Iron Single Atom Catalyzed Quinoline Synthesis

Article abstract of DOI:10.1002/adma.202101382

The production of high-value chemicals by single-atom catalysis is an attractive proposition for industry owing to its remarkable selectivity. Successful demonstrations to date are mostly based on gas-phase reactions, and reports on liquid-phase catalysis are relatively sparse owing to the insufficient activation of reactants by single-atom catalysts (SACs), as well as, their instability in solution. Here, mechanically strong, hierarchically porous carbon plates are developed for the immobilization of SACs to enhance catalytic activity and stability. The carbon-based SACs exhibit excellent activity and selectivity (≈68%) for the synthesis of substituted quinolines by a three-component oxidative cyclization, affording a wide assortment of quinolines (23 examples) from anilines and acetophenones feedstock in an efficient, atom-economical manner. Particularly, a Cavosonstat derivative can be synthesized through a one-step, Fe₁-catalyzed cyclization instead of traditional Suzuki coupling. The strategy is also applicable to the deuteration of quinolines at the fourth position, which is challenging by conventional methods. The synthetic utility of the carbon-based SAC, together with its reusability and scalability, renders it promising for industrial scale catalysis.

Full text of DOI:10.1002/adma.202101382

ReseaRch aRticle
www.advmat.de
Iron Single Atom Catalyzed Quinoline Synthesis
Zhongxin Chen, Jingting Song, Xinwen Peng,* Shibo Xi, Jia Liu, Wenhui Zhou, Runlai Li,
Rile Ge, Cuibo Liu, Haisen Xu, Xiaoxu Zhao, Haohan Li, Xin Zhou, Lu Wang, Xing Li,
Linxin Zhong, Alexandre I. Rykov, Junhu Wang, Ming Joo Koh, and Kian Ping Loh*
has seen limited progress owing to their
The production of high-value chemicals by single-atom catalysis is an attrac-
tive proposition for industry owing to its remarkable selectivity. Successful
demonstrations to date are mostly based on gas-phase reactions, and reports
on liquid-phase catalysis are relatively sparse owing to the insufficient activa-
tion of reactants by single-atom catalysts (SACs), as well as, their instability
in solution. Here, mechanically strong, hierarchically porous carbon plates
are developed for the immobilization of SACs to enhance catalytic activity
and stability. The carbon-based SACs exhibit excellent activity and selectivity
(≈68%) for the synthesis of substituted quinolines by a three-component
oxidative cyclization, affording a wide assortment of quinolines (23 examples)
from anilines and acetophenones feedstock in an efficient, atom-economical
manner. Particularly, a Cavosonstat derivative can be synthesized through a
one-step, Fe1-catalyzed cyclization instead of traditional Suzuki coupling. The
strategy is also applicable to the deuteration of quinolines at the fourth posi-
tion, which is challenging by conventional methods. The synthetic utility of
the carbon-based SAC, together with its reusability and scalability, renders it
promising for industrial scale catalysis.
instability in solution and insufficient
activation of reactants by single metal
sites under ambient conditions.[4,5] Con-
sequently, applications of SACs in organic
synthesis were limited to certain hydro-
genations,[6,7] oxidations,[8,9] and C C

bond formations.[10] Very recently, we have
reported the first SAC-catalyzed prepara-
tion of pharmaceuticals (Lonidamine etc.,
and their 15N-labeled analogues) by selec-
tive hydrogenation to E-hydrazones and
subsequent cyclization using Pt1/CeO2 cat-
alyst.[11] We have also developed the late-
stage functionalization of pharmaceuticals
(Tamiflu) by chemoselective oxidation
of sulfides using Co1-in-MoS2 catalyst.[12]
Despite excellent functional group toler-
ance and synthetic utility in both cases,
the scope is limited by the use of complex
starting materials (i.e., carboxylic esters
mediated α-diazoesters synthesis and
multifunctionalized sulfides), and the
inaccessibility to synthesize multi-ring system as the reactions
1. Introduction
mainly involve simple hydrogenation/oxidation.[11,12]
The past decade has witnessed an explosion of research in
single-atom catalysts (SACs) on account of their maximized
atom utilization efficiency, remarkable turnover frequency, and
selectivity in gas-phase reactions and (photo)electrocatalysis.[1–3]
Despite the advances in gas-phase reactions, the liquid-phase
transformation of chemicals into high-value products by SACs
Quinolines, a major class of heterocycles, are widely occur-
ring in natural and synthetic products with diverse pharmaco-
logical and physical properties.[13,14] Among the many methods
to synthesize quinolines, the classical Friedländer condensation
of an aromatic 2-amino-substituted carbonyl compound with
another substituted carbonyl derivative is one of the simplest
Dr. Z. Chen, J. Song, Dr. J. Liu, Prof. R. Li, Prof. C. Liu, Prof. H. Xu, Dr. X. Zhao,
H. Li, Dr. X. Zhou, L. Wang, Dr. X. Li, Prof. M. J. Koh, Prof. K. P. Loh
Department of Chemistry
3 Science Drive 3, Singapore 117543, Singapore
E-mail: chmlohkp@nus.edu.sg
Dr. S. Xi
Institute of Chemical and Engineering Sciences
Agency for Science, Technology and Research (A*STAR)
1 Pesek Road, Jurong Island, Singapore 627833, Singapore
W. Zhou, Dr. R. Ge, Prof. A. I. Rykov, Prof. J. Wang
Center for Advanced Mössbauer Spectroscopy
Dalian Institute of Chemical Physics
Chinese Academy of Sciences
J. Song, L. Wang, Prof. K. P. Loh
Joint School of National University of Singapore and Tianjin University
International Campus of Tianjin University
Binhai New City, Fuzhou 350207, China
Dalian, Liaoning 116023, China
Prof. X. Peng, Prof. L. Zhong
Prof. C. Liu
State Key Laboratory of Pulp and Paper Engineering
South China University of Technology
Guangzhou 510641, China
Department of Chemistry
Institute of Molecular Plus
School of Science
E-mail: fexwpeng@scut.edu.cn
Tianjin University
Tianjin 300072, China
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adma.202101382.
DOI: 10.1002/adma.202101382
2101382 (1 of 11)
© 2021 Wiley-VCH GmbH
Adv. Mater. 2021, 2101382

