Methane-to-Methanol on Mononuclear Copper(II) Sites Supported on Al2O3: Structure of Active Sites from Electron Paramagnetic Resonance**

Article abstract of DOI:10.1002/anie.202105307

The selective conversion of methane to methanol remains one of the holy grails of chemistry, where Cu-exchanged zeolites have been shown promote this reaction under stepwise conditions. Over the years, several active sites have been proposed, ranging from mono-, di- to trimeric Cu^II. Herein, we report the formation of well-dispersed monomeric Cu^II species supported on alumina using surface organometallic chemistry and their reactivity towards the selective and stepwise conversion of methane to methanol. Extensive studies using various transition alumina supports combined with spectroscopic characterization, in particular electron paramagnetic resonance (EPR), show that the active sites are associated with specific facets, which are typically found in γ- and η-alumina phase, and that their EPR signature can be attributed to species having a tri-coordinated [(Al₂O)CuIIO(OH)]^? T-shape geometry. Overall, the selective conversion of methane to methanol, a two-electron process, involves two monomeric Cu^II sites that play in concert.

Full text of DOI:10.1002/anie.202105307

Angewandte
Research Articles
Chemie
How to cite:
International Edition: doi.org/10.1002/anie.202105307
German Edition: doi.org/10.1002/ange.202105307
Methane-to-Methanol
Methane-to-Methanol on Mononuclear Copper(II) Sites Supported on
Al2O3: Structure of Active Sites from Electron Paramagnetic
Resonance**
Jordan Meyet, Anton Ashuiev, Gina Noh, Mark A. Newton, Daniel Klose, Keith Searles,
Alexander P. van Bavel, Andrew D. Horton, Gunnar Jeschke, Jeroen A. van Bokhoven,* and
Christophe Copꢀret*
Abstract: The selective conversion of methane to methanol
remains one of the holy grails of chemistry, where Cu-
exchanged zeolites have been shown promote this reaction
under stepwise conditions. Over the years, several active sites
have been proposed, ranging from mono-, di- to trimeric CuII.
Herein, we report the formation of well-dispersed monomeric
CuII species supported on alumina using surface organome-
tallic chemistry and their reactivity towards the selective and
stepwise conversion of methane to methanol. Extensive studies
using various transition alumina supports combined with
spectroscopic characterization, in particular electron para-
magnetic resonance (EPR), show that the active sites are
associated with specific facets, which are typically found in g-
and h-alumina phase, and that their EPR signature can be
attributed to species having a tri-coordinated [(Al2O)CuIIO-
(OH)]ꢀ T-shape geometry. Overall, the selective conversion of
methane to methanol, a two-electron process, involves two
monomeric CuII sites that play in concert.
CH4 is highly desirable, this reaction remains challenging due
to the low reactivity of CH4 compared to that of CH3OH
which leads to facile generation of over-oxidized and un-
wanted products (e.g. CO, CO2).[1–4] In nature, methane
monooxygenases (MMOs, with either Fe or Cu active sites)
are highly efficient at performing this selective oxidation.[5]
The particulate form of this enzyme (pMMO) possesses
several copper-containing sites for which the exact structure
and nuclearity are still under debate.[6–8] Initial studies
ascribed the active site as a dimeric copper species present
in the PmmoB subunit (CuB, Figure 1a).[9] A more recent
report, supported by electron paramagnetic resonance (EPR)
spectroscopy, however, has shown the exclusive presence of
mononuclear sites in pMMO; the CuC site located in the
PmmoC subunit has been assigned to the metal binding site
responsible for hydrocarbon binding and oxidation (Fig-
ure 1a).[8]
In parallel, promising processes based on inorganic
materials have been developed in recent years. These
processes rely on chemical looping, which decouples an
aerobic (O2) oxidation step from the CH4 activation step, and
therefore curtails over-oxidation of the methoxy species
formed.[10,11] The most-studied materials within the chemical
looping paradigm are copper-exchanged zeolites. Different
types of active centers have been proposed based on
spectroscopic characterization.[12] These proposals include
a dinuclear mono-m-oxo [Cu-O-Cu]2+ center[13] (Figure 1a)
and a trinuclear [Cu3O3]2+ core.[14] In addition, monomeric Cu
sites [CuOH]+ in chabazite,[15] and two monomers present in
mordenite have been proposed based on density functional
theory (DFT) calculations.[16] Recently, the formation of
Introduction
The direct and selective conversion of CH4 to CH3OH
could transform the petrochemical industry, by enabling on-
site conversion of CH4 on a scale much smaller than is
currently viable. Existing routes, which are both energy-
intensive and indirect, require the generation of syngas (H2/
CO) through CH4 reforming, which can then be converted to
CH3OH. Cheap and abundant CH4 is the main constituent of
natural gas (70–90%), but its transportation in liquid form is
energy demanding and comes at additional cost. Although the
development of catalytic processes for the partial oxidation of
[*] J. Meyet, A. Ashuiev, Dr. G. Noh, Dr. M. A. Newton, Dr. D. Klose,
Dr. K. Searles, Prof. Dr. G. Jeschke, Prof. Dr. J. A. van Bokhoven,
Prof. Dr. C. Copꢀret
[**] A previous version of this manuscript has been deposited on
a preprint server (https://doi.org/10.26434/chemrxiv.13295879.v2).
