Selective Electrocatalytic Reduction of Oxygen to Hydroxyl Radicals via 3-Electron Pathway with FeCo Alloy Encapsulated Carbon Aerogel for Fast and Complete Removing Pollutants

Article abstract of DOI:10.1002/anie.202101804

We reported the selective electrochemical reduction of oxygen (O₂) to hydroxyl radicals (^.OH) via 3-electron pathway with FeCo alloy encapsulated by carbon aerogel (FeCoC). The graphite shell with exposed -COOH is conducive to the 2-

Full text of DOI:10.1002/anie.202101804

Angewandte
Research Articles
Chemie
How to cite: Angew. Chem. Int. Ed. 2021, 60, 10375–10383
International Edition: doi.org/10.1002/anie.202101804
German Edition: doi.org/10.1002/ange.202101804
Electrocatalysis
Selective Electrocatalytic Reduction of Oxygen to Hydroxyl Radicals
via 3-Electron Pathway with FeCo Alloy Encapsulated Carbon Aerogel
for Fast and Complete Removing Pollutants
Fan Xiao, Zining Wang, Jiaqi Fan, Tetsuro Majima, Hongying Zhao,* and Guohua Zhao*
Abstract: We reported the selective electrochemical reduction
of oxygen (O2) to hydroxyl radicals (COH) via 3-electron
pathway with FeCo alloy encapsulated by carbon aerogel
(FeCoC). The graphite shell with exposed -COOH is con-
ducive to the 2-electron reduction pathway for H2O2 generation
stepped by 1-electron reduction towards to COH. The electro-
catalytic activity can be regulated by tuning the local electronic
environment of carbon shell with the electrons coming from the
inner FeCo alloy. The new strategy of COH generation from
electrocatalytic reduction O2 overcomes the rate-limiting step
over electron transfer initiated by reduction-/oxidation-state
cycle in Fenton process. Fast and complete removal of
ciprofloxacin was achieved within 5 min in this proposed
system, the apparent rate constant (kobs) was up to 1.44 Æ
0.04 minÀ1, which is comparable with the state-of-the-art
advanced oxidation processes. The degradation rate almost
remains the same after 50 successive runs, suggesting the
satisfactory stability for practical applications.
be selectively reduced to H2O2 and H2O via 2- and 4-electron
reduction reactions [Eq. (1) and (2), respectively].[2]
O2
O2
2 H
4 H
2 eÀ ! H2O2
4 eÀ ! 2 H2O
1
2
Very recently, the direct H2O2 synthesis through ORR
over carbonaceous materials has gained increasing attention
in environmental remediation,[3] avoiding the transport,
storage and handling of bulk H2O2 to site.[4] Additionally,
homogeneous ferrous irons (Fe2+) and heterogeneous solids
can efficiently catalyze the electrogenerated H2O2 into COH.
This is a variant of Huron-Dow process named as electro-
Fenton (EF) [Eq. (3)],[5] producing abundant COH in ambient
condition. However, the one-time depleted H2O2 reactant and
continuous consumption of metal species make EF process
unsustainable with poor recyclability and secondary pollu-
tion. Meanwhile, the degradation rate in EF process is
severely restricted by the rate-limiting step over the reduction
of Fe3+ to Fe2+ [Eq. (4)]. Unfortunately, these problems are
difficult to be solved by conventional approaches.
Introduction
With the increased utilization of organic compounds in
industry and daily life, the consequent growing discharges of
persistent organic pollutants irreversibly affect the environ-
ment and potentially threaten human health as well. Ad-
vanced oxidation processes (AOPs) have been deeply studied
for removing recalcitrant organic contaminants through the
generation of reactive oxygen species (ROS).[1] It is well
known that hydroxyl radicals (COH) possess the highest
oxidation potential (2.80 V) among ROS that can efficiently
and unselectively oxidize most of organic pollutants. From the
economic and environmental points of view, O2 is an ideal
oxidant since it is abundant and cheap. Numerous investiga-
tions concerning about oxygen reduction reaction (ORR)
have been carried out. In electrochemical system, O2 would
H2O2
Fe3
Fe2
H
!
OH
Fe3
H2O
C
3
4
eÀ ! Fe2
Inspired by the concept that electron penetration through
carbon layer from active metal core to the carbon layer can
stimulate electrocatalytic activity of carbon surface in a vari-
ety of catalytic reactions,[6] we turn our research interest to the
COH production during electrocatalytic reduction of O2. The
remaining key point is how to simultaneously trigger the
selective electrocatalytic reduction of O2 and H2O2 but
without the contribution of Fenton reaction. In this case,
carbonaceous electrode with graphite shell can be considered
as a promising material to promote the electrocatalytic
activity and prevent the core sites from reacting with external
solution. However, the electrocatalytic mechanism of con-
verting O2 to COH yet remains underexplored, which is not
only critical in helping realize a multi-electron oxygen
reduction process, but provides fundamental insight into the
development of novel technologies in environmental reme-
diation.
[*] F. Xiao, Z. N. Wang, J. Q. Fan, H. Y. Zhao, G. H. Zhao
Shanghai Key Lab of Chemical Assessment and Sustainability, School
of Chemical Science and Engineering, Tongji University
1239 Siping Road, Shanghai 200092 (China)
E-mail: hyzhao@tongji.edu.cn
g.zhao@tongji.edu.cn
T. Majima
Herein, we propose a new strategy of designing FeCo
alloy-encapsulated carbon aerogel (FeCoC) as working
electrode directly, conducting the electrochemical reduction
of O2 to COH via overall 3-electron pathway [Eq. (6)].
Carbon-based aerogel is one of particular interest since it
The institute of Scientific and Industrial Research, Osaka University
Mihogaoka 8-1, Ibaraki, Osaka, 567-0047 (Japan)
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.202101804.
Angew. Chem. Int. Ed. 2021, 60, 10375 –10383
ꢀ 2021 Wiley-VCH GmbH
10375

Angewandte
Research Articles
Chemie
can be shaped into desired dimensions according to the actual
requirements. The encapsulated FeCo alloy with optimal
metal doping ratio favors the in situ formation of oxygen-
enriched functional groups, benefiting the 2-electron ORR.
Meanwhile, The electrons from encapsulated FeCo alloy
endow the carbon shell with the activity of electrocatalytic
reducing H2O2 to COH via 1-electron pathway [Eq. (5)],[7]
whereby all problems originating from the continuing H2O2
supply, poor catalyst recyclability, and accumulation of metal-
containing sludge in traditional Fenton reaction are solved.[6b]
Moreover, investigating the mechanism of transient ROS
formation and resolving the rate-limiting step of classic EF
reaction can boost the removal of various pollutants.
