Synthesis of Cu Single Atoms Supported on Mesoporous Graphitic Carbon Nitride and Their Application in Liquid-Phase Aerobic Oxidation of Cyclohexene

Article abstract of DOI:10.1021/acscatal.1c01468

Different loadings of Cu single atoms were anchored on a graphitic carbon nitride (g-C3N4) matrix using a two-step thermal synthesis method and applied in liquid-phase cyclohexene oxidation under mild conditions using molecular O2 as the oxidizing agent. The oxidation state of Cu was determined to be Cu+, which is in linear coordination with two neighboring nitrogen atoms at a distance of 1.9 ?. The catalyst with 0.9 wt % Cu pyrolyzed at 380 °C was found to exhibit the best catalytic performance with the highest conversion up to 82% with an allylic selectivity of 55%. It also showed high reusability over four catalytic runs without any detectable Cu leaching. Cyclohexene oxidation followed first-order kinetics with an apparent activation energy of 66.2 kJ mol-1. The addition of hydroquinone as a radical scavenger confirmed that cyclohexene oxidation proceeds via a radical mechanism. Time-resolved in situ attenuated total reflection infrared (ATR-IR) spectroscopy was carried out to qualitatively monitor the cyclohexene oxidation pathways. The comparison with the homogeneous analogue Cu(I) iodide indirectly verified the linearly N-coordinated single Cu(I) species to be the active sites for cyclohexene oxidation.

Full text of DOI:10.1021/acscatal.1c01468

pubs.acs.org/acscatalysis
Research Article
Synthesis of Cu Single Atoms Supported on Mesoporous Graphitic
Carbon Nitride and Their Application in Liquid-Phase Aerobic
Oxidation of Cyclohexene
Julia Buker, Xiubing Huang,* Johannes Bitzer, Wolfgang Kleist, Martin Muhler, and Baoxiang Peng*
̈
Cite This: ACS Catal. 2021, 11, 7863−7875
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Different loadings of Cu single atoms were anchored on a graphitic
carbon nitride (g-C3N4) matrix using a two-step thermal synthesis method and
applied in liquid-phase cyclohexene oxidation under mild conditions using
molecular O2 as the oxidizing agent. The oxidation state of Cu was determined to
be Cu+, which is in linear coordination with two neighboring nitrogen atoms at a
distance of 1.9 Å. The catalyst with 0.9 wt % Cu pyrolyzed at 380 °C was found to
exhibit the best catalytic performance with the highest conversion up to 82% with
an allylic selectivity of 55%. It also showed high reusability over four catalytic runs
without any detectable Cu leaching. Cyclohexene oxidation followed first-order
kinetics with an apparent activation energy of 66.2 kJ mol−1. The addition of
hydroquinone as a radical scavenger confirmed that cyclohexene oxidation proceeds via a radical mechanism. Time-resolved in situ
attenuated total reflection infrared (ATR-IR) spectroscopy was carried out to qualitatively monitor the cyclohexene oxidation
pathways. The comparison with the homogeneous analogue Cu(I) iodide indirectly verified the linearly N-coordinated single Cu(I)
species to be the active sites for cyclohexene oxidation.
KEYWORDS: cyclohexene oxidation, 2-cyclohexene-1-hydroperoxide, ATR-IR, copper single atoms, EXAFS
challenging due to the presence of two active bonds in the
molecule (Scheme 1).7−10 The oxidation at the C−H bond
initiates the allylic oxidation leading to the formation of the
intermediate 2-cyclohexene-1-hydroperoxide.
1. INTRODUCTION
Selective oxidation of hydrocarbons is of high importance for the
chemical industry as it provides access to many high value-added
fine chemicals.1,2 Over 90% of the globally produced chemicals
are manufactured by catalytic processes3 because catalysis is the
most economical and energy-saving mode to achieve high
conversion and selectivity with low consumption of resources.
Nevertheless, many oxidation processes established in the
chemical industry are still using toxic oxidizing agents in
stoichiometric amounts leading to the coproduction of
substantial amounts of environmentally unfriendly and un-
wanted greenhouse gases such as N2O as well as inorganic solid
waste.2 For example, adipic acid has high relevance in the
production of synthetic fibers, nylon 6,6, polyurethanes and
plasticizers. Its current annual production volume is estimated to
be 3.5 million tons,2 but it is still produced by oxidizing K/A oil
with stoichiometric amounts of HNO3, the production of which
leads to large quantities of N2O.1,2,4 Therefore, processes using
heterogeneous catalysts and more environmentally friendly
oxidizing agents are of high interest.
This hydroperoxide can be oxidized to 2-cyclohexene-1-one
or decomposed to 2-cyclohexene-1-ol, which can also be further
oxidized to 2-cyclohexene-1-one. In addition, 2-cyclohexene-1-
hydroperoxide can react with 2-cyclohexene-1-one, yielding 7-
oxabicyclo[4.1.0]-heptane-2-one.9 When oxidation takes place
at the CC double bond, cyclohexene oxide is formed. At the
same time, this epoxide is also the product of the reaction of 2-
cyclohexene-1-hydroperoxide with another cyclohexene mole-
cule, indicating that the allylic and epoxidation pathways cannot
be completely separated from one another. By hydrolysis of
cyclohexene oxide, cyclohexane-1,2-diol is formed, which is the
key intermediate of liquid-phase cyclohexene oxidation in adipic
acid production.12 Moreover, other reaction products of
cyclohexene oxidation are also important building blocks for
An alternative production route is liquid-phase cyclohexene
oxidation, which enables the formation of cyclohexane-1,2-diol
as the key product to produce adipic acid by subsequent ring-
opening.5,6 Cyclohexene oxidation can be carried out using
molecular O2, which is the most attractive oxidizing agent as it is
abundant, inexpensive, and absolutely harmless to the environ-
ment.1,7 However, selective cyclohexene oxidation is rather
Received: March 31, 2021
Revised: June 7, 2021
Published: June 15, 2021
https://doi.org/10.1021/acscatal.1c01468
© 2021 American Chemical Society
ACS Catal. 2021, 11, 7863−7875
7863

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
degree of conversion of 86% and excellent selectivity of 97% to
phenol in benzene oxidation.31 However, most of the metal
species doped in g-C3N4 in the reported literature are embedded
in the bulk structure of C3N4, which cannot be easily accessed by
reactants. To improve the metal atom efficiency and stability, the
single-metal atoms should be attached on the surface with
suitable interaction with the support.
Scheme 1. Reaction Network of Cyclohexene Oxidation with
O2 as the Oxidizing Agent7,9−11
In the present work, we report for the first time on a series of
catalysts consisting of Cu single atoms anchored on g-C3N4
using copper disodium ethylenediaminetetraacetic acid (Cu-
EDTA) as a precursor by a two-step thermal synthesis method,
which was further applied in liquid-phase cyclohexene oxidation
with molecular O2 under mild conditions. The catalysts were
characterized in detail by X-ray diffraction (XRD), N2
physisorption, inductively coupled plasma (ICP), transmission
electron microscopy (TEM), energy-dispersive spectrometry
(EDS), X-ray photoelectron spectroscopy (XPS), and X-ray
absorption spectroscopy (XAS) measurements. The influence of
the Cu precursor, the Cu amount, and the pyrolysis temperature
on the catalytic properties was studied. Kinetic investigations
and a reusability test were performed for the best catalyst. The
effects of H2O and the addition of a radical scavenger were also
investigated. In situ attenuated total reflection infrared (ATR-
IR) spectroscopy was used to monitor cyclohexene oxidation
and the complex reaction network. The catalytic activity of a
homogeneous analogue catalyst (i.e., CuI) was compared to
identify the active sites.
the synthesis of bulk and fine chemicals, comprising agro-
chemicals, pharmaceuticals, fragrances, and polymers.10,12
To obtain an increased selectivity despite the wide variety of
products of cyclohexene oxidation, it is of great importance to
find a suitable catalyst. Recently, single-atom catalysts have
attracted extensive attention as they provide maximum
efficiency of atom utilization.13 The prerequisite is to obtain
highly reactive and stable single atoms on an appropriate
support, which suppresses their agglomeration under reaction
conditions. Graphitic carbon nitride (g-C3N4) was found to be a
suitable support for single atoms because it enables the efficient
anchoring of isolated metal atoms at the nitrogen coordination
centers, for example, into the sixfold cavities of the g-C3N4
matrix (Scheme S1).13−15 Tri-s-triazine was observed to be the
elementary building block of g-C3N4, and ideal g-C3N4
exclusively consists of carbon and nitrogen atoms with a C/N
ratio of 0.75.16 However, until now, no ideally condensed
structure has been synthesized, and the as-grown polymers are
2. RESULTS AND DISCUSSION
2.1. Catalyst Characterization. Cu single atoms were
anchored on g-C3N4 via a two-step synthesis method. First, g-
C3N4 was synthesized using Pluronic P123 as a soft-template to
achieve high specific surface areas.31,39 In the second step,
suitable amounts of ethylenediaminetetraacetic acid copper(II)
disodium salt (Cu-EDTA) were used to anchor Cu single atoms
on the porous g-C3N4 targeting at 0.5, 1.0, 1.5, and 2.0 wt % Cu
loading. To study the effect of the pyrolysis temperature after the
incorporation of Cu into the g-C3N4 matrix, the samples were
thermally treated at 380 or 500 °C, leading to two temperature
series each including four different Cu loadings (Table 1). The
samples are labeled “xCu-y”, where x is the nominal Cu amount
in wt % and y is the pyrolysis temperature in °C. ICP-MS
measurements were performed to determine the actual Cu
loading in the samples. For all samples, ICP-MS analysis shows a
lower actual Cu loading compared with the nominal amount of
the Cu precursor (Table 1).
Figure 1 shows the XRD patterns of the g-C3N4 support and
its Cu-containing derivatives. For all samples, distinct reflections
at 27.5° were observed, which belong to the (002) lattice plane
of g-C3N4.21,25,40 Additionally, the in-plane structural packing
motif (100) at 12.8° can be weakly observed for all samples.40
The absence of reflections at high angles indicates the lack of
parallel stacking of large numbers of layers.22 The observed
slight broadening of the (002) reflections with similar intensities
for the Cu-containing samples suggests that the incorporation of
Cu atoms does not lead to structural changes in g-C3N4.
Moreover, no further reflections corresponding to crystalline Cu
species were detected, indicating the absence of large Cu
nanoparticles. Thus, Cu is most likely to be chemically
coordinated to the N atoms in the g-C3N4.40
rich in structural and surface defects rendering g-C3N4
a
promising support for heterogeneous catalysis.17 Due to the sp2-
hybridized C−N bonds, g-C3N4 has a superior tolerance toward
high temperatures, and the single-layer stackings based on van
der Waals interactions make g-C3N4 insoluble in common
solvents such as water, alcohols, dimethylformamide, and
acetonitrile.16,18−21
Metal-doped g-C3N4 has already been successfully applied in
several photocatalytic reactions such as H2 evolution and CO2
reduction.22−30 For example, Zhang et al. reported a bimetallic
Ni−Cu cocatalyst supported on g-C3N4 synthesized by a
template-free preassembly strategy.31 Lu et al. studied photo-
catalytic water splitting over Cu-doped g-C3N4 and observed a
remarkable increase in anodic photocurrents.32 Gong et al.
found Cu-doped mesoporous g-C3N4 synthesized by one-step
polymerization of melamine to be 2 times more active for
photocatalytic methyl orange degradation than the undoped g-
C3N4.33 In addition, the application in electrocatalysis has been
studied in detail.34 For instance, Dai et al. investigated the
electrochemical NO reduction over Cu-decorated g-C3N4 and
observed that the catalyst efficiently suppressed the competing
hydrogen evolution reaction.14 A similar catalyst synthesized by
one-step pyrolysis by Sarkar et al. was found to be highly efficient
in the oxygen evolution reaction.35 Recent work has also shown
that doped g-C3N4 catalysts are suitable for alkane, olefin, or
alcohol oxidation reactions.36−38 Zhao et al. fabricated Cu single
sites on g-C3N4 by a preassembly strategy, which achieved a high
The nitrogen adsorption−desorption isotherms of the g-C3N4
support and the representative Cu-containing sample 1.5Cu-380
are shown in Figure S1. Both samples exhibit type IV isotherms
7864
https://doi.org/10.1021/acscatal.1c01468
ACS Catal. 2021, 11, 7863−7875