www.advancedsciencenews.com
www.advmat.de
and most effective approaches.[13,14] Owing to the requirement
for specialized substrates (ortho-substituted aniline) and the
lack of multifunctional capability, researchers have devised
the three-component annulation reaction of readily available
materials, including transition metal-catalyzed Povarov reac-
tion using arylamine, aldehyde, and a third component (alkene,
alkyne, etc.).[15,16] The conundrum is that homogeneous reac-
tions usually entail tedious separation of catalyst residues,
whereas the use of heterogeneous catalysts leads to undesired
side reactions.[17,18] There is therefore a compelling need to
develop SACs that enable efficient access to valuable quinoline
derivatives with high selectivity and are easily separable from
the reaction mixtures.
coordination (metal = Fe, Co, Ni, or Cu), which was found to
exhibit excellent activity and selectivity (≈68%) for three-com-
ponent oxidative cyclization to synthesize 6-methoxy-2-phe-
nylquinoline, outperforming the benchmark Pd/C, Fe/C, and
homogeneous FeCl3 catalysts. Density functional theory (DFT)
calculations show that the Fe–N4 coordination centers present
lower reaction barriers to condensation and product desorption
than Fe–C2N2 and Fe–C1N3 catalytic centers, thus affording a
higher reaction efficiency.
2. Results
Since conventional liquid phase reactions are limited by high
activation energy for the activation of reactants,[1–3,19] a hierar-
chically structured support for SAC is favorable for high turn-
over frequency by maximizing the exposure of metallic sites to
reactants. However, the preparation of SAC usually requires a
pyrolysis step, which inevitably collapses the pores or channels
in the porous support supporting the SAC.[1,20] Natural wood
possesses abundant channels and can be utilized as a mechani-
cally robust scaffold for constructing hierarchically porous sup-
port for catalysts via scalable top-down approaches.[21–23] Using
natural woods with built-in channels is cost-effective and con-
venient compared to bottom-up synthesis of hierarchical scaf-
folds from molecular precursors.[24] This allows direct fabrica-
tion of catalyst body without the addition of porogens, binders,
fillers, or other additives.[25] We have previously demonstrated
the capability of nitrogen-doped carbonized wood plate as a
monolithic, “breathable” cathode for Zn–air batteries.[26,27] In
this regard, natural wood with excellent mechanical stability
and abundant built-in channels can be a good candidate as cost-
effective, scalable scaffold.
2.1. Carbon-Based, Hierarchically Porous Single-Atom Catalysts
We have employed a modified top-down approach as shown in
Figure 1 to immobilize SAC on hierarchically porous carbons.
Natural birch wood was treated with metal ion precursors
(FeCl3, CoCl2, NiCl2, or CuCl2) by a similar pretreatment pro-
cess in pulp and paper industry to form a black liquor. During
this pretreatment, metal ions will strongly bind to the oxygen-
ated groups in lignin and break the β-O-4 ether bonds.[28–30]
The hydrothermal treated wood was then carbonized at high
temperature with extrinsic nitrogen source (urea) to convert
into stable SACs (denoted as Fe-T SACs where T represents
the pyrolysis temperature). The addition of N source (urea) is
vital to the SAC preparation because significant metal particle
aggregation is observed in the control sample without urea in
Figure S2, Supporting Information. Such top-down process is
highly scalable, producing >20 g of carbon-based SACs with
0.8 wt% of metal in a single batch experiment. The hierar-
chically porous SAC is also mechanically strong, supporting
> 50 times of its weight in Figure S1, Supporting Information.
We first examined the morphology of such carbon-based
SACs by scanning electron microscopy (SEM). As shown in
Figure 2A and Figure S3, Supporting Information, natural
birch wood has a highly orientated macroporous structure
Herein, we report the construction of a robust SACs plat-
form by immobilizing metal ion precursors in natural wood,
following by the carbonization to convert into stable SACs. The
carbon-based SACs was processed to optimize the metal–N4
Figure 1. Schematic of the preparation of carbon-based single-atom catalysts.
Adv. Mater. 2021, 2101382
2101382 (2 of 11)
© 2021 Wiley-VCH GmbH

www.advancedsciencenews.com
www.advmat.de
Figure 2. Hierarchically porous structure of Fe-800 SAC. A,B) SEM images showing the macroporous channels of: A) raw wood and B) Fe-800 SAC
along the vertical direction. C) SEM image of Fe-800 SAC along the parallel direction. D) Atomic-resolution STEM-HAADF image of Fe-800 SAC.
E) XRD patterns of various samples. F) N2 adsorption/desorption isotherm at 77 K and G) pore-size distribution of Fe-800 SAC. Scale bars: A) 500 µm;
B,C) 50 µm; D) 2 nm.
with most of the macropores having diameters of 10–20 µm
from its aligned cellulose fibers and some macropores with
>100 µm from its vessels. Such hierarchically porous struc-
ture is beneficial for the diffusion of liquid into the inner
cell walls.[31] The hydrothermal treatment and pyrolysis
processes did not affect the intrinsic macropores, which
were well-retained in Figure 2B,C and Figures S4–S6, Sup-
porting Information. Apart from the macroporous struc-
ture, Fe-800 SAC exhibits a typical type-IV behavior with a
Brunauer–Emmett–Teller (BET) surface area of 164 m2 g−1 and
a prominent pore size of 3.8 nm in the N2 gas adsorption/
desorption isotherms in Figure 2F,G. The presence of uniformly
distributed N and Fe on the carbon skeleton was confirmed
by elemental mapping in Figure S7, Supporting Information.
The N and Fe mass loading were determined as 7.3–4.4% and
0.6–0.8% by elemental analysis (EA) and inductively coupled
plasma (ICP) at various pyrolysis temperatures in Table S1,
Supporting Information. The atomic dispersion of transition
metal in wood-derived catalysts was characterized by scanning
transmission electron microscopy in annular dark-field mode
(STEM-ADF). Because of abundant N ligands, transition metal
single atoms such as Fe1, Co1, Ni1, or Cu1 can be readily immo-
bilized on the carbonized plates,[29] whose uniform distribu-
tion is observed as bright spots in Figure 2D and Figures S8
and S9. The doping configuration of N is also verified by X-ray
photoelectron spectroscopy (XPS) in Figure S12. Two char-
acteristic peaks at 398.5 and 400.8 eV can be assigned to the
pyridinic N and graphitic N, respectively.[26,32] The presence
of defective sites and functional groups for the immobiliza-
tion of single atom is further confirmed by Raman, Fourier
transform infrared (FT-IR), and solid-state NMR spectra in
Figured S10–S13, Supporting Information.
The electronic and coordination structure of iron in carbon-
based SAC were validated by X-ray absorption near-edge struc-
ture (XANES) and extended X-ray absorption fine structure
(EXAFS).[33,34] The Fe K-edge XANES spectrum of Fe-800 SAC
exhibits a white line intensity similar to iron phthalocyanine
(FePc) in Figure 3A, indicating an oxidation state of +2 after
in situ reduction by pyrolysis. This is supported by a com-
prehensive analysis on the pre-edge feature, edge location
and white line intensity in Figure S14 and Table S2, Sup-
porting Information, as well as, Fe2p XPS and Fe L2,3-edge
XAS spectra in Figure S12, Supporting Information. Both
calculated and experimental XANES curves in Figure 3B and
Figure S15, Supporting Information, confirms a pyridinic type
Fe–N4 structure in our samples, where Fe–C1N3 and Fe–C2N2
species may co-exist.[11,35] Other typical Fe–N–C structures is
excluded due to mismatch of the simulation and experiment.
Unlike typical signals of FeCl3, the Fourier-transformed EXAFS
spectrum of FeCl3-wood exhibits a prominent Fe-O peak at
≈1.53 Å in Figure 3C. This is a result of transformation of the
iron salt to Fe–O during hydrothermal treatment. Fe-800 SAC
also exhibits a characteristic Fe–N peak at 1.44 Å, which is
fitted with a coordination number of 4 in Table S3, Supporting
Information. No metallic Fe-Fe peak at 2.21 Å is observed for
Fe-800 SAC, revealing that Fe1 exists as isolated single atoms
in carbon-based SACs, consistent with the HAADF-STEM
images and the absence of a metal particle peak at ≈44° in the
XRD spectrum. Wavelet transforms have also been performed
in Figure 3D,E and Figure S16, Supporting Information, to
2101382 (3 of 11)
© 2021 Wiley-VCH GmbH
Adv. Mater. 2021, 2101382