O2
2 H
eÀ
2 eÀ ! H2O2
1a
5
H2O2
O2
!
!
OH
OH
OHÀ
OHÀ
C
carbon shell
3 eÀ
6
C
2 H
Results and Discussion
Properties of FeCo Alloy-Encapsulated Carbon Aerogel (FeCoC)
Electrode
Figure 1. a) Synthetic route and chemical reaction pathway of FeCoC.
b) HRTEM images of Fe0.5Co0.5C. Inset: a photography of the FeCoC
electrode. c) TEM images of Fe0.5Co0.5C. d–f) Energy dispersive X-ray
spectroscopy elemental mappings of Fe0.5Co0.5C. g) XRD patterns and
h) XPS spectra of pure carbon aerogel and Fe0.5Co0.5C. i) 57Fe Mçssba-
uer spectrum of Fe0.5Co0.5C. Scale bar: b) 5 nm, c) 40 nm, and d–
f) 30 nm.
As schematically illustrated in Figure 1a, FeCoC was
prepared through a copolymerization method by controlling
the introduction of ferric-/cobalt- acetylacetonate during the
synthesis of phenolic polymer aerogel from resorcinol and
formaldehyde. The obtained FeCoC (exemplified with
Fe0.5Co0.5C here, see Table 1 for the synthesized series of
electrodes) was directly used as working electrode with
adjustable size. The high-resolution transmission electron
microscopy (HRTEM) image (Figure 1b) revealed that the
lattice fringe of well-defined shell was 0.34 nm, matching with
the C (002) plane of graphitic carbon, whereas the interplanar
spacing of encapsulated nanoparticles was 0.20 nm, corre-
sponding to the (110) plane of FeCo alloy (Figure 1c).
Moreover, the energy dispersive X-ray spectroscopy elemen-
tal mapping in Figures 1d–f explicated that Fe (yellow) and
Co (magenta) atoms were perfectly correlated with each
other, further authenticating the existence of FeCo alloy
structure. The element map of O (red) uniformly distributed
with C (green). Furthermore, only graphitic carbon shell
(JCPDS. No. 01-1578) and FeCo alloy (JCPDS No. 65-4131)
were detected on Fe0.5Co0.5C electrode by X-ray diffraction
(XRD) measurement (Figure 1g).[8] Notably, according to the
results of X-ray photoelectron spectroscopy (XPS) measure-
ment (Figure 1h and Table 1), the estimated oxygen content
for Fe0.5Co0.5C (27.3%) was higher than that of pure carbon
aerogel (4.1%), indicating the encapsulated FeCo alloy was
conducive to the stabilization of oxygen-rich groups on
graphite carbon shell. 57Fe Mçssbauer spectroscopy is a power-
ful atom-selective technique to reveal the local structure of Fe
atoms (Figure 1i, and the related Mçssbauer parameters were
listed in Table S4). Fe0.5Co0.5C exhibited a dominant resonant-
lines sextet to be attributed to the formation of FeCo alloy
(Center shift: 0.03 mmsÀ1; H: 35.2 T), and the spectral area of
FeCo was up to 78.0%. An additional magnetically split
sextet (13.1%) attributed to FexO (CS: 0.40 mmsÀ1; H: 48.3
T) was due to the surface oxidation of partial Fe when
exposed in the air. Besides, a singlet and a quadrupole split
doublet were observed at the center of the spectrum. The
singlet was assigned to g-Fe (CS: À0.09 mmsÀ1), accounting
for 3.3%, while the doublet was ascribed to FeCx (CS:
0.50 mmsÀ1) owing to the reaction of iron and carbon
substrate during the calcination process at a high
temperature (9508C). The spectral area of FeCx was
ca. 5.0%. Based on the results, majority of Fe atoms
in Fe0.5Co0.5C electrode are coordinated with Co,
existing in the form of FeCo alloy. The graphite
shells prevent the direct contact between inner FeCo
alloy and external solution during wastewater treat-
ment, as evidenced by spin-trapping electron para-
magnetic resonance (EPR) analysis in the following
section.
Table 1: Textural properties and element contents of as-prepared electrodes.
Electrode
ICP [wt.%]
Fe
Atom ratio of Mass conc. of SBET [m2 gÀ1
]
Fe/Co
O [wt%]
Co
–
FeC
1.23Æ0.04
–
1.5
1.2
0.6
–
4.71
4.98
27.3
4.75
4.31
395
295
287
293
313
Fe0.7Co0.5
Fe0.5Co0.5
Fe0.5Co0.7
CoC
C
C
C
0.71Æ0.03 0.48Æ0.02
0.55Æ0.03 0.47Æ0.02
0.48Æ0.02 0.78Æ0.03
–
1.25Æ0.05
10376 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 10375 –10383

Angewandte
Research Articles
Chemie
AOPs Performance of Selective Electrochemical O2 Activation
generation of ROS by using 5,5-dimethyl-1-pyrroline-N-oxide
(DMPO) and 5-tert-butoxycarbonyl-5-methyl-1-pyrroline-N-
Antibiotics are extensively used for the treatment and
prevention of infectious diseases. Ciprofloxacin (CIP), a syn-
thetic third-generation fluoroquinolone antibiotic, is selected
as a model contaminant to evaluate the electrocatalytic
performance of synthesized electrodes in converting O2 to
COH. As depicted in Figure 2a, only 5.0 Æ 1.2% (the values in
this work are expressed as mean Æ standard deviation. See
equation S1 for the detailed information) CIP was degraded
with pure carbon aerogel, which was possibly due to the
electrochemical adsorption effect. 100.0% CIP removal was
astonishingly achieved in 5 min by using Fe0.5Co0.5C. Pseudo-
first-order plot for the CIP removal was obtained with the
apparent rate constant (kobs) of 1.44 Æ 0.04 minÀ1 with corre-
lation coefficient (R2) = 0.990. While the contribution of
physical adsorption and electrochemical adsorption to the
CIP removal were inappreciable 2.7 Æ 0.1% and 8.1 Æ 0.2%
in 5 min, respectively (Figure S1), indicating that the satis-
factory CIP degradation was mainly induced by heteroge-
neous catalytic reaction with Fe0.5Co0.5C cathode in the
presence of O2.
oxide (BMPO) as the trapping agents to observe the
À
characteristic peak signals of DMPO-COH and BMPO-CO2
.
In the electrocatalytic process using Fe0.5Co0.5C as cathode in
the presence of O2 (denoted as Fe0.5Co0.5C/E(O2)), a high
intensity of the particular quartet line with an intensity ratio
of 1:2:2:1 corresponding to the DMPO-COH adduct was
observed (Figure 2c).[9] No DMPO-COH signal appeared in
Fe0.5Co0.5C/E(N2) when replacing O2 with N2, indicating that
COH are originated from the electrochemical reduction of O2.