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
The morphology of the samples was investigated by
transmission electron microscopy (TEM). Figure 2A,B shows
high-angle annular dark field scanning transmission electron
microscopy (HAADF-STEM) images of the 1.5Cu-380 sample.
An irregular platelet-like structure is observed, and no Cu
nanoparticles or clusters are found for all investigated samples
(Figure S2), implying that Cu is mainly coordinated to the
support as single atoms in agreement with the XRD results
(Figure 1). Energy dispersive X-ray (EDX) mapping confirms
the uniform distribution of Cu, C, and N (Figure 2D−F), which
additionally supports the presumption that Cu is embedded in
the g-C3N4 matrix.
XP spectra of the g-C3N4 support and the samples 0.5Cu-380,
1.5Cu-380, 2.0Cu-380, and 1.5Cu-500 are shown in Figure S3,
revealing the presence of small amounts of Cu for the metal-
containing samples. Figure 3 shows spectra of 1.5Cu-380 in the
C 1s, N 1s, and Cu 2p regions. In the C 1s region, the main peak
at 288.1 eV was identified as sp2 carbon atoms bound to N in an
aromatic ring (CN−C), while the weak peak at 286.4 eV was
assigned to sp3-bound carbon (C−N).32,33,43 In the N 1s region
(Figure 3B), the main peak at 398.8 eV originates from sp2-
hybridized N (CN−C) and the peak located at 399.6 eV was
identified as tertiary N bound to C and the peak at 401.3 eV was
caused by π-excitations.24 The Cu 2p spectrum (Figure 3C)
shows two symmetrical peaks located at 932.8 and 952.7 eV
attributed to Cu 2p3/2 and Cu 2p1/2, respectively.24,33 No
satellite peak was recorded for all samples, indicating the absence
of Cu2+ species. The differentiation of metallic Cu and Cu(I)
species is rather challenging due to the overlapping binding
energies and similar splitting of 20 eV. Therefore, the use of the
modified Auger parameter allows a more accurate assignment at
least for the 2.0Cu-380 sample, where a Cu LMM Auger peak
was recorded at a kinetic energy of 914.3 eV. The modified
Auger parameter is derived to amount to 1847.1 based on a Cu
2p3/2 binding energy of 932.8 eV corresponding to Cu−N−C,44
which indicates that Cu is present as Cu+ species incorporated
into the g-C3N4 structure. This can be ascribed to the reduction
of Cu−EDTA to Cu+ species during pyrolysis.
Table 1. Cu Loadings Determined by ICP-MS Measurements
and Derived Specific Surface Areas, Total Pore Volumes, and
Pore Diameters for the g-C3N4 Support and Its Cu-
Containing Derivatives Based on N2 Physisorption
Measurements
sample
name
target Cu
amount
actual Cu
amount
specific
surface area
total pore
volume
pore
diameter
[wt %]
[wt %]
[m2 g−1
]
[cm3 g−1
]
[nm]
g-C3N4
0.5Cu-
380
0.0
0.5
0.00
0.34
63.3
61.6
0.367
0.338
34.5
25.9
1.0Cu-
380
1.5Cu-
380
2.0Cu-
380
0.5Cu-
500
1.0Cu-
500
1.5Cu-
500
2.0Cu-
500
1.0
1.5
2.0
0.5
1.0
1.5
2.0
0.84
0.93
1.61
0.36
0.64
0.93
1.79
61.0
54.1
58.9
59.0
57.3
52.5
55.8
0.340
0.484
0.395
0.424
0.454
0.375
0.451
46.6
43.9
58.6
40.1
28.7
24.1
31.8
The surface concentrations of C, N, Cu, and O in the g-C3N4
support and the samples 0.5Cu-380, 1.5Cu-380, 2.0Cu-380, and
1.5Cu-500 were determined as well (Table S1). For the g-C3N4
support, the C/N ratio is calculated to amount to 0.76, which
corresponds to the ratio of 0.75 of defect-free carbon nitrides. A
minor increase in the C/N ratio up to 0.87 for 2.0Cu-380 was
observed by the incorporation of Cu, confirming the
coordination of Cu ions by N atoms.
Figure 1. XRD patterns of the g-C3N4 support and its Cu-containing
derivatives.
The different types of C 1s peaks are summarized in Table S2.
In the g-C3N4 support, 96 atom % of the C atoms are sp2-
hybridized. With increasing Cu amount, the concentration
decreases to 85.5 atom % in 2.0Cu-380, whereas the N−C and
C−C species increase from 2.2 to 8.9 atom % and 1.8 to 5.6 atom
%, respectively. These results suggest that the anchoring of Cu
atoms in the support matrix leads to structural changes of the
aromatic system toward sp3-hybridized N−C and C−C species,
but the graphitic carbon nitride structure is still the dominant
phase. Moreover, the pyrolysis temperature has a negligible
influence on the existing C bonds, as all concentrations of 1.5Cu-
500 are similar to those of 1.5Cu-380. For the N atoms,
approximately 60% are part of the aromatic system, too, whereas
roughly 20% are present as C3−N species, which are assumed to
be the free coordination sites for the Cu species (Table S3).45
X-ray absorption spectroscopy (XAS) is a commonly used
method to obtain detailed information about the oxidation state
with H3 hysteresis loops, indicating a mesoporous material with
slit-shaped pores.33,41 The specific surface area derived from the
Brunauer−Emmett−Teller (BET) method is 63.3 m2 g−1 for g-
C3N4 (Table 1). With increasing Cu amount, a slight decrease in
the specific surface area was observed, leading to the lowest
surface area of 52.2 m2 g−1 for 2.0Cu-380. No strong influence of
the different pyrolysis temperatures in the second synthesis step
on the specific surface area as well as the total pore volume and
the pore diameter derived from the Barrett−Joyner−Halenda
(BJH) method was observed. The total pore volumes varied
between 0.338 and 0.484 cm3 g−1 accompanied by pore
diameters ranging from 24.1 to 58.6 nm (Table 1). Figure S1
shows the corresponding BJH pore size distribution curves,
indicating a broad distribution of the pore size for all catalysts.
This observation can be explained by the fact that C3N4 itself is
not a highly porous material and the pores are formed between
the C3N4 layers.42
7865
https://doi.org/10.1021/acscatal.1c01468
ACS Catal. 2021, 11, 7863−7875

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
Figure 2. HAADF-STEM images (A−C) and EDX mapping (D−F) of as-synthesized 1.5Cu-380.
Figure 3. XP spectra of the C 1s (A), N 1s (B), and Cu 2p (C) regions of as-synthesized 1.5Cu-380.
Figure 4. XANES (A) and Fourier-transformed EXAFS spectra (B) of as-synthesized 1.5Cu-380, 1.5Cu-500, and 2.0Cu-380 in comparison to the Cu
foil. Note that the Cu foil is multiplied by a factor of 0.2 for better visualization in B. Furthermore, no phase shift corrections have been applied, and the
real distances are expected to be ≈0.4 Å larger.
7866
https://doi.org/10.1021/acscatal.1c01468
ACS Catal. 2021, 11, 7863−7875