www.advancedsciencenews.com
www.advmat.de
Figure 3. Coordination environment of Fe-800 SAC. A) Fe K-edge XANES spectra and B) comparison with the calculated spectrum using the pyridinic
Fe–N4 model (inset, outer carbon atoms are omitted for the sake of clarity). C) FT-EXAFS spectra of various catalysts. The dotted lines represent the
FT fitting curves. D) The wavelet transform (WT) contours of Fe K-edge EXAFS spectra for Fe-800 SAC and E) Fe foil. F) Room-temperature 57Fe Möss-
bauer spectrum of the 57Fe-enriched Fe-800 SAC.
distinguish Fe–N from iron oxides. It is seen that Fe-800 SAC
exhibits an intensity maximum at 3.8 Å−1 from the Fe–N–C
coordination but without the appearance of an Fe–Fe signal at
7.5 Å−1 in Fe foil or 5.5–6.0 Å−1 in iron oxides. The secondary
maxima of Fe-800 SAC at 3.8 Å−1 (2.2 Å in R space) is attributed
to the second shell coordination of Fe–C in the proposed pyri-
dinic Fe–N4 structure. Control experiment without the addition
of urea suggests the oxygen-containing lignins cannot stabilize
the Fe single atoms upon high-temperature treatment.[28,36]
Mössbauer spectroscopy is a powerful tool for the iden-
tification of iron species in a quantitative manner, particu-
larly for those species with similar coordination environment
but different electronic states.[9,37,38] As shown in the room-
temperature 57Fe Mössbauer spectra in Figure 3F, Figures S17
and S18 and Table S4, Supporting Information, the pro-
posed pyridinic Fe–N4 coordination (53–72%, D1) is found to
be dominant in all samples (except for un-pyrolyzed FeCl3-
wood with an Fe–O coordination), which is consistent with
the XANES and EXAFS analysis. This D1 doublet with small
isomer shift (IS, ≈0.3) and quadrupole-splitting (QS, ≈1.3)
is typically assigned to the low-spin state of ferrous in the
square-planar coordination (i.e., pyridinic Fe–N4) in previous
reports on the Fe–N–C catalysts.[9] The absence of sextet
and singlet in Mössbauer spectra indicate that zero-valent
metal particle is not present. Meanwhile, small fractions of
other high-spin (Fe2+ with S = 2, D2/D4 doublets) and
middle-spin (Fe2+ or Fe3+ with S = 1 or 3/2, D3 doublet) species
are observed in the Fe SACs. The relatively large QS values
in two high-spin species indicate a complicated coordination
environment of Fe–N4 analogues with one or two axial ligands
to form a deformed octahedral structure.[37]
2.2. Oxidative Cyclization to Synthesize Multifunctional
Quinolines
As mentioned above, the hierarchically porous structure in our
carbon-based SACs maximizes the exposure of active sites to
reactants.[26,33] We chose the simple oxidative dehydrogenation
of tetrahydroquinoline (THQ) as a model reaction, which has
been previously promoted by other SAC catalysts.[39,40] As
shown in Figure S19(A), Supporting Information, nearly quan-
titative conversion of THQ and above 85% selectivity of quino-
line could be obtained using Fe-800 SAC at 120 °C for 48 h,
which is slightly lower than the benchmark performance of
10% Pd/C. The oxidative efficiency varies significantly with the
nature of transition metals. The replacement of Fe by Co fur-
ther improved the selectivity to ≈100%, while Ni-800 SAC and
Adv. Mater. 2021, 2101382
2101382 (4 of 11)
© 2021 Wiley-VCH GmbH

www.advancedsciencenews.com
www.advmat.de
Cu-800 SAC gave poor or even no performance. The use of
commercial 10% Fe/C, Fe2O3, or FeCl3 failed to oxidize THQ
due to the absence of activation site for O2.[41] This was fur-
ther confirmed by a survey of various Fe-, Co-, and Ni-based
catalysts where only SACs exhibited reactivity toward oxidative
dehydrogenation (except for 10% Pd/C, CoCl2, and cobalt fer-
rocene (CoFc)). Control experiments revealed that both catalyst
and oxidant (O2) are essential for this transformation.[42] The
yield of quinoline is also proportional to the catalyst loading.
More importantly, we observed a strong dependence of cata-
lytic activity on the pyrolysis temperature of Fe, Co, and Ni
SACs in Figure S19, Supporting Information. This indicates
that the coordination environment has strong influence on
catalytic performance, which can be regulated by the replace-
ment of four-coordinated N atoms with one or two C atoms,
together with Fe aggregations at higher pyrolysis temperatures
as indicated in the EXAFS spectra in Figures S20 and S21,
Supporting Information.[36,43] The different chemical environ-
ment of doped N species at various pyrolysis temperature may
give rise to differing catalytic activity, as reported in previous
studies.[44] However, regardless of the pyrolysis temperatures,
we obtained closely similar pyridinic structure of Fe-SACs,
as judged from FT-IR spectra, Raman spectra, SSNMR, XRD,
and XPS characterizations of the Fe-SACs in Figures S11–S13,
Supporting Information.
In view of the difficulty in accessing THQ subunits, a more
practical approach using readily available substrates would be
more attractive for the synthesis of complex, valuable quino-
lines.[13,14] As shown in Figure 4A, a three-component, oxidative
cyclization reaction employing feedstock chemicals aniline,
acetophenone and dimethyl sulfoxide (DMSO) as the starting
materials was pursued.[15] The formation of the desired product
Figure 4. Oxidative cyclization toward the synthesis of quinoline. A) Model reaction of the three-component quinoline synthesis. B) Catalyst screening
on the selectivity toward quinoline synthesis (in red), and C) chemical structures of starting materials and possible products. Selectivity was analyzed
by GC-MS.
2101382 (5 of 11)
© 2021 Wiley-VCH GmbH
Adv. Mater. 2021, 2101382