Previous studies have suggested that H2O2 can be electro-
generated via 2-electron ORR pathway over carbon-based
materials.[5a,10] The concentration of accumulated H2O2 with
pure carbon aerogel (C) was 1350 Æ 38.9 mM in 2 h, while
almost no H2O2 was detected with Fe0.5Co0.5C (Figure S2). In
order to clarify whether FeCoC is involved in the catalytic
decomposition of H2O2, a clear comparison of generating COH
was carried out by using Fe0.5Co0.5C cathode with externally-
supplied H2O2. Very weak DMPO-COH signal was observed in
Fe0.5Co0.5C/H2O2, suggesting that inner FeCo alloy was
perfectly protected by outer graphite shell to impede the
Fenton reaction. However, strong DMPO-COH signal ap-
peared in presence of externally-supplied H2O2 with applied
electric current in N2 atmosphere (Fe0.5Co0.5C/H2O2/E(N2)),
suggesting the COH generation came from electrocatalytic
reduction of H2O2. Iron-site-shielding test was further con-
ducted by employing the chelation agent 1,10-phenanthroline
in Fe0.5Co0.5C/E(O2) system to investigate whether inner FeCo
alloy was involved in Fenton reaction. There was no inhibition
for COH generation after adding 10 mmolLÀ1 1,10-phenan-
throline. However, in FeCo alloy and externally-supplied
H2O2 system, COH production from Fenton-like reaction was
completely depressed (Figure S3). Moreover, inner FeCo
alloy was tightly encapsulated by graphite shell ever after the
treatment with 0.1 molLÀ1 H2SO4 at 808C for 8 h (Figure S4),
and the DMPO-COH signals were comparable to that with
Fe0.5Co0.5C (Figure S5). Besides, no COH was produced with
pure C in a series of control experiments including C/E(O2),
C/H2O2, and C/H2O2/E(N2). Thus, it is reasonable to indicate
that, in Fe0.5Co0.5C/E(O2), O2 was efficiently reduced to COH
through electrocatalytic reduction way on the graphite carbon
shell enriched by electrons coming from inner FeCo alloy.
This is completely different from traditional Fenton reaction
with poor electron transfer initiated by reduction-/oxidation-
state cycle.
To Figure out the active sites for catalytic degradation
efficiency, EPR analysis was conducted to identify the
Radical quenching experiments were conducted to iden-
tify the active species generated during electrochemical
activation of O2, using Tert-butanol (TBA) and superoxide
dismutase (SOD) to scavenge COH and CO2À, respectively. As
exhibited in Figures 2d and S7, the addition of TBA led to
a significant decrease of kobs from 1.44 Æ 0.04 minÀ1 to 0.05 Æ
0.02 minÀ1, demonstrating that CIP was eliminated through
the surface catalytic oxidation by COH. Additionally, the CIP
degradation efficiency decreased significantly with kobs of
0.07 Æ 0.02 minÀ1 in the presence of SOD, even no BMPO-
Figure 2. a) The CIP degradation of C/E(O2) and Fe0.5Co0.5C/E(O2).
Insets: the corresponding degradation rate constant of Fe0.5Co0.5
cathode and structural models of pure carbon aerogel (up) and
C
Fe0.5Co0.5C (down). b) The cycle stability experiments within Fe0.5Co0.5
C
for the CIP degradation. c) Spin-trapping EPR spectra for detection of
COH using 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) as trapping agent
in different systems. d) Comparison of the reaction rate under different
quenching conditions. e) Comparison of kobs and J among various
electrodes (Fe0.5Co0.5C, C/Fe2+, CeO2/RGO,[11] Ce0.75Zr0.25O2/RGO,[12]
MnCo2O4-CF,[13] Fr-ErGO,[14] Ni-CF[15]) for the CIP removal in electro-
catalytic degradation processes.
À
CO2 signal was detected during EPR analysis. These results
À
manifested that O2 was firstly reduced to CO2 intermediate
via 1-electron pathway, which followed by subsequent reduc-
Angew. Chem. Int. Ed. 2021, 60, 10375 –10383
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 10377

Angewandte
Research Articles
Chemie
tion to H2O2 and then to COH. In order to verify the stability
and durability of Fe0.5Co0.5C electrode, we conducted 50 cy-
cles of degradation experiments. As expected, the electro-
catalytic performance almost remained as high as 95.5% after
50 successive runs, causing only 3.5% electrocatalytic activity
loss than the first run (99.8%) (Figure 2b). Furthermore, the
generated intermediates and possible degradation pathway of
CIP during the cyclic experiments were also investigated in
Table S5 and Figure S11. The morphology, crystalline phases
and surface oxygen groups of Fe0.5Co0.5C were well preserved
(Figures S12–15), and the negligible metal leaching of Fe and
Co ions were 0.037 Æ 0.002 and 0.050 Æ 0.001 mgLÀ1, respec-
tively with Fe0.5Co0.5C after the degradation process,
indicating its excellent catalytic stability and reusability.
There is no contribution of homogeneous Fenton-like reac-
tion with released metal ions for COH generation and CIP
removal (Figure S16).
Fe0.5Co0.5C possessed a properly high Fermi level (EF) of
encapsulated metal, leading to lower metal work function
(Wm). In general, the normalized electron transfer (DQET
)
from encapsulated metal to outer carbon layer could be
promoted with lower Wm, and thereby increased the oxygen
binding energy of outer carbon layer for obtaining abundant
oxygen-contained functional groups.