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.202105307.
Department of Chemistry and Applied Biosciences, ETH Zurich
Vladimir-Prelog-Weg 1–5, 8093 Zꢁrich (Switzerland)
E-mail: ccoperet@ethz.ch
ꢂ 2021 The Authors. Angewandte Chemie International Edition
published by Wiley-VCH GmbH. This is an open access article under
the terms of the Creative Commons Attribution Non-Commercial
License, which permits use, distribution and reproduction in any
medium, provided the original work is properly cited and is not used
for commercial purposes.
Prof. Dr. J. A. van Bokhoven
Laboratory for Catalysis and Sustainable Chemistry, Paul Scherrer
Institute
5232 Villigen (Switzerland)
E-mail: jeroen.vanbokhoven@chem.ethz.ch
Dr. A. P. van Bavel, Dr. A. D. Horton
Shell Global Solutions International B.V.
Grasweg 31, 1031 HW Amsterdam (The Netherlands)
&&&&
ꢂ 2021 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
These are not the final page numbers!

Angewandte
Research Articles
Chemie
Figure 1. a) Proposed dinuclear (left) and mononuclear (right) active sites in the enzyme Pmmo (green frame) and the equivalent (in terms of
copper nuclearity) analogues proposed for Cu supported on oxide supports (blue frame); b) The strategy used to generate monomeric CuII sites
in a pure alumina environment and the subsequent reaction with CH4 to form hydroxy and methoxy species.
paired monomeric [CuOH]+ in zeolite Omega has been
reported using anomalous X-ray diffraction.[17] However,
despite major efforts, the active site structures remain highly
debated. This deficit notwithstanding, substantial improve-
ments in CH3OH yield per chemical looping cycle have been
achieved to reach the theoretical limit of CH3OH for two CuII
sites.[18] However, state-of-the-art zeolite-based materials
remain economically unviable for commercial applications
because of an insufficient amount of reactive sites per unit
mass and the long cycle times associated with this step-wise
process.[19,20] Thus, alternate materials must be explored for
this challenging transformation. While the discovery of more
efficient materials is of great importance to overcome some
limitations faced by zeolites materials, the identification and
characterization of the reactive centers is required to provide
a structure/reactivity relationship as a guideline to improve
material performance. Among possible alternative materials
explored, copper supported on silica is shown to produces
CH3OH,[21–23] however the obtained yield is low and the active
sites are ill-defined (small CuO clusters).
CuII sites (Figure 1b) even if the reactivity of CuII centers in
a pure aluminum environment is yet to be explored.
Herein we describe the formation of well-dispersed CuII
sites on g-alumina using the SOMC/TMP approach, and
extend the scope of the study to different transition alumina
(q, d, a) to understand the influence of the support on the
formation of reactive centers. Further analysis of the best
performing material by EPR spectroscopy and reactivity with
CO as probe molecule was then used to identify the structure
of the active sites. In particular, in situ EPR spectroscopy
allows for the spectroscopic identification of reactive CuII
surface sites involved in the selective partial oxidation of CH4.
These reactive sites display a unique EPR signature, which
can be observed for CuII on g-, h- and q-Al2O3 but not for non-
defective a-Al2O3. Combining X-ray absorption spectroscopy
(XAS), advanced EPR methods and computational modeling
enabled the assignment of the reactive monomeric CuII sites
on alumina as a tri-coordinated [(Al2O)CuO(OH)]ꢀ species.
Recently, we showed that proximal monomeric CuII sites Results and Discussion
on alumina in an aluminosilicate environment, that resembled
this of zeolites, could convert CH4 to CH3OH.[24] The material
was generated via the surface organometallic chemistry
(SOMC) and thermolytic molecular precursor (TMP) ap-
proach,[25] allowing the formation of highly-dispersed CuII
centers with a coordination sphere dictated by the choice of
tailored molecular precursor. However, the mixed (Si/Al)
surface environment arising from the use of CuII siloxide
precursors, necessary to reproduce the chemical environment
found in zeolites, also led to the formation of multiple sites
and prevented to obtain any detailed structural character-
ization of the reactive CuII sites.
The copper aluminate molecular precursor,[26] [Cu(k2-
Al(OtBu)4)2] (1) (see Supporting Information, Section 3,
Figures S1–S5 for the full characterization of 1) was grafted
on g-Al2O3 (“g-a”) dehydroxylated under vacuum at varied
temperatures (denoted by X in degree Celsius in “g-aX”): g-
a
500 (OH density, V= 2.4 OHnmꢀ2), g-a700 (V= 0.9 OHnmꢀ2)
and g-a800 (V= 0.4 OHnmꢀ2) (see also Supporting Informa-
tion, Figures S6 and S7), to yield the corresponding grafted
materials 1-g-aX. These grafted materials were characterized
by UV–Vis, diffuse reflectance infrared (DRIFTS), EPR and
X-ray absorption near-edge spectroscopies (XANES) (Sup-
porting Information, Section 4, Figures S7–S9). All 1-g-aX00
materials show similar spectroscopic features, indicating the
generation of similar surface species after grafting irrespec-
tive of the initial dehydroxylation temperature for g-Al2O3.