www.advancedsciencenews.com
www.advmat.de
(6-methoxy-2-phenylquinoline) together with possible side
products (imine and amide from aldol condensation as well
as azo-compound and phenazine from the homo-coupling of
arylamine) were monitored by gas chromatography-mass spec-
trometry (GC-MS) in Figure 4A. As illustrated in Figure 4B
and Figure S22, Supporting Information, our SACs gave the
highest GC selectivity toward quinoline derivatives among all
catalysts examined (≈68%, 44%, 42%, and 54% for Fe, Co, Ni,
and Cu-800 SAC with 10 mol% catalyst loading). In contrast,
commercial 10% Pd/C and 10% Fe/C catalyst only generated
trace amounts of quinoline (<15%), while homogeneous FeCl3
gave a moderate selectivity of 30%. The observation that iron
oxides were catalytically active (15% and 34% for Fe2O3 and
Fe3O4) means that an oxidation states between +2 and +3 is
beneficial for such oxidative cyclization. As shown in Table S5,
Supporting Information, temperature has a profound influence
on the catalytic efficiency (entries 1–6). Lowering the reaction
temperature to 80 or 100 °C prevented the thermolysis of DMSO
to trigger the reaction. Further elevating the temperature to
140 °C failed to improve the yield. Since DMSO is also serving
as reactant, replacement by other solvents (including HCHO
and methanol as alternative proton donor) will diminish the
reaction activity in entries 7–11. The existence of oxygen
also has a crucial role in forming the quinoline product by
the oxidative dehydrogenation in a radical pathway (entries
12 and 13).[41] Meanwhile, the use of higher loading catalysts in
entries 16–18 has no significant improvement to the reaction. The
reaction order of the catalyst is almost one (linear) with a much
lower slope compared to THQ dehydrogenation (6.8 vs 17.5),
indicating an analogous reaction pathway to that shown in
Figure S23, Supporting Information. To probe the influence of
porous channel on liquid-phase reactions, control experiment
using our Fe-800 SAC as monolith reactor has been conducted
in entry 19, where the reaction occurs with a lower yield of 39%.
This indicates the mass diffusion kinetics of reactant is not
restricted by the macroporous structure.[22] Other SAC catalysts
are also catalytically active as revealed in entries 20–22, sug-
gesting the universal applicability of such reaction.
The pyrolysis temperature had a profound influence on the
catalytic efficiency, where Fe-800 SAC with the typical Fe–N4
coordination was found to be the most effective with a TOF value
(0.42 h−1) comparable to homogeneous catalysts for similar three-
component synthesis of quinolines in Table S6, Supporting Infor-
mation. The influence of coordination environment on catalytic
activity[36] can be seen from the generation of imine as an inter-
mediate in the reaction with Fe-600 SAC, owing to insufficient
activation of acetophenone. At a lower temperature of 600 °C,
the Fe–O coordination in FeCl3-wood is not be fully converted to
the desired Fe–N coordination, leading to a decrease in catalytic
activity.[36,43] As far as we know, this is the first study highlighting
the synthetic utility of SACs in complex three-component reac-
tions, paving the way toward the production of high-value chemi-
cals in an efficient, atom-economical manner. For comparison
with traditional approaches, we have also verified the cata-
lytic efficiency of Fe-800 SAC in the modified, two-component
Friedländer synthesis in Figure S24, Supporting Information,
where > 95% GC yield of 2-phenylquinoline was obtained.[45]
To assess the generality of our protocol, various combina-
tions of substituted anilines and acetophenones were examined
in Figure 5A. For instance, multifunctional quinolines with
methoxy (1b), alkyl (1a and 1d), halogens (1e–1h), and meth-
ylthio (1c) moieties at the 6-position can be efficiently con-
structed (46–87% yield) by using the corresponding anilines.
The electronic attributes had a strong influence on the outcome
of reaction, where electronic-deficient anilines generally give a
lower activity to form quinoline.[15] Substituted acetophenones
with synthetically valuable functionalities such as methoxy
(2a), methylthio (2b), nitro (2i), nitrile (2j), halogens (2d–2g),
and trifluoromethyl (2h) can be well-tolerated by this oxidative
cyclization protocol (52–78% yield) regardless of the electronic
attributes. Interestingly, modification at 3-position of quinoline
can be realized by using propiophenone despite of a lower yield
(2k, 27%). Our protocol is also applicable to those heterocycles
such as pyridine (2i and 2m), furan (2n), and thiophene (2o),
which is typically challenging in the Suzuki coupling owing to
strong poisoning effect to the catalytically active metal centers
and requires a high loading of noble metal catalysts.[10]
The synthetic utility of Fe1-catalyzed oxidative cyclization was
further highlighted by the deuteration at the fourth position
of quinoline with low-cost deuteration reagent (DMSO-d6).[46]
Five different representative examples with methoxy-, bromo-,
methylthio-, and trifluoromethyl-units at the 2- or 6-position of
quinolines were examined in Figure 5B. Similarly, all of the sub-
strates can be deuterated with similar yields to the undeuterated
analogues (42–73%). Furthermore, we explored the synthesis
of a Cavosonstat derivative (a F508del-CFTR mutation treat-
ment medicine) and its 4-deuterated analogue in Figure 5C.[47]
Despite lower yields owing to the multifunctional nature of
substrate and possible side reactions, the desired compounds
(3a, 35%; 3a', 32%) can be obtained by a one-step, Fe1-catalyzed
cyclization rather than traditional palladium-catalyzed Suzuki
coupling between specified quinoline and phenyl boronic ester.
A kinetic study was also performed in Figure 6A to provide
mechanistic insights. An induction period was observed in the
initial 6 hours owing to the multi-step pathway toward quino-
line formation, where the intermediate (imine) was predomi-
nately detected by GC-MS during this period. Further extending
the reaction period to 42 h did not result in a better yield com-
pared with the optimized condition. Finally, the Fe-800 SAC
can be recycled eight times without any decrease in conver-
sion (>95%) and selectivity (>60%) in Figure 6B. The additional
HAADF-STEM images, 57Fe Mössbauer and EXAFS charac-
terization in Figure 6, and Figures S25 and S26, Supporting
Information, confirm the intact single-atom nature of the spent
catalyst. In particular, quantitative analysis by 57Fe Mössbauer
spectrum suggests no obvious change in the catalyst com-
position in D1–D4 species in Table S4. ICP testing of the
supernatant of reaction mixture and the recovered catalyst also
confirmed no metal leaching after reaction completion.[12]
As shown in Figure S18, Supporting Information, the pyri-
dinic Fe–N4 species (D1) is believed to be catalytically active
toward oxidative cyclization in view of moderate activity in
Fe-600/700 SAC. The high-spin species (D2 and D4) are gen-
erally inactive because of their saturated coordination. The
catalytic activity is strongly correlated to the percentage of the
middle-spin species (D3), where the most active Fe-800 SAC
also possesses the highest content of D3 (≈19%). This is in