Intrinsic Mechanism of COH Generation in Oxygen
Electrochemical Reduction System
In order to investigate the intrinsic roles for converting O2
to COH in electrocatalytic process, a series of FexCoyC (x:y =
0.7:0.5, 0.5:0.5, 0.5:0.7) as well as pure FeC and CoC were
fabricated. As expected, all FeCoC electrodes exhibited
higher degradation efficiency than FeC and CoC (Fig-
ure S21). Remarkably, the highest CIP removal (100.0%)
was achieved for Fe0.5Co0.5C electrode with an optimal Fe/Co
ratio of 1.0, while the CIP removal in 5 min were 85.1 Æ 1.6%,
72.9 Æ 1.2%, 63.2 Æ 1.4% and 42.4 Æ 0.8% for Fe0.7Co0.5C,
Fe0.5Co0.7C, FeC and CoC electrodes, respectively. The X-ray
absorption near edge structure (XANES) measurement was
conducted to further explore the chemical state and coordi-
nation environment of doped metal species at the atomic
level. Taking Fe0.5Co0.5C as an example, the Fe K-edge profile
of Fe0.5Co0.5C in XANES was close to that of Fe foil, but far
from that of iron oxides, manifesting that the valence state of
Fe in Fe0.5Co0.5C was very close to 0. Simultaneously, the
absorption edge positions of Fe0.7Co0.5C and Fe0.5Co0.7C at the
Fe K-edge almost overlapped with that of Fe0.5Co0.5C,
revealing that the existence form of Fe species did not change
with the altered ratio of Fe/Co (Figure 3a). Furthermore, the
Fourier transform-extended X-ray absorption fine structure
(FT-EXAFS) spectra of FexCoyC at Fe K-edge are illustrated
in Figure S22. All FexCoyC showed a prominent peak at
2.15 Š that interpreted as a Fe-Co contribution, demonstrat-
ing that atomically dispersed Fe atom existed in the form of
Fe-Co coordination. In addition, further structural configu-
ration of FexCoyC was obtained from Raman spectra (Fig-
ure 3b). The intensity ratio (ID/IG) of the D band (1360 cmÀ1)
and G band (1600 cmÀ1) is considered as a significant
parameter that reflects the defect-site and graphitic structure
of carbon-based electrodes.[17] The values of ID/IG were 0.58
and 0.70 for FeC and CoC, respectively, indicating a high
To further evaluate the electrocatalytic performance, the
CIP removal within Fe0.5Co0.5C cathode was assessed by
comparing kobs and applied current density (J, mAcmÀ2) with
previously reported catalysts. As shown in Figure 2e and
Table S6, kobs were 0.02, 0.01, 0.03, 0.22 and 0.04 minÀ1 in
traditional electrochemical advanced oxidation processes
(EAOPs)
with
CeO2/RGO,[11]
Ce0.75Zr0.25O2/RGO,[12]
MnCo2O4-CF,[13] Fc-ErGO[14] and Ni-CF[15] as cathodes,
respectively. Moreover, kobs in homogeneous EF system (C/
Fe2+) with carbon aerogel as cathode and Fe2+ as Fenton
catalyst were 0.11 Æ 0.01 minÀ1 under the same conditions.
Remarkably, Fe0.5Co0.5C cathode had the kobs of 1.44 Æ
0.04 minÀ1 to be 7–144 times higher than other catalysts.
Additionally, the lowest current density (J = 2.5 mAcmÀ2)
was applied with Fe0.5Co0.5C to achieve high degradation
efficiency. The specific energy consumption of 0.16 kWhgÀ1
TOC was required for complete removal of CIP with
Fe0.5Co0.5C, which only accounted for 1.5–8.9% of the energy
consumption in traditional EAOPs (Figure S17). The removal
of dissolved organic carbon (DOC) and defluorination
efficiency were explored as shown in Table S7, Figures S19
and S20. Meanwhile, compared to the traditional electrodes,
the electrochemical stability is much higher with Fe0.5Co0.5C
during the wastewater purification. These advantages endow
Fe0.5Co0.5C electrode with great prospect in practical applica-
tions.
The textural properties and element contents in different
cathodes are illustrated in Table 1. The total concentration of
doped metal in series of cathodes was regulated to 0.97– degree of graphitization. By comparison, much higher values
1.31 wt.%. The Fe/Co ratio varied from 0.6–1.5 with the
increasing amount of doped Fe from 0.48 Æ 0.02 to 0.71 Æ
0.03 wt.%. The as-prepared electrodes were nominated as
FeC, Fe0.7Co0.5C, Fe0.5Co0.5C, Fe0.5Co0.7C and CoC, respective-
ly, according to the ratio of mass concentration of doped Fe
and Co. Fe0.5Co0.5C had the smaller specific surface area of
287 m2 gÀ1 than FeC, Fe0.7Co0.5C, Fe0.5Co0.7C and CoC with
395, 295, 293 and 313 m2 gÀ1, respectively, indicating that the
astonishing CIP degradation efficiency within Fe0.5Co0.5C was
not ascribed to the CIP adsorption, in great accordance with
the results in Figure S1. Noticeably, the content of surface
oxygen in Fe0.5Co0.5C was highest. According to the observa-
tion in literatures,[7a,16] when the Fe/Co ratio was 1.0,
of ID/IG for FexCoyC (0.94, 0.92, 0.93 for Fe0.7Co0.5C,
Fe0.5Co0.5C, Fe0.5Co0.7C, respectively) solidly confirmed the
formation of encapsulated FeCo alloy resulted in a relatively
lower graphitization degree. Additionally, the ideal electrical
conductivity of synthesized FexCoyC with different ID/IG was
conducted as shown in Figures 3c and S24. The conductivity
values for FeC, Fe0.7Co0.5C, Fe0.5Co0.5C, Fe0.5Co0.7C and CoC
were 2.52 Æ 0.05, 2.80 Æ 0.08, 4.19 Æ 0.12, 3.24 Æ 0.14 and
3.12 Æ 0.16 SmmÀ1, respectively. The carbon layers not only
protect the inner FeCo alloy, but also act as conductor for
electron transfer from inner alloy to out carbon shell. In
general, the electrocatalytic activity depends on how much
the carbon shell is activated by inner FeCo alloy and how
10378 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 10375 –10383

Angewandte
Research Articles
Chemie
highest among FeC (0.09 wt%), Fe0.7Co0.5C (0.29 wt%),
Fe0.5Co0.7C (0.25 wt%) and CoC (0.04 wt%) (Figure 3e).
The effect of oxygen functional groups on the electro-
catalytic performance of oxygen reduction was initially
assessed by a standard method via a rotating ring-disk
electrode (RRDE). As presented in Figure S27, both FeC
and CoC provided a low ring current output, suggesting a less
favorable 2-electron pathway with the calculated H2O2
selectivity of 19.7% and 18.4%, respectively, while the
corresponding electron transfer number were 3.67 and 3.60
(Figure S28). As an alternative, Fe0.5Co0.5C with optimal Fe/
Co ratio and abundant -COOH groups exhibited the highest
H2O2 selectivity of 70.7% with the electron transfer number
of 2.6. A positive relationship was obtained between -COOH
content and H2O2 selectivity of 2eÀ ORR (Figure 3e).
Additionally, the exposed -COOH on Fe0.5Co0.5C was re-
moved through chemical reduction with hydrazine hydrate
(N2H4·H2O), denoted as N2H4-Fe0.5Co0.5C. As expected, both
the activity and selectivity of 2eÀ ORR was obviously
declined after the elimination of -COOH group (Figure S29).