The presence of CuII was confirmed by both EPR and the
We thus reasoned that the formation of well-dispersed
CuII in a pure alumina environment, using a copper-aluminate
molecular precursor, would enable the identification of the
active site structure while keeping the high reactivity of these
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
ꢂ 2021 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
www.angewandte.org
&&&&
These are not the final page numbers!

Angewandte
Research Articles
Chemie
observation in the electronic spectra of the d-d transition
characteristic of CuII species. However, the presence of a pre-
edge feature at 8983 eV in the XANES, characteristic of CuI
species, indicates the partial reduction of Cu (Figure S9), as
previously observed when different CuII precursors were
grafted on alumina.[24]
Thermal activation of these 1-g-aX materials under a flow
of dry synthetic air at 5008C leads to the exclusive generation
of organic-free CuII surface species (1500-g-aX), as demon-
strated by IR, UV–Vis, EPR and XANES (Section 4,
Figures S10–S12). Possible auto-reduction of the Cu center
under thermal activation, as previously reported for zeolitic
materials,[27,28] can be excluded as no pre-edge features for CuI
could be observed in the XANES spectra (Figure S12).
Continuous wave (CW) X-band EPR analysis of the activated
material (Figure S12) shows the presence of paramagnetic
CuII species in a nearly axial environment (gk > g? > ge) with
well-resolved Cu hyperfine coupling, which confirms the
presence of monomeric CuII species well-dispersed on the
alumina support.[24] Comparison of the CW EPR spectra
amongst the activated Cu generated on the partially dehy-
droxylated alumina at different temperatures (vide infra),
allows the identification of at least two different Cu species
based on the observed parallel transitions (Ak and gk): Ak
ꢁ 370 MHz and gk ꢁ 2.39 for one site (denoted site I) and Ak
ꢁ 440 MHz and gk ꢁ 2.33 for the remainder of the signal
(denoted site II, but which is composed of multiple species;
vide infra) (Figure S12).
Figure 2. a) In situ X-band EPR spectra (298 K) for 1500-g-a500 before
and after reaction under 6 bar of CH4 for 30 min at 2008C b) X-band
EPR difference spectrum (298 K) before–after reaction.
The ratio between these two sites is highly dependent on
the initial dehydroxylation temperature of the alumina
support prior to grafting. A dehydroxylation temperature of
5008C (1500-g-a500) yields predominantly site I, whereas
dehydroxylation at 8008C (1500-g-a800) gives primarily site II.
The activated materials, 1500-g-aX, were reacted under 6
bars of CH4 at 2008C for 30 min and isolated under inert
conditions for further spectroscopic characterization.
DRIFTS analysis after reaction shows the appearance of
two bands at 2957 and 2853 cmꢀ1 assigned to the asymmetric
and symmetric stretch of methoxy surface species, respec-
tively (Figure S13).[29]
absence of Cu sites under similar conditions does not yield
any CH3OH products.[24]
To understand the origin of the observed reactivity, the
reaction with CH4 was monitored by in situ EPR spectroscopy
(Figure 2a). For all materials, the initial spectra include
a mixture of sites, while after reaction only the signal
associated with site I completely disappears, leaving a broad
signal consisting of a distribution of unreactive Cu monomers
described as site II (Figures 2a, S15,S16). The partial dis-
appearance of the EPR signal can be rationalized by the
reduction of reactive CuII species to EPR-silent CuI. Fur-
thermore spin quantification, obtained from double integrals
of the signal before and after reaction, is consistent with the
expected number of electrons (2eꢀ for CH4 to CH3OH)
required for the selective reaction with CH4 (Figure 3a).[24]
This CuII/CuI redox process is further evidenced by Cu K-edge
XANES analysis of material 1500-g-a500 before and after
reaction, which shows the emergence of a CuI pre-edge
feature (Figure S17) similar to previous reports for zeolitic-[31]
and alumina-based systems.[24] Thus, the EPR signal of the
reactive CuII species can be obtained by subtracting the
spectra between the reacted and unreacted materials, reveal-
ing the spectroscopic signature of the reactive monomeric
CuII site (site I). In contrast to many spectroscopic techniques,
which characterize both reactive and spectator CuII sites, this
EPR signature belongs exclusively to the reactive centers
involved in the redox reaction.
This assignment was confirmed by 13C solid-state nuclear
magnetic resonance (SSNMR) of 1500-g-a500 reacted with
13CH4, where two distinct peaks are observed in the 13C NMR
spectra: a major broad peak at 48.3 ppm along with a minor
peak at 64 ppm, assigned, respectively to surface methoxy/
methanol (CH3O(H)-) and small amounts of dimethyl ether
(DME) (Figure S14).[30] DME is the only side product
observed in 1500-g-a500 under reaction conditions according
to NMR and infrared spectroscopy.
