An iron-nitrogen doped carbon and CdS hybrid catalytic system for efficient CO2photochemical reduction

Article abstract of DOI:10.1039/d0cc07692a

Iron porphyrin and carbon black (CB) were utilized to fabricate an iron-nitrogen doped carbon (Fe-N-C) catalyst to create a new heterogeneous catalytic system with CdS to drive CO₂reduction to CO under UV/vis light (AM 1.5G) irradiation. The system delivers a high CO production yield of 111 mmol g_cat^?1and a large turnover number (TON) of 1.22 × 10³in 8 h with a selectivity of 85%, all of which are competitive with state-of-the-art systems. The mechanism of the system was investigated by experimental and theoretical methods indicating that the high affinity between the iron active center and the *COOH intermediate facilitates the brilliant catalytic performance. This work provides a new direction for constructing heterogeneous CO₂photoreduction systems.

Full text of DOI:10.1039/d0cc07692a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
An iron–nitrogen doped carbon and CdS hybrid
catalytic system for efficient CO2 photochemical
reduction†
Cite this: Chem. Commun., 2021,
57, 2033
Received 24th November 2020,
Accepted 25th January 2021
Jin-Han Guo, Xiao-Yao Dao and Wei-Yin Sun
*
DOI: 10.1039/d0cc07692a
rsc.li/chemcomm
Iron porphyrin and carbon black (CB) were utilized to fabricate an catalytic performance.2 Combining the advantages of both mole-
iron–nitrogen doped carbon (Fe–N–C) catalyst to create a new cular and heterogeneous catalysts, a family of metal–nitrogen doped
heterogeneous catalytic system with CdS to drive CO2 reduction to carbon (M–N–C) catalysts such as Fe–N–C12–14 and Ni–N–C15–17 are
CO under UV/vis light (AM 1.5G) irradiation. The system delivers a promising for CO2 reduction. They are effective and selective (up to
high CO production yield of 111 mmol gcat and a large turnover B99%) for CO2 reduction even in aqueous solution.12,17 Also, as
À1
number (TON) of 1.22 Â 103 in 8 h with a selectivity of 85%, all of heterogeneous catalysts with high stability, their large-scale applica-
which are competitive with state-of-the-art systems. The mechanism tion is a certain prospect.15
of the system was investigated by experimental and theoretical
Among M–N–C catalysts fabricated by various precursors,
methods indicating that the high affinity between the iron active zeolitic imidazolate frameworks (ZIFs) derived Fe–N–C catalysts
center and the *COOH intermediate facilitates the brilliant catalytic exhibit outstanding electrocatalytic performance due to their
performance. This work provides a new direction for constructing pyrrolic N–Fe structure.12,18 In this work, iron porphyrin and
heterogeneous CO2 photoreduction systems.
carbon black (CB) were utilized to fabricate iron–nitrogen
doped carbon catalysts based on the following considerations.