During the 3-electron pathway, the 2-electron oxygen
reductive H2O2 stepped by 1-electron electrocatalytic reduc-
tion towards to COH. It is noteworthy that Fe0.5Co0.5C cathode
with the optimal Fe/Co ratio exhibited the highest intensity of
the DMPO-COH signals (Figure 3 f). The Fe0.5Co0.5C with the
efficient ability of COH generation indeed resulted in high
degradation efficiency with highest kobs of CIP removal
(Figure S21). However, the similar intensities were detected
for all FexCoyC/H2O2/E(N2) with same external H2O2 (Fig-
ure S30), validating that the electron transfer ability was
independent from Fe/Co ratio. Notably, in the absence of
applied electric field, the DMPO-COH signals in FexCoyC/
H2O2/N2 were negligibly weak, suggesting the electrocatalytic
reduction of H2O2 via 1-electron pathway. Figure S31 clearly
illustrates why 3-electron oxygen reduction occurs for FeCoC
but not for FeCo alloy and pure carbon aerogel. Furthermore,
the electrocatalytic performance for producing COH remained
constant in pH range of 3.0–11.0 in Fe0.5Co0.5C/E(O2) (Fig-
ure S32).
Figure 4a displays the voltammograms for Fe0.5Co0.5C
electrode at rotation rates from 400 to 2500 rpm. As expected,
well-defined sigmoidal shapes are obtained. The electron
transfer number (n) during ORR is calculated to be approx-
imately 2.8 from the Koutecky–Levich (K–L) plots with
potential from À0.25 to À0.6 V vs. SCE (Figure 4b). Oper-
ando pressure test has been conducted to investigate the
fundamental mechanism of electrocatalytic oxygen reduc-
tion.[19] Figure 4c shows the O2 pressure response on ORR
cells in Na2SO4 electrolytes at pH 3. Fe0.5Co0.5C follows the
3.1 e/O2 coefficient during electrocatalytic reduction of oxy-
gen, in line with the theoretical value of 3.0 e/O2 in
Equation (6). It is clear that the pressure in the cell can
stabilize after repeated galvanostatic ORR periods, which
indicates that Fe0.5Co0.5C electrode has good stability. The
electrocatalytic reduction of H2O2 is also validated by linear
sweep voltammetry (LSV) (Figure 4d), and the reduction of
H2O2 occurs at a more positive potential than oxygen
reduction. All of these results support the overall 3-electron
oxygen reduction reaction pathway on Fe0.5Co0.5C, referring
Figure 3. a) Fe K-edge XANES of FexCoyC, standard Fe foil, Fe2O3, and
Fe3O4 samples. b) Raman spectra of FexCoyC. c) The electrical conduc-
tivity of FexCoyC. d) C-K edge NEXAFS of FeC, Fe0.5Co0.5C and CoC.
e) Comparison of selectivity of H2O2 and the content of -COOH groups
from XPS analysis for FexCoyC electrodes. f) EPR spectra for detection
of COH in FexCoyC/E(O2) systems.
much the surface is enriched with electrons coming from inner
FeCo alloy.[7a,18] The ideal conductivity of Fe0.5Co0.5C can
accelerate the electron transfer process and thus enrich the
electron density on graphite shell.
Recent studies demonstrated that the oxygen functional
groups can enhance the electrochemical reactivity during
ORR process. To investigate the “depth profile” information
about the electronic structure of the designed electrodes, the
near-edge X-ray absorption fine structure (NEXAFS) was
applied. As shown in Figure 3d, a much more intense peak
centered at 288.5 eV observed in Fe0.5Co0.5C electrode was
assigned to oxygen-containing function group (-COOH),[17b]
owing to the promoted 1s!p* resonances after the doping of
Fe and Co into the carbon matrix at the same time.
Correspondingly, a sharp and intense peak at 531.3 eV
=
ascribed to the p* of -C O also demonstrated that the
coexistence of Fe and Co caused more intense 1s!p*
resonance and thus increased -COOH (Figure S25). This
feature was found to be strengthened with co-doping of Fe
and Co and lower degree of graphitization, which was
consistent with the Raman spectral results. The identification
of oxygen-based sites was further revealed by X-ray photo-
electron spectroscopy (XPS). The O1s spectra of FexCoyC in
Figure S26 were decomposed to different oxygen groups
including C-O-C ( ꢀ 533.8 eV), C-O-M ( ꢀ 532.4 eV), -COOH
( ꢀ 531.6 eV) and O-M ( ꢀ 530.8 eV) in carbon matrix. The
surface -COOH content of 3.09 wt% for Fe0.5Co0.5C was the
Angew. Chem. Int. Ed. 2021, 60, 10375 –10383
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 10379

Angewandte
Research Articles
Chemie
( ꢀ 778.9 eV) and Co 2p1/2 ( ꢀ 794.0 eV) in CoC (Figure 5b)
corresponding to Co2+ exhibited 0.74 eV and 1.39 eV lower
shifts than Co 2p in Fe0.5Co0.5C, suggesting that the valence of
Co changed to a slightly higher Co(2+d)+, and the negative
charge was partly transferred from Co to Fe.[20]
It has been proposed that FeCo alloy in FeCoC electrode
was perfectly protected by the outer graphite shell, leading to
prominent electrocatalytic durability. In order to verify this
perspective, XANES measurement was performed for
Fe0.5Co0.5C cathode before and after the CIP degradation,
denoted as fresh/used Fe0.5Co0.5C. Interestingly, as shown in
Figure 5c (left), the line position (absorption edge) of used
Fe0.5Co0.5C, shifting slightly to lower energy than fresh
Fe0.5Co0.5C, was located between Fe foil and fresh Fe0.5Co0.5C,
suggesting the valence of iron species in used Fe0.5Co0.5C was
more negative and closer to 0. Similarly, the XANES spectra
of fresh and used Fe0.5Co0.5C at Co K-edge revealed that the
valence of Co remained the same after the elimination of
contaminants according to the completely overlapped ab-
sorption edge line of fresh and used Fe0.5Co0.5C in Figure 5c
(right). The valence of Fe was also examined with the Fe K-
edge XANES. The linear-fit curves were obtained from
standard Fe foil, Fe2O3 and Fe3O4 with different Fe oxidation
states from 0 to + 3. Noteworthily, the valence of Fe species in
used Fe0.5Co0.5C was calculated as + 0.11, which was more
negative than that of fresh Fe0.5Co0.5C (+ 0.29) (Figure 5d).
When Fe0.5Co0.5C was employed, COH can be efficiently
generated from oxygen via 3-electron pathway. During the
electrocatalytic reduction process, the graphite carbon shell
was activated and enriched by electrons from encapsulated
FeCo alloy.[21] Simultaneously, the electrons from external
circuit would be fast provided to inner FeCo alloy through
carbon framework.
