadened and dampened secondꢀshell sigꢀ
nal. After Heꢀactivation, the third maximum at ca. 3.2 Å, well
evident after O2ꢀactivation, becomes barely detectable.
2.2 Impact of activation conditions on the XAS spec-
troscopic fingerprints
In the previous section, we demonstrated that temperature,
oxygen partial pressure, and activation time have an impact on
the methanol productivity of the 0.5CuꢀSSZꢀ13(12). We charꢀ
acterized the material by XAS after four different pretreatꢀ
ments, namely (i) standard O2ꢀactivation at 500 °C, (ii) activaꢀ
tion in He at 500 °C, (iii, iv) activation in He at 500 °C folꢀ
lowed by reaction with O2, either at 500 °C or at 200 °C. The
productivity was determined by contacting the sample with
1000 mbar of CH4 at 200 °C for 60 min, followed by steamꢀ
assisted extraction of the products at 135 °C in a laboratory
fixed bed reactor in order to identify Cuꢀspeciation ↔ perforꢀ
mance relationships.
The productivities observed after the two different pretreatꢀ
ments are 0.008 molCH3OH/molCu (4.7 µmol/g) and 0.082
molCH3OH/molCu (50.8 µmol/g) for He and O2 activation,
respectively (Figure 2c), clearly indicating a positive correlaꢀ
tion between methanol yield and the existence of CuII species
in the activated materials. Instead, a low productivity is seen
when the pretreatment is carried out in the absence of O2.
The characteristic XANES features detected after the differꢀ
ent activation protocols can be interpreted based on previous
Cu Kꢀedge XANES studies in Cuꢀzeolites32, 40ꢀ41, 45, 47, 54ꢀ57 and
Cuꢀcontaining enzymes.58ꢀ60 XAS for the O2ꢀactivated material
at 500 °C (Figure 2, red curves) indicates a virtually pure CuII
state.40ꢀ41, 47 In the XANES, a pronounced risingꢀedge shoulder
is observed in the 8985–8990 eV range, assigned to 1s → 4p
transitions in threeꢀ/fourꢀfold coordinated CuII sites. The weak
preꢀedge peak, arising from the dipoleꢀforbidden 1s → 3d
transition in d9 CuII ions, is also well evident. Within the specꢀ
tral resolution of our data, no evidence of CuI species is found
after the O2ꢀactivation step.
The k2ꢀweighted, phase uncorrected, Fourier transform of
the EXAFS spectrum (hereafter FTꢀEXAFS) collected at 500
°C after O2ꢀactivation shows three well defined maxima, peakꢀ
ing at ca. 1.4, 2.4, and 3.2 Å. These are seen throughout the
entire stepwise methane conversion process, with modificaꢀ
tions in their intensity (see Section 2.3). EXAFS fitting (see
SI, Section 5.1) reveals that the first maximum derives from
CuꢀO single scattering (SS) paths involving two framework
(Ofw) and one extraꢀframework (Oef) oxygen in the first coorꢀ
dination shell of Cu, at (1.97 ± 0.04) Å and (1.86 ± 0.05) Å,
respectively. In line with previous EXAFS studies on CuꢀSSZꢀ
13, the second maximum is assigned to the SS path involving
a secondꢀshell Al atom (Tfw) in frameworkꢀcoordinated Cuꢀ
species,45, 47, 61 here refined at CuꢀTfw = (2.72 ± 0.02) Å. Difꢀ
ferently to what was proposed for CuꢀMOR31 and CuꢀZSMꢀ
5,57 no significant contribution from CuꢀCu SS paths is found
in this Rꢀspace range for CuꢀSSZꢀ13. Conversely, considering
To decouple the highꢀtemperature treatment in inert leading
to dehydration and selfꢀreduction of copper from the reaction
with O2 resulting in the oxidation of CuI species, we reacted
the Heꢀactivated sample with O2 for 30 min at 500 and 200 °C
(bordeaux and blue curves/bars in Figure 2aꢀc).
The oxidation at 500 °C (subsequent to He activation at the
same temperature) yields very similar XAS spectra compared
to the sample activated directly in O2. In addition, the producꢀ
tivity after the two different activations is almost equivalent,
only differing by ca. 10 %. Nonetheless, as it can be clearly
noted by looking at the first derivative of the XANES specꢀ
trum, a small amount of CuI is still present after the subseꢀ
quent oxidation. The amount of the remaining CuI is ca. 10%
of the total copper, which is in good agreement with the slightꢀ
ly lower productivity. After both O2ꢀactivation and oxidation
at 500°C, a higher third shell component is observed compared
to the Heꢀonly activated sample.
Carrying out the reaction with O2 at 200 °C (after He activaꢀ
tion at 500 ºC) results in the same relative CuII/CuI abundance
as observed at 500 °C: a dominant CuII state is formed, with a
residual fraction of ca. 10% CuI, indicating that the same fracꢀ
tion of CuI species formed during inert treatment react with
oxygen even at lower temperature. Although XAS evidences
4
ACS Paragon Plus Environment

Page 5 of 17
Journal of the American Chemical Society
that the fraction of CuII species formed upon reꢀoxidation at
the highꢀR region of FTꢀEXAFS spectra, highlighted by light gray
1
2
3
4
boxes in the main panels. (c) Normalized methanol productivity
(molCH3OH/molCu) over the 0.5CuꢀSSZꢀ13(12) in correspondꢀ
ence of each preꢀtreatment, followed by CH4 activation at 200 °C
for 60 min using 1000 mbar CH4 feed, and H2O assisted extracꢀ
tion at 135 °C.
500 °C and 200 °C is equivalent, the methanol yield of 0.046
molCH3OH/molCu for the low temperature oxidation is 45%
lower. Hence, low temperature oxidation alters the balance
between active and inactive CuII species.
5
6
7
8
9
Further comparing the XAS spectra of the material contactꢀ
ed with O2 at 500 °C and 200 °C, significant differences in the
average coordination environment of the CuII component are
observed. In particular, oxidation at 200 °C yields a higher inꢀ
tensity of the XANES whiteꢀline peak around 8998 eV and of
the firstꢀshell peak in the FTꢀEXAFS (beyond what is exꢀ
pected due to different acquisition temperature, see SI, Section
5.2 for details). Both these observations indicate a higher relaꢀ
tive fraction of fourꢀfold coordinated CuII sites formed when
CuI is reacted with O2 at 200 °C. The observed differences in
coordination of CuII complexes, during direct O2ꢀactivation
and oxidation of a preꢀreduced sample point to a temperatureꢀ
dependent coordination preference: threeꢀcoordinated species
are favored at 500 °C but undergo a partial conversion to fourꢀ
coordinated species at 200 °C. If the reaction between CuI and
O2 is directly conducted at 200 °C, the preference towards
fourꢀcoordinated CuII complexes appears more pronounced.
The lower productivity observed in the latter conditions sugꢀ
gests that the tridentate coordination mode of CuII centers,
with firstꢀshell ligation to two Ofw and one Oef atom, is most
favorable for the active Cu site responsible of methane to
methanol conversion in CuꢀSSZꢀ13. Such a coordination mode
is the one characteristic of Z[CuIIOH] species, which are curꢀ
rently considered as the prototypes for redoxꢀactive frameꢀ
workꢀinteracting Cuꢀsites in CuꢀSSZꢀ13. Their presence after
O2ꢀactivation at 400 °C has been ascertained by in situ FTIR40,
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
47, 67
detecting their υ(OH) fingerprint band at 3656 cmꢀ1 and
their high stability in the proximity of 1Al sites in the 8r has
47
been confirmed by DFT.43, Nevertheless, from the results
discussed above, it is difficult to conceive that the Z[CuIIOH]
species are directly responsible for the CH4 conversion, as it
has been suggested theoretically.35 In fact, significant populaꢀ
tions of Z[CuIIOH] species have been evidenced after both He
and O2 activation at 250 °C, i.e. when dehydration is completꢀ
ed and before the selfꢀreduction process initiates (see also Secꢀ
tion 2.5).40, 64, 67 However, these kinds of preꢀtreatments do not
yield significant methanol productivity (SI, Section 4).
2.3 Operando XAS of the complete process
We employed operando XAS to follow the oxidation state
and average coordination of the Cu ions in the material during
each step of the methane to methanol conversion over 0.5Cuꢀ
SSZꢀ13(12). Figure 3 shows the XANES and FTꢀEXAFS
spectra collected after each step of the process, namely O2ꢀ
activation at 500 °C, He flush after cooling to 200 °C in O2,
CH4ꢀactivation and H2Oꢀassisted extraction at 200 °C. Each
FTꢀEXAFS spectrum was fitted as described in SI, Section
5.3, to evaluate average Cu coordination environment during
the stepwise process. The procedure adopted for operando
XAS experiments was adjusted to balance between the optiꢀ
mized process conditions and an effective exploitation of the
beamꢀtime. The exact experimental conditions as well as qualꢀ
itative Mass Spectroscopy (MS) data collected during XAS
experiment can be found in the SI. Such conditions resulted in
an overall productivity of 0.100 molCH3OH/molCu measured
in a fixed bed reactor.