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
and the local chemical environment of metal centers. We
recorded XAS spectra for 1.5Cu-380, 1.5Cu-500, and 2.0Cu-
380. Based on the X-ray absorption near edge structure
(XANES) spectra (Figure 4), the oxidation states of the copper
centers in all three materials were determined to be Cu+.46,47
The edge energy E0 was shifted to slightly higher energies
compared to the reference Cu foil (8979 Cu0 vs 8981.8 eV
samples). For Cu2+, larger shifts to ∼8985 eV and a pre-edge
peak close to 8978 eV (1s → 3d dipole forbidden transition)
would have been expected.47 Due to the d10 configuration of
Cu+, no 1s → 3d transition, and thus no pre-edge peak was
observed. The presence of Cu+ is also indicated by the strong
intensity of the peak at 8983 eV, which is characteristic of Cu+ in
a linear coordination mode and results from a 1s → 4p
transition.46,48 The clearly present +1 oxidation state of the Cu
centers is particularly remarkable since other publications of Cu
on C3N4 featured an average oxidation state between +1 and
+2.31,49 The XANES and the Fourier-transformed extended X-
ray absorption fine structure (EXAFS) spectra (Figure 4B) of all
three samples differed only slightly. In particular for 1.5Cu-380,
the amplitude of the oscillations in the XANES spectra seemed
to be dampened compared with the other two materials.
However, the recorded spectra in transmission mode did not
show this difference, suggesting that this dampening resulted
from self-absorption effects. The lower intensity of 1.5Cu-380 in
the Fourier-transformed EXAFS spectra might also result from
self-absorption effects. Therefore, the Cu loading and the
pyrolysis temperature did not seem to have a significant
influence on the oxidation state.
Scheme 2. Proposed Structure for the Cu Ions Incorporated
into the g-C3N4 Matrix Based on XAS Results
a
a
The Cu ions are linearly coordinated between two N atoms of two g-
C3N4 layers (Cu: red, N: blue, C: gray).
peroxide (Figure S7B). By comparison, Cu-containing g-C3N4
catalysts led to a significantly improved catalytic activity. Figures
S5 and S6 show cyclohexene conversion and product selectivity
as a function of time over the different catalysts thermally treated
at 380 and 500 °C, respectively. For all catalysts, a strong
increase in cyclohexene conversion over time was recorded,
resulting in high degrees of conversion between 64.7 and 82.3%
after 6 h. The use of 1,2-dichlorobenzene as an internal standard
for GC analysis revealed carbon balances between 91.2 and
93.4% for all reactions independent of the degree of cyclohexene
conversion. The missing amounts are assumed to be the loss of
the vapor-phase reactants and products during sampling. All
catalysts show high activity toward 2-cyclohexene-1-hydro-
peroxide decomposition as its initially high selectivity decreases
over time, leading to the formation of the desired reaction
products. As hydroperoxide is the key intermediate of
cyclohexene oxidation, its decomposition plays a crucial role
in catalytic conversion. Figures 5 and S8 depict cyclohexene
conversion and product selectivity over the four catalysts with
different amounts of Cu thermally treated at 380 and 500 °C
after 1 and 6 h of reaction, respectively. For the catalysts
pyrolyzed at 500 °C (Figure 5B), low degrees of conversion were
observed, amounting to 12.8 and 17.3% over 0.5Cu-500 and
The first shell of the Fourier-transformed EXAFS spectra was
found at 1.5 Å (note that no phase shift corrections have been
applied and the real distances are expected to be ≈0.4 Å larger),
which is expected for light backscatterers like C, N, or O that are
coordinated to Cu.
The shells at distances >2 Å displayed only a small intensity,
which would be expected for single site Cu centers immobilized
on g-C3N4. The comparison of the XANES and the Fourier-
transformed EXAFS spectra to literature references confirmed
that no Cu0, Cu2O, or CuO was present.
We performed structure fitting based on the Fourier-
transformed EXAFS spectra to determine the number of
backscatterers in the first coordination sphere around the Cu
centers (Figure S4 and Table S4). Since it is likely that Cu
coordinates to nitrogen, we assumed N as the nearest neighbors,
but we cannot unambiguously distinguish C, N, or O due to their
similar backscattering properties. In accordance with the linear
coordination that was identified from the XANES spectra,
approximately two N neighbors were determined by the
structure fitting (Table S4). Any additional shell of Cu
backscatterers from Cu0, Cu2O, or CuO did not result in
meaningful fitting results.
Based on these results, we propose the Cu+ species being
coordinated to the C3N4 support via two N neighbors in a linear
coordination mode as demonstrated in Scheme 2.
2.2. Influence of the Cu Amount, Thermal Treatment
Temperature, and Cu Precursor. Cyclohexene oxidation
over the as-synthesized Cu-g-C3N4 catalysts was investigated
using molecular O2 as the oxidizing agent. A blank experiment
showed only 36.0% cyclohexene conversion and a high
hydroperoxide selectivity of 60.0% after 6 h (Figure S7A). g-
C3N4 led to a similar degree of conversion and even higher
hydroperoxide selectivity above 70%, suggesting that the g-C3N4
support is not able to decompose the intermediate hydro-
Figure 5. Influence of the Cu amount and pyrolysis temperature on
cyclohexene oxidation after 1 h. Cyclohexene oxidation over the
samples pyrolyzed at 380 °C (A) and over the samples pyrolyzed at 500
°C (B).
7867
https://doi.org/10.1021/acscatal.1c01468
ACS Catal. 2021, 11, 7863−7875

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
1.5Cu-500, respectively. For all catalysts, similar product
selectivities were observed, where 2-cyclohexene-1-hydroper-
oxide is the main product with a selectivity of roughly 60%. The
selectivity to allylic and epoxidation products is nearly constant,
independent of the Cu amount incorporated into the catalysts
pyrolyzed at 500 °C. After 6 h of reaction, 2-cyclohexene-1-
hydroperoxide was still the main reaction product of cyclo-
hexene oxidation with a selectivity of roughly 27%.
Compared with the corresponding sample 0.5Cu-500
pyrolyzed at 500 °C, cyclohexene oxidation over 0.5Cu-380
resulted in a similar conversion and product selectivity (Figure
5A). With increasing Cu amount up to 1.5Cu-380, cyclohexene
conversion significantly increased, whereas the selectivity to 2-
cyclohexene-1-hydroperoxide decreased by 31.2%, indicating a
faster hydroperoxide decomposition. This results in increasing
both allylic and epoxidation selectivity by 15.1 and 14.7%,
respectively.
Cyclohexene oxidation over the catalysts pyrolyzed at 380 °C
is clearly improved compared with the catalysts thermally
treated at 500 °C. The best catalyst, 1.5Cu-380, enables a high
cyclohexene conversion of 82.3% and a rapid decomposition of
2-cyclohexene-1-hydroperoxide leading to a low selectivity of
12.8% after 6 h. 2-cyclohexene-1-one becomes the main product
of cyclohexene oxidation with 24.6% selectivity, followed by
cyclohexane-1,2-diol with 17.8%.
The product selectivity is rather similar for all catalysts
thermally treated at 380 °C, suggesting that the active Cu centers
are mainly involved in the decomposition of 2-cyclohexene-1-
hydroperoxide, which is strongly affected by the increasing Cu
amount up to a loading of 1.5 wt %. The slight decrease of
hydroperoxide decomposition ability over the 2.0Cu-380
catalyst is assumed to result from the enhanced C/N ratio
(Table S1) and the lower concentration of C3−N free
coordination sites (Table S3). In conclusion, pyrolysis at the
lower temperature of 380 °C enables higher degrees of
cyclohexene conversion, faster hydroperoxide decomposition,
and higher selectivity to allylic products.
In addition to the pyrolysis temperature, the influence of the
Cu precursor was studied using Cu(NO3)2·3H2O instead of
Cu−EDTA (Figure S10).24 The characterization of the nitrate-
based catalyst 1.5Cu-380-NO3 also indicated successful Cu
anchoring in the g-C3N4 matrix and a high Cu distribution
(Figure S9). However, cyclohexene oxidation over 1.5Cu-380-
NO3 resulted in a significantly lower degree of cyclohexene
conversion of only 58.4% compared with 82.9% over 1.5Cu-380.
Moreover, a strong increase in hydroperoxide selectivity was
observed for 1.5Cu-380-NO3 resulting in lower product
selectivities, especially for cyclohexane-1,2-diol.
2.3. Kinetic Investigations over 1.5Cu-380. The best
catalyst, 1.5Cu-380, was selected to perform kinetic studies on
cyclohexene oxidation. The continuous increase in cyclohexene
conversion as a function of temperature is presented in Figures
6A and S12A−C. A high degree of conversion of 94.9% was
already reached after 4 h at 100 °C. With increasing temperature,
a strikingly lower hydroperoxide selectivity was observed, which
decreased from 50.0% after 6 h at 60 °C to 0.0% at 100 °C,
confirming a complete decomposition of the reaction
intermediate at the elevated temperature. The fast hydro-
peroxide decomposition resulted in higher product selectivities,
which increased by 8.7, 7.7, 9.6, and 6.0% for 2-cyclohexene-1-
one, cyclohexene oxide, 7-oxabicyclo-[4.1.0]heptane-2-one, and
cyclohexane-1,2-diol, respectively. Only 2-cyclohexene-1-ol
selectivity was not affected, indicating its further oxidation to
Figure 6. Kinetic investigations of cyclohexene oxidation over 1.5Cu-
380. The degree of cyclohexene conversion as a function of time at
different temperatures (A) and the Arrhenius plot based on first-order
kinetics (B).
2-cyclohexene-1-one (Scheme 1). The increase in selectivity and
cyclohexene conversion with increasing temperature from 60 to
100 °C led to an increased total product yield of oxidation
products by 35.4−68.8% after 6 h. For all temperatures, an
increase of the ketone/alcohol ratio over time was observed
(Figure S12D). Furthermore, the ketone/alcohol ratio was
increased with increasing temperature suggesting a selectivity
shift to 2-cyclohexene-1-one at elevated temperatures. This
observation indicates the promotion of the further oxidation of
2-cyclohexene-1-ol to 2-cyclohexene-1-one or 2-cyclohexene-1-
hydroperoxide into 2-cyclohexene-1-one (Scheme 1). The ratio
of allylic reaction products including 2-cyclohexene-1-one, 2-
cyclohexene-1-ol, and 7-oxabicyclo[4.1.0]heptane-2-one to the
epoxidation products cyclohexene oxide and cyclohexane-1,2-
diol also shows a significant temperature dependence (Figure
S12D). This ratio strongly decreases with increasing temper-
ature, indicating the preferential formation of epoxidation
products at higher temperatures. Therefore, higher reaction
temperatures were found to promote the formation of
epoxidation products and to shift the allylic selectivity toward
2-cyclohexene-1-one.
7868
https://doi.org/10.1021/acscatal.1c01468
ACS Catal. 2021, 11, 7863−7875