good agreement with selective oxidation of the C H bond by
Adv. Mater. 2021, 2101382
2101382 (6 of 11)
© 2021 Wiley-VCH GmbH

www.advancedsciencenews.com
www.advmat.de
Figure 5. Synthetic utility of SAC-catalyzed oxidative cyclization. A) General reaction in the color scheme. A wide assortment of multifunctional qui-
nolines containing halogens, nitro, nitrile, sulfides, and heterocycles can be accessed with the Fe-800 SAC catalyst. The reaction was performed with
0.1 mmol of substituted acetophenone, 0.3 mmol of substituted aniline, 10 mol% of catalyst in 1 mL of DMSO at 120 °C under O2 for 24 h. B) Synthesis
of 4-deuterated multifunctional quinolines by the same strategy with DMSO-d6 as deuteration reagent. C) One-step synthesis of Cavosonstat building
block and its 4-deuterated analogue using our Fe-800 SAC catalyst.
Fe–N–C catalyst, where the middle-spin species is 33 times
more active than high-spin species and ≈1 order of magni-
tude more active than the pyridinic Fe–N4 species.[9] Although
the detailed configuration of this middle-spin species remains
unclear due to its low percentage in the Fe–N–C catalysts,
they are most probably Fe2+ or Fe3+ with a spin-state of S = 1
or 3/2.[9,37,38] Considering the importance of the coordina-
tion atoms on the catalytic performance, considerable efforts
have been devoted to tailor the type of coordination atoms,
coordination number and second- or higher-coordination
shells in recent years.[48,49] The structure of the SAC is dynamic
in the liquid phase. For instance, as demonstrated by operando
57Fe Mössbauer spectroscopy, high-spin Fe–N4 species suffer
from significant catalytic degradation owing to the switching
of oxidation state (Fe3+/Fe2+) under ORR operating conditions
in fuel cells, while the low or middle spin species are more
durable and remain constant (Fe2+).[50]
We have conducted DFT calculations to identify the reac-
tion pathway of oxidative cyclization, and the influence of the
coordination environment on reaction efficiency.[40,51] Pyridinic
Fe–N4 and analogous Fe–C1N3, Fe–C2N2 structures are employed
as models due to their abundance in Fe–N–C catalysts. DFT
calculations and experimental evidence on the activation of
O2 by Fe–N4 SAC and its analogues have been detailed in pre-
vious studies on the oxygen reduction, benzene oxidation and
oxidative dehydrogenation of THQ; these were reported to
2101382 (7 of 11)
© 2021 Wiley-VCH GmbH
Adv. Mater. 2021, 2101382

www.advancedsciencenews.com
www.advmat.de
Figure 6. Kinetic study and stability measurement. A) Monitoring of the reaction under standard conditions over time. B) Cycling stability of Fe-800
SAC for the synthesis of 6-methoxy-2-phenylquinoline by oxidative cyclization. C) Atomic-resolution STEM-HAADF image of the recycled Fe-800 SAC.
D) Room-temperature 57Fe Mössbauer spectrum of the recycled catalyst. The reaction was performed with 0.1 mmol of acetophenone, 0.3 mmol of
p-anisidine, 10 mol% of catalyst in 1 mL of DMSO at 120 °C under O2 for 24 h. Scale bar: 2 nm.
follow a spin crossover pathway or a redox process involving
the appearance of high-valence metal species.[38,52] As illus-
trated in the proposed catalytic cycle in Figure 7 and models in
Figures S28–S30, Supporting Information, the preferred
adsorption of acetophenone on metallic center is the initial step
of the reaction; this is consistent with the wide substrate scope
of acetophenone regardless of its electronic attributes compared
to aniline. The activation of carbonyl by SAC has been proven by
Li et al. to be an exothermic step.[53] The activated acetophenone
is then reacted with the fragment of DMSO under its catalytic
thermolysis to form the demethylthioated intermediate[15] with a
minimum activation barrier (≈0.5 eV) over the Fe–N4 or Fe–C1N3
Figure 7. Proposed mechanism by DFT calculation. Energy profiles of the oxidative cyclization of acetophenone and p-anisidine over Fe SACs with the
Fe–N4, Fe–C1N3, or Fe–C2N2 coordinations. Inset shows the optimized configurations of reactants, intermediates, and product on an Fe–N4 catalyst. The
color scheme used: crimson for Fe, blue for N, gray for C, red for O, and white for H. Outer carbon atoms in the model are omitted for the sake of clarity.
Adv. Mater. 2021, 2101382
2101382 (8 of 11)
© 2021 Wiley-VCH GmbH