Figure 2. XAS characterization of the 0.5CuꢀSSZꢀ13(12) after
different preꢀtreatment and impact of the preꢀtreatment on producꢀ
tivity. (a) Main panel: Cu Kꢀedge XANES spectra; left inset:
backgroundꢀsubtracted CuII 1s → 3d preꢀedge peak; right inset:
first derivative of the XANES spectra reported in the main panel
in the risingꢀedge region. Labels (1) and (2) refer to the risingꢀ
edge 1s → 4p peaks characteristic of CuI and CuI/CuII ions, reꢀ
spectively. (b) Magnitude (left) and imaginary part (right) of the
FTꢀEXAFS spectra for the 0.5CuꢀSSZꢀ13(12) after each preꢀ
treatment, obtained by Fourier transforming the k2χ(k) curves reꢀ
ported Figure S10a (SI) in the (2.4−10.8) Åꢀ1 range and using the
same color code as in part (a). The insets report a magnification of
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 17
The XAS spectrum of O2ꢀactivated 0.5CuꢀSSZꢀ13(12) at
agreement with the formation of a fraction of mobile CuII
1
2
3
4
5
6
7
8
500 °C has been already described in detail in Section 2.2, and
it is reported in Figure 3 (red curves) for comparison. After the
highꢀtemperature O2ꢀactivation step, the sample was cooled
down to 200 °C in O2 and then flushed with He for 60 min.
The XAS spectra collected at this stage depict the actual conꢀ
dition of the material before interaction with CH4. From
XANES, monitoring both the 1s→3d preꢀedge peak and the
risingꢀedge region, it is evident that the oxidation state of Cu
remains unchanged. Nonetheless, XAS evidences an increased
coordination number for the Oef subꢀshell, from 1.0 ± 0.5 to
1.3 ± 0.5. The modifications observed after cooling to 200 °C
are very similar to what was noted in Section 2.2, when, comꢀ
paring the XAS spectra of the material after oxidation at 500
and 200 °C. Indeed, a significant increase in the XANES
whiteꢀline region is observed, accompanied by a 27% increase
in the intensity of the firstꢀshell peak in the FTꢀEXAFS specꢀ
trum with respect to the O2ꢀactivated state at 500 °C. Such
spectral changes are beyond what would be expected solely
due to the decreased thermal contribution to DebyeꢀWaller
(DW) factors when lowering the temperature from 500 to 200
°C (see SI, Section 5.2). Concomitantly, the secondꢀ and thirdꢀ
shell regions are less perturbed, suggesting that, on average,
the coordination mode to the zeolite framework is retained and
that multimeric moieties formed during highꢀtemperature O2ꢀ
activation are still intact.
aquoꢀcomplexes, 24% from LCF. Conversely, the thirdꢀshell
maximum is largely intact after 60 min in 10% H2O/He flow at
200 °C. Based on these evidences, we propose that introducꢀ
tion of water during the methanol extraction provokes a deꢀ
tachment of Cuꢀions from redoxꢀinert positions. If multiple
cycles of the process are run, the mobile CuIIꢀaquo
complexes40, 47, 68 formed during extraction are then thermally
decomposed and able to form new active sites during the O2ꢀ
activation step of the subsequent cycle, justifying the observed
performance enhancement (see Figure 1h).
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
Similarly to what has been previously reported for Cuꢀ
MOR,31ꢀ32 loading of methane at 200 °C on CuꢀSSZꢀ13 results
in the partial reduction of CuII to CuI, as indicated by the deꢀ
velopment of a risingꢀedge peak around 8983 eV. In parallel,
the 1s→ 3d peak undergoes a ca. 35% intensity decrease with
respect to what was observed after He flush at 200 °C, conꢀ
firming the reduction upon interaction with CH4. As described
in SI Section 5.4, we applied linear combination fit (LCF)
analysis to obtain quantitative insights into Cuꢀspeciation durꢀ
ing interaction with CH4. LCF evidences the formation of 27%
CuI species, together with a small fraction (ca. 7%) of hydratꢀ
ed CuII complexes (see Figure 3d). In the FTꢀEXAFS (Figure
3b,c), interaction with CH4 causes a slight decrease in the inꢀ
tensity of the first maximum, in line with a diminished firstꢀ
shell coordination number in the newly formed CuI species
(NO(fw) lowers from 2.0 to 1.6 after methane loading), whereas
in the range from 2.0 to 3.5 Å only minor changes can be obꢀ
served. The reduction of Cuꢀspecies after methane activation
can, on the one hand, result from the formation of the reactive
intermediate, which, according to Sushkevich et al.,6 could
involve the stabilization of methoxy species on the active sites
and the formation of new Brønsted sites. On the other hand,
CuI formation can be connected to the formation of COx speꢀ
cies from O2ꢀdonating Cuꢀsites, prompting the CH4 oxidation.
Noteworthy, oxidation of CH4 to COx would be reasonably
accompanied by production of H2, reported to induce a comꢀ
plete CuII to CuI reduction in both CuꢀSSZꢀ1347 and Cuꢀ
MOR.32 H2 has been indeed detected by MS during operando
XAS, together with overꢀoxidation products (SI, Section 5.8).
Finally, H2Oꢀassisted methanol extraction (Figure 3, light
blue curves) promotes the formation of a significant fraction
of hydrated CuII species, yielding an intensity increase and
shape change in the XANES whiteꢀline region, together with
an enhancement of the firstꢀshell intensity in the FTꢀEXAFS
spectra with respect to the CH4ꢀactivation step. Importantly, at
the end of the extraction step, the wellꢀdefined secondꢀshell
peak appears to be partially degraded, which is in excellent
Figure 3. Operando XAS after each step of the methane converꢀ
sion process: #1 O2ꢀactivation at 500 °C; #2 He flush after cooling
6
ACS Paragon Plus Environment

Page 7 of 17
Journal of the American Chemical Society
to 200 °C in O2 ; #3 CH4ꢀactivation at 200 °C ; #4 H2Oꢀassisted
CH3OH extraction at 200 °C. (a) Main panel: Cu Kꢀedge XANES
1
spectra 0.5CuꢀSSZꢀ13(12) after each process step; left inset:
backgroundꢀsubtracted CuII 1s → 3d preꢀedge peak; right inset:
first derivative of the XANES spectra reported in the main panel
in the risingꢀedge region. Labels (1) and (2) refer to the risingꢀ
edge 1s → 4p peaks characteristic of CuI and CuI/CuII ions, reꢀ
spectively. (b, c) Magnitude (b) and imaginary part (c) of the FTꢀ
EXAFS spectra of 0.5CuꢀSSZꢀ13(12) after each process step, obꢀ
tained by Fourier transforming the k2χ(k) curves reported Figure
S10b (SI) in the (2.4−10.8) Åꢀ1 range. (d) Relative fractions of
frameworkꢀinteracting CuII, hydrated CuII and CuI species during
steps #3 and step #4 derived from linear combination fit (LCF) of
XANES spectra reported in part (a).
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
We also note that, albeit diminished with respect to the CH4ꢀ
activation step, a certain fraction of CuI is still present in the
system: LCF indicates 14% of residual CuI at the end of the
extraction step. By calculating the difference between the relaꢀ
tive fractions of CuI present in the system after methane loadꢀ
ing and extraction steps, we find 13% of total Cu to be reꢀ
oxidized during extraction, in good agreement with the correꢀ
sponding productivity of 0.100 molCH3OH/molCu. This corꢀ
relation suggests that these Cuꢀspecies which undergo reducꢀ
tion when exposed to CH4 and are reꢀoxidized when contacted
with water, directly participate in the reaction.
2.4 Cu-SSZ-13 materials with different compositions
Having explored the effect of the different process parameꢀ
ters on the methanol yield and Cuꢀspeciation for the 0.5Cuꢀ
SSZꢀ13(12), we extended our investigation to CuꢀSSZꢀ13
samples with different elemental compositions (i.e. Cu:Al and
Si:Al ratios).
Figure 4. Productivity and selectivity for the methane to methanol
conversion over CuꢀSSZꢀ13 with different compositional characꢀ
teristics evaluated utilizing the optimized process conditions (i.e.
O2 activation at 500 °C for 480 min in 1000 mbar O2, CH4 activaꢀ
tion at 200 °C for 360 min using 1000 mbar CH4 feed, H2O assistꢀ
ed extraction at 200 °C) (a) normalized methanol productivity
(molCH3OH/molCu) as a function of Cu:Al ratio for CuꢀSSZꢀ13
samples with Si:Al=12, (b) normalized methanol productivity
(molCH3OH/molCu) as a function of Si:Al ratio for CuꢀSSZꢀ13
samples with Cu:Al=0.5.
Using the optimized process conditions (SI, Table S3), we
evaluated samples with Cu:Al from 0.14 to 0.49, and constant
Si:Al of 12 with respect to methanol productivity. Increasing
the Cu content results in higher productivity per gram of samꢀ
ple, while selectivity remains constant at ca. 87 %. Additional
insights can be obtained considering the normalized methanol
productivity, reported in Figure 4a. Interestingly, at Cu:Al ~
0.15 the output is significantly lower than for the higherꢀ
loading samples. Further increasing the Cu:Al ratio to ~ 0.5
only results in a slight increase in the normalized methanol
yield, which reaches 0.172 molCH3OH/molCu. Hence, the
fraction of active species depends on the Cuꢀloading.