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
Figure 7. Reusability test for cyclohexene oxidation over 1.5Cu-380. The degree of conversion and selectivity as a function of time (A), STEM image
(B), and EDX mappings (C−E) of the spent 1.5Cu-380 catalyst after four cycles.
The time curves of cyclohexene oxidation at 60, 80, and 100
°C were used for Arrhenius analysis (Figure 6A). The linearized
plot for first-order reaction kinetics is well suited for linear
regression (R2 > 0.995), suggesting that cyclohexene oxidation
over 1.5Cu-380 follows first-order reaction kinetics (Figure
S12E). The Arrhenius analysis results in an apparent activation
energy of 66.2 kJ mol−1 (Figure 6B).
numerous reaction paths impede selective cyclohexene
oxidation to one desired product, hence the search for highly
selective catalysts is rather challenging. To clarify the different
reaction pathways of cyclohexene oxidation, the reaction was
investigated by in situ ATR-IR spectroscopy, which enables the
time-resolved monitoring of cyclohexene oxidation and product
formation. Characteristic vibrations of cyclohexene and several
possibly formed reaction products were assigned by recording
spectra of the pure compounds (Table S6). Until now, the
assignment of the intermediate 2-cyclohexene-1-hydroperoxide
has not been possible yet, as its characteristic vibrations are
found to be very weak and overlapping with several vibrations of
cyclohexene and its oxidation products.50 We succeeded in
synthesizing a solution of 2-cyclohexene-1-hydroperoxide with a
high concentration of 65% (Figure S14A) and recorded the
spectra of the mixture dissolved in acetonitrile (Figure S14B).
The addition of three times each 1 mL of the 2-cyclohexene-1-
hydroperoxide solution resulted in similar spectra with
increasing intensities. 2-Cyclohexene-1-one (1685 cm−1) and
2-cyclohexene-1-ol (1064 cm−1) as well as two unknown signals
at 1036 and 949 cm−1 were identified. To determine the
characteristic signal of the hydroperoxide, triphenylphosphine
(PPh3) was added to selectively decompose 2-cyclohexene-1-
hydroperoxide into 2-cyclohexene-1-ol. With increasing PPh3
amount added to the mixture, the signals at 1036 and 949 cm−1
vanished, whereas the characteristic signal of 2-cyclohexene-1-ol
increases suggesting these signals to correspond to 2-cyclo-
hexene-1-hydroperoxide. The intensity of the signal assigned to
2-cyclohexene-1-one was retained, confirming the selective
decomposition of 2-cyclohexene-1-hydroperoxide into 2-cyclo-
hexene-1-ol.
2.4. Reusability Test. The reusability test over 1.5Cu-380
was performed to study the catalyst stability in terms of activity
and selectivity toward cyclohexene oxidation as well as the
stability of the anchored Cu single atoms in the g-C3N4 matrix.
After four reaction cycles, the degree of cyclohexene conversion
was still stable amounting to 78.2, 75.8, 75.9, and 76.3% for the
consecutive runs (Figure 7A). To investigate its stability, an
XRD pattern of the spent catalyst was recorded after four
reaction runs. An identical pattern compared with the fresh
catalyst was obtained, providing no indication of structural
changes of the g-C3N4 support or formation of large crystalline
Cu species (Figure S13A). ICP-MS measurements excluded the
presence of Cu in the solution, confirming that there is no Cu
leaching due to the strong coordination of the Cu ions to the N
atoms in the g-C3N4 matrix. The recorded XP spectrum of the
spent 1.5Cu-380 catalyst reveals similar surface concentrations
of Cu, C, O, and N (Table S5) as well as equal peak positions for
the Cu 2p1/2 and 2p3/2 signals, which would be expected to shift
to higher binding energies if Cu2+ species were present (Figure
S13). In addition, the STEM image does not indicate structural
changes of the catalyst or the formation of Cu nanoparticles
either (Figure 7B). EDX mapping still reveals a uniform
elemental distribution of Cu, C, and N on the catalyst surface
(Figure 7C−E) so that XRD, ICP, STEM, EDX, and XPS
measurements confirm the high stability of the catalyst observed
over four consecutive runs of cyclohexene oxidation.
Based on this knowledge, in situ ATR-IR spectroscopy of
cyclohexene oxidation was carried out. The reaction progress
was continuously monitored by recording the spectra every 2
min. The reaction was initiated by pressurizing the heated
solution with molecular O2 and the corresponding spectrum was
used for background correction. Figure S15 depicts the recorded
2.5. Time-Resolved In Situ ATR-IR Spectroscopy. As
shown in the previous investigations, cyclohexene oxidation
follows a complex reaction network, which can lead to the
formation of several oxidation products (Scheme 1). The
7869
https://doi.org/10.1021/acscatal.1c01468
ACS Catal. 2021, 11, 7863−7875

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
spectra of cyclohexene oxidation over 1.5Cu-380 over 6 h. A
significant decrease of the C−H bending vibration at 725 cm−1
was recorded, demonstrating the consumption of cyclohexene
during cyclohexene oxidation. A strong increase in the intensity
was observed for the stretching vibration of a carbonyl group at
1694 cm−1, which corresponds to the formation of 2-
cyclohexene-1-one. In addition, the signals at 1036 and 952
cm−1 assigned to 2-cyclohexene-1-hydroperoxide increased over
time. Moreover, higher intensities over time were recorded for
the stretching vibrations at 1096, 1075, and 1010 cm−1, which
indicate the formation of cyclohexene oxide, cyclohexane-1,2-
diol, and 2-cyclohexene-1-ol, respectively. Figure 8A shows the
18.2%. In addition, in situ ATR-IR spectroscopy showed that
hydroperoxide is the first product that is formed in cyclohexene
oxidation, confirming the proposed reaction network (Scheme
1). The formation of all other reaction products started within
the first 30 min, and the highest reaction rate was observed for 2-
cyclohexene-1-one, which was formed with the second highest
yield over the whole investigated time period indicating the
successful oxidation of 2-cyclohexene-1-hydroperoxide to 2-
cyclohexene-1-one (Scheme 1). In agreement with the
quantitative GC analysis, the vibrations at 1010 cm−1 assigned
to 2-cyclohexene-1-ol stays at the same intensity after 1 h,
whereas the band of 2-cyclohexene-1-one arises, consequently
suggesting 2-cyclohexene-1-ol to be further oxidized to 2-
cyclohexene-1-one. This observation was previously discussed
by Maschmeyer et al., who found an equilibrium concentration
of cyclohexanol in cyclohexane oxidation.51 Moreover, the
consecutive pathway from cyclohexene oxide to cyclohexane-
1,2-diol can be confirmed by these results, as cyclohexene oxide
is formed first followed by the formation of cyclohexane-1,2-
diol, whereas the cyclohexene oxide yield remained constant
(Figure 8).
2.6. Mechanistic Investigations. The influence of H2O on
cyclohexene oxidation is studied since H2O is formed in several
oxidation reaction steps as a byproduct of cyclohexene oxidation
and, in addition, H2O is consumed by the hydrolysis of
cyclohexene oxide to form cyclohexane-1,2-diol (Scheme 1).
Therefore, 2 mL of H2O was added to the standard solution.
Figure S16A shows cyclohexene oxidation over 1.5Cu-380 in the
absence of H2O, whereas Figure S16B shows cyclohexene
conversion and selectivity over time with H2O addition. The
reaction rate of cyclohexene oxidation is drastically lowered by
H2O addition, as cyclohexene conversion decreased from 82.9 to
64.2% after 6 h. The addition of H2O led to a significantly
enhanced 2-cyclohexene-1-hydroperoxide selectivity of 67.2%
after 1 h and 34.9% after 6 h, while in the absence of H2O, initial
hydroperoxide selectivity was determined to be 39.5 and 15.2%
after 6 h. Consequently, the initial selectivity to cyclohexane-1,2-
diol, 7-oxabicyclo[4.1.0]-heptane-2-one, and 2-cyclohexene-1-
one was strongly decreased by 12.3, 9.0, and 8.8%, respectively.
Meanwhile, 2-cyclohexene-1-ol selectivity remained constant
leading to the assumption that the 2-cyclohexene-1-hydro-
peroxide decomposition step to form 2-cyclohexene-1-ol is not
affected by the presence of H2O, whereas the reaction steps
where H2O is produced as the byproduct are significantly
suppressed (Scheme 1). Likewise, the cyclohexene oxide
selectivity is similar over the whole period of time and
independent of the addition of H2O. As the overall catalytic
activity is decreased when adding H2O, it can be concluded that
H2O probably blocks the active Cu centers of the catalyst.
Although H2O is needed for the hydrolysis of cyclohexene oxide
to cyclohexane-1,2-diol, a lower selectivity to the diol was
observed, suggesting that hydrolysis catalyzed by active Cu sites
was also blocked by H2O. Moreover, the presence of H2O
suppresses reactions where H2O is produced as the byproduct.
To verify whether cyclohexene oxidation over 1.5Cu-380 also
proceeds via a radical mechanism as found earlier, 20 mol %
hydroquinone was added to the solution acting as a radical
scavenger. Even after 6 h, no cyclohexene conversion was
observed, indicating that cyclohexene oxidation completely
proceeds via a radical mechanism.10 Cyclohexene conversion in
the absence of any catalyst was found to result in the same data
compared with the reaction in the presence of the g-C3N4
Figure 8. Quantitative trends of the assigned vibrations recorded
during in situ ATR spectroscopy (A) and product yields of cyclohexene
oxidation during in situ ATR spectroscopy quantified by GC analysis
(B).
time-resolved relative trends of the assigned compounds, which
again demonstrate the consumption of cyclohexene and the
formation of the corresponding reaction products. In addition to
ATR spectroscopy, offline samples were analyzed by GC to
determine the formation of reaction products quantitatively
(Figure 8B). 2-Cyclohexene-1-hydroperoxide was examined to
be the main product of cyclohexene oxidation with a yield of
7870
https://doi.org/10.1021/acscatal.1c01468
ACS Catal. 2021, 11, 7863−7875