www.advancedsciencenews.com
www.advmat.de
go beyond 140 °C) Hydrothermal treated wood sticks were taken out,
thoroughly washed by water, ethanol until the supernatant became
colorless and dried at 80 °C overnight before socking into an urea
aqueous solution for another 24 h. After that, 14 g of FeCl3/urea treated
wood sticks were loaded into a quartz boat (20 × 10 × 2 cm3), transferred
to a tube furnace under 50 sccm of Ar gas, and then heated to 800 °C at
5 °C min−1 and maintained for 2 h. The Fe loading on carbonized wood
(≈0.8 wt%) was determined by ICP-OES. The Fe loading can be further
increased to ≈2.5 wt% by using 0.5 m FeCl3 ethanol/conc. HCl mixture,
respectively. In the case of Co1, Ni1, or Cu1-SACs, a 0.05 m CoCl2, NiCl2,
or CuCl2 ethanol/conc. HCl mixture were used. Control samples without
the addition of FeCl3 or urea were also prepared.
Oxidative Dehydrogenation of Tetrahydroquinoline Using Fe1-Single-Atom
Catalyst: Typically, Fe1-SAC (5 mol% Fe) was loaded in a vial with 2 mL of
mesitylene, 0.1 mmol of 1,2,3,4-tetrahydro-quinoline (THQ, 12.55 µL) and
stirred at 120 °C for 48 h in air.[39] After reaction, the solid catalyst was
recovered by centrifugation and washed with ethyl acetate for 3 times.
The organic substance was combined and concentrated under reduced
pressure. After completely removal of mesitylene by evaporation, the
crude product was analyzed by GC-MS.
The Construction of Substituted Quinolines by Oxidative Cyclization
Using Fe1-Single-Atom Catalysts: Typically, Fe-800 SAC (10 mol% Fe) was
loaded in Schlenk tube with 1 mL of DMSO, 0.1 mmol of acetophenone
(11.66 µL), and 0.3 mmol of p-anisidine (37.96 mg). The mixture was
stirred at 120 °C for 24 h under O2 atmosphere (O2 balloon).[15] After
reaction, the solid catalyst was recovered by centrifugation and washed
with ethyl acetate for 3 times. 10 mL of brine was added to the organic
substance, and the resulting mixture was extracted with 5 mL of ethyl
acetate (3×). The combined organic layer was washed with 10 mL of brine
for another 3 cycles and dried over anhydrous Na2SO4. After completely
removal of ethyl acetate by evaporation, the crude product was analyzed
by GC-MS. To obtain the 4-deuterated derivatives, 1 mL of DMSO-d6
was used with other identical conditions. For stability measurement, the
solid catalyst was recovered by centrifugation and thoroughly washed by
DMSO, ethyl acetate and dried at 80 °C overnight before re-dispensing
for another cycle.
catalyst. A similar pathway to homogeneous catalysis may
be operational, in which acetophenone reacts with sulfe-
nium cation to form phenyl vinyl ketone, and amino group
serves as proton donor to the decomposition of DMSO in
the control experiments in Figure S27, Supporting Infor-
mation. This further promotes the exothermic, co-activa-
tion of aniline and subsequent condensation to form the
imine intermediate, which is a detectable side product in
GC-MS. Such condensation process is calculated to be highly
dependent on the coordination environment of the cata-
lyst, where Fe–C2N2 exhibits a relatively high energy barrier
(≈1 eV) compared to others. The following cyclization and
aromatization to afford the quinoline structure are rela-
tively rapid in term of thermodynamics, as-proven by the
excellent performance of SACs toward the oxidative dehy-
drogenation of THQ in Figure S19, Supporting Informa-
tion.[39] Meanwhile, the desorption process of quinoline
product from the catalyst surface is sluggish and could
be rate-determining step owing to its strong π-π interac-
tion (poisoning) with the catalyst, especially in the case of
Fe–C2N2 and Fe–C1N3. This can be attributed to the charge-
deficient nature of the Fe center after the replacement of N by
one or two C, leading to a stronger interaction with external
heteroatoms (such as, N in quinoline).[36] Based on the above
DFT calculations, we propose the Fe–N4 coordination with
lower reaction barriers would be the primarily active center for
this three-component reaction.
3. Conclusion
We have successfully synthesized a mechanically strong,
hierarchically porous support for SACs by immobilizing
transition metal atoms (M = Fe, Co, Ni, Cu) in natural wood
by a top-down pyrolysis approach. Such carbon-supported
SACs exhibited excellent catalytic activities and selectivity
toward the oxidative dehydrogenation of THQ, as well as, the
three-component, oxidative cyclization to afford 6-methoxy-
2-phenylquinoline, outperforming the benchmark Pd/C, Fe/C,
and homogeneous FeCl3 catalysts. The superior performance of
our SACs is attributed to the optimized metal–N4 coordination
with relatively weak product desorption energy, as well as, max-
imized exposure of active sites to reactants by the hierarchically
porous carbon plates.
Material Characterization: The following equipment was used: STEM
(JEOL ARM200F equipped with ASCOR probe corrector at 80 kV), SEM
(JEOL JSM-6701F), TEM (JEM 2010F, 200 kV), XPS (AXIS UltraDLD,
monochromatic Al Ka), XRD (Bruker D8), TGA (Discovery 5500), ICP-OES
(Perkin Elmer Avio 500, ppm level accuracy), GC-MS (Agilent 5975 C
inert MSD with triple-axis detector), MS (Bruker MicroTOF-QII), NMR
(Bruker AV400), BET (Quantachrome Instruments Autosorb-iQ, 77K),
Raman (WITec Alpha 300R), FT-IR (Bruker vertex 80v), SSNMR
(9T Bruker Advance Neo 400 MHz, wide-bore,
4 mm probe).
XANES/EXAFS: SAC samples were cut into a dimension of 1 × 1 cm2
and placed onto sample holder. Measurements were carried out at the
Singapore Synchrotron Light Source, X-ray Absorption Fine Structure for
Catalysis beamline.[54] Data analysis and simulation were carried out on
Athena, Artemis, and Hephaestus (Version 0.9.23).[55]
Room-Temperature 57Fe Mössbauer Spectroscopy: 57Fe-enriched samples
were prepared by dissolving 100 mg of 57Fe foil (Isoflex) in 4.2 mL of
concentrated HCl, 3.4 mL of water, and 1.0 mL of 30% H2O2 with stirring
at 75 °C until complete dissolution. 4.8 mL of the above 57FeCl3 clear
solution was added to 16 mL of ethanol for the hydrothermal treatment
of raw wood at 140 °C for 2h. 57Fe-enriched SACs were prepared by
socking in urea solution, drying, and subsequent annealing at various
temperatures. The 57Fe Mössbauer measurements were performed with
a proportional counter and a Topologic 500A spectrometer with 57Co
(Rh) as a γ-ray radioactive source. An Fe foil was used to calibrate the
velocity.
4. Experimental Section
Synthesis of Fe1-Single-Atom Catalysts: Natural wood sticks (birch)
were purchased from Daiso (Singapore), cut into rectangular shape
(5 × 1 × 0.2 cm3 along the growth direction) and dried at 100 °C for 24 h
to evaporate all the moisture. Prior to use, 30 g of dried wood sticks were
extracted in 225 mL of an ethanol/toluene mixture (1:2, v/v) at 75 °C for
24 h and then washed with ethanol (3 ×) and acetone (1×) before drying
at 75 °C to remove unwanted waxes, fats, and resins. 14 g of extracted
wood sticks were then immersed into 20 mL of a 0.05 m FeCl3 ethanol/
conc. HCl solution (4:1, v/v) in a 50 mL Teflon-lined stainless-steel
autoclave. After socking at room temperature for 2 h, the autoclave
was sealed and heated at 140 °C for 2 h in an oven and then cooled
down to room temperature naturally. (Caution: Corrosive ethanol/conc.
HCl mixture was used and the hydrothermal temperature should not
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
2101382 (9 of 11)
© 2021 Wiley-VCH GmbH
Adv. Mater. 2021, 2101382