After extraction with water at 1208C, the 1500-g-a500
material shows the highest CH3OH productivity among all
samples with 0.12 mol CH3OHmolꢀ1 of Cu. The CH3OH
yields for the other materials are lower, with 1500-g-a800
yielding only half the amount of CH3OH (0.06 mol
CH3OHmolꢀ1 of Cu; Supporting Information, Table 1). A
recycling study was also performed using 1-g-a700 calcined at
7008C (1700-g-a700) showing similar CH3OH yields before and
after recycling with ca. 0.09 CH3OHmolꢀ1 of Cu (Supporting
Information, Table S1). The reaction of g-Al2O3 in the
The maximum ratio of reduced CuII, determined from the
double integrals of the signals before and after reaction, was
observed for 1500-g-a500 with 32% of CuI formed for a CH3OH
&&&&
www.angewandte.org
ꢂ 2021 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
These are not the final page numbers!

Angewandte
Research Articles
Chemie
Figure 3. a) Reducibility of the monomeric Cu site determined by double integral of the EPR signal for the initial material less that of the reacted
material (in % of Cu) versus CH3OH yield obtained after reaction with CH4 in mol CH3OHmolꢀ1 of Cu, for the different Cu supported transition
aluminas tested. Deviation from the theoretical limit (dashed line), indicating the two electrons required for the formation of CH3OH, provides
information on the material selectivity toward CH3OH, and b) X-band difference EPR spectra (298 K) before–after reaction of 1500-z-a500 supported
on various transitional alumina displaying identical reactive CuII monomeric site I.
yield of 0.12 molCH3OHmolꢀ1 of Cu (Figure 3a). Additional
information regarding the amount of CuII centers reduced
upon reaction with CH4 was obtained from linear combina-
tion analysis (LCA) of the XANES spectra for 1500-g-a500; it
corresponds to ca. 44% of CuI formed upon reaction with
methane. The higher proportion of reduced Cu centers
obtained from LCA analysis compared to the one obtained
by EPR spectroscopy suggests the presence of a small amount
of EPR silent reactive CuII centers, possibly associated with
the presence of small antiferromagnetically coupled CuII-oxo
clusters.[32]
better understand the formation and identity of this reactive
monomeric Cu species. Transition aluminas provide different
anchoring sites on the support by exposing different low-
index planes and facets, although the nature of specific facets
remains a matter of debate.
The g- and h-Al2O3 phases are of particular interest
because of their high Lewis acidity and similar reactivity.[33] In
fact, both are able to coordinate N2 with a blue shifted IR
band at 2355 cmꢀ1, assigned to adsorption at tri-coordinated
Al (AlIII) for g-Al2O3.[34] Such highly Lewis acidic under-
coordinated sites are absent on a-Al2O3 surface.
To further investigate the presence of small multimeric
CuII-oxo centers, the material was characterized by EXAFS
spectroscopy. Cu K-edge EXAFS analysis of 1500-g-a500 by
(Figure S18, Table S2) indicates an average of 3 O in the first
coordination sphere of the Cu center and, crucially, the
absence of a Cu-Cu path consistent with the presence of
mostly CuII monomers. Comparing the first shell between the
initial and reacted states does not reveal significant changes in
the local Cu–O coordination sphere (Figure S18, Table S2).
Despite the absence of Cu-O-Cu scattering in the deconvo-
lution of the EXAFS, the presence of minor contributions of
multimeric CuII species cannot be excluded.
Further interpretation of the EXAFS is, however, limited
by a relatively reduced dataset and low signal-to-noise ratio.
Moreover, EXAFS averages over all (both active and
inactive) Cu species present; and, in these cases these
different contributions cannot be disentangled with any
degree of precision or specificity.
Similar grafting and activation procedures were applied
for these different transition aluminas, leading to the gen-
eration of well-dispersed CuII species (Figures S19–S24)
supported on Al2O3, 1500-z-aX (where “z” denotes the phase
of the transition alumina employed). Post reaction with CH4,
the surface speciation for each of these materials, was again
probed using DRIFTS (Figure S25). Material reactivity was
assessed by in situ EPR spectroscopy, coupled with CH3OH
quantification (Figures S26–S30, Table S3).
In all cases, except for 1500-a-a500, only sites I and II were
observed in EPR spectra of the activated material (Figur-
es S26–S29). Upon reaction with CH4, only site I disappeared.
For 1500-a-a500, CH3OH could not be quantified, and the
low intensity EPR signal observed for CuII (Figure S30), likely
due to the very low specific surface area (3 m2 gꢀ1) and the
small amounts of supported Cu, is unperturbed by reaction
with CH4. The CH3OH yield measured for all other materials,
however, correlates with the disappearance of EPR signal
(Figure 3a) as was previously observed in the case of 1500-g-aX.
The signal disappearance correspond to the reduction of
reactive CuII sites (sites I), leaving only unreactive CuII
species (sites II).
By contrast, EPR spectroscopy permits more selective
probing of the reactive site (vide supra). Taking advantage of
the specificity of this spectroscopy, a wider range of transi-
tional alumina supports was explored (h-, q- and a-Al2O3) to
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
ꢂ 2021 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
www.angewandte.org
&&&&
These are not the final page numbers!