As revealed in the above experimental results, abundant
-COOH functional groups were formed on the surface of
Fe0.5Co0.5C. To explore the electrocatalytic mechanism of O2
reduction, density functional theory (DFT) calculations based
on a three-dimensional graphite sheet with FeCoC were
carried out (Figures 6a–e). The models of FeCo alloy
encapsulated graphene sheet structure with -COOH and C-
O-C groups are exhibited in Figure S33. The models of
different ORR processes on FeCoC electrode are exhibited in
Figure S34. The computational details were provided in the
Supporting Information. The 2-electron ORR proceeds via 1-
electron oxygen reduction to COOH [Eq. (7)] and then
subsequent 1-electron reduction of COOH to H2O2 [Eq. (8)].
Thus, the adsorption binding energy of COOH is the key
descriptor for controlling and modulating the electrocatalytic
activity in 2-electron ORR process.[22]
Figure 4. a) Linear-sweep voltammograms of Fe0.5Co0.5C measured on
a rotating disk electrode at different rotation speeds at pH 3. b) The
Koutecky–Levich curves at various potentials. c) Operando pressure
measurement of Fe0.5Co0.5C during the electrocatalytic reduction of
oxygen at pH 3. d) Steady-state voltammograms of electrocatalytic
oxygen reduction at pH 3 on Fe0.5Co0.5C before (black c) and after
the addition of 4”10À3 molLÀ1 H2O2 (red a).
to the 2-electron oxygen reductive H2O2 intermediate simul-
taneously stepped by 1-electron electrocatalytic reduction
toward COH.
The insight of the efficient electrocatalytic performance of
Fe0.5Co0.5C was further investigated. NEXAFS spectra of Fe
and Co L-edge provided direct evidence of charge transfer
between Fe and Co. As shown in Figure 5a, the Fe L-edge
binding energies of Fe0.5Co0.5C at 708.1 and 720.5 eV corre-
sponding to Fe 2p3/2 and Fe 2p1/2 of Fe2+ exhibited 0.53 and
1.42 eV shifts to lower compared to FeC with 708.6 and
721.9 eV, respectively, indicating less positive charge of Fe2+:
Fe(2Àd)+ (d, close to 0). The binding energies of Co 2p3/2
eÀ
!
OOH
7
8
C
O2
H
C
OOH
H
eÀ ! H2O2
The *OOH adsorption energies were calculated for
FeCoC respectively with -COOH and C-O-C model struc-
tures. The obtained results were summarized in Figure 6a in
the form of free energy diagram at the equilibrium potential
Figure 5. a) Fe L-edge NEXAFS of Fe0.5Co0.5C and FeC. b) Co L-edge
NEXAFS of Fe0.5Co0.5C and CoC. c) Fe K-edge and Co K-edge XANES of
fresh and used Fe0.5Co0.5C, standard Fe foil, Fe2O3 and Fe3O4 samples.
d) Linear fitting curve of fresh and used Fe0.5Co0.5C, and reference
materials derived from the corresponding Fe K-edge XANES spectra.
(U0
= 0.7 V). An ideal catalyst should have a flat free
O2=H2O2
10380 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 10375 –10383

Angewandte
Research Articles
Chemie
and (13), the kinetic inclination of COH reduced from *H2O2
could be acquired via the actual barriers of forming COH and
OHÀ under a certain potential. Therefore, we have deter-
mined the eventually generated species by comparing the
calculated free energies of producing COH and OHÀ from
*H2O2 according to the thermodynamic formulas [Eqs. (16)
and (17), see Supporting Information for more details]. As
shown in Figure 6 f, DG for COH generation process [Eqs. (12)
and (13), DG(COH) = À3.02 eV] is much lower than DG-
(H2O2) desorption process [Eq. (14), DG(H2O2) = À0.91 eV],
indicating that the former pathway is more thermodynami-
cally-favorable. Accordingly, the 3-electron pathway can be
expressed as Equation (15). These results of theoretical
calculations are consistent with COH simultaneously and
generously released from the surface active sites in
Fe0.5Co0.5C/E(O2) system [Eq. (14)].
*
*
*
*
*
OH
H2O2
OH
eÀ
!
OHÀ
12
13
14
15
Figure 6. a) Free energy diagram for O2 reduction to H2O2. Typical
noble metal catalysts (Au, Pt and Pd) were compared with the as-
prepared Fe0.5Co0.5C with -COOH and C-O-C groups. Top and side
views of COOH-adsorbed configurations on FeCoC with b,d) -COOH
and c,e) C-O-C groups, respectively. f) The thermodynamic activity of
O2 reduction over FeCoC with -COOH.
OHÀ
!
OH
OHÀ
*
C
*
H2O2
!
H2O2
3 eÀ
O2
2 H
!
OH
OHÀ
*
C
where the asterisk (*) denotes the surface active site.
energy diagram at this potential.[23] Both stronger and weaker
binding to *OOH would bring additional overpotential,
causing a bottleneck for the reduction of *OOH to H2O2.[24]
Notably, for Fe0.5Co0.5C with -COOH it was 0.13 eV downhill
DGIV DG OHÀ
G OHÀ À G H2O2
eU
C
G OH
16
17
C
DGV
DGVI DG OH
G OHÀ
1:9 eV À eU
to form *OOH at U0
, suggesting that the lowest
O2=H2O2
where G(COH), G(OHÀ), and G(H2O2) refer to free energies
of the generated COH, OHÀ, and H2O2, respectively. U is the
electrode potential vs. SCE.
thermodynamic overpotential of 0.13 V was required to drive
the reaction. However, catalysts such as Au,[25] Pt,[25] Pd,[10a]
CN[10c] respectively with higher thermodynamic overpotential
of 0.39, 0.40, 0.16, 0.54 V were not sufficiently active to
motivate the 2-electron ORR.
The free energy diagrams of selective electrochemical
reduction of O2 to COH via 3-electron pathway over Conclusion
Fe0.5Co0.5C with surface -COOH were exhibited in Figure 6 f.
During 3-electron pathway, O2 is firstly adsorbed on the
surface active sites (*) of Fe0.5Co0.5C [Eq. (9)], then *O2 is
reduced and protonated to COOH as the intermediate
[Eq. (10)], and H2O2 is subsequently generated and chemi-
cally bound with * to form *H2O2 [Eq. (11)]. The correspond-
ing Gibbs free energy differences of these three steps were
DGI (1.18 eV), DGII (0.57 eV) and DGIII (À0.58 eV), indicat-
ing that O2 adsorption is speculated as the rate-limiting step
during the whole electrocatalytic process, and a good O2
adsorption capability was benefit for 2-electron ORR.