www.advancedsciencenews.com
www.advmat.de
[7] L. Lin, S. Yao, R. Gao, X. Liang, Q. Yu, Y. Deng, J. Liu, M. Peng,
Z. Jiang, S. Li, Y.-W. Li, X.-D. Wen, W. Zhou, D. Ma, Nat. Nano-
technol. 2019, 14, 354.
[8] Y. Zhu, W. Sun, J. Luo, W. Chen, T. Cao, L. Zheng, J. Dong, J. Zhang,
M. Zhang, Y. Han, C. Chen, Q. Peng, D. Wang, Y. Li, Nat. Commun.
2018, 9, 3861.
[9] W. Liu, L. Zhang, X. Liu, X. Liu, X. Yang, S. Miao, W. Wang, A. Wang,
T. Zhang, J. Am. Chem. Soc. 2017, 139, 10790.
[10] Z. Chen, E. Vorobyeva, S. Mitchell, E. Fako, M. A. Ortuño, N. López,
S. M. Collins, P. A. Midgley, S. Richard, G. Vilé, J. Pérez-Ramírez,
Nat. Nanotechnol. 2018, 13, 702.
[11] C. Liu, Z. Chen, H. Yan, S. Xi, K. M. Yam, J. Gao, Y. Du, J. Li,
X. Zhao, K. Xie, H. Xu, X. Li, K. Leng, S. J. Pennycook, B. Liu,
C. Zhang, M. J. Koh, K. P. Loh, Sci. Adv. 2019, 5, eaay1537.
[12] Z. Chen, C. Liu, J. Liu, J. Li, S. Xi, X. Chi, H. Xu, I.-H. Park, X. Peng,
X. Li, W. Yu, X. Liu, L. Zhong, K. Leng, W. Huang, M. J. Koh,
K. P. Loh, Adv. Mater. 2020, 32, 1906437.
Acknowledgements
K.P.L acknowledges NRF-CRP grant “Two-Dimensional Covalent
Organic Framework: Synthesis and Applications,” grant number
NRF-CRP16-2015-02, funded by National Research Foundation, Prime
Minister’s Office, Singapore. X.P. would like to thank the National
Science Foundation of China (31971614, 21736003), Guangdong Natural
Science Funds for Distinguished Young Scholar (2016A030306027),
Guangdong Natural Science Funds (2017A030313130), Guangzhou
Science and Technology Funds (201904010078), State Key Laboratory of
Pulp and Paper Engineering and Fundamental Research Funds for the
Central Universities. J.W. would like to thank the International Partnership
Program of Chinese Academy of Sciences (No. 121421KYSB 20170020).
The authors appreciate the near-edge X-ray adsorption measurement by
Dr. Caozheng Diao at Singapore Synchrotron Light Source (SSLS), Soft
X-ray—ultraviolet (SUV) beamline and the SSNMR measurement by Dr.
Gustavo G. Shimamoto at NUS CMMAC Laboratories.
[13] J. Marco-Contelles, E. Pérez-Mayoral, A. Samadi, M. d. C. Carreiras,
E. Soriano, Chem. Rev. 2009, 109, 2652.
[14] V. V. Kouznetsov, L. Y. V. Méndez, C. M. M. Gómez, Curr. Org.
Chem. 2005, 9, 141.
[15] Y. Liu, Y. Hu, Z. Cao, X. Zhan, W. Luo, Q. Liu, C. Guo, Adv. Synth.
Catal. 2018, 360, 2691.
Conflict of Interest
The authors declare no conflict of interest.
[16] S. Mahato, A. Mukherjee, S. Santra, G. V. Zyryanov, A. Majee,
Org. Biomol. Chem. 2019, 17, 7907.
[17] S. M. Prajapati, K. D. Patel, R. H. Vekariya, S. N. Panchal,
H. D. Patel, RSC Adv. 2014, 4, 24463.
[18] L.-H. Li, Z.-J. Niu, Y.-M. Liang, Chem. - Asian J. 2020, 15, 231.
[19] L. Liu, A. Corma, Chem. Rev. 2018, 118, 4981.
[20] Y. Yang, Y. Yang, Z. Pei, K.-H. Wu, C. Tan, H. Wang, L. Wei,
A. Mahmood, C. Yan, J. Dong, S. Zhao, Y. Chen, Matter 2020, 3,
1442.
[21] T. Li, Y. Zhai, S. He, W. Gan, Z. Wei, M. Heidarinejad, D. Dalgo,
R. Mi, X. Zhao, J. Song, J. Dai, C. Chen, A. Aili, A. Vellore, A. Martini,
R. Yang, J. Srebric, X. Yin, L. Hu, Science 2019, 364, 760.
[22] Y. Wang, G. Sun, J. Dai, G. Chen, J. Morgenstern, Y. Wang, S. Kang,
M. Zhu, S. Das, L. Cui, L. Hu, Adv. Mater. 2017, 29, 1604257.
[23] A. Husbands, S. Cranford, Matter 2019, 1, 1092.
[24] Z.-L. Yu, N. Yang, L.-C. Zhou, Z.-Y. Ma, Y.-B. Zhu, Y.-Y. Lu, B. Qin,
W.-Y. Xing, T. Ma, S.-C. Li, H.-L. Gao, H.-A. Wu, S.-H. Yu, Sci. Adv.
2018, 4, eaat7223.
Author Contributions
Z.C. and J.S. contributed equally to this work. Z.C. and J.S. conceived
the research, synthesized the materials, conducted catalytic reactions,
and wrote the draft under the supervision of X.P. and K.P.L. TEM,
STEM characterization, and data analysis were conducted by Z.C., X.Z.,
and X.Z.; X-ray adsorption measurements and data processing were
conducted by S.X.; Mössbauer spectra and data analysis were performed
by W.Z., R.G., A.I.R., and J.W.; SEM measurements were conducted
by J.S. and R.L.; X.P. and L.Z. performed the XPS measurements; J.L.,
R.L., C.L., H.X., H. L., L.W., X.L., and M.J.K. assisted with materials
characterization and data analysis. All authors discussed and
commented on the manuscript.
Data Availability
The data that support the findings of this study are available from the
corresponding author upon reasonable request.
[25] S. Mitchell, N.-L. Michels, J. Pérez-Ramírez, Chem. Soc. Rev. 2013,
42, 6094.
[26] X. Peng, L. Zhang, Z. Chen, L. Zhong, D. Zhao, X. Chi, X. Zhao,
L. Li, X. Lu, K. Leng, C. Liu, W. Liu, W. Tang, K. P. Loh, Adv. Mater.
2019, 31, 1900341.
Keywords
[27] Z. Chen, H. Zhuo, Y. Hu, H. Lai, L. Liu, L. Zhong, X. Peng, Adv.
Funct. Mater. 2020, 30, 1910292.
hierarchically porous structure, organic catalysis, oxidative cyclization,
single-atom catalysts, three-component reaction
[28] Y. Ohashi, T. Watanabe, ACS Omega 2018, 3, 16271.
[29] H. Zhou, S. Hong, H. Zhang, Y. Chen, H. Xu, X. Wang, Z. Jiang,
S. Chen, Y. Liu, Appl. Catal., B 2019, 256, 117767.
[30] W. Chen, X.-W. Peng, L.-X. Zhong, R.-C. Sun, ACS Sustainable Chem.
Eng. 2015, 3, 1366.
Received: February 18, 2021
Revised: May 1, 2021
Published online:
[31] Y. Huang, Y. Chen, X. Fan, N. Luo, S. Zhou, S.-C. Chen, N. Zhao,
C. P. Wong, Small 2018, 14, 1801520.
[1] A. Wang, J. Li, T. Zhang, Nat. Rev. Chem. 2018, 2, 65.
[2] X. Cui, W. Li, P. Ryabchuk, K. Junge, M. Beller, Nat. Catal. 2018, 1,
385.
[32] J. Sun, S. E. Lowe, L. Zhang, Y. Wang, K. Pang, Y. Wang, Y. Zhong,
P. Liu, K. Zhao, Z. Tang, H. Zhao, Angew. Chem., Int. Ed. 2018, 57,
16511.
[3] Y. Pan, C. Zhang, Z. Liu, C. Chen, Y. Li, Matter 2020, 2, 78.
[4] H. Yan, C. Su, J. He, W. Chen, J. Mater. Chem. A 2018, 6, 8793.
[5] L. Zhang, Y. Ren, W. Liu, A. Wang, T. Zhang, Natl. Sci. Rev. 2018, 5,
653.
[6] L. Liu, M. Lopez-Haro, C. W. Lopes, C. Li, P. Concepcion,
L. Simonelli, J. J. Calvino, A. Corma, Nat. Mater. 2019, 17, 866.
[33] Z. Chen, C. Liu, X. Zhao, H. Yan, J. Li, P. Lyu, Y. Du, S. Xi, K. Chi,
X. Chi, H. Xu, X. Li, W. Fu, K. Leng, S. J. Pennycook, S. Wang,
K. P. Loh, Adv. Mater. 2019, 31, 1804763.
[34] Z. Chen, K. Leng, X. Zhao, S. Malkhandi, W. Tang, B. Tian, L. Dong,
L. Zheng, M. Lin, B. S. Yeo, K. P. Loh, Nat. Commun. 2017, 8,
14548.
Adv. Mater. 2021, 2101382
2101382 (10 of 11)
© 2021 Wiley-VCH GmbH