Angewandte
Research Articles
Chemie
Figure 4. a) X-Band EPR spectra (103 K) for 1500-h-a500 before and after reaction under 6 bar of CH4 for 30 min at 2008C, and b) X-band EPR
difference spectrum (103 K) for 1500-h-a500 before–after reaction (solid line) and the resulting simulation (dashed line) parametrizing the spectrum
of the active species.
ꢀ
Greater reactivity is observed for 1500-h-a500 (0.16 mol
CH3OHmolꢀ1 Cu), compared to that for 1500-g-a500 (0.12 mol
CH3OHmolꢀ1 Cu), which could be rationalize by higher
Lewis acidity for h-a500 compared to g-a500, and the presence
of greater amounts of defective undercoordinated aluminum
(AlIII) surface sites. These AlIII, and their relative amounts, are
evidenced by the area of the band of coordinated N2 observed
in the IR spectra of the dehydroxylated Al2O3 (Figure S19).
This enhanced reactivity suggests that Cu sites are located in
close proximity to surface AlIII defective sites.[35–37]
Low-temperature CW EPR, and simulation of the EPR
parameters, was performed for 1500-h-a500, which possesses the
highest fraction of reactive site I. The EPR spectra of the
activated and reacted material for 1500-h-a500 (Figure 4a) are
similar to those observed for 1500-g-a500. The resulting differ-
ence spectra, corresponding to the reactive, nearly axial,
monomeric CuII species observed on all the materials studied,
can be simulated with the g tensor parameters g (site I) =
[2.0625 2.0787 2.3923], and hyperfine coupling ACu (MHz) =
[30 38 400] (Figure 4b), while the inactive sites (site II)
correspond to at least two distinct species (Supporting
Information, Figure S31). Double electron-electron reso-
nance (DEER) measurements of 1500-g-a500 before and after
reaction with CH4 (Figure S32) indicates a homogeneous 3D
distribution of the CuII centers, indicating that spatial
proximity between the CuII centers is not necessary for their
reactivity.
associated with stable bicarbonate (HOCO2 ) surface inter-
mediates formed during the oxidation of CO, as previously
reported for the interaction of CO2 with g-Al2O3.[38] The same
experiment monitored by EPR spectroscopy results in partial
disappearance of site I along with appearance of a new EPR
signal (Figure 5b & Figure S34).
Difference spectra confirm the involvement of the reac-
tive monomeric CuII site for the one-electron oxidation of CO
ꢀ
to a HOCO2 C radical anion. Similar reactivity was observed
for 1500-h-Al2O3–500 under identical conditions (Figure S35).
The new signal, which has an effective g value of 2.0026,
possesses an effective super-hyperfine coupling of 50 MHz, as
determined from simulation (Figure 5c). The presence of 11
lines for the radical species formed can be explained by
hyperfine interaction between the radical center with two
equivalent Al nuclei (Aeff(27Al) = 50 MHz). Similar super-
hyperfine couplings were observed by Gafurov et al. upon
contacting g-Al2O3 with anthraquinone, which were associ-
ated to a tetra-coordinated AlIV pair.[39]
This experiment with CO further supports the assignment
of a unique reactive monomeric CuII species (site I) for the
selective conversion of CH4 to CH3OH. From the reactivity
observed during the one-electron oxidation of carbon mon-
oxide, one can probe the chemical surrounding of the active
center. The formation of a bicarbonate radical along with the
observed hyperfine coupling constant is consistent with the
presence of a hydroxyl group and two equivalent AlIV in the
coordination sphere of the reactive CuII center. We further
investigated the reactive species (site I) in 1500-h-a500, which
contains the highest amount of reactive CuII, with hyperfine
sublevel correlation spectroscopy (HYSCORE).[40] Note that
the inactive species (site II) in fact did not show characteristic
hyperfine couplings in X-band HYSCORE spectra (see
Figure S37) such that this method allows for selective
observation of the 1H and 27Al couplings of the reactive
species.
To gain further insight into the reactive center identity,
1500-g-a500 was contacted with CO as a probe molecule and
monitored by IR and EPR (Figures S33 & S34). After
introduction of 110 mbar of CO, followed by evacuation,
new features are observed by IR in the range of 1700–
1200 cmꢀ1. The difference between the reacted and activated
spectra reveal the appearance of bands at 3611 cmꢀ1 (nOH),
1647 cmꢀ1 (ns OCO), 1479 cmꢀ1 (ns OCO), 1443 cmꢀ1 (ns OCO),
1233 cmꢀ1 (dOH) (Figure 5a & Figure S32). These bands are
&&&&
www.angewandte.org
ꢂ 2021 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
These are not the final page numbers!

Angewandte
Research Articles
Chemie
Figure 5. a) FTIR and b) X-band EPR spectra (298 K) of 1500-g-a500 before (blue) and after (red) reaction with 110 mbar of CO at room temperature
and the corresponding difference spectrum (black). c) Subtracted EPR signal of the carbonate radical formed upon reaction with CO (black) and
simulation (dashed red) [geff =2.0026 and Aeff =50 MHz].