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
support, confirming Cu as an active site for cyclohexene
oxidation (Figure S7).
3. CONCLUSIONS
Cu single ions with uniform distribution were successfully
anchored on mesoporous g-C3N4 with different loadings via
chemical coordination to C3−N species using Cu−EDTA as the
precursor. It was confirmed by XRD, TEM, and EDX that the Cu
ions were uniformly distributed over the support, and their
incorporation did not change the structure of the g-C3N4 matrix.
XPS and XANES measurements determined the oxidation state
of the Cu centers to be Cu+. The distance between the Cu
centers and their direct neighbors was determined to be 1.9 Å,
which can be assigned to Cu presumably coordinated to N. The
linear coordination of Cu was proposed based on the EXAFS
spectra, indicating two neighbors coordinated to the Cu centers.
An improved catalytic activity for the cyclohexene oxidation
using molecular oxygen as the oxidizing agent was found for the
Cu-g-C3N4 catalysts thermally treated at 380 °C. A lower
pyrolysis temperature of 380 °C resulted in higher degrees of
conversion and enhanced hydroperoxide decomposition leading
to higher allylic selectivities. Interestingly, the selectivity to
epoxidation products remained constant. Cu−EDTA was found
to be a suitable Cu precursor for the synthesis of Cu single atoms
anchored on g-C3N4, as the alternative use of Cu(NO3)2
revealed a significantly lower degree of cyclohexene conversion
as well as a significantly decreased cyclohexane-1,2-diol
selectivity. Kinetic investigations over the best catalyst, 1.5Cu-
380, determined cyclohexene oxidation to follow first-order
reaction kinetics with an apparent activation energy of 66.2 kJ
mol−1. The catalyst was also found to be highly reusable. The
addition of H2O resulted in a drastic decrease of conversion due
to the blocking of the active Cu+ centers by H2O. By adding the
radical scavenger hydroquinone, we confirmed that cyclohexene
oxidation proceeds via a radical mechanism. Time-resolved in
situ ATR-IR spectroscopy was employed to monitor the reaction
progress and confirmed the complex reaction network
spectroscopically. The comparison with the homogeneous
catalyst Cu(I) iodide identifies the Cu(I) species linearly
coordinated to two N atoms as the active sites with a high
catalytic activity due to their coordinatively unsaturated nature.
Cyclohexene oxidation over Cu(I) iodide resulted in similar
catalytic data identifying the single Cu atom material as a
promising heterogeneous catalyst for cyclohexene oxidation,
enabling significantly easier catalyst separation and reusability.
Our work provides an easy strategy to prepare supported
transition metal single atoms on g-C3N4 with high catalytic
activity and excellent stability.
XPS and EXAFS measurements identified the oxidation state
of the Cu species to be 1+. For this reason, Cu(I) iodide as a
homogeneous analogue of our heterogeneous Cu catalyst was
used for cyclohexene oxidation to determine the activity of
Cu(I) species in homogeneous solution (Figure S17). Cu(I) is
known to form bidentate complexes in acetonitrile (Scheme S2),
and the Cu ion is linearly coordinated to two N atoms,52 which is
rather similar to our proposed single-atom catalyst (Scheme 2).
The initial reaction rates of cyclohexene oxidation and
hydroperoxide decomposition over the homogeneous catalyst
were significantly enhanced. However, only a slightly higher
degree of cyclohexene conversion of 87.8% was observed over
Cu(I) iodide compared with 1.5Cu-380 after 6 h. In addition,
the product selectivities using the homogeneous and heteroge-
neous catalysts are nearly equal. These results indicate that
Cu(I) species are mainly involved in the decomposition of the
reaction intermediate 2-cyclohexene-1-hydroperoxide. More
importantly, the comparable catalytic activity of the Cu(I)
bidentate complexes in Cu(I) iodide to our Cu single-atom
catalyst 1.5Cu-380 verifies the presence of Cu(I) species and the
proposed linearly coordinated Cu(I) species with two N atoms
indirectly.
Despite the anchoring of the Cu atoms on the g-C3N4 support,
these catalysts enable a similarly high degree of interaction with
the substrate as homogeneous catalysts, as shown by the
catalytic data. The high catalytic activity of Cu(I) species can be
attributed to its coordinatively unsaturated nature. The high
catalytic activity of coordinatively unsaturated sites (CUS) in
single-atom catalysts has been investigated by Zhang et al., who
studied the selective oxidation of C−H bonds with tert-
butylhydroperoxide over atomically dispersed FeNx species.53
They found the unsaturated Fe(III)N5 site to be the most active
with a high tendency for homolytic cleavage of the tert-
butylhydroperoxide-forming radicals. Investigations on CO
oxidation over a defect-engineered HKUST-1 metal−organic
̈
framework by Woll et al. demonstrated the excellent catalytic
activity of defect Cu+.54 Bao et al. studied the catalytic activity of
exposed Cu−N active sites within graphene toward oxygen
reduction for zinc−air batteries.55 Based on density functional
theory (DFT) calculations, the Cu(I)−N2 site was examined to
reveal a superior catalytic activity toward ORR in comparison
with Cu−N3 and Cu−N4 due to its high potential to bind and
activate O2. Moreover, Cu(I) species were found to be highly
active for decomposition of hydroperoxides into radicals56
according to the Haber−Weiss mechanism, which is an
important feature in cyclohexene oxidation (Scheme 1). An
example of catalyzed oxidation reactions by single Cu centers in
nature is the dopamine-β-monooxygenase (DβM), which
catalyzes the C−H bond activation by preforming a Cu(I)−
DβM complex. This complex then reacts with O2 by forming a
Cu hydroperoxide, which hydroxylates the substrate.57,58 All of
these examples show that Cu(I) species are highly active toward
C−H bond activation and hydroperoxide decomposition, which
are the key steps of cyclohexene oxidation. The herein presented
single Cu atom catalyst revealed excellent stability toward
cyclohexene oxidation with nearly no loss in catalytic activity
compared with the homogeneously catalyzed reaction. Thus, we
succeeded in heterogenizing single Cu(I) species, which
facilitates their separation and reuse. Consequently, this material
is a promising catalyst for combining the advantages of
homogeneous and heterogeneous catalysis.
4. EXPERIMENTAL SECTION
4.1. Materials. For Cu−C3N4 synthesis, Pluronic P123,
melamine (99%), sulfuric acid (95−98%), Cu−EDTA (>97%),
and Cu(NO3)2·3H2O (99%) were purchased from Sigma-
Aldrich. Methanol in analytical reagent grade was bought from
Fisher Chemicals.
For cyclohexene oxidation, cyclohexene (99%), 2-cyclo-
hexene-1-one (98%), 2-cyclohexene-1-ol (95%), cyclohexene
oxide (98%), 7-oxabicyclo[4.1.0]heptane-2-one (98%), cyclo-
hexane-1,2-diol (98%), and 1,2-dichlorobenzene (99%) were
purchased from Sigma Aldrich. Acetonitrile in analytical reagent
grade was bought from Fisher Chemicals. All reagents were
employed without further purification.
4.2. Catalyst Synthesis. For the synthesis of g-C3N4, 5 g of
Pluronic P123 and 25 g of melamine were dissolved in 100 mL of
deionized H2O. The solution was stirred at 100°C. After 1 h, 10
7871
https://doi.org/10.1021/acscatal.1c01468
ACS Catal. 2021, 11, 7863−7875