In order to interpret the normalized methanol yields derived
from the samples with different Si:Al as a function of the Cu
speciation in the preꢀtreated materials, we applied XAS. In
Section 2.2 we have shown how O2ꢀactivation and highꢀ
temperature reaction with O2 of the Heꢀactivated state result in
almost identical performance, thus yielding equivalent populaꢀ
tions of active Cuꢀsites. This knowledge allows us to decouple
the dehydration, selfꢀreduction and reꢀoxidation steps and obꢀ
tain an inꢀdepth understanding of the underlying redox mechaꢀ
nisms. Figure 5aꢀc shows the XANES and FTꢀEXAFS spectra
of Heꢀactivated CuꢀSSZꢀ13 samples with fixed Cu:Al ratio of
0.5 and Si:Al ratios of 5, 15, 29, namely 0.5CuꢀSSZꢀ13(5),
0.5CuꢀSSZꢀ13(15) and 0.5CuꢀSSZꢀ13(29). The corresponding
spectra collected on the same materials after subsequent oxidaꢀ
tion at 500 °C are reported in Figure 5dꢀf.
We also tested CuꢀSSZꢀ13 materials with similar Cu:Al raꢀ
tios (~ 0.5) at Si:Al ratios varying from 5 to 29. Not surprisꢀ
ingly, the methanol yields increase monotonically with the
amount Cu per gram of sample, reaching 130 µmol/g for the
Si:Al = 5 sample. However, normalization of these results
with respect to the Cu loading leads to a rather different picꢀ
ture (Figure 4b), with the Si:Al = 12 sample showing the highꢀ
est productivity among the four tested materials, equal to
0.172 molCH3OH/molCu. In this comparison, the sample with
the highest aluminum content reaches only a productivity of
0.112 molCH3OH/molCu. Similar observations were reported
also by Wulfers et al., evidencing a poorer performance for
CuꢀSSZꢀ13 zeolites with low Si:Al ratios.15 For samples with
Si:Al ratios of 12 and 15 we measured very similar normalized
productivity, in line with the similar compositional characterꢀ
istics. Conversely, increasing Si:Al to 29 results in a decrease
of the productivity, down to 0.131 molCH3OH/molCu.
The XANES spectra of the Heꢀactivated samples always
show the presence of a certain fraction of CuI, which however
increases in the following order: Si:Al = 15 > 29 > 5. These
results are in line with previous reports, demonstrating that the
SSZꢀ13 framework provides two different cation sites, with
markedly different redox barrier, namely redoxꢀactive 1Al
sites in the 8r and redox inert 2Al sites in the 6r.47, 69ꢀ70
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 17
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 5. (a) Main panel: XANES spectra collected on 0.5CuꢀSSZꢀ13(5) (red curves), 0.5CuꢀSSZꢀ13(15) (black curves) and 0.5CuꢀSSZꢀ
13(29) (blue curves), with fixed Cu:Al = 0.5 and Si:Al = 5, 15 and 29, respectively, after Heꢀactivation at 500 °C; left inset: backgroundꢀ
subtracted preꢀedge 1s → 3d peak; right inset: first derivative of the XANES spectra reported in the main panel in the risingꢀedge region.
(b, c) Magnitude (b) and imaginary part (c) of the FTꢀEXAFS spectra for the Heꢀactivated samples reported in part (a), obtained by Fourier
transforming the k2χ(k) curves reported Figure S10c,d (SI) in the (2.4−10.8) Åꢀ1 range. (dꢀf) XAS spectra for 0.5CuꢀSSZꢀ13(5), 0.5Cuꢀ
SSZꢀ13(15) and 0.5CuꢀSSZꢀ13(29) as parts (aꢀc) but collected after reꢀoxidation in O2 at 500 °C of the Heꢀactivated samples.
Table 1. Fractions of framework-interacting CuII and CuI species evaluated by XANES LCF analysis for samples 0.5Cu-
SSZ-13(5), 0.5Cu-SSZ-13(15) and 0.5Cu-SSZ-13(29) after He-activation together with correspondent LCF R-factors. Total
fraction of Cu species at 1Al sites is also reported, and compared with values extracted from the compositional phase dia-
gram from ref. 47
Fraction of frameworkꢀinteracting Cuꢀspecies
Sample
LCF
2Al Z2CuII
0.51 ± 0.03
0.08 ± 0.03
0.01 ± 0.04
1Al Z[CuIIOH]
0.00 ± 0.02
0.13 ± 0.03
0.35 ± 0.04
1Al ZCuI
Total 1Ala
0.47 ± 0.03
0.92 ± 0.04
0.99 ± 0.05
Total 1Alb
~ 0.50
(He Act. 500 °C)
Rꢀfactor
0.5CuꢀSSZꢀ13(5)
0.5CuꢀSSZꢀ13(15)
0.5CuꢀSSZꢀ13(29)
a This work.
1·10ꢀ4
3·10ꢀ4
6·10ꢀ4
0.47 ± 0.01
0.79 ± 0.01
0.64 ± 0.01
~ 0.85
~ 0.95
b Values extracted from the compositional phase diagram reported by Paolucci et al. in ref. 47
For an accurate quantification of the Cuꢀspeciation, we have
taken advantage of the results of our recent work,64 where adꢀ
vanced multivariate curve resolution (MCR) methods were
applied to extract the theoretical XANES signatures of ‘pure’
Cuꢀspecies formed in Heꢀactivated CuꢀSSZꢀ13 as a function of
the Si:Al ratio. These pure XANES curves, previously conꢀ
nected with wellꢀdefined local geometries of Cuꢀions by multiꢀ
component EXAFS fits,64 have been employed here as referꢀ
ences to assess Cuꢀspeciation from the spectra reported in Figꢀ
ure 5a by linear combination fit (LCF) analysis (see SI, Secꢀ
tion 5.5 for additional details). As expected, Cuꢀspeciation in
the Heꢀactivated state at 500 °C is dominated by frameworkꢀ
interacting Cuꢀspecies, including 1Al sites in their oxidized,
Z[CuIIOH], and reduced, ZCuI, form as well as 2Al Z2CuII
sites. The contributions from mobile CuII complexes are alꢀ
most negligible (< 3% total Cu) for all samples. The fractions
of frameworkꢀinteracting CuII and CuI species evaluated by
LCF analysis are reported in Table 1.
From Table 1 it is evident that for Si:Al = 5 a substantial
contribution of redoxꢀinert 2Al Z2CuII sites is present (51%
8
ACS Paragon Plus Environment

Page 9 of 17
Journal of the American Chemical Society
total Cu), as also reflected in the higher intensity of both first
proportional to the relative abundance of CuI sites available for
highꢀtemperature reaction with O2 to a set of CuII Oy species
1
2
3
4
5
6
7
8
and second maximum in the corresponding FTꢀEXAFS specꢀ
trum with respect to the other Heꢀactivated samples (Figure
5b,c). The sample with Si:Al = 15 shows optimal reducibility:
LCF reveals 79% of ZCuI species and around 10% of both reꢀ
sidual Z[CuIIOH] species and Z2CuII sites in 6r. Concomitantꢀ
ly, the FTꢀEXAFS indicates a majority of twofoldꢀcoordinated
sites, and a dampened/broadened secondꢀshell peak, in relation
with higher mobility of CuI cations and heterogeneity in ZCuI
siting.64, 66 Finally, the sample with Si:Al = 29 shows intermeꢀ
diate reducibility, with 64% of ZCuI and 35% of Z[CuIIOH]
surviving in their oxidized state; at such low Alꢀloading, 2Al
Z2CuII are almost undetectable.
x
among which the active sites should be searched. Thereby, it is
also clear that redoxꢀinert 2Al Z2CuII site are inactive toward
the conversion of methane. Although our recent study64 sugꢀ
gests that a complete ‘2Al saturation’ scenario is unlikely,
there is solid evidence in the literature that points towards a
preferential population of 2Al sites at low Cuꢀloading.47, 70, 73
We can therefore also link the lower normalized productivity
observed in lowꢀloading CuꢀSSZꢀ13 samples (Cu:Al ~ 0.15) at
Si:Al = 12 (see
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 4a) with a higher relative abundance of Z2CuII speꢀ
cies.
These findings fit well into the larger picture developed so
far in the literature about the influence of composition on Cuꢀ
speciation in CuꢀSSZꢀ13 materials prepared by solution ion
exchange. As it can be also observed in Table 1, the total fracꢀ
tions of Cuꢀspecies hosted at 1Al sites determined here by
XANES LCF analysis are in excellent agreement with the corꢀ
respondent values extracted from the theoretical compositional
phase diagram reported by Paolucci et al.47 The presence of
two Cu docking sites yielding Cuꢀspecies with markedly difꢀ
ferent redox properties also provides the basis to rationalize
the two peaks observed during H2ꢀTPR experiments on Cuꢀ
SSZꢀ13 materials.70 Focusing on the compositional impact,
70
Gao et al. systematically analyzed several CuꢀSSZꢀ13 zeoꢀ
lites by TPR with Si:Al ratios of 6, 12 and 35, and Cu:Al ratiꢀ
os in the 0.005ꢀ0.44 range. Two reduction peaks at ∼230 °C
(∼250 °C for the Si:Al = 35 samples) and at ∼380 °C, are obꢀ
served throughout the compositional series, assigned to reducꢀ
tion of partially hydrated CuII and [CuIIOH]+ ions within CHA
cages, and of CuII hosted at 2Al sites in 6R. The trends obꢀ
served in the intensity of the reduction peaks as a function of
the composition corroborates that CuII ions at 2Al Z2CuII sites
possess a significantly higher binding energy to the CHA
framework. TPR results also indicate that their fraction deꢀ
creases by increasing either the Si:Al ratio or the Cu:Al ratio
in the material, in line with the values reported in Table 1.