Photocatalytic conversion of CO2 into gaseous fuel precursors Metalloporphyrins,7,8 a series of macrocyclic complexes with
or liquid fuels is a promising strategy to reduce CO2 emission to inherent M–N4–C pyrrolic nitrogen coordinated structures, are
alleviate the reverse effect of global warming.1 A photocatalytic desired precursors for the M–N–C catalysts.19,20 In addition,
system needs to accomplish two tasks: first, the photosensitizer commercially available CB was chosen due to its porosity and
absorbs the sunlight and generates activated electrons; second, conductivity that ensure mass transfer and charge mobility during
the catalyst transfers the activated electrons into the CO2 the catalysis.15,19,21 Incorporating metal meso-tetraphenyl-
molecule.2 Although in some systems, these two tasks could porphyrin (TPPM, M = FeCl and Ni, Fig. S1, ESI†) on the surface
be accomplished by one species,3,4 the hybrid systems of CB and pyrolyzing at 800 1C, M–N–C catalysts with surface
composed of photosensitizers and catalysts are more favorable supported active centers can be easily achieved, denoting as
due to their high efficiency and diversity.2
M–CBs (M = Fe and Ni, see Fig. 1a and ESI† for details). We
Many catalysts have been developed for effective CO2 reduction optimized the amount of iron porphyrin utilizing different weight
by photo- and electrocatalysis. Molecular catalysts are mainly metal content of TPPFeCl. The results of linear scan voltammetry (LSV)
coordination compounds fabricated with nitrogen-containing and partial CO current density ( jCO) at À0.9 V vs. SHE in the CO2
ligands, such as polypyridine,5 phthalocyanine6 and porphyrin.7,8 saturated 0.5 M KHCO3 aqueous solution show that 20% weight
These catalysts are productive and selective for CO2 reduction. content of the TPPFeCl is the best (Fig. S2, ESI†). Thus, we
However, the stability issues and the difficulties for separation and selected 20% as the weight ratio of iron porphyrin in the
reuse prevent their large-scale application.6,9 On the other hand, preparation of Fe–CB, which is the same as Ni–CB.17
heterogeneous catalysts such as semiconductors,2 silver10 and
Although promising for photochemical CO2 reduction,
gold11 are stable and easy for separation and reuse with lower investigations on M–N–C catalysts in photocatalysis are
limited.2,13,14 The ZIFs-derived Fe–N–C catalysts were inte-
Coordination Chemistry Institute, State Key Laboratory of Coordination Chemistry,
grated with a noble metal-based molecular photosensitizer
[Ru(bpy)3]2+ in acetonitrile/water mixed solvent to produce CO
School of Chemistry and Chemical Engineering, Nanjing National Laboratory of
Microstructures, Collaborative Innovation Center of Advanced Microstructures,
under light irradiation.13,14 However, a cheap, stable and water-
Nanjing University, Nanjing 210023, China. E-mail: sunwy@nju.edu.cn
compatible photosensitizer is preferable to be integrated with the
M–CB catalysts to avoid the usage of noble-metal-containing
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
d0cc07692a
This journal is The Royal Society of Chemistry 2021
Chem. Commun., 2021, 57, 2033À2036 | 2033

View Article Online
Communication
ChemComm
Fig. 2 (a) Mott–Schottky plot of CdS. (b) Total current density, jCO, and
FECO at varied applied potentials around À0.9 V vs. SHE for Fe–CB and
Ni–CB. (c) EIS Nyquist plot of the catalysts at À0.9 V vs. SHE. (d) PL spectra
of CdS and the mixture of CdS and Fe–CB. (e) Photocurrent measurements
of the catalytic systems. (f) Proposed mechanism of the Fe–CB + CdS
hybrid system for CO2 photochemical reduction.
Fig. 1 (a) Schematic illustration of preparation process for Fe–CB (hydro-
gen atoms on the porphyrin molecules were omitted for clarity). (b) Typical
SEM image of Fe–CB. (c) Fe 2p region XPS spectra of Fe–CB and TPPFeCl.
(d) Core level XPS spectra of N 1s region of Fe–CB.
photosensitizer and organic solvents. Also, the heterogeneous
photosensitizers are advantageous in the application due to easy
separation and reuse. CdS is one of the most widely applied
heterogeneous photosensitizers due to its excellent photochemical
properties despite its toxicity.2,22,23 The conduction band of CdS
(ECB, BÀ0.9 V vs. SHE, pH = 7.0) is negative enough to promote the
reduction of CO2 to CO (À0.53 V vs. SHE, pH = 7.0).2,24,25 Thus, we
intended to construct new heterogeneous catalytic systems by using
M–CB catalysts with CdS as photosensitizer.