In summary, a new strategy of selective electrochemical
reduction of O2 to COH via 3-electron pathway is demon-
strated over integral FeCoC electrode. Surface -COOH
groups on graphite shell act as the key role to induce the
formation of H2O2 via the 2-electron ORR. Simultaneously,
H2O2 intermediate would be electrocatalytic reduced to COH
via 1-electron on the graphite shell enriched with electrons
coming from inner FeCo alloy. This is completely different
from traditional Fenton reaction. The proposed strategy
overcomes the restriction of rate-limiting step for reducing
oxidized metal ions and requirement of acidic pH condition.
100.0% removal of CIP is obtained in 5 min with kobs of 1.44 Æ
0.04 minÀ1. The high degradation efficiency of FeCoC/E(O2)
system maintains even after 50 cycles. This work provides
a novel strategy for forming COH from electrochemical
reducing O2 without using foreign reagents, which represents
a fundamental breakthrough towards the generation of COH
in advanced oxidation processes.
*
*
*
*
O2
O2
!
9
10
11
*
C
O2
2 H
eÀ
!
OOH
H
eÀ
!
H2O2
*
C
OOH
H
Once H2O2 is formed via 2-electron pathway, exploring
whether the generated H2O2 is desorbed from the surface of
Fe0.5Co0.5C or further reduced to produce COH and OHÀ via 1-
electron pathway is necessary.[26] According to Equations (12)
Angew. Chem. Int. Ed. 2021, 60, 10375 –10383
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 10381

Angewandte
Research Articles
Chemie
Zhang, X. Bao, C. Li, Angew. Chem. Int. Ed. 2014, 53, 7023 –
7027; Angew. Chem. 2014, 126, 7143 –7147.
Acknowledgements
[7] a) D. Deng, L. Yu, X. Chen, G. Wang, L. Jin, X. Pan, J. Deng, G.
Sun, X. Bao, Angew. Chem. Int. Ed. 2013, 52, 371 –375; Angew.
Chem. 2013, 125, 389 –393; b) H. Su, H. Feng, Q. Zhao, X.
Zhang, J. Sun, Y. He, S. Huang, T. Huang, J. Zhong, D. Wu, B.
Ren, J. Am. Chem. Soc. 2020, 142, 1341 –1347.
[8] X. Zhu, T. Jin, C. Tian, C. Lu, X. Liu, M. Zeng, X. Zhuang, S.
Yang, L. He, H. Liu, S. Dai, Adv. Mater. 2017, 29, 1704091.
[9] L. Lyu, D. Yan, G. Yu, W. Cao, C. Hu, Environ. Sci. Technol.
2018, 52, 4294 –4304.
[10] a) Q. Chang, P. Zhang, A. H. B. Mostaghimi, X. Zhao, S. R.
Denny, J. H. Lee, H. Gao, Y. Zhang, H. Xin, S. Siahrostami, J.
Chen, Z. Chen, Nat. Commun. 2020, 11, 2178; b) Z. Lu, G. Chen,
S. Siahrostami, Z. Chen, K. Liu, J. Xie, L. Liao, T. Wu, D. Lin, Y.
Liu, T. F. Jaramillo, J. K. Norskov, Y. Cui, Nat. Catal. 2018, 1,
156 –162; c) S. Chen, Z. Chen, S. Siahrostami, D. Higgins, D.
Nordlund, D. Sokaras, T. R. Kim, Y. Liu, X. Yan, E. Nilsson, R.
Sinclair, J. K. Norskov, T. F. Jaramillo, Z. Bao, J. Am. Chem. Soc.
2018, 140, 7851 –7859; d) C. W. B. Bezerra, L. Zhang, K. Lee, H.
Liu, A. L. B. Marques, E. P. Marques, H. Wang, J. Zhang,
Electrochim. Acta 2008, 53, 4937 –4951.
The authors acknowledge funding from the National Natural
Science Foundation P. R. China (Project No. 22076142,
21537003, 21677106, 22076140), National Key Basic Research
Program of China (2017YFA0403402), National Natural
Science Foundation of China (U1932119), the Science &
Technology Commission of Shanghai Municipality
(14DZ2261100), the Fundamental Research Funds for the
Central Universities. The authors thank the staff of the 11B
beamline at the Shanghai Synchrotron Radiation Facility
(SSRF) for data collection. This work was also supported by
Shanghai Large Scientific Facilities Center.
Conflict of interest
The authors declare no conflict of interest.
Keywords: carbon aerogel ·FeCo alloy ·hydroxyl radical ·
oxygen reduction reaction ·water purification
[11] Y. Li, J. Han, B. Xie, Y. Li, S. Zhan, Y. Tian, J. Electroanal. Chem.
2017, 784, 6 –12.
[12] Y. Li, Y. Li, B. Xie, J. Han, S. Zhan, Y. Tian, Environ. Sci. Nano
2017, 4, 425 –436.
[13] X. Mi, Y. Li, X. Ning, J. Jia, H. Wang, Y. Xia, Y. Sun, S. Zhan,
Chem. Eng. J. 2019, 358, 299 –309.
[14] G. Divyapriya, I. Nambi, J. Senthilnathan, Chemosphere 2018,
209, 113 –123.
[15] X. Liu, W. Li, Y. Wang, G. Zhou, Y. Wang, C. He, G. Wang, Y.
Mu, Environ. Sci. Technol. 2019, 53, 6927 –6936.
[16] a) H. Feng, J. Yu, L. Tang, G. Zeng, W. Tang, J. Wang, T. Luo, B.
Peng, B. Song, L. Wang, C. Liang, Environ. Sci. Technol. 2019,
53, 13878 –13887; b) C. H. Choi, H. K. Lim, M. W. Chung, J. C.
Park, H. Shin, H. Kim, S. I. Woo, J. Am. Chem. Soc. 2014, 136,
9070 –9077; c) Y. Ha, S. Kang, K. Ham, J. Lee, H. Kim, J. Phys.
Chem. Lett. 2019, 10, 3109 –3114; d) Q. Mo, W. Zhang, L. He, X.
Yu, Q. Gao, Appl. Catal. B 2019, 244, 620 –627; e) W. Ran, J.
Dong, T. Sun, J. Chen, L. Xu, ChemistrySelect 2019, 4, 7728 –
7733.
[17] a) Y. Dong, Q. Zhang, Z. Tian, B. Li, W. Yan, S. Wang, K. Jiang,
J. Su, C. W. Oloman, E. L. Gyenge, R. Ge, Z. Lu, X. Ji, L. Chen,
Adv. Mater. 2020, 32, 2001300; b) W. Wang, L. Shang, G. Chang,
C. Yan, R. Shi, Y. Zhao, G. I. N. Waterhouse, D. Yang, T. Zhang,
Adv. Mater. 2019, 31, 1808276.