Figure 6. Correlation between the normalized productivity of Cuꢀ
SSZꢀ13 samples with Si:Al = 5, 15, 29 and their reducibility in the
Heꢀactivated state, quantified through the relative fraction on
ZCuI from LCF of the XANES spectra reported in Figure 5a. The
best fit line through the experimental points is reported in grey.
So far, we have not answered the question about the nucleꢀ
arity of CuII Oy species derived from the reꢀoxidation of ZCuI,
x
and whether the nuclearity correlates with productivity in Cuꢀ
SSZꢀ13. Unfortunately, the EXAFS signature of such species
is only found in the highꢀR range, where the technique sensiꢀ
tivity inherently decreases (especially for highꢀtemperature
measurements on complex, multiꢀcomponent systems). Moreꢀ
over, CuꢀCu contributions prevail over the more abundant Cuꢀ
O ones only in high kꢀrange, where the increased noise could
significantly affect the FTꢀEXAFS signal. To further assess
this aspect, we thus employed Raman spectroscopy as a comꢀ
plementary technique in order to identify fingerprints correlatꢀ
Figure 5dꢀf shows how reꢀoxidation in O2 at 500 °C effecꢀ
tively restores a largely dominant CuII state, irrespectively of
the composition. The characteristic XAS fingerprints of 2Al
Z2CuII sites47, 64 remain a distinctive feature at Si:Al = 5, while
the reꢀoxidized 0.5CuꢀSSZꢀ13(29) and 0.5CuꢀSSZꢀ13(15)
samples show a very similar XANES. However, we observe
significant differences in the FTꢀEXAFS at Si:Al = 15 and 29.
In particular, at Si:Al = 29, a lower firstꢀshell peak is obꢀ
served, together with an overall deterioration of the EXAFS
signal, connected with a higher heterogeneity in the CuII comꢀ
ponent. The absence of a wellꢀdefined third maximum in the
FTꢀEXAFS at Si:Al = 29 suggest that the formation of multiꢀ
nuclear oxo or peroxo CuII species is hampered in this sample.
The development of such EXAFS feature appears to be inꢀ
versely correlated with the Si:Al ratio (see SI, Figure S11), in
line with the statistical availability of suitable docking sites
with AlꢀAl separation < 10 Å71ꢀ72
ed to multimeric CuII Oy species, as described in Section 2.5.
x
2.5 Complementary Results from Raman and FTIR and
Spectroscopy
Raman spectroscopy is a highly sensitive technique to detect
the presence of multinuclear CuII Oy species in zeolites, due to
x
the fact that resonance conditions towards these moieties can
be achieved by proper excitation.25, 60, 74 The Raman spectrum
of the O2 activated 0.5CuꢀSSZꢀ13(15) sample (Figure 7a, red
line) shows a series of new bands which are absent in the hyꢀ
drated sample (Figure 7a, black line), revealing the presence
of several oxygenꢀactivated CuII species.
The bands at 510, 580 and 830 cmꢀ1 were also observed by
Ipek et.al25 for activated CuꢀSSZꢀ13 and assigned to [Cu(transꢀ
µꢀ1,2ꢀO2)Cu]2+ species. In addition, we observe strong Raman
Correlating the LCF results reported in Table 1 with the perꢀ
formance of the investigated Cuꢀzeolites with Si:Al = 5, 15,
29, we found a linear correlation between the normalized
productivity and the sample reducibility in the Heꢀactivated
state, quantified through the fraction of ZCuI. The productiviꢀ
tyꢀreducibility correlation is highlighted in Figure 6, demonꢀ
strating that the normalized methanol productivity is directly
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 17
bands between ~1000ꢀ1200 cmꢀ1 (exhibiting maxima at 1100
and 1155 cmꢀ1) for the activated sample, in agreement with the
the most intense feature of the SSZꢀ13 framework (485 cmꢀ1) and
vertically shifted for sake of visualization.
1
•
presence of [CuIIO2 ]+ superoxo species.74 The broadness of
2
3
4
5
6
7
8
9
To further complement the characterization of relevant Cuꢀ
species for the conversion of methane to methanol, we chose
Infrared (IR) spectroscopy, to monitor the O2ꢀactivation step to
400 °C. In the 3800 to 3500 cmꢀ1 range, the spectra exhibit
common features related to ν(OꢀH) modes of silanols at (3737
cmꢀ1) as well as Brønsted sites at 3611 cmꢀ1 and 3584 cmꢀ1.40ꢀ41
Another distinct band around 3650 cmꢀ1 gradually decreases as
the temperature increases. In previous studies, such band has
been identified as the fingerprint of [CuIIOH]+ species in Cuꢀ
SSZꢀ13.40 The spectrum collected at 400 °C shows a clear deꢀ
pletion in intensity of the specific feature which decreases furꢀ
ther after 150 mins at that temperature. As discussed in the
previous sections, such species are unlikely to be the active
species for the methane activation. However, the data confirms
that high temperature combined with prolonged exposure time
in oxidative atmosphere depletes the population of [CuIIOH]+
sites. This trend is in agreement with the picture obtained from
XAS and Raman spectroscopy, indicating that during O2 actiꢀ
vation at high temperature [CuIIOH]+ species further evolve,
acting as precursors for new species, most probably the active
sites, as will be discussed in the following section.
the bands, suggests that the Cuꢀsuperoxo species are located at
various Cu sites, as well as in different coordination symmeꢀ
tries (endꢀon vs sideꢀon).75ꢀ76 The presence of [Cuꢀ(µꢀO)ꢀCu]2+
species is not easy to access, due to the very intense band at
580 cmꢀ1, which is the range where one would expect the
symmetric metalꢀoxo stretch of such species.25 However, a
shoulder at ~620 cmꢀ1 adumbrates their presence. In this conꢀ
text, Ipek et.al observed a fast intensity decrease of the band at
580 cmꢀ1 with time, which they ascribed to small amounts of
water leaking into the system and the high sensitivity of the
[Cu(transꢀµꢀ1,2ꢀO2)Cu]2+ species towards hydrolysis. Interestꢀ
ingly, we found [Cu(transꢀµꢀ1,2ꢀO2)Cu]2+ to be sensitive to the
measurements conditions, such as different sampling times
and laser powers (Figure 7b). By increasing the energy at the
sample through higher laser power and/or longer sampling
time the intensity of the 580 cmꢀ1 band increases. At very low
laser power (~0.1 mW, Figure 7b, purple line) or at short exꢀ
posure at ~1 mW (Figure 7b, pink line), the latter is diminꢀ
ished significantly whereas a band at 620 cmꢀ1 (assigned to the
[Cuꢀ(µꢀO)ꢀCu]2+ species) 25 becomes visible. In the case of the
spectrum of the activated sample (Figure 7a, red line), we
aimed for a highꢀquality spectrum and therefore applied a
sampling time of 120 min at laser output power of ~1mW. The
high energy input therefore explains the high intensity of the
band at 580 cmꢀ1, overshadowing the much weaker band at
620 cmꢀ1 (observed as a shoulder in this case).
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
Since the samples were sealed in capillaries at ~550 °C (see
SI, Section 1.5), we can exclude a hydrationꢀrelated effect as
25
suggested by Ipek et.al. We suspect in our case a temperaꢀ
tureꢀrelated effect, due to laser heating, as the cause for the
observed changes in the spectrum. However, such behavior
requires a careful evaluation and will be the object of future
investigation.
Figure 8. O2 activation of sample 0.5CuꢀSSZꢀ13(15) monitored by
in situ FTIR spectroscopy in the 250–400°C range, in the ν(OH)
region. The dotted line highlights the progressive decrease in the
intensity of the band at ca. 3650 cmꢀ1 (fingerprinting Z[CuIIOH])
as temperature increases.
3. THE FORMATION OF ACTIVE CU SPECIES
Figure 7. (a) Raman spectra of 0.5CuꢀSSZꢀ13(15) in its hydrated
(black line) and activated (red line) forms. The spectral contribuꢀ
We have shown that the availability of CuI sites able to react
with O2 at high temperature has a key impact on the methanol
productivity. The reducibility of CuꢀSSZꢀ13, and, in general,
the resulting Cu speciation, is highly dependent on the compoꢀ
sition, in terms of Si:Al and Cu:Al ratios. Optimal methanol
productivity is associated with optimal reducibility, so that the
compositional characteristics of the material have to be tuned
in order to maximize the relative fraction of Cu sites efficientꢀ
ly undergoing selfꢀreduction during activation. As depicted in
tions from the observed CuII Oy moieties are highlighted:
x
[Cu(transꢀµꢀ1,2ꢀO2)Cu]2+ (green filling); [Cuꢀ(µꢀO)ꢀCu]2+ (orange
• +
filling); and [CuIIO2 ] (lightꢀblue filling). (b) Raman spectra of
the activated sample, collected at different exposure times and
excitation laser powers. The spectra on activated 0.5CuꢀSSZꢀ
13(15) have been collected with the sample inside sealed capillarꢀ
ies (see SI, section 1.5). All the spectra have been normalized to
10
ACS Paragon Plus Environment

Page 11 of 17
Journal of the American Chemical Society
Scheme 1a and further corroborated by a number of previous operation temperature. As discussed in Section 2.3, cooling
studies,40, 42, 45, 47, 64, 69ꢀ70, 73, 77 this condition is met at intermediꢀ
ate Si:Al ratios (~ 12−15) and high Cuꢀloading (~ 0.5). Inꢀ
stead, low Si:Al and low Cuꢀloading promote large populaꢀ
tions of redoxꢀinert 2Al Z2CuII, trapping Cu sites in stable, inꢀ
active positions in the 6r. However, when the Si:Al ratio inꢀ
creases beyond the optimum (~ 30), although the relative fracꢀ
tion of 1Al sites is maximized, the redox potential of such speꢀ
cies is diminished, suggesting that the efficiency of the selfꢀ
reduction process depends on the Brønsted acid site density in
the material. Z[CuIIOH] complexes at 1Al sites efficiently unꢀ
dergoing selfꢀreduction (middle panel in Scheme 1a) are thus
identified as the precursors to the active sites responsible for
the methane to methanol conversion in CuꢀSSZꢀ13.
down to methane loading temperature (i.e. 200 °C) during op-
erando XAS experiments, we observe modifications in both
XANES and FTꢀEXAFS spectra pointing to an increased coꢀ
ordination number in the CuꢀOef shell at lower temperature.