guarantees Fe–CB’s efficient catalytic performance. Besides, PXRD
and SEM inspection of CdS verified the successful synthesis of CdS
following a facile procedure with a ball-like morphology and a
particle size of B200 nm (Fig. S7, ESI†).22,26
To validate the viability of our hybrid system, we need to
ensure that the photogenerated electrons can drive the reduction
of CO2 on the catalyst while the holes can be quenched effectively
by the sacrificial agent.2 First, the flat-band potential (Efb, corres-
Ni–CB was characterized in our previous report,17 and here we
describe the characterization of Fe–CB. The scanning electron
microscope (SEM) image shows that Fe–CB consists of balls cross-
linked with each other to form a 3D porous structure (Fig. 1c and
Fig. S3, ESI†). The surface of Fe–CB is rough and guarantees a
high specific surface area of 100.2 m2 gÀ1 (Fig. S4, ESI†). The powder
X-ray diffraction (PXRD) pattern of Fe–CB only exhibits two pro-
nounced graphite diffraction peaks, while no metallic phase was
detected (Fig. S5, ESI†). X-ray photoelectron spectroscopy (XPS) was
utilized to determine the chemical composition and elemental states
of Fe–CB catalyst (Fig. 1c, d and Fig. S6, ESI†). The XPS survey
spectrum reveals an iron atomic content of 0.30 at% (1.3 wt%).
Compared with the lower weight content of iron measured by
inductively coupled plasma mass spectrum (ICP-MS) (0.51 wt%),
most of the Fe atoms exist on the surface of the catalyst, ensuring
the availability of the active centers during the catalysis.17 In
addition, the binding energy of Fe 2p3/2 in Fe–CB catalysts is
B710.4 eV (Fig. 1c), which is B0.6 eV lower than that in the
TPPFeCl (B711.0 eV), verifying that the iron was partially reduced
and exists in a valence state similar with that in the previously
reported Fe–N–C catalysts.12,18 The high-resolution XPS N 1s spec-
trum of Fe–CB catalyst was deconvoluted into pyrrolic N–Fe, pyr-
idinic N–Fe, pyrrolic N and graphite N species.12 Besides, no
pyridinic N was detected. The pyrrolic N–Fe (43.8 at%) and pyridinic
N–Fe species (14.3 at%) were confirmed in Fe–CB. Compared with
the higher content of pyrrolic N–Fe structure (68 at%) and the lower
content of pyridinic N–Fe species (2 at%) in the reported ZIF-derived
Fe3+–N–C catalyst,12 the pyrrolic N–Fe structure in iron porphyrin
precursor was partially transformed to pyridinic N–Fe structure in
the pyrolysis process. The high content of pyrrolic N–Fe species
27
ponding to the ECB
)
of CdS is evaluated by the Mott–Shottcky
test to be À0.9 V vs. SHE (Fig. 2a), indicating that catalysts need to
have an onset potential for CO2 reduction of less negative than
À0.9 V vs. SHE.5,27 Thus, we measured the electrocatalytic proper-
ties of the M–CBs in a CO2 saturated 0.5 M KHCO3 aqueous
solution (pH = 7.3). The catalysts’ onset potentials for CO2 to CO
conversion were determined to be À0.70 and À0.82 V vs. SHE for
Fe–CB and Ni–CB, respectively,17 verified that both the catalysts
are qualified to accomplish the hybrid system. Besides, compared
with Ni–CB, Fe–CB delivers much higher partial current densities
( jCO) and faradaic efficiencies (FECO) for CO (Fig. 2b), indicating
that Fe–CB could be a more efficient catalyst under the excitement
of CdS. In addition, the high FECO of 97% at 0.35 V overpotential
on Fe–CB is competitive with the ZIF-derived Fe–N–C catalysts
while outperformed other Fe–N–C materials (Table S1, ESI†),
further illustrated the advantage of the porphyrin precursor used
in the synthetic procedure.12 Also, the electrochemical impedance
spectra (EIS) indicated the resistance of charge transfer (Rct)
between the Fe–CB and substrates is much smaller than Ni–CB
(Fig. 2c). Also, we tested the photoluminescence (PL) spectra of
the CdS and Fe–CB + CdS mixture to investigate the electron
transfer between the CdS and Fe–CB. In Fig. 2d, the individual
CdS exhibits an emission peak at 537 nm wavelength under
420 nm excitation, while the Fe–CB + CdS composite sample
displays a sharp decrease of PL emission, implying photogener-
ated electron transfer between Fe–CB and CdS.27,28 In addition,
we performed time-resolved PL decay spectra to quantify the
average carrier lifetimes (Fig. S8, ESI†). The fitting results revealed