Figure 6. a) X-band echo detected spectra recorded at 10 K for materials before (1500-h-a500, red) and after reaction with CH4 (1500-h-a500 reacted,
blue). The dashed lines indicate the field positions used to record the HYSCORE spectra. b) X-band HYSCORE spectrum of 1500-g-Al2O3–500 at 10 K
recorded at the field of 326.6 mT, sum of three spectra with t=[280 312 400] ns. Spectral intensity along the antidiagonal of (ꢀ+) quadrants is
1
caused by phase cycling imperfection. Antidiagonal lines in the HYSCORE spectra indicate nuclear frequencies of H (dashed) and 27Al (dotted).
The experimental spectrum is displayed in blue to yellow and the simulated 1H ridge in red. c) DFT calculated structure of the reactive
tricoordinated [(Al2O)CuO(OH)]ꢀ site present on the (110) facet of g-Al2O3. d) A possible 2 electron mechanism for CH4 activation with 2
[(Al2O)CuO(OH)]ꢀ through hydrogen abstraction followed reaction of the methyl radical.
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
ꢂ 2021 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
www.angewandte.org
&&&&
These are not the final page numbers!

Angewandte
Research Articles
Chemie
The HYSCORE spectrum of the reactive species, mea-
sured at the magnetic field position corresponding to the
maximum intensity of the echo-detected EPR spectrum
(Figure 6a), reveals the presence of coupled 1H nuclei in
a CuI surface methoxy or CH3OH species via reaction with
a second CuII site (Figure 6d).
close proximity to the CuII center, together with a distribution Conclusion
of 27Al couplings (Figure 6b). The HYSCORE spectrum of
the reactive species was simulated (Figure 6b & Figure S36),
Overall, this study shows that alumina is able to stabilize
assuming the presence of two equivalent 27Al nuclei, with the
highly reactive isolated CuII sites. EPR spectroscopy is shown
here to be a particularly powerful characterization method,
allowing the exclusive detection of the reactive monomeric
sites involved in the partial oxidation of CH4. The detection of
a specific EPR signature for such active sites, [(Al2O)CuO-
(OH)]ꢀ, which adopt a T-shape structure, is particularly
noteworthy as it opens ways to tailor materials that can
stabilize such type of structures. We are currently investigat-
ing alternative supports that could stabilize such species with
the aim to increase the density of reactive sites.
related hyperfine and quadrupole couplings of aiso
=
3.39 MHz, adip = [ꢀ0.88 0.03 0.86] MHz, P = ꢀ10.03 MHz
(h = 0.185). These parameters are quite close to both 27Al
hyperfine and quadrupole tensors, estimated for the molec-
ular precursor 1 (see Supporting Information, Section 3). This
indicates that the Al atoms of the reactive species possess
similar symmetry and coordinative surrounding as in 1. The
1H hyperfine coupling in 1500-h-a500 was simulated to be aiso
=
ꢀ0.5 ꢂ 0.5 MHz,
a
dip = [ꢀ4.75 ꢂ 0.25; ꢀ4.75 ꢂ 0.25; 9.5 ꢂ
0.5] MHz, which corresponds to a Cu-H distance of 2.57 ꢂ
0.05 ꢀ within the point-dipole approximation. Such proton
distance is attributed to the presence of hydroxyl group bound
to the CuII active sites.
Acknowledgements
The presence of a single, unique EPR signature for all
reactive CuII centers (site I) formed on different transition
aluminas allows us to refine the geometrical configuration
and surrounding environment of this center through combin-
ing the experimental evidences derived for the 1500-g-Al2O3–500
and 1500-h-Al2O3–500 materials. The CuII active sites have the
following characteristics: i) they possess a coordinated hy-
droxyl group, ii) they are in close proximity to 2 equivalent
AlIV sites, as shown for g- and h-Al2O3 and iii) their formation
is correlated to the presence of highly Lewis acidic AlIII sites.
Such features are strongly reminiscent of the structural
characteristics of the (110) facet of g-Al2O3,[41] or the related
structures formed at the edge between the (110) and (100)
facets.[42] We therefore explored possible structures for the
active sites that would feature a characteristic, nearly-axial
EPR signature, as well as reasonable formation energies.
These CuII species were modelled by adsorbing a CuO
fragment onto various partially hydroxylated configurations
of the (110) facet of g-Al2O3 (1 H2O per unit cell or
3 OHnmꢀ2 similar to the experimental OH density for g-
Al2O3-500-2.5 OHnmꢀ2—see Supporting Information, Sec-
tion 8, Figures S38–40). Among these, one model meets all
aforementioned experimentally determined requirements
(see also Figure S39, model s1a_1). This model corresponds
to a tri-coordinated CuII-oxo site coordinated to two equiv-
alent AlIV and an -OH bound to the so-called AlIII with a Cu—
H distance of 2.76 ꢀ (Figure 6c). The Cu—H distance is
slightly longer than the experimental value (2.76 ꢀ compared
to 2.57 ꢀ), likely a result of the shallow potential energy
surface. In particular, the T-shape geometry at Cu (Figure 6c)
is consistent with the EPR signature of the reactive species
featuring a nearly axialg tensor.
J.M. and M.A.N. thank Shell Global Solutions International
B.V. for financial support. We acknowledge Elettra Sincro-
trone Trieste for providing access to its synchrotron radiation
facilities and we thank Giuliana Aquilanti, Luca Olivi and
Danilo Oliveira de Souza for assistance in using XAFS
beamline. J.M. acknowledges Dr. Z. Berkson and Dr. D.