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
mL of sulfuric acid solution (H2SO4/H2O=1:1 volume) was
added dropwise, and a white precipitate was obtained. The
precipitate was filtered off after the solution was cooled down to
room temperature and dried in a hot-air oven at 80 °C overnight.
The obtained powder was heat-treated in Ar at 600 °C for 4 h in
a quartz tube. Afterward, it was calcined at 550 °C for 2 h.
For the preparation of Cu−C3N4 catalysts, 0.5 g of C3N4 was
dispersed in 40 mL of methanol. A calculated amount of Cu−
EDTA was added for obtaining 0.5, 1.0, 1.5, and 2.0 wt % Cu
loadings, and the dispersion was stirred for several hours at room
temperature until the solvent was evaporated. The obtained
powder was dried at 80 °C overnight and thermally treated
under inert He flow for 2 h at 380 and 500 °C. For comparison, a
catalyst with 1.5 wt % Cu on g-C3N4 was synthesized using
Cu(NO3)2·3H2O as the precursor and the same procedure as
described above.
4.3. X-ray Diffraction (XRD) Measurements. X-ray
diffraction measurements were carried out with a Bruker D8
Discover X-ray diffractometer. As the X-ray source, Kα radiation
with a wavelength of 1.5401 Å was used. The patterns were
recorded in a 2θ range from 20 to 60 with a scan step size of 0.02
° using a continuous scan mode with a scan time of 1 s per step
and a rotation of 5 rpm. Evaluation of the recorded diffraction
pattern was conducted using High Score Plus software equipped
with access to the International Centre for Diffraction Data
(ICDD) database.
4.4. N2 Physisorption Experiments. N2 physisorption
measurements were performed at 77 K in a BEL-mini apparatus.
The as-prepared powders were first degassed at 120 °C under
vacuum for 6 h to remove adsorbed water. The specific surface
areas were determined from the adsorption isotherms using the
BET method. The pore volume and the pore size distribution
were obtained by applying the BJH method.
4.5. High-Resolution Transmission Electron Micros-
copy (HR-TEM). High-resolution and high-angle annular dark
field scanning transmission electron microscopy (HR-TEM,
HAADF-STEM) and energy dispersive X-ray spectroscopy
(EDX) characterization was performed using a probe-side
aberration-corrected JEM-2200FS (JEOL, Akishima, Japan)
with an acceleration voltage of 200 kV. For sample preparation,
nanoparticles were dispersed in ethanol for 5 min with ultrasonic
treatment, and the dispersions of the nanoparticles were added
dropwise on a carbon-coated gold grid for TEM measurements.
4.6. Inductively Coupled Plasma Mass Spectroscopy
(ICP-MS). ICP-MS measurements were carried out in a Thermo
Fisher iCAP RQ using a KED (kinetic energy discrimination)
cell and a quadrupole mass spectrometer. For digestion, 10 mg
of the sample was treated with 4 mL of HNO3 and 0.5 mL of
HClO4 in a microwave system.
double crystal monochromator was used for the measurements
at the Cu K-edge. The beam current was 100 mA with a ring
energy of 6.08 GeV. The samples were measured in glass
capillaries without dilution. All spectra were recorded as
continuous scans in both fluorescence and transmission
modes at ambient temperature and pressure in the range of
−150 to 1000 eV around the edge in 180 s. The fluorescence
data were used for the evaluation due to the better quality. For
calibration, a copper foil was measured as a reference
simultaneously with the samples.
The data treatment was performed using the Demeter
software package.60 To compensate for the oversampling of
the continuous scan mode, the data points of the obtained
spectra were reduced with the help of the “rebin”-function of
Athena software (edge region: −50 to +50 eV; pre-edge grid: 5
eV; XANES grid: 0.5 eV; EXAFS grid: 0.05 Å−1). For data
evaluation, a Victoreen-type polynomial was subtracted from the
spectrum to remove the background using Athena software. The
first inflection point was taken as edge energy E0. No phase shift
corrections have been applied. The EXAFS analysis was
performed using Artemis software. Prior to the fitting procedure,
2
the amplitude reduction factor S0 was determined for the Cu
reference foil and used as a fixed parameter for all materials.
4.9. Catalytic Oxidation Reactions. Oxidation reactions
were carried out in a 100 mL autoclave reactor equipped with a
Teflon liner (Parr Instrument). About 25 mg of the catalyst was
dispersed in 30 mL of acetonitrile. Cyclohexene (20 mmol) and
4.5 mmol of 1,2-dichlorobenzene as the internal standard for GC
analysis were added. The autoclave was purged with oxygen
three times and pressurized to 10 bar. Subsequently, the reaction
mixture was heated to 80 °C. The reaction was initiated by
switching on the stirrer at 75 °C. Equally, stirring was started at 5
°C below the set temperature for oxidation reactions performed
at different temperatures. Control experiments were carried out
by varying the stirring speed using 400, 500, 600, and 700 rpm to
investigate the influence of mass transfer (Figure S11). It can be
observed that there is no further increase in conversion at 600
and 700 rpm, suggesting the absence of external mass transfer
resistance with respect to the overall reaction rate. Therefore, a
stirring speed of 600 rpm was chosen for the catalytic reactions.
Samples were taken through an online sampling system after
1, 2, 4, and 6 h. For GC analysis, two samples of 1.5 mL were
taken after filtering off the catalyst with a syringe filter (200 nm).
One sample was purely analyzed while the other was treated with
2 mmol triphenylphosphine (PPh3) to decompose the formed 2-
cyclohexene-1-hydroperoxide into the corresponding alcohol 2-
cyclohexene-1-ol. The hydroperoxide amount was calculated by
subtraction of the detected alcohol amounts. The deviation of
the GC analysis is below 0.5%. The experiments are
reproducible, leading to a typical deviation below 2%.
4.7. X-ray Photoelectron Spectroscopy (XPS). XPS was
performed in an ultrahigh vacuum setup equipped with a
Gammadata Scienta SES 2002 analyzer. The spectra were
obtained at a pass energy of 200 eV at a base pressure of 6 ×
10−10 mbar with monochromatic Al Kα radiation (1486.3 eV,
14.5 kV, 45 mA). The C 1s peak at 284.8 eV was used as the
reference. The deconvolution of the spectra was performed
using CasaXPS software with Shirley-type background sub-
traction and symmetric and asymmetric Gaussian−Lorentzian
line shapes.
4.8. X-ray Absorption Spectroscopy (XAS). XAS experi-
ments were performed at PETRA III Extension beamline P65
(energy range: 4−44 keV) at DESY (Deutsches Elektronen-
synchrotron) in Hamburg (Germany).59 A Si(311) C-type
For homogeneously catalyzed cyclohexene oxidation, 1.1 mg
of Cu(I) iodide (0.0059 mmol corresponding to 1.5% Cu on g-
C3N4) was used under otherwise standard conditions.
To test the reusability of the catalyst, four reaction runs were
carried out under standard conditions. After each run, the
catalyst was separated by centrifugation, washed three times
with 5 mL of acetonitrile and dried overnight at room
temperature. The catalyst amount decreased from an initial 25
mg, to 23, 20, and 18 mg in the second, third, and fourth run,
respectively.
For the investigation of cyclohexene oxidation in the presence
of H2O, 2 mL of H2O was added to the initial standard solution,
whereas all other reaction conditions remained unchanged.
7872
https://doi.org/10.1021/acscatal.1c01468
ACS Catal. 2021, 11, 7863−7875

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
To verify a radical mechanism, 20 mol % hydroquinone
relative to the molar amount of cyclohexene was added to the
standard solution, and all other reaction conditions were kept
constant, too.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acscatal.1c01468.
■
sı
4.10. Synthesis of 2-Cyclohexene-1-hydroperoxide.
About 750 mg of 2,2′-azobis-2-dimethylpropionitrile as the
radical initiator, 1 mL of 1,2-dichlorobenzene as the internal
standard for GC analysis, and 30 mL of cyclohexene were filled
in a 50 mL two-neck round bottom flask equipped with a reflux
condenser. The solution was heated to 40 °C and bubbled with
20 mL min−1 O2 under stirring for 48 h. Afterward, the
temperature was set to 30 °C for an additional 4 h. The obtained
colorless, highly viscous solution was diluted with MeCN and
analyzed by GC analysis. Its composition was determined to be
65.0% 2-cyclohexene-1-hydroperoxide, 10.5% 2-cyclohexene-1-
one, 9.5% 7-oxabicyclo[4.1.0]-heptane-2-one, 4.9% cyclohex-
ane-1,2-diol, 4.7%, 2-cyclohexene-1-ol, 4.4% cyclohexene oxide,
and 1.4% cyclohexene.
4.11. Time-Resolved In Situ ATR-IR Spectroscopy. In
situ ATR-IR spectroscopy was carried out with a ReactIR 15
from Mettler Toledo equipped with a diamond internal
reflection element (IRE), a AgX 6 mm × 1.5 m fiber, and a
liquid nitrogen cooled mercury−cadmium−telluride (MCT)
detector. The spectral range of 3000−650 cm−1 was recorded
with a resolution of 4 cm−1.
For detection of a characteristic signal of 2-cyclohexene-1-
hydroperoxide, 30 mL of MeCN was filled in a beaker equipped
with the IR probe. A background spectrum was recorded, which
was used for background subtraction. Then, 1 mL of the
synthesized 2-cyclohexene-1-hydroperoxide mixture was added
three times and the corresponding spectra were recorded.
Afterward, 1 mL solution of 1 g of triphenylphosphine (PPh3)
dissolved in 6 mL of MeCN was added six times, to selectively
decompose the 2-cyclohexene-1-hydroperoxide stepwise into
the corresponding alcohol 2-cyclohexene-1-ol.
For the in situ ATR-IR study, 125 mg of 1.5Cu-380 was
suspended in 150 mL of acetonitrile. About 100 mmol of
cyclohexene and 22.5 mmol of 1,2-dichlorobenzene as the
internal standard for GC analysis were added. The reaction
solution was filled in a Teflon liner of a 350 mL autoclave
(Berghof) equipped with an ATR-IR probe (Mettler Toledo).
The reactor was flushed three times with nitrogen and
pressurized to 3 bar. It was heated in an inert atmosphere to
80 °C under stirring with 350 rpm. After the temperature was
stabilized, the oxidation reaction was initiated by pressurizing
the reactor with oxygen to 20 bar. The reaction progress was
monitored by recording the spectra every 2 min. Meanwhile,
samples were taken through an online sampling system after 0, 1,
2, 3, 4, and 6 h. For GC analysis, two samples of 1.5 mL were
taken after the catalyst was filtered off using a syringe filter (200
nm) and analyzed by GC, as previously described, without and
with PPh3 treatment.
HAADF-TEM images of Cu single-atom catalysts;
regional C 1s, N 1s, and Cu 2p XP spectra; surface
composition and relative abundance of species derived
from XP spectra; EXAFS and Fourier-transformed
EXAFS spectra with fittings; catalytic data of cyclohexene
oxidation over Cu single-atom catalysts as a function of
time; effect of temperature on cyclohexene oxidation;
XRD patterns of the recycled 1.5Cu-380 catalyst;
synthesis of 2-cyclohexene-1-hydroperoxide; ATR-IR
spectra of 2-cyclohexene-1-hydroperoxide; and time-
resolved in situ ATR-IR spectra of cyclohexene oxidation
(PDF)
AUTHOR INFORMATION
Corresponding Authors
■
Xiubing Huang − Laboratory of Industrial Chemistry, Ruhr
University Bochum, 44780 Bochum, Germany; School of
Materials Science and Engineering, University of Science and
Technology Beijing, 100083 Beijing, China; orcid.org/
0000-0002-3779-0486; Email: xiubinghuang@ustb.edu.cn
Baoxiang Peng − Laboratory of Industrial Chemistry, Ruhr
University Bochum, 44780 Bochum, Germany; Max Planck
̈
Institute for Chemical Energy Conversion, 45470 Mulheim an
der Ruhr, Germany; orcid.org/0000-0002-0972-3534;
Email: baoxiang.peng@techem.rub.de
Authors
Julia Bu
̈
ker − Laboratory of Industrial Chemistry, Ruhr
University Bochum, 44780 Bochum, Germany
Johannes Bitzer − Laboratory of Industrial Chemistry, Ruhr
University Bochum, 44780 Bochum, Germany; Department of
Chemistry, TU Kaiserslautern, 67663 Kaiserslautern,
Germany
Wolfgang Kleist − Department of Chemistry, TU
Kaiserslautern, 67663 Kaiserslautern, Germany;
orcid.org/0000-0002-9364-9946
Martin Muhler − Laboratory of Industrial Chemistry, Ruhr
University Bochum, 44780 Bochum, Germany; Max Planck
̈
Institute for Chemical Energy Conversion, 45470 Mulheim an
der Ruhr, Germany; orcid.org/0000-0001-5343-6922
Complete contact information is available at:
https://pubs.acs.org/10.1021/acscatal.1c01468
Author Contributions
All authors have given approval to the final version of the
manuscript.
Notes
The authors declare no competing financial interest.
4.12. Gas Chromatography. Gas chromatography analysis
was carried out in a 7820 A GC from Agilent Technologies. It
was equipped with an Agilent DB-XLB column (30 m × 180 μm
× 0.18 μm) and an FID detector. The injection volume was set
to 0.5 μL with a split ratio of 75:1, a split flow of 30 mL min−1,
and an inlet temperature of 260 °C. The column was first kept at
80 °C for 5 min. Subsequently, the oven was heated to 170 °C
with a rate of 15 °C min−1. Afterward, it was heated with a ramp
of 30 °C min−1 up to 300 °C to avoid deposits of PPh3 in the
column. The end temperature was kept for 1 min.
ACKNOWLEDGMENTS
This research was funded by the Deutsche Forschungsgemein-
schaft (DFG, German Research Foundation, Projektnummer
■
388390466TRR 247). J. Bu
̈
ker acknowledges financial
support as a fellow of the Evonik Stiftung. X.H. acknowledges
financial support from Fundamental Research Funds for the
Central Universities (QNXM20210016). The authors acknowl-
edge DESY (Hamburg, Germany), a member of the Helmholtz
Association HGF, for the provision of experimental facilities.
7873
https://doi.org/10.1021/acscatal.1c01468
ACS Catal. 2021, 11, 7863−7875