The XAS data are consistent with a partial temperatureꢀ
induced transition from endꢀon to sideꢀon O2 binding modes,
resulting in threefold and fourfold coordinated CuIIꢀsites, reꢀ
spectively. We observe the same spectroscopic fingerprints,
but more pronounced, after reꢀoxidation of the preꢀreduced
sample at 200 °C, which also leads to a significantly lower
methanol yield compared to reꢀoxidation at 500 °C. These obꢀ
servations suggest that a tridentate frameworkꢀinteracting CuII
site with endꢀon binding mode of an O2ꢀderived ligand, as it
seen e.g. in species 3a, 4a or 6, represents the most favorable
site geometry for methane conversion.
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
In Scheme 1b we give an overview of the possible routes for
the formation of active CuII Oy species, which we will criticalꢀ
x
ly discuss by combining the results presented in the previous
sections with key findings from literature. Complementary,
Table 2 reports a selection of spectroscopic fingerprints for the
Cuꢀspecies presented in Scheme 1b. It is worth to note that
Scheme 1b only considers CuI and CuII species, since the forꢀ
mation and the involvement of CuIII moieties, e.g. bis(ꢁꢀoxo)
dicopper(III), have been ruled out based on the absence of any
CuIII evidence either from XAS60,78 and from Raman79 in all the
experimental conditions monitored in the present study.
Apart from the formation of CuxOy species through the reacꢀ
tion of O2 with CuI sites, internal pathways have also been
proposed in the literature, based on experimental evidences
and theoretical studies25, 31, 35, 62ꢀ63, 82 (Scheme 1b, bottom part).
Such pathways explain the residual productivity observed after
preꢀtreatment in inert. If the OH• released during the selfꢀ
reduction of a first Z[CuIIOH] is able to reach another
Z[CuIIOH] complex, a ZCuIIO• (5) could be formed together
with an H2O molecule.35, 62ꢀ63 Albeit species (5) has been preꢀ
viously indicated as one of the most effective in methane actiꢀ
vation, 34, 83ꢀ84 to the best of our knowledge this highly reactive
intermediate has never been detected in Cuꢀzeolites, hence its
existence and role remain speculative.
The homolytic cleavage of the Cu−OH bond in the
Z[CuIIOH] precursor (1 in Scheme 1b), yielding a reduced
ZCuI site (2) and an HO• radical is envisaged as an initial key
step towards the active species formation. Temperatureꢀ
dependent IR data (Figure 8) evidence that this step is also ocꢀ
curring to some extent in O2 flux. The methaneꢀconverting
DFT studies actually support the selfꢀorganization of monoꢀ
nuclear oxygenated/hydroxylated cations in zeolite frameꢀ
works towards oxygenꢀbridged binuclear complexes (black
dashed arrows in Scheme 1b). Computational analysis predicts
energetic benefits as high as 400 kJ mol−1 associated with such
selfꢀorganization phenomena, in connection with the stabilizaꢀ
tion of underꢀcoordinated metal centers with excessively basic
terminal O/OH ligands.82 Interestingly, tridentate ZCuIIO•
complexes in CuꢀMFI are predicted to dimerize into either
bis(ꢁꢀoxo)ꢀdicopper or ꢁꢀ(η2:η2)ꢀperoxo dicopper cores for
two different docking sites with Al−Al separations of 5.48 and
8.72 Å, respectively.82 Similarly, condensation of two
Z[CuIIOH] complexes at a suitable distance will result in the
formation of a H2O molecule and a monoꢀ(ꢁꢀoxo) dicopper
core (6 in Scheme 1b), currently indicated as the active site for
direct methanol conversion in CuꢀZSMꢀ5 and CuꢀMOR6, 31ꢀ32,
CuII Oy species can be primarily formed by direct reaction beꢀ
x
tween O2 and the available ZCuI sites (Scheme 1b, top part).
This is supported by our observation that introduction of O2
after Heꢀactivation at 500 °C promptly restores a largely domꢀ
inant CuII state rather irrespectively of the tested reꢀoxidation
temperature: ~ 90% total Cu is found as CuII both carrying out
the reꢀoxidation at 500 and at 200 °C. Starting from isolated
ZCuI sites, CuII moieties possibly formed under these condiꢀ
tions include ZCuIIO2 superoxo complexes, with either endꢀ
•
on Cu(η1ꢀO2) or side on Cu(η2ꢀO2) binding modes, species
(3a) and (3b), respectively. Multimeric peroxoꢀCuII species
such as ZCuII(O2)CuIIZ (4) can be also formed. The probability
to stabilize such multimeric moieties depends on the proximity
of Al atoms in the zeolite framework: Al−Al separations up to
~10 Å are envisaged to enable suitable docking sites for
neighboring Cu ions. As may be observed in Scheme 1b, havꢀ
ing two ZCuI satisfying the aforementioned criterion, it is posꢀ
sible that they react with O2 to ZCuII(O2)CuIIZ species, e.g.
forming [Cu(transꢀµꢀ1,2ꢀO2)Cu]2+ (4a) or [Cu(µꢀη2:η2ꢀ
O2)Cu]2+ complexes (4b).
34, 74
Highꢀtemperature treatment in O2ꢀdeficent atmosphere
has been proposed to favour desorption of the bridging oxygen
from the oxocationic species,81, 85 leaving two Cu(I) sites and
an Oef vacancy. In this context, Smeets et al.81 demonstrated
that first a ꢁꢀ(η2:η2)ꢀperoxo dicopper species is formed by reꢀ
acting O2 at RT with a preꢀreduced CuꢀZSMꢀ5 zeolite, which
is then transformed into monoꢀ(ꢁꢀoxo) dicopper upon thermalꢀ
lyꢀdriven oxygen elimination from 175 °C upwards (orange
dashed orange arrow in Scheme 1b). Our Raman data indicate
that the monoꢀ(ꢁꢀoxo) dicopper core (6) is also present in airꢀ
activated CuꢀSSZꢀ13, as seen in Figure 7 and shown by Ipek
et.al.25 Beside bisꢀµꢀoxo dicopper species, Cuꢀtrimers have
been also proposed to form in O2ꢀactivated CuꢀMOR.12 Moreꢀ
over, from DFT, clusters of even higher nuclearity have been
suggested as favorable sites for the methane conversion.33ꢀ34
Nonetheless, combining our XAS and Raman results on Cuꢀ
SSZꢀ13, no evidence for CuxOy species with x > 2 is obtained.
Indeed, EXAFS analysis supports the formation of multinuꢀ
clear CuII species in CuꢀSSZꢀ13 with average CuꢀCu distances
of ~ 3.4 Å, involving ca. 30% of Cu sites in the O2ꢀactivated
material. As summarized in Table 2, with Raman we were able
•
to identify the spectroscopic fingerprints of both ZCuIIO2 suꢀ
peroxo complexes in different coordination modes (3) and of
[Cu(transꢀµꢀ1,2ꢀO2)Cu]2+ (4a). Conversely, species 4b was
previously identified in CuꢀZSMꢀ5,81 but has not been detected
by Raman neither by us nor in the recent study by Ipek et al.25
Importantly, the average coordination mode, sideꢀonꢀ or
endꢀonꢀlike, in such species is found to be influenced by the
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 17
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
Scheme 1. (a) Rationalization of the effect of composition (Cu:Al and Si:Al ratios) on the productivity for the methane to
methanol conversion over Cu-SSZ-13. (b) Possible mechanisms for the formation of active CuII Oy moieties from precursor
x
1Al Z[CuIIOH] sites, considering CuII Oy formation from reaction of self-reduced ZCuI sites with O2 (top part) and from in-
x
ternal pathways (bottom part). Selected spectroscopic fingerprints for the Cu-species (1)-(6) presented in the scheme, are
summarized in Table 2, comparing previous literature and experimental observations in the present work. Species indicated
with grey labels have been previously detected in Cu-zeolites (4b) or proposed as intermediates (5) but are not directly ob-
served in this study on Cu-SSZ-13 materials.