Mance for their help conducting and analyzing the NMR
experiments.
Conflict of interest
The authors declare no conflict of interest.
Keywords: alumina · copper · EPR spectroscopy ·
methane-to-methanol · surface organometallic chemistry
[1] J. P. Lange, K. P. De Jong, J. Ansorge, P. J. A. Tijm, Stud. Surf.
Sci. Catal. 1997, 107, 81 – 86.
[2] M. Ahlquist, R. J. Nielsen, R. A. Periana, W. A. Goddard, J. Am.
Chem. Soc. 2009, 131, 17110 – 17115.
[3] M. Ravi, M. Ranocchiari, J. A. van Bokhoven, Angew. Chem.
Int. Ed. 2017, 56, 16464 – 16483; Angew. Chem. 2017, 129, 16684 –
16704.
[4] A. A. Latimer, A. Kakekhani, A. R. Kulkarni, J. K. Nørskov,
ACS Catal. 2018, 8, 6894 – 6907.
[5] T. J. Lawton, A. C. Rosenzweig, J. Am. Chem. Soc. 2016, 138,
9327 – 9340.
[6] S. I. Chan, V. C. C. Wang, J. C. H. Lai, S. S. F. Yu, P. P. Y. Chen,
K. H. C. Chen, C. L. Chen, M. K. Chan, Angew. Chem. Int. Ed.
2007, 46, 1992 – 1994; Angew. Chem. 2007, 119, 2038 – 2040.
[7] R. Balasubramanian, S. M. Smith, S. Rawat, L. A. Yatsunyk,
T. L. Stemmler, A. C. Rosenzweig, Nature 2010, 465, 115 – 119.
[8] M. O. Ross, F. MacMillan, J. Wang, A. Nisthal, T. J. Lawton,
B. D. Olafson, S. L. Mayo, A. C. Rosenzweig, B. M. Hoffman,
Science 2019, 364, 566 – 570.
Because the partial oxidation of CH4 to CH3OH is a two-
electron process, two CuII monomers must be involved since
the CuII sites are reduced to CuI according to XANES and
EPR spectroscopy. This process is feasible via an H-atom
abstraction process on the hydroxyl or the oxo bound to CuII,
generating an OH and a methyl radical that will generate
[9] R. L. Lieberman, A. C. Rosenzweig, Nature 2005, 434, 177 – 182.
[10] E. V. Starokon, M. V. Parfenov, L. V. Pirutko, S. I. Abornev, G. I.
Panov, J. Phys. Chem. C 2011, 115, 2155 – 2161.
&&&&
www.angewandte.org
ꢂ 2021 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
These are not the final page numbers!

Angewandte
Research Articles
Chemie
[11] M. H. Groothaert, P. J. Smeets, B. F. Sels, P. A. Jacobs, R. A.
[27] H. J. Jang, W. K. Hall, J. L. DꢃItri, J. Phys. Chem. 1996, 100,
9416 – 9420.
[28] V. L. Sushkevich, J. A. Van Bokhoven, Chem. Commun. 2018,
54, 7447 – 7450.
[29] T. P. Beebe, J. E. Crowell, J. T. Yates, J. Chem. Phys. 1990, 92,
5119 – 5126.
[30] A. Comas-Vives, M. Schwarzwꢁlder, C. Copꢂret, P. Sautet, J.
Phys. Chem. C 2015, 119, 7156 – 7163.
[31] M. A. Newton, A. J. Knorpp, A. B. Pinar, V. L. Sushkevich, D.
Palagin, J. A. Van Bokhoven, J. Am. Chem. Soc. 2018, 140,
10090 – 10093.
[32] A. Godiksen, F. N. Stappen, P. N. R. Vennestrøm, F. Giordanino,
S. B. Rasmussen, L. F. Lundegaard, S. Mossin, J. Phys. Chem. C
2014, 118, 23126 – 23138.
[33] G. Busca, Adv. Catal. 2014, 57, 319 – 404.
[34] D. S. Maciver, H. H. Tobin, R. T. Barth, J. Catal. 1963, 2, 485 –
497.
[35] S. A. Zubkov, V. Y. Borovkov, S. G. Gagarin, V. B. Kazansky,
Chem. Phys. Lett. 1984, 107, 337 – 340.
Schoonheydt, J. Am. Chem. Soc. 2005, 127, 1394 – 1395.
[12] M. A. Newton, A. J. Knorpp, V. L. Sushkevich, D. Palagin, J. A.
Van Bokhoven, Chem. Soc. Rev. 2020, 49, 1449 – 1486.
[13] J. S. Woertink, P. J. Smeets, M. H. Groothaert, M. A. Vance, B. F.
Sels, R. A. Schoonheydt, E. I. Solomon, Proc. Natl. Acad. Sci.
USA 2009, 106, 18908 – 18913.
[14] S. Grundner, M. A. C. Markovits, G. Li, M. Tromp, E. A. Pidko,
E. J. M. Hensen, A. Jentys, M. Sanchez-Sanchez, J. A. Lercher,
Nat. Commun. 2015, 6, 7546.