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
(16) Gong, Y.; Li, M.; Li, H.; Wang, Y. Graphitic Carbon Nitride
Polymers: Promising Catalysts or Catalyst Supports for Heterogeneous
Oxidation and Hydrogenation. Green Chem. 2015, 17, 715−736.
(17) Nilforoushan, S.; Ghiaci, M.; Hosseini, S. M.; Laurent, S.; Muller,
R. N. Selective Liquid Phase Oxidation of Ethyl Benzene to
Acetophenone by Palladium Nanoparticles Immobilized on a g-
C3N4 −rGO Composite as a Recyclable Catalyst. New J. Chem. 2019,
43, 6921−6931.
(18) Muniandy, L.; Adam, F.; Mohamed, A. R.; Iqbal, A.; Rahman, N.
R. A. Cu2+ Coordinated Graphitic Carbon Nitride (Cu-g-C3N4)
Nanosheets from Melamine for the Liquid Phase Hydroxylation of
Benzene and VOCs. Appl. Surf. Sci. 2017, 398, 43−55.
(19) Wang, Y.; Wang, X.; Antonietti, M. Polymeric Graphitic Carbon
Nitride as a Heterogeneous Organocatalyst: From Photochemistry to
Multipurpose Catalysis to Sustainable Chemistry. Angew. Chem., Int. Ed.
2012, 51, 68−89.
(20) Payra, S.; Saha, A.; Banerjee, S. On Water Cu@g-C3N4
Catalyzed Synthesis of NH-1,2,3-Triazoles via [2+3] Cycloadditions of
Nitroolefins/Alkynes and Sodium Azide. ChemCatChem 2018, 10,
5468−5474.
(21) Yang, J.; Zhang, H.; Chen, B.; Tang, H.; Li, C.; Zhang, Z.
Fabrication of the g-C3N4/Cu Nanocomposite and Its Potential for
Lubrication Applications. RSC Adv. 2015, 5, 64254−64260.
(22) Inagaki, M.; Tsumura, T.; Kinumoto, T.; Toyoda, M. Graphitic
Carbon Nitrides (g-C3N4) with Comparative Discussion to Carbon
Materials. Carbon 2019, 141, 580−607.
(23) Li, F.; Han, G.-F.; Noh, H.-J.; Kim, S.-J.; Lu, Y.; Jeong, H. Y.; Fu,
Z.; Baek, J.-B. Boosting Oxygen Reduction Catalysis with Abundant
Copper Single Atom Active Sites. Energy Environ. Sci. 2018, 11, 2263−
2269.
Parts of this research were carried out at PETRA III and the
authors would like to thank Dr. Edmund Welter for assistance in
using P65 (I-20191191). They also thank Dr. Heidelmann
(Interdisiplinary Center for Analytics on the Nanoscale
(ICAN), University of Duisburg-Essen) for transmission
electron microscopy measurements. The authors acknowledge
Mariia Merko and Philipp Schwiderowski (Chair of Industrial
Chemistry, Ruhr-University Bochum) for ICP-MS and XPS
measurements, respectively.
REFERENCES
■
(1) Ameur, N.; Bachir, R.; Bedrane, S.; Choukchou-Braham, A. A
Green Route to Produce Adipic Acid on TiO2 -Fe2O3 Nano-
composites. J. Chin. Chem. Soc. 2017, 64, 1096−1103.
(2) Ghosh, S.; Acharyya, S. S.; Adak, S.; Konathala, L. N. S.; Sasaki, T.;
Bal, R. Selective Oxidation of Cyclohexene to Adipic Acid over Silver
Supported Tungsten Oxide Nanostructured Catalysts. Green Chem.
2014, 16, 2826.
(3) de Vries, J. G.; Jackson, S. D. Homogeneous and Heterogeneous
Catalysis in Industry. Catal. Sci. Technol. 2012, 2, 2009.
(4) Damm, M.; Gutmann, B.; Kappe, C. O. Continuous-Flow
Synthesis of Adipic Acid from Cyclohexene using Hydrogen Peroxide
in High-Temperature Explosive Regimes. ChemSusChem 2013, 6, 978−
982.
(5) Rozhko, E.; Raabova, K.; Macchia, F.; Malmusi, A.; Righi, P.;
Accorinti, P.; Alini, S.; Babini, P.; Cerrato, G.; Manzoli, M.; Cavani, F.
Oxidation of 1,2-Cyclohexanediol to Adipic Acid with Oxygen: A Study
Into Selectivity-Affecting Parameters. ChemCatChem 2013, 5, 1998−
2008.
(6) Rogers, O.; Pattisson, S.; Engel, R. V.; Jenkins, R. L.; Whiston, K.;
Taylor, S. H.; Hutchings, G. J. Adipic Acid Formation from
Cyclohexanediol using Platinum and Vanadium Catalysts: Elucidating
the Role of Homogeneous Vanadium Species. Catal. Sci. Technol. 2020,
10, 4210−4218.
(24) Shi, G.; Yang, L.; Liu, Z.; Chen, X.; Zhou, J.; Yu, Y. Photocatalytic
Reduction of CO2 to CO over Copper Decorated g-C3N4 Nanosheets
with Enhanced Yield and Selectivity. Appl. Surf. Sci. 2018, 427, 1165−
1173.
(25) Dai, Y.; Gu, Y.; Bu, Y. Modulation of the Photocatalytic
Performance of g-C3N4 by Two-Sites Co-Doping Using Variable
Valence Metal. Appl. Surf. Sci. 2020, 500, No. 144036.
(26) Khan, M. A.; Mutahir, S.; Wang, F.; Zhen, H.; Lei, W.; Xia, M.;
Ouyang, Y.; Muhmood, T. Synthesis of Environmentally Encouraged,
Highly Robust Pollutants Reduction 3-D System Consisting of Ag/g-
C3N4 and Cu-Complex to Degrade Refractory Pollutants. J. Photochem.
Photobiol., A 2018, 364, 826−836.
(27) Lan, H.; Tang, Y.; Zhang, X.; You, S.; Tang, Q.; An, X.; Liu, H.;
Qu, J. Decomplexation of Cu(II)-EDTA over Oxygen-Doped g-C3N4:
An Available Resource Towards Environmental Sustainability. Chem.
Eng. J. 2018, 345, 138−146.
(28) Zhang, P.; Song, T.; Wang, T.; Zeng, H. Effectively Extending
Visible Light Absorption with a Broad Spectrum Sensitizer for
Improving the H2 Evolution of In-Situ Cu/g-C3N4 Nanocomponents.
Int. J. Hydrogen Energy 2017, 42, 14511−14521.
(29) Zhao, Z.; Sun, Y.; Dong, F. Graphitic Carbon Nitride Based
Nanocomposites: A Review. Nanoscale 2015, 7, 15−37.
(30) Zheng, Y.; Lin, L.; Wang, B.; Wang, X. Graphitic Carbon Nitride
Polymers toward Sustainable Photoredox Catalysis. Angew. Chem., Int.
Ed. 2015, 54, 12868−12884.
(31) Zhang, T.; Zhang, D.; Han, X.; Dong, T.; Guo, X.; Song, C.; Si,
R.; Liu, W.; Liu, Y.; Zhao, Z. Preassembly Strategy To Fabricate Porous
Hollow Carbonitride Spheres Inlaid with Single Cu-N3 Sites for
Selective Oxidation of Benzene to Phenol. J. Am. Chem. Soc. 2018, 140,
16936−16940.
(32) Li, Z.; Kong, C.; Lu, G. Visible Photocatalytic Water Splitting and
Photocatalytic Two-Electron Oxygen Formation over Cu- and Fe-
Doped g-C3N4. J. Phys. Chem. C 2016, 120, 56−63.
(7) da Silva, F. P.; Gonca̧ lves, R. V.; Rossi, L. M. Magnetically
Recoverable Copper Oxide Catalysts for Aerobic Allylic Oxidation of
Cyclohexene. J. Mol. Catal. A: Chem. 2017, 426, 534−541.
(8) Cai, Z.-Y.; Zhu, M.-Q.; Chen, J.; Shen, Y.-Y.; Zhao, J.; Tang, Y.;
Chen, X.-Z. Solvent-Free Oxidation of Cyclohexene over Catalysts Au/
OMS-2 and Au/La-OMS-2 with Molecular Oxygen. Catal. Commun.
2010, 12, 197−201.
(9) Buker, J.; Alkan, B.; Fu, Q.; Xia, W.; Schulwitz, J.; Waffel, D.; Falk,
̈
T.; Schulz, C.; Wiggers, H.; Muhler, M.; Peng, B. Selective Cyclohexene
Oxidation with O2, H2O2 and Tert-Butyl Hydroperoxide Over Spray-
Flame Synthesized LaCo1−xFexO3 nanoparticles. Catal. Sci. Technol.
2020, 10, 5196−5206.
(10) Denekamp, I. M.; Antens, M.; Slot, T. K.; Rothenberg, G.
Selective Catalytic Oxidation of Cyclohexene with Molecular Oxygen:
Radical Versus Nonradical Pathways. ChemCatChem 2018, 10, 1035−
1041.
(11) Rogers, O.; Pattisson, S.; Macginley, J.; Engel, R.; Whiston, K.;
Taylor, S.; Hutchings, G. The Low Temperature Solvent-Free Aerobic
Oxidation of Cyclohexene to Cyclohexane Diol over Highly Active Au/
Graphite and Au/Graphene Catalysts. Catalysts 2018, 8, No. 311.
(12) Cao, Y.; Yu, H.; Peng, F.; Wang, H. Selective Allylic Oxidation of
Cyclohexene Catalyzed by Nitrogen-Doped Carbon Nanotubes. ACS
Catal. 2014, 4, 1617−1625.
(13) Ji, S.; Chen, Y.; Wang, X.; Zhang, Z.; Wang, D.; Li, Y. Chemical
Synthesis of Single Atomic Site Catalysts. Chem. Rev. 2020, 120,
11900−11955.
(14) Wu, Q.; Wei, W.; Lv, X.; Wang, Y.; Huang, B.; Dai, Y. Cu@g-
C3N4: An Efficient Single-Atom Electrocatalyst for NO Electro-
chemical Reduction with Suppressed Hydrogen Evolution. J. Phys.
Chem. C 2019, 123, 31043−31049.
(15) Ong, W.-J.; Tan, L.-L.; Ng, Y. H.; Yong, S.-T.; Chai, S.-P.
Graphitic Carbon Nitride (g-C3N4)-Based Photocatalysts for Artificial
Photosynthesis and Environmental Remediation: Are We a Step Closer
To Achieving Sustainability? Chem. Rev. 2016, 116, 7159−7329.
(33) Le, S.; Jiang, T.; Zhao, Q.; Liu, X.; Li, Y.; Fang, B.; Gong, M. Cu-
Doped Mesoporous Graphitic Carbon Nitride for Enhanced Visible-
Light Driven Photocatalysis. RSC Adv. 2016, 6, 38811−38819.
(34) Zou, X.; Silva, R.; Goswami, A.; Asefa, T. Cu-Doped Carbon
Nitride: Bio-Inspired Synthesis of H2-Evolving Electrocatalysts Using
7874
https://doi.org/10.1021/acscatal.1c01468
ACS Catal. 2021, 11, 7863−7875