12
ACS Paragon Plus Environment

Page 13 of 17
Journal of the American Chemical Society
Table 2. Selected spectroscopic fingerprints for the Cu species presented in Scheme 1, derived from previous literature on
Cu-zeolites and/or Cu model compounds and experimental data reported in the present work. FTIR frequencies and Ra-
man shifts are reported in cm-1.
1
2
3
4
5
6
Relevant FTIR, Raman and XAS spectroscopic fingerprints
Cu species in Scheme 1
Previous literature [ref.
]
This work
FTIR: ν(OꢀH) ~ 3650 [40ꢀ41, 47
]
FTIR: ν(OꢀH) 3650
XAS: CuII, 3Ca
XAS: CuII, 3C [40ꢀ41, 47
]
(1) Z[CuIIOH]
7
8
9
N2ꢀFTIR: ZCuIꢀN2 2300, 2992 [64, 67
]
COꢀFTIR: ZCuI(CO) 2155; ZCuI(CO)2
2178, 2148; ZCuI(CO)3 2194, 2169 (sh),
2134 (sh) [67]
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
(2) ZCuI
XAS: CuI, 2C
XAS: CuI, 2C [40, 47, 64
]
(3a) [Cu(η1ꢀO2)]+
(3b) [Cu(η2ꢀO2)]+
Raman: ν(OꢀO) 1121, ν(CuꢀO) 472 [74ꢀ75
Raman: ν(OꢀO) 1043, ν(CuꢀO) 554 [74, 76
]
Raman: ν(OꢀO) broad bands with
maxima at 1100 and 1155
XAS: CuII 3C/4C
•
(3) ZCuIIO2
]
Raman:
ν(OꢀO)
ν(CuꢀO)
580
830,
507,
(laserꢀ
Raman: ν(OꢀO) 837, ν(CuꢀO) 510, 580
[25]; ν(OꢀO) 832, ν(CuꢀO) 561 [25, 80
(4a)[Cu(transꢀµꢀ1,2ꢀO2)Cu]2+
(4b) [Cu(µꢀη2:η2ꢀO2)Cu]2+
(6) [Cuꢀ(µꢀO)ꢀCu]2+
]
(4) ZCuII(O2)CuIIZ
XAS:
sensitive)
CuII 3C/4C;
Cu−Cu 3.42 Å
Raman: ν(OꢀO) 736, ν(CuꢀO) 269 [81];
ν(OꢀO) 763, ν(CuꢀO) 284 [80];
Not observed in [25]
Raman:
Not observed
(5) ZCuIIO•
Not identified
Not identified
Raman: ν(CuꢀO) 616ꢀ617 [25];
XAS: CuII 3C; Cu−Cu ~ 2.9 Å in
CuꢀMOR [31] and CuꢀZSMꢀ5 [57]
Raman: ν(CuꢀO) 618
(6) ZCuIIOCuIIZ
XAS: CuII 3C; Cu−Cu 3.42 Å
a 2C, 3C,4C = 2ꢀ, 3ꢀ, 4ꢀfold coordinated Cu species with Oef and Ofw atoms the first coordination sphere, respectively.
Based on these considerations, the residual CuII component
4. CONCLUSIONS
observed after Heꢀactivation of CuꢀSSZꢀ13 with Si:Al in the
12ꢀ15 range could be ascribed to a mixture of inactive Z2CuII
The individual process steps in the direct conversion of meꢀ
sites in 6r and active ZCuIIOCuIIZ cores still retaining their
thane to methanol over CuꢀSSZꢀ13 materials have been invesꢀ
tigated in detail and relationships between productivity and
Cuꢀspeciation have been established. Highꢀtemperature treatꢀ
ment at 500 °C in oxygen and a prolonged activation time posꢀ
itively impact the productivity, as a result of an increased forꢀ
mation of reactive CuII species. We evidence how high partial
pressure of CH4 at 200 °C is required to achieve saturation of
the active centers with reactive intermediates, and illustrate the
possibility of isothermal extraction of methanol. Employing
optimized process conditions, we are able to achieve a methaꢀ
nol productivity equal of 0.172 molCH3OH/molCu (107
µmol/g). Intriguingly, four cycles of the process drive the
productivity to 0.202 molCH3OH/molCu (125 µmol/g). This
increase is related to the waterꢀassisted mobilization of the
metal centers during the extraction step, driving higher fracꢀ
tions of Cu into the active framework positions.
oxo bridge, which would account for the corresponding nonꢀ
zero yield. These internal pathways to CuII Oy moieties can, in
x
principle, occur also during O2ꢀactivation in parallel to reacꢀ
tions of newly formed CuI sites. In this case, the O2ꢀrich atꢀ
mosphere is expected to limit the aforementioned oxygen deꢀ
sorption phenomena, thus enriching the population of ZCuIIꢀ
OCuIIZ cores. In CuꢀSSZꢀ13, these latter species have so far
been assumed only to form in 8r containing 2 Al atoms, correꢀ
sponding to DFTꢀoptimized CuꢀCu distances in the 2.6ꢀ3.1 Å
range, depending on the number of Si atoms separating the
two Al sites.36, 72 The longer CuꢀCu distances found here by
EXAFS, more similar to what predicted for monoꢀ(ꢁꢀoxo) diꢀ
copper in the 12r of CuꢀMOR (CuꢀCu ~ 3.5 Å72) could suggest
these species to form also by bridging two Cu centers distinctꢀ
ly located in 8r and 6r,as previously found from DFT for a
ZCuIIO2CuIIZ complex.25, 71
Summarizing Scheme 1, Z[CuIIOH] precursors can evolve
to a set of triꢀcoordinated CuII moieties either from direct reacꢀ
tion of selfꢀreduced ZCuI with O2, giving [Cu(η1ꢀO2)]+ superꢀ
oxo or [Cu(transꢀµꢀ1,2ꢀO2)Cu]2+ peroxo complexes, or from
internal pathways, resulting e.g. in monoꢀ(ꢁꢀoxo) dicopper
cores. The abundance of multiꢀcopper species is definitely deꢀ
termined by the availability and distribution of Al in the zeoꢀ
lite.
Our results support triꢀcoordinated frameworkꢀinteracting
CuII centers, with two Ofw and one Oef ligands in the first metal
coordination sphere, as the most favorable sites for methane
conversion. We found this coordination mode to be dominant
at 500 °C, whereas at 200 °C fourꢀcoordinated CuII species apꢀ
pear to be also present, resulting in a lower productivity. By
coupling the XAS spectroscopic insights with the results from
activity measurements, we were able to exclude the direct inꢀ
volvement of Z[CuIIOH] complexes. These species progresꢀ
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 17
8. Lieberman, R. L.; Rosenzweig, A. C., Nature 2005, 434, 177ꢀ182.
sively deplete during exposure to high temperature oxidative
treatment, and are proposed to be precursor to the active speꢀ
cies.
9. Balasubramanian, R.; Smith, S. M.; Rawat, S.; Yatsunyk, L. A.;
Stemmler, T. L.; Rosenzweig, A. C., Nature 2010, 465, 115ꢀ9.
10. Tinberg, C. E.; Lippard, S. J., Acc. Chem. Res. 2011, 44, 280ꢀ8.
11. Groothaert, M. H.; Smeets, P. J.; Sels, B. F.; Jacobs, P. A.;
Schoonheydt, R. A., J. Am. Chem. Soc. 2005, 127, 1394ꢀ1395.
12. Grundner, S.; Markovits, M. A.; Li, G.; Tromp, M.; Pidko, E. A.;
Hensen, E. J.; Jentys, A.; SanchezꢀSanchez, M.; Lercher, J. A., Nat.
Commun. 2015, 6, 7546.
13. Alayon, E. M. C.; Nachtegaal, M.; Kleymenov, E.; van Bokhoven, J.
A., Microporous Mesoporous Mater. 2013, 166, 131ꢀ136.
14. Beznis, N. V.; Weckhuysen, B. M.; Bitter, J. H., Catal. Lett. 2010,
138, 14ꢀ22.
15. Wulfers, M. J.; Teketel, S.; Ipek, B.; Lobo, R. F., Chem. Commun.
2015, 51, 4447ꢀ4450.
16. Kalamaras, C.; Palomas, D.; Bos, R.; Horton, A.; Crimmin, M.;
Hellgardt, K., Catal. Lett. 2016, 146, 483ꢀ492.
17. Hammond, C.; Forde, M. M.; Ab Rahim, M. H.; Thetford, A.; He, Q.;
Jenkins, R. L.; Dimitratos, N.; LopezꢀSanchez, J. A.; Dummer, N. F.;
Murphy, D. M.; Carley, A. F.; Taylor, S. H.; Willock, D. J.; Stangland, E.
E.; Kang, J.; Hagen, H.; Kiely, C. J.; Hutchings, G. J., Angew. Chem. Int.
Ed. 2012, 51, 5129ꢀ33.
18. Shan, J.; Huang, W.; Nguyen, L.; Yu, Y.; Zhang, S.; Li, Y.; Frenkel,
A. I.; Tao, F. F., Langmuir 2014, 30, 8558ꢀ69.
19. Krisnandi, Y. K.; Putra, B. A. P.; Bahtiar, M.; Zahara; Abdullah, I.;
Howe, R. F., Procedia Chem. 2015, 14, 508ꢀ515.