[15] A. R. Kulkarni, Z. J. Zhao, S. Siahrostami, J. K. Nørskov, F.
Studt, ACS Catal. 2016, 6, 6531 – 6536.
[16] V. L. Sushkevich, D. Palagin, J. A. van Bokhoven, Angew. Chem.
Int. Ed. 2018, 57, 8906 – 8910; Angew. Chem. 2018, 130, 9044 –
9048.
[17] A. Knorpp, A. B. Pinar, C. Baerlocher, L. B. McCusker, N.
Casati, M. A. Newton, S. Checchia, J. Meyet, D. Palagin, J. A.
van Bokhoven, Angew. Chem. Int. Ed. 2021, 60, 5854 – 5858;
Angew. Chem. 2021, 133, 5918 – 5922.
[18] D. K. Pappas, A. Martini, M. Dyballa, K. Kvande, S. Teketel,
K. A. Lomachenko, R. Baran, P. Glatzel, B. Arstad, G. Berlier,
et al., J. Am. Chem. Soc. 2018, 140, 15270 – 15278.
[19] J. P. Lange, V. L. Sushkevich, A. J. Knorpp, J. A. Van Bokhoven,
Ind. Eng. Chem. Res. 2019, 58, 8674 – 8680.
[20] Z. R. Jovanovic, J. P. Lange, M. Ravi, A. J. Knorpp, V. L.
Sushkevich, M. A. Newton, D. Palagin, J. A. van Bokhoven, J.
Catal. 2020, 385, 238 – 245.
[21] H. V. Le, S. Parishan, A. Sagaltchik, H. Ahi, A. Trunschke, R.
Schomꢁcker, A. Thomas, Chem. Eur. J. 2018, 24, 12592 – 12599.
[22] S. E. Bozbag, P. Sot, M. Nachtegaal, M. Ranocchiari, J. A.
Van Bokhoven, C. Mesters, ACS Catal. 2018, 8, 5721 – 5731.
[23] X. Wang, N. M. Martin, J. Nilsson, S. Carlson, J. Gustafson, M.
Skoglundh, P. A. Carlsson, Catalysts 2018, 8, 545.
[24] J. Meyet, K. Searles, M. A. Newton, M. Wçrle, A. P. van Bavel,
A. D. Horton, J. A. van Bokhoven, C. Copꢂret, Angew. Chem.
Int. Ed. 2019, 58, 9841 – 9845; Angew. Chem. 2019, 131, 9946 –
9950.
[36] R. Wischert, C. Copꢂret, F. Delbecq, P. Sautet, Chem. Commun.
2011, 47, 4890 – 4892.
[37] R. Wischert, P. Laurent, C. Copꢂret, F. Delbecq, P. Sautet, J. Am.
Chem. Soc. 2012, 134, 14430 – 14449.
[38] J. Szanyi, J. H. Kwak, Phys. Chem. Chem. Phys. 2014, 16, 15117 –
15125.
[39] M. R. Gafurov, I. N. Mukhambetov, B. V. Yavkin, G. V. Mamin,
A. A. Lamberov, S. B. Orlinskii, J. Phys. Chem. C 2015, 119,
27410 – 27415.
[40] P. Hçfer, A. Grupp, H. Nebenfꢄhr, M. Mehring, Chem. Phys.
Lett. 1986, 132, 279 – 282.
[41] M. Digne, P. Sautet, P. Raybaud, P. Euzen, H. Toulhoat, J. Catal.
2002, 211, 1 – 5.
[42] A. T. F. Batista, D. Wisser, T. Pigeon, D. Gajan, F. Diehl, M.
Rivallan, L. Catita, A.-S. Gay, A. Lesage, C. Chizallet, et al., J.
Catal. 2019, 378, 140 – 143.
[25] C. Copꢂret, Nat. Energy 2019, 4, 1018 – 1024.
[26] M. Veith, K. Valtchev, V. Huch, Inorg. Chem. 2008, 47, 1204 –
1217.
Manuscript received: April 19, 2021
Version of record online: && &&
,
&&&&
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
ꢂ 2021 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
www.angewandte.org
&&&&
These are not the final page numbers!

Angewandte
Research Articles
Chemie
Research Articles
Methane-to-Methanol
Well-dispersed CuII sites generated on
transition alumina using a molecular
approach were employed in the stepwise
partial oxidation of methane to methanol.
The reactive CuII centers could be identi-
fied spectroscopically due to their unique
EPR signature. A structure for these
unusual CuII monomers, working in pairs,
was proposed based on advanced spec-
troscopic characterization, reactivity
studies, and DFT modeling.
J. Meyet, A. Ashuiev, G. Noh,
M. A. Newton, D. Klose, K. Searles,
A. P. van Bavel, A. D. Horton, G. Jeschke,
J. A. van Bokhoven,*
C. Copꢀret*
&&&& — &&&&
Methane-to-Methanol on Mononuclear
Copper(II) Sites Supported on Al2O3:
Structure of Active Sites from Electron
Paramagnetic Resonance
&&&&
www.angewandte.org
ꢂ 2021 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
These are not the final page numbers!