www.advancedsciencenews.com
www.advmat.de
[35] J. Meng, J. Li, J. Liu, X. Zhang, G. Jiang, L. Ma, Z.-Y. Hu, S. Xi,
Y. Zhao, M. Yan, P. Wang, X. Liu, Q. Li, J. Z. Liu, T. Wu, L. Ma, ACS
Cent. Sci. 2020, 6, 1431.
[36] Y. Pan, Y. Chen, K. Wu, Z. Chen, S. Liu, X. Cao, W.-C. Cheong,
T. Meng, J. Luo, L. Zheng, C. Liu, D. Wang, Q. Peng, J. Li, C. Chen,
Nat. Commun. 2019, 10, 4290.
[45] K. Motokura, T. Mizugaki, K. Ebitani, K. Kaneda, Tetrahedron Lett.
2004, 45, 6029.
[46] T. Pirali, M. Serafini, S. Cargnin, A. A. Genazzani, J. Med. Chem.
2019, 62, 5276.
[47] K. W. Southern, S. Patel, I. P. Sinha, S. J. Nevitt, Cochrane Database
Syst. Rev. 2018, 8, 1465.
[37] U. I. Kramm, M. Lefèvre, N. Larouche, D. Schmeisser, J.-P. Dodelet,
J. Am. Chem. Soc. 2014, 136, 978.
[38] X. Li, C.-S. Cao, S.-F. Hung, Y.-R. Lu, W. Cai, A. I. Rykov, S. Miao,
S. Xi, H. Yang, Z. Hu, J. Wang, J. Zhao, E. E. Alp, W. Xu, T.-S. Chan,
H. Chen, Q. Xiong, H. Xiao, Y. Huang, J. Li, T. Zhang, B. Liu, Chem
2020, 6, 3440.
[39] S. Wei, Y. Wang, W. Chen, Z. Li, W.-C. Cheong, Q. Zhang, Y. Gong,
L. Gu, C. Chen, D. Wang, Q. Peng, Y. Li, Chem. Sci. 2020, 11, 786.
[40] Y. Han, Z. Wang, R. Xu, W. Zhang, W. Chen, L. Zheng, J. Zhang,
J. Luo, K. Wu, Y. Zhu, C. Chen, Q. Peng, Q. Liu, P. Hu, D. Wang,
Y. Li, Angew. Chem., Int. Ed. 2018, 57, 11262.
[41] D. Xu, H. Zhao, Z. Dong, J. Ma, ChemCatChem 2019, 11, 5475.
[42] J. Li, G. Liu, X. Long, G. Gao, J. Wu, F. Li, J. Catal. 2017, 355, 53.
[43] Y.-N. Gong, L. Jiao, Y. Qian, C.-Y. Pan, L. Zheng, X. Cai, B. Liu,
S.-H. Yu, H.-L. Jiang, Angew. Chem., Int. Ed. 2020, 59, 2705.
[44] J. Zhang, C. Zheng, M. Zhang, Y. Qiu, Q. Xu, W.-C. Cheong, W. Chen,
L. Zheng, L. Gu, Z. Hu, D. Wang, Y. Li, Nano Res. 2020, 13, 3082.
[48] J. Yang, W. Li, D. Wang, Y. Li, Small Struct. 2021, 2, 2000051.
[49] X. Li, H. Rong, J. Zhang, D. Wang, Y. Li, Nano Res. 2020, 13,
1842.
[50] J. Li, M. T. Sougrati, A. Zitolo, J. M. Ablett, I. C. Ogˇuz, T. Mineva,
I. Matanovic, P. Atanasov, Y. Huang, I. Zenyuk, A. Di Cicco,
K. Kumar, L. Dubau, F. Maillard, G. Dražić, F. Jaouen, Nat. Catal.
2021, 4, 10.
[51] L. Wang, M.-X. Chen, Q.-Q. Yan, S.-L. Xu, S.-Q. Chu, P. Chen, Y. Lin,
H.-W. Liang, Sci. Adv. 2019, 5, eaax6322.
[52] H.-T. Lien, S.-T. Chang, P.-T. Chen, D. P. Wong, Y.-C. Chang, Y.-R. Lu,
C.-L. Dong, C.-H. Wang, K.-H. Chen, L.-C. Chen, Nat. Commun.
2020, 11, 4233.
[53] L. Kuai, Z. Chen, S. Liu, E. Kan, N. Yu, Y. Ren, C. Fang, X. Li, Y. Li,
B. Geng, Nat. Commun. 2020, 11, 48.
[54] Y. Du, Y. Zhu, S. Xi, P. Yang, H. O. Moser, M. B. H. Breese,
A. Borgna, J. Synchrotron Radiat. 2015, 22, 839.
[55] B. Ravel, M. Newville, J. Synchrotron Radiat. 2005, 12, 537.
2101382 (11 of 11)
© 2021 Wiley-VCH GmbH
Adv. Mater. 2021, 2101382