that the average lifetime of Fe–CB + CdS and CdS are 2.57 and
1.87 ns, respectively. The results indicated that the hybrid system
2034 | Chem. Commun., 2021, 57, 2033À2036
This journal is The Royal Society of Chemistry 2021

View Article Online
ChemComm
Communication
can effectively prevent the photoexcited charge carriers from
For comparison, Ni–CB catalyst and commercial CB were
recombining, which displays a photogenerated electron transfer also tested as the catalysts in the system, and the photocatalytic
between the two components.27,28
Next, we turned to explore the photo-oxidation part of the system. Fig. S12, ESI†). As depicted in Fig. 3b, only 9.5 mmol gcat of
The valence band (EVB) of CdS was estimated using a UV-vis diffuse- CO was produced when 1 mg Ni–CB is utilized as the catalyst
performance of bare CdS was also evaluated (Fig. 3bÀa1nd
reflectance spectroscopy (UV-vis-DRS). CdS has an absorption in the during the 8 h process, corresponding to a TON of 97 and a
À1
visible light region (up to B580 nm), suggesting the feasibility for production rate of only 1.2 mmol gcat
h
À1. Nevertheless, the
solar-light utilization (Fig. S9a, ESI†). The Tauc-plot derived from the H2 yield reached 20 mmol gcatÀ1, which led to a low CO
UV-vis-DRS indicates that CdS has a band-gap (Eg) of 2.45 eV selectivity of 32%. Besides, when utilizing 1 mg of CB in
(Fig. S9b, ESI†). Combining the Mott–Schottky measurement the replacement of Fe–CB, the performance was even worse,
À1
(Fig. 2a), the EVB was determined to be 1.55 V vs. SHE. Furthermore, delivered only 2.0 mmol gcatÀ1 of CO together with 21 mmol gcat
photocurrent tests were conducted to measure the photo-oxidation of H2, which is slightly better than the performance of bare CdS due
of the hybrid system in the presence of triethanolamine (TEOA), a to the better charge mobility in the CB.22
harmless and efficient sacrificial donor commonly used for
The origination of photochemical produced CO was con-
photochemical processes.4,27 As illustrated in Fig. 2e, pure CdS firmed by a labelling experiment with 13CO2 in the absence of
photosensitizer can photo-oxidize TEOA efficiently under solar- KHCO3 (Fig. S13, ESI†).17 Also, no CO was detected when the
light illumination, delivered a high photocurrent density of reaction is conducted under a N2 atmosphere. Besides, no CO
B1.4 mA cmÀ2. Besides, the mixture of CdS and Fe–CB also shows and H2 were detected under dark conditions or in the absence
prominent photocurrent density slightly lower than that of pure of CdS, which emphasized the vital function of the CdS photo-
CdS, which might be contributing to the low photocurrent response sensitizer. More importantly, no apparent deactivation was
of the CB based materials. This is illustrated in the photocurrent test observed after three cycles of photocatalysis (Fig. S14, ESI†).
of pure Fe–CB, which exhibits negligible photocurrent compared The excellent stability is ascribed to the inherent structural
with CdS containing samples (Fig. 2e). All the above results ensure stability of the Fe–N–C structure and CdS in the presence of
that the hybrid system composed of M–CB, CdS and TEOA TEOA,22 while the high flat-band potential of CdS is protective
is capable of CO2 photochemical reduction, and a proposed mecha- for the catalyst from deactivation.12,16
nism is depicted in Fig. 2f: CdS is excited by light irradiation to
generate an electron–hole pair, the generated electrons transfer to mance over Ni–CB, we performed DFT simulations to draw the
the catalyst to reduce CO2 while TEOA quenches the holes. catalytic pathways (Table S3 and Fig. S15–S17, ESI†). As exhibited in
To investigate the origin of Fe–CB’s superior catalytic perfor-
Photoreduction of CO2 catalyzed by Fe–CB was performed Fig. 4a, the energy diagrams for CO evolution illustrate that the first
using a suspension of 1 mg of Fe–CB and 10 mg CdS dispersed electron and proton transfer process generating the *COOH species
in an aqueous solution of 10 wt% TEOA and 0.5 M KHCO3. The is the rate-determining step for the evolution of CO. Only 0.63 eV of
time-dependent yields of CO and H2 after 8 h of UV/vis light energy barrier is required to reduce CO2 to *COOH on the Fe–N–C.