20. Bozbag, S. E.; Alayon, E. M. C.; Pecháček, J.; Nachtegaal, M.;
Ranocchiari, M.; van Bokhoven, J. A., Catal. Sci. Technol. 2016, 6, 5011ꢀ
5022.
21. Alayon, E. M.; Nachtegaal, M.; Ranocchiari, M.; van Bokhoven, J. A.,
Chem. Commun. 2012, 48, 404ꢀ6.
22. Grundner, S.; Luo, W.; SanchezꢀSanchez, M.; Lercher, J. A., Chem.
Commun. 2016, 52, 2553ꢀ6.
23. Narsimhan, K.; Iyoki, K.; Dinh, K.; RomanꢀLeshkov, Y., ACS Cent.
Sci. 2016, 2, 424ꢀ9.
1
2
3
4
5
6
7
8
Via operando XAS we directly probed the CuII to CuI reꢀ
duction of 27% of the Cu sites during methane loading. Howꢀ
ever, a certain fraction of the CuI formed during interaction
with methane likely originates from pathways leading to overꢀ
oxidation products. Indeed, during H2Oꢀassisted extraction, we
observed the reꢀoxidation of only about half of the previously
reduced Cu sites (13% total Cu), identified as the actual reacꢀ
tive sites.
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
Furthermore, we explored the correlations between eleꢀ
mental composition (Cu:Al and Si:Al ratios) in CuꢀSSZꢀ13
and methanol productivity. The combination of testing and
XAS measurements clearly evidences a positive linear correlaꢀ
tion between the methanol productivity and the reducibility of
the Cu centers. The high redox inertness of 2Al Z2CuII sites in
6r, favored at low Si:Al and low Cuꢀloading, results in comꢀ
pletely inactive Cu sites for the methane to methanol converꢀ
sion. Instead, intermediate Si:Al ratios (~ 12−15) and high Cuꢀ
loading (~ 0.5) lead to a high population of Z[CuIIOH] sites,
ultimately resulting in high methanol yields per Cu.
ASSOCIATED CONTENT
The Supporting Information is available free of charge via the Inꢀ
ternet at http://pubs.acs.org. Material Synthesis; Laboratory Charꢀ
acterization; Experimental Details; XAS Analysis; Linear Combiꢀ
nation Fitting;
24. Ipek, B.; Lobo, R. F., Chem. Commun. 2016, 52, 13401ꢀ13404.
25. Ipek, B.; Wulfers, M. J.; Kim, H.; Göltl, F.; Hermans, I.; Smith, J. P.;
Booksh, K. S.; Brown, C. M.; Lobo, R. F., ACS Catal. 2017, 7, 4291ꢀ
4303.
26. Sheppard, T.; Daly, H.; Goguet, A.; Thompson, J. M., ChemCatChem
2016, 8, 562ꢀ570.
27. Xu, J.; Armstrong, R. D.; Shaw, G.; Dummer, N. F.; Freakley, S. J.;
Taylor, S. H.; Hutchings, G. J., Cat. Today 2016, 270, 93ꢀ100.
28. Tsai, M.ꢀL.; Hadt, R. G.; Vanelderen, P.; Sels, B. F.; Schoonheydt, R.
A.; Solomon, E. I., J. Am. Chem. Soc. 2014, 136, 3522ꢀ3529.
29. Sheppard, T.; Hamill, C. D.; Goguet, A.; Rooney, D. W.; Thompson,
J. M., Chem. Commun. 2014, 50, 11053ꢀ5.
30. Le, H. V.; Parishan, S.; Saaltchik, A.; Gobel, C.; Schlesiger, C.;
Malzer, W.; Trunschke, A.; Schomacker, R.; Thomas, A., ACS Catal.
2017, 7, 1403ꢀ1412.
31. Alayon, E. M.; Nachtegaal, M.; Bodi, A.; Ranocchiari, M.; van
Bokhoven, J. A., Phys Chem Chem Phys 2015, 17, 7681ꢀ93.
32. Alayon, E. M. C.; Nachtegaal, M.; Bodi, A.; van Bokhoven, J. A., ACS
Catal. 2014, 4, 16ꢀ22.
33. Palagin, D.; Knorpp, A. J.; Pinar, A. B.; Ranocchiari, M.; van
Bokhoven, J. A., Nanoscale 2017, 9, 1144ꢀ1153.
34. Tomkins, P.; Ranocchiari, M.; van Bokhoven, J. A., Acc. Chem. Res.
2017, 50, 418ꢀ425.
35. Kulkarni, A. R.; Zhao, Z.ꢀJ.; Siahrostami, S.; Nørskov, J. K.; Studt, F.,
ACS Catal. 2016, 6, 6531ꢀ6536.
36. Mahyuddin, M. H.; Staykov, A.; Shiota, Y.; Miyanishi, M.;
Yoshizawa, K., ACS Catal. 2017, 3741ꢀ3751.
37. Brandenberger, S.; Krocher, O.; Tissler, A.; Althoff, R., Catal. Rev.
2008, 50, 492ꢀ531.
38. Deka, U.; LezcanoꢀGonzalez, I.; Weckhuysen, B. M.; Beale, A. M.,
ACS Catal. 2013, 3, 413ꢀ427.
39. Beale, A. M.; Gao, F.; LezcanoꢀGonzalez, I.; Peden, C. H. F.; Szanyi,
J., Chem. Soc. Rev. 2015, 44, 7371ꢀ7405.
40. Borfecchia, E.; Lomachenko, K. A.; Giordanino, F.; Falsig, H.; Beato,
P.; Soldatov, A. V.; Bordiga, S.; Lamberti, C., Chem. Sci. 2015, 6, 548ꢀ
563.
41. Giordanino, F.; Borfecchia, E.; Lomachenko, K. A.; Lazzarini, A.;
Agostini, G.; Gallo, E.; Soldatov, A. V.; Beato, P.; Bordiga, S.; Lamberti,
C., J. Phys. Chem. Lett. 2014, 5, 1552ꢀ9.
AUTHOR INFORMATION
Corresponding Authors
* Elisa Borfecchia: elisa.borfecchia@unito.it
* Stian Svelle: stian.svelle@kjemi.uio.no
* Pablo Beato: pabb@topsoe.com
ACKNOWLEDGMENT
We are grateful to A. Longo, M. Monte Caballero and D. Motta
Meira for the competent and friendly support during our XAS exꢀ
periments at BM26A and BM23 beamlines of ESRF. DKP and
MD acknowledge the iCSI (industrial Catalysis Science and Innoꢀ
vation) Centre of Researchꢀbased Innovation, which receives fiꢀ
nancial support from the Norwegian Research Council under conꢀ
tract no. 237922. EB acknowledges Innovation Fund Denmark
(Industrial postdoc n. 5190ꢀ00018B). CL, KAL and IP
acknowledge the Megaꢀgrant of the Russian Federation Governꢀ
ment to support scientific research at the Southern Federal Uniꢀ
versity, No. 14.Y26.31.0001. C. Negri and L. Altavilla are
acknowledged for support in FTIR and Raman data collection,
respectively.
REFERENCES
1. Gesser, H. D.; Hunter, N. R.; Prakash, C. B., Chem. Rev. 1985, 85, 235ꢀ
244.
2. Lunsford, J. H., Cat. Today 2000, 63, 165ꢀ174.
3. Saha, D.; Grappe, H. A.; Chakraborty, A.; Orkoulas, G., Chem. Rev.
2016, 116, 11436ꢀ11499.
4. Horn, R.; Schlögl, R., Catal. Lett. 2015, 145, 23ꢀ39.
5. da Silva, M. J., Fuel Process. Technol. 2016, 145, 42ꢀ61.
6. Sushkevich, V. L.; Palagin, D.; Ranocchiari, M.; van Bokhoven, J. A.,
Science 2017, 356, 523ꢀ527.
7. Tomkins, P.; Mansouri, A.; Bozbag, S. E.; Krumeich, F.; Park, M. B.;
Alayon, E. M.; Ranocchiari, M.; van Bokhoven, J. A., Angew. Chem. Int.
Ed. 2016, 55, 5467ꢀ71.
14
ACS Paragon Plus Environment

Page 15 of 17
Journal of the American Chemical Society
42. Gao, F.; Walter, E. D.; Karp, E. M.; Luo, J.; Tonkyn, R. G.; Kwak, J.
H.; Szanyi, J.; Peden, C. H. F., J. Catal. 2013, 300, 20ꢀ29.
43. Janssens, T. V. W.; Falsig, H.; Lundegaard, L. F.; Vennestrøm, P. N.
R.; Rasmussen, S. B.; Moses, P. G.; Giordanino, F.; Borfecchia, E.;
Lomachenko, K. A.; Lamberti, C.; Bordiga, S.; Godiksen, A.; Mossin, S.;
Beato, P., ACS Catal. 2015, 5, 2832ꢀ2845.
44. Tyrsted, C.; Borfecchia, E.; Berlier, G.; Lomachenko, K. A.; Lamberti,
C.; Bordiga, S.; Vennestrøm, P. N. R.; Janssens, T. V. W.; Falsig, H.;
Beato, P.; PuigꢀMolina, A., Catal. Sci. Technol. 2016, 6, 8314ꢀ8324.
45. Lomachenko, K. A.; Borfecchia, E.; Negri, C.; Berlier, G.; Lamberti,
C.; Beato, P.; Falsig, H.; Bordiga, S., J. Am. Chem. Soc. 2016, 138,
12025ꢀ8.