ACS Catalysis
pubs.acs.org/acscatalysis
Research Article
Graphitic Carbon Nitride (g-C3N4) as a Host Material. Appl. Surf. Sci.
2015, 357, 221−228.
(35) Sarkar, S.; Sumukh, S. S.; Roy, K.; Kamboj, N.; Purkait, T.; Das,
M.; Dey, R. S. Facile One Step Synthesis of Cu-g-C3N4 Electrocatalyst
Realized Oxygen Reduction Reaction with Excellent Methanol
Crossover Impact and Durability. J. Colloid Interface Sci. 2020, 558,
182−189.
of Atomically Dispersed Fe−N−C Catalyst for Selective Oxidation of
the C−H Bond. J. Am. Chem. Soc. 2017, 139, 10790−10798.
(54) Wang, W.; Sharapa, D. I.; Chandresh, A.; Nefedov, A.; Heißler,
S.; Heinke, L.; Studt, F.; Wang, Y.; Wöll, C. Interplay of Electronic and
Steric Effects to Yield Low-Temperature CO Oxidation at Metal Single
Sites in Defect-Engineered HKUST-1. Angew. Chem., Int. Ed. 2020, 59,
10514−10518.
(55) Wu, H.; Li, H.; Zhao, X.; Liu, Q.; Wang, J.; Xiao, J.; Xie, S.; Si, R.;
Yang, F.; Miao, S.; et al. Highly Doped and Exposed Cu(I)−N Active
Sites Within Graphene Towards Efficient Oxygen Reduction for Zinc−
Air Batteries. Energy Environ. Sci. 2016, 9, 3736−3745.
(56) Kochi, J. K. The Mechanism of the Copper Salt Catalysed
Reactions of Peroxides. Tetrahedron 1962, 18, 483−497.
(57) Klinman, J. P. Mechanisms Whereby Mononuclear Copper
Proteins Functionalize Organic Substrates. Chem. Rev. 1996, 96, 2541−
2562.
(58) Fontecave, M.; Pierre, J.-L. Oxidations by Copper Metal-
loenzymes and Some Biomimetic Approaches. Coord. Chem. Rev. 1998,
170, 125−140.
(59) Welter, E.; Chernikov, R.; Herrmann, M.; Nemausat, R. A
Beamline for Bulk Sample X-Ray Absorption Spectroscopy at the High
Brilliance Storage Ring PETRA III. AIP Conf. Proc. 2019, No. 040002.
(60) Ravel, B.; Newville, M. ATHENA, ARTEMIS, HEPHAESTUS:
Data Analysis for X-Ray Absorption Spectroscopy Using IFEFFIT. J.
Synchrotron Radiat. 2005, 12, 537−541.
(36) Wang, L.; Zhu, Y.; Yang, D.; Zhao, L.; Ding, H.; Wang, Z. The
Mixed Marriage of Copper and Carbon Ring-g-C3N4 Nanosheet: A
Visible-Light-Driven Heterogeneous Fenton-Like Catalyst. Appl. Surf.
Sci. 2019, 488, 728−738.
(37) Chen, X.; Zhang, J.; Fu, X.; Antonietti, M.; Wang, X. Fe-g-C3N4-
Catalyzed Oxidation of Benzene to Phenol Using Hydrogen Peroxide
and Visible Light. J. Am. Chem. Soc. 2009, 131, 11658−11659.
(38) Marcinkowski, M. D.; Darby, M. T.; Liu, J.; Wimble, J. M.; Lucci,
F. R.; Lee, S.; Michaelides, A.; Flytzani-Stephanopoulos, M.;
Stamatakis, M.; Sykes, E. C. H. Pt/Cu Single-Atom Alloys as Coke-
Resistant Catalysts for Efficient C-H Activation. Nat. Chem. 2018, 10,
325−332.
(39) Yan, Q.; Zhao, C.; Zhang, L.; Hou, Y.; Wang, S.; Dong, P.; Lin, F.;
Wang, Y. Facile Two-Step Synthesis of Porous Carbon Nitride with
Enhanced Photocatalytic Activity Using a Soft Template. ACS
Sustainable Chem. Eng. 2019, 7, 3866−3874.
(40) Ding, Z.; Chen, X.; Antonietti, M.; Wang, X. Synthesis of
Transition Metal-Modified Carbon Nitride Polymers for Selective
Hydrocarbon Oxidation. ChemSusChem 2011, 4, 274−281.
(41) ALOthman, Z. A Review: Fundamental Aspects of Silicate
Mesoporous Materials. Materials 2012, 5, 2874−2902.
(42) Yan, H. Soft-Templating Synthesis of Mesoporous Graphitic
Carbon Nitride with Enhanced Photocatalytic H2 Evolution Under
Visible Light. Chem. Commun. 2012, 48, 3430−3432.
(43) Jin, Z.; Zhang, L. Performance of Ni-Cu Bimetallic Co-Catalyst g-
C3N4 Nanosheets for Improving Hydrogen Evolution. J. Mater. Sci.
Technol. 2020, 49, 144−156.
(44) Biesinger, M. C. Advanced Analysis of Copper X-Ray
Photoelectron Spectra. Surf. Interface Anal. 2017, 49, 1325−1334.
(45) Hu, S.; Qu, X.; Bai, J.; Li, P.; Li, Q.; Wang, F.; Song, L. Effect of
Cu(I)−N Active Sites on the N2 Photofixation Ability over Flowerlike
Copper-Doped g-C3N 4 Prepared via a Novel Molten Salt-Assisted
Microwave Process: The Experimental and Density Functional Theory
Simulation Analysis. ACS Sustainable Chem. Eng. 2017, 5, 6863−6872.
(46) Ene, A. B.; Bauer, M.; Archipov, T.; Roduner, E. Adsorption of
Oxygen on Copper in Cu/HZSM5 Zeolites. Phys. Chem. Chem. Phys.
2010, 12, 6520−6531.
(47) Drake, I. J.; Fujdala, K. L.; Baxamusa, S.; Bell, A. T.; Tilley, T. D.
Effects of Precursor Composition on the Local Structure of Cu
Dispersed on Mesoporous Silica: A Detailed X-ray Absorption
Spectroscopy Study. J. Phys. Chem. B 2004, 108, 18421−18434.
(48) Lamble, G.; Moen, A.; Nicholson, D. G. Structure of the
Diamminecopper(I) Ion in Solution. An X-Ray Absorption Spectro-
scopic Study. Faraday Trans. 1994, 90, No. 2211.
(49) Xiao, X.; Gao, Y.; Zhang, L.; Zhang, J.; Zhang, Q.; Li, Q.; Bao, H.;
Zhou, J.; Miao, S.; Chen, N.; Wang, J.; Jiang, B.; Tian, C.; Fu, H. A
Promoted Charge Separation/Transfer System from Cu Single Atoms
and C3N4 Layers for Efficient Photocatalysis. Adv. Mater. 2020, 32,
No. e2003082.
(50) Shreve, O. D.; Heether; Knight, H. B.; Swern, D. Infrared
Absorption Spectra of Some Hydroperoxides, Peroxides, and Related
Compounds. Anal. Chem. 1951, 23, 282−285.
(51) Nowotny, M.; Pedersen, L. N.; Hanefeld, U.; Maschmeyer, T.
Increasing the Ketone Selectivity of the Cobalt-Catalyzed Radical
Chain Oxidation of Cyclohexane. Chem. - Eur. J. 2002, 8, 3724−3731.
(52) Liang, H.-C.; Kim, E.; Incarvito, C. D.; Rheingold, A. L.; Karlin,
K. D. A Bis-Acetonitrile Two-Coordinate Copper (I) Complex:
Synthesis and Characterization of Highly Soluble B(C6F5) 4-Salts of
[Cu(MeCN)2]+ and [Cu(MeCN)4]+. Inorg. Chem. 2002, 41, 2209−
2212.
(53) Liu, W.; Zhang, L.; Liu, X.; Liu, X.; Yang, X.; Miao, S.; Wang, W.;
Wang, A.; Zhang, T. Discriminating Catalytically Active FeNx Species
7875
https://doi.org/10.1021/acscatal.1c01468
ACS Catal. 2021, 11, 7863−7875