[1] a) H. Kim, J. Lim, S. Lee, H. H. Kim, C. Lee, J. Lee, W. Choi,
Environ. Sci. Technol. 2019, 53, 2918 –2925; b) X. Shen, F. Xiao,
H. Zhao, Y. Chen, C. Fang, R. Xiao, W. Chu, G. Zhao, Environ.
Sci. Technol. 2020, 54, 4564 –4572; c) P. Cao, X. Quan, K. Zhao,
S. Chen, H. Yu, Y. Su, Environ. Sci. Technol. 2020, 54,
12662 –12672; d) Y. Chen, G. Zhang, H. Liu, J. Qu, Angew.
Chem. Int. Ed. 2019, 58, 8134 –8138; Angew. Chem. 2019, 131,
8218 –8222.
[2] a) Y. Wang, R. Shi, L. Shang, G. Waterhouse, J. Zhao, Q. Zhang,
L. Gu, T. Zhang, Angew. Chem. Int. Ed. 2020, 59, 13057 –13062;
Angew. Chem. 2020, 132, 13157 –13162; b) M. Wang, N. Zhang,
Y. Feng, Z. Hu, Q. Shao, X. Huang, Angew. Chem. Int. Ed. 2020,
59, 14373 –14377; Angew. Chem. 2020, 132, 14479 –14483; c) Y.
Wang, Y. Lai, L. Song, Z. Zhou, J. Liu, Q. Wang, X. Yang, C.
Chen, W. Shi, Y. Zheng, M. Rauf, S. Sun, Angew. Chem. Int. Ed.
2015, 54, 9907 –9910; Angew. Chem. 2015, 127, 10045 –10048.
[3] a) M. Melchionna, P. Fornasiero, M. Prato, Adv. Mater. 2019, 31,
1802920; b) C. Wang, Y. Gu, S. Wu, H. Yu, S. Chen, Y. Su, Y.
Guo, X. Wang, H. Chen, W. Kang, X. Quan, Environ. Sci.
Technol. 2020, 54, 1920 –1928.
[4] a) H. W. Kim, M. B. Ross, N. Kornienko, L. Zhang, J. Guo, P.
Yang, B. D. McCloskey, Nat. Catal. 2018, 1, 282 –290; b) Y. Liu,
X. Quan, X. Fan, H. Wang, S. Chen, Angew. Chem. Int. Ed. 2015,
54, 6837 –6841; Angew. Chem. 2015, 127, 6941 –6945; c) Y. J. Sa,
J. H. Kim, S. H. Joo, Angew. Chem. Int. Ed. 2019, 58, 1100 –1105;
Angew. Chem. 2019, 131, 1112 –1117.
[5] a) Y. Jiang, P. Ni, C. Chen, Y. Lu, P. Yang, B. Kong, A. Fisher, X.
Wang, Adv. Energy Mater. 2018, 8, 2307; b) W. Jiang, X. Xia, J.
Han, Y. Ding, M. R. Haider, A. Wang, Environ. Sci. Technol.
2018, 52, 9972 –9982; c) Y. Chen, H. Chen, J. Li, L. Xiao, Water
Res. 2019, 152, 181 –190; d) K. Liu, J. Yu, H. Dong, J. C. S. Wu,
M. R. Hoffmann, Environ. Sci. Technol. 2018, 52, 12667 –12674;
e) J. Ji, Q. Yan, P. Yin, S. Mine, M. Matsuoka, M. Xing, Angew.
Chem. Int. Ed. 2021, 60, 2903 –2908; Angew. Chem. 2021, 133,
2939 –2944.
[18] L. Yu, D. Deng, X. Bao, Angew. Chem. Int. Ed. 2020, 59, 15294 –
15297; Angew. Chem. 2020, 132, 15406 –15409.
[19] a) K. Chen, G. Huang, J. Ma, J. Wang, D. Yang, X. Yang, Y. Yu,
X. Zhang, Angew. Chem. Int. Ed. 2020, 59, 16661 –16667; Angew.
Chem. 2020, 132, 16804 –16810; b) T. Liu, Z. Liu, G. Kim, J. T.
Frith, N. Garcia-Araez, C. P. Grey, Angew. Chem. Int. Ed. 2017,
56, 16057 –16062; Angew. Chem. 2017, 129, 16273 –16278.
[20] J. Yang, Y. Xiao, Q. Zhao, G. Zhang, R. Wang, G. Teng, X. Chen,
M. Weng, D. He, S. Mu, Y. Lin, F. Pan, Nano Energy 2019, 59,
443 –452.
[21] X. Li, X. Huang, S. Xi, S. Miao, J. Ding, W. Cai, S. Liu, X. Yang,
H. Yang, J. Gao, J. Wang, Y. Huang, T. Zhang, B. Liu, J. Am.
Chem. Soc. 2018, 140, 12469 –12475.
[22] C. Tang, Y. Jiao, B. Shi, J. Liu, Z. Xie, X. Chen, Q. Zhang, S.
Qiao, Angew. Chem. Int. Ed. 2020, 59, 9171 –9176; Angew.
Chem. 2020, 132, 9256 –9261.
[6] a) X. Liu, L. Wang, P. Yu, C. Tian, F. Sun, J. Ma, W. Li, H. Fu,
Angew. Chem. Int. Ed. 2018, 57, 16166 –16170; Angew. Chem.
2018, 130, 16398 –16402; b) P. Ren, D. Deng, X. Bao, Angew.
Chem. Int. Ed. 2015, 54, 2100 –2104; Angew. Chem. 2015, 127,
2128 –2132; c) X. Zheng, J. Deng, N. Wang, D. Deng, W. H.
[23] Q. Zhang, X. Tan, N. M. Bedford, Z. Han, L. Thomsen, S. Smith,
R. Amal, X. Lu, Nat. Commun. 2020, 11, 4181.
[24] X. Song, N. Li, H. Zhang, L. Wang, Y. Yan, H. Wang, L. Wang, Z.
Bian, ACS Appl. Mater. Interfaces 2020, 12, 17519 –17527.
10382 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 10375 –10383

Angewandte
Research Articles
Chemie
[25] S. Siahrostami, A. Verdaguer-Casadevall, M. Karamad, D.
Deiana, P. Malacrida, B. Wickman, M. Escudero-Escribano,
E. A. Paoli, R. Frydendal, T. W. Hansen, I. Chorkendorff, I. E.
Stephens, J. Rossmeisl, Nat. Mater. 2013, 12, 1137 –1143.
[26] J. M. Nol, A. Latus, C. Lagrost, E. Volanschi, P. Hapiot, J. Am.
Chem. Soc. 2012, 134, 2835 –2841.
Manuscript received: February 4, 2021
Accepted manuscript online: February 19, 2021
Version of record online: March 22, 2021
Angew. Chem. Int. Ed. 2021, 60, 10375 –10383
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 10383