46. Godiksen, A.; Vennestrøm, P. N. R.; Rasmussen, S. B.; Mossin, S.,
Top. Catal. 2016, 60, 13ꢀ29.
47. Paolucci, C.; Parekh, A. A.; Khurana, I.; Di Iorio, J. R.; Li, H.;
Albarracin Caballero, J. D.; Shih, A. J.; Anggara, T.; Delgass, W. N.;
Miller, J. T.; Ribeiro, F. H.; Gounder, R.; Schneider, W. F., J Am Chem
Soc 2016, 138, 6028ꢀ48.
48. Paolucci, C.; Verma, A. A.; Bates, S. A.; Kispersky, V. F.; Miller, J.
T.; Gounder, R.; Delgass, W. N.; Ribeiro, F. H.; Schneider, W. F., Angew.
Chem. Int. Ed. 2014, 53, 11828ꢀ33.
49. Bordiga, S.; Groppo, E.; Agostini, G.; van Bokhoven, J. A.; Lamberti,
C., Chem. Rev. 2013, 113, 1736ꢀ1850.
50. Mathon, O.; Beteva, A.; Borrel, J.; Bugnazet, D.; Gatla, S.; Hino, R.;
Kantor, I.; Mairs, T.; Munoz, M.; Pasternak, S.; Perrin, F.; Pascarelli, S.,
J. Synchrot. Radiat. 2015, 22, 1548ꢀ1554.
51. Nikitenko, S.; Beale, A. M.; van der Eerden, A. M. J.; Jacques, S. D.
M.; Leynaud, O.; O'Brien, M. G.; Detollenaere, D.; Kaptein, R.;
Weckhuysen, B. M.; Bras, W., J. Synchrot. Radiat. 2008, 15, 632ꢀ640.
52. Kim, Y.; Kim, T. Y.; Lee, H.; Yi, J., Chem. Commun. 2017, 53, 4116ꢀ
4119.
53. Park, M. B., Ahn, S. H., Ranocchiari, M. and van Bokhoven, J.,
ChemCatChem. 2017, doi 10.1002/cctc.201700768.
54. Llabrés i Xamena, F. X.; Fisicaro, P.; Berlier, G.; Zecchina, A.;
Palomino, G. T.; Prestipino, C.; Bordiga, S.; Giamello, E.; Lamberti, C., J.
Phys. Chem. B 2003, 107, 7036ꢀ7044.
55. Prestipino, C.; Berlier, G.; Xamena, F.; Spoto, G.; Bordiga, S.;
Zecchina, A.; Palomino, G. T.; Yamamoto, T.; Lamberti, C., Chem. Phys.
Lett. 2002, 363, 389ꢀ396.
56. Lamberti, C.; Bordiga, S.; Salvalaggio, M.; Spoto, G.; Zecchina, A.;
Geobaldo, F.; Vlaic, G.; Bellatreccia, M., J. Phys. Chem. B 1997, 101,
344ꢀ360.
57. Groothaert, M. H.; van Bokhoven, J. A.; Battiston, A. A.;
Weckhuysen, B. M.; Schoonheydt, R. A., J. Am. Chem. Soc. 2003, 125,
7629ꢀ7640.
58. Kau, L. S.; Spirasolomon, D. J.; Pennerhahn, J. E.; Hodgson, K. O.;
Solomon, E. I., J. Am. Chem. Soc. 1987, 109, 6433ꢀ6442.
59. Sano, M.; Komorita, S.; Yamatera, H., Inorg. Chem. 1992, 31, 459ꢀ
463.
71. Falsig, H.; Vennestrøm, P. N. R.; Moses, P. G.; Janssens, T. V. W.,
Top. Catal. 2016, 59, 861ꢀ865.
1
2
3
4
5
6
7
8
72. Vilella, L.; Studt, F., Eur. J. Inorg. Chem. 2016, 2016, 1514ꢀ1520.
73. Bates, S. A.; Verma, A. A.; Paolucci, C.; Parekh, A. A.; Anggara, T.;
Yezerets, A.; Schneider, W. F.; Miller, J. T.; Delgass, W. N.; Ribeiro, F.
H., J. Catal. 2014, 312, 87ꢀ97.
74. Woertink, J. S.; Smeets, P. J.; Groothaert, M. H.; Vance, M. A.; Sels,
B. F.; Schoonheydt, R. A.; Solomon, E. I., Proc. Natl. Acad. Sci. U.S.A.
2009, 106, 18908ꢀ13.
75. Chen, P.; Root, D. E.; Campochiaro, C.; Fujisawa, K.; Solomon, E. I.,
J. Am. Chem. Soc. 2003, 125, 466ꢀ74.
76. Maiti, D.; Fry, H. C.; Woertink, J. S.; Vance, M. A.; Solomon, E. I.;
Karlin, K. D., J. Am. Chem. Soc. 2007, 129, 264ꢀ5.
77. Gao, F.; Walter, E. D.; Kollar, M.; Wang, Y.; Szanyi, J.; Peden, C. H.
F., J. Catal. 2014, 319, 1ꢀ14.
78. DuBois, J. L.; Mukherjee, P.; Collier, A. M.; Mayer, J. M.; Solomon,
E. I.; Hedman, B.; Stack, T. D. P.; Hodgson, K. O., J. Am. Chem. Soc.
1997, 119, 8578ꢀ8579.
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
79. Henson, M. J.; Mukherjee, P.; Root, D. E.; Stack, T. D. P.; Solomon,
E. I., J. Am. Chem. Soc. 1999, 121, 10332ꢀ10345.
80. Baldwin, M. J.; Ross, P. K.; Pate, J. E.; Tyeklar, Z.; Karlin, K. D.;
Solomon, E. I., J. Am. Chem. Soc. 1991, 113, 8671ꢀ8679.
81. Smeets, P. J.; Hadt, R. G.; Woertink, J. S.; Vanelderen, P.;
Schoonheydt, R. A.; Sels, B. F.; Solomon, E. I., J. Am. Chem. Soc. 2010,
132, 14736ꢀ8.
82. Pidko, E. A.; Hensen, E. J. M.; van Santen, R. A., Proc. R. Soc. A-
Math. Phys. Eng. Sci. 2012, 468, 2070ꢀ2086.
83. Gagnon, N.; Tolman, W. B., Acc Chem Res 2015, 48, 2126ꢀ31.
84. Shiota, Y.; Yoshizawa, K., Inorg Chem 2009, 48, 838ꢀ45.
85. Jang, H. J.; Hall, W. K.; dItri, J., J. Phys. Chem. 1996, 100, 9416ꢀ
9420.
60. Solomon, E. I.; Heppner, D. E.; Johnston, E. M.; Ginsbach, J. W.;
Cirera, J.; Qayyum, M.; KieberꢀEmmons, M. T.; Kjaergaard, C. H.; Hadt,
R. G.; Tian, L., Chem. Rev. 2014, 114, 3659ꢀ3853.
61. Paolucci, C.; Di Iorio, J. R.; Ribeiro, F. H.; Gounder, R.; Schneider,
W. F., Adv. Catal. 2016, 59, 1ꢀ107.
62. Larsen, S. C.; Aylor, A.; Bell, A. T.; Reimer, J. A., J. Phys. Chem.
1994, 98, 11533ꢀ11540.
63. Palomino, G. T.; Fisicaro, P.; Bordiga, S.; Zecchina, A.; Giamello, E.;
Lamberti, C., J. Phys. Chem. B 2000, 104, 4064ꢀ4073.
64. Martini, A.; Borfecchia, E.; Lomachenko, K. A.; Pankin, I.; Negri, C.;
Berlier, G.; Beato, P.; Falsig, H.; Lamberti, C.; Bordiga, S., Chem. Sci.
2017, doi: 10.1039/c7sc02266b.
65. Andersen, C. W.; Borfecchia, E.; Bremholm, M.; Jorgensen, M. R. V.;
Vennestrom, P. N. R.; Lamberti, C.; Lundegaard, L. F.; Iversen, B. B.,
Angew. Chem. Int. Ed. 2017, 56, 10367ꢀ10372.
66. Göltl, F.; Sautet, P.; Hermans, I., Cat. Today 2016, 267, 41ꢀ46.
67. Giordanino, F.; Vennestrom, P. N.; Lundegaard, L. F.; Stappen, F. N.;
Mossin, S.; Beato, P.; Bordiga, S.; Lamberti, C., Dalton Trans. 2013, 42,
12741ꢀ61.
68. McEwen, J. S.; Anggara, T.; Schneider, W. F.; Kispersky, V. F.;
Miller, J. T.; Delgass, W. N.; Ribeiro, F. H., Cat. Today 2012, 184, 129ꢀ
144.
69. Kwak, J. H.; Zhu, H. Y.; Lee, J. H.; Peden, C. H. F.; Szanyi, J., Chem.
Commun. 2012, 48, 4758ꢀ4760.
70. Gao, F.; Washton, N. M.; Wang, Y. L.; Kollar, M.; Szanyi, J.; Peden,
C. H. F., J. Catal. 2015, 331, 25ꢀ38.
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 17
1
2
3
For Table of Contents Only
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
16
ACS Paragon Plus Environment

Page 17 of 17
Journal of the American Chemical Society
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
324x176mm (96 x 96 DPI)
ACS Paragon Plus Environment