(AM 1.5G) irradiation are shown in Fig. 3a, CO was the primary In comparison, the formation of *COOH on the Ni–N–C requires a
product in the gas phase with a yield of 111 mmol gcatÀ1 and a high energy barrier of 1.82 eV, which retards the catalytic process.
TON of 1.22 Â 103 based on the Fe–CB catalyst. The CO Thus Fe–CB has better activity over Ni–CB. The energy barriers for
production rate was as high as 13.9 mmol gcatÀ1 hÀ1. H2 was alÀso1
observed in the catalytic system with a yield of 19 mmol gcat
(Fig. S10, ESI†). Besides, the 1H NMR spectrum confirmed that no
formic acid was produced (Fig. S11, ESI†), revealing a high
selectivity for CO of 85% of the system during the 8 h catalytic
process. This highly efficient CO2 reduction performance of our
hybrid system is exceptionally competitive with the state-of-the-art
heterogeneous systems for CO2 reduction (Table S2, ESI†).
Fig. 4 (a) Free energy diagram for the CO2 reduction to CO at E = 0 V vs.
RHE on Fe–N–C and Ni–N–C. (b) Projected crystal orbital Hamilton popula-
tions (pCOHP) for Fe–C and Ni–C bonds in the *COOH intermediate
structure and corresponding integrated values (IpCOHP). Calculated electron
Fig. 3 (a) Photochemical CO and H2 production as a function of time of
the Fe–CB + CdS hybrid system. (b) Comparison of CO and H2 yield of
varied photocatalytic systems.
density differences for Fe–N–C (c) and Ni–N–C (d) coordinated with *COOH
intermediate. Electron accumulation and depletion are represented by yellow
and cyan, respectively.
This journal is The Royal Society of Chemistry 2021
Chem. Commun., 2021, 57, 2033À2036 | 2035

View Article Online
Communication
ChemComm
H2 evolution on the catalysts were also calculated to be 0.72 and References
1.65 eV for Fe–CB and Ni–CB, respectively. Although Fe–CB is a
1 M. Robert, ACS Energy Lett., 2016, 1, 281–282.
better catalyst for H2 evolution than Ni–CB, its low energy barrier for
*COOH generation ensures the excellent selectivity for CO evolution
even in the presence of CdS, which is also a good photocatalyst for
H2 evolution (Fig. 3b).22
2 X. Li, J. Yu, M. Jaroniec and X. Chen, Chem. Rev., 2019, 119,
3962–4179.
3 X. Li, Y. Sun, J. Xu, Y. Shao, J. Wu, X. Xu, Y. Pan, H. Ju, J. Zhu and
Y. Xie, Nat. Energy, 2019, 4, 690–699.
4 Y. Liu, J.-H. Guo, X.-Y. Dao, X.-D. Zhang, Y. Zhao and W.-Y. Sun,
Chem. Commun., 2020, 56, 4110–4113.
To gain more specific insight into the influence of two
different metals in the catalysis, we performed the crystal
orbital Hamilton population (COHP) analysis to confirm the
bonding strength of the two structures (Fig. 4b). The integrated
projected COHP (IpCOHP) summed between Fe–C and Ni–C
bond below the Fermi level in the *COOH structure are À3.09
and À1.84 eV, respectively, indicating more stable coordination
of Fe–C than Ni–C. Besides, the deformation charge density
calculated based on the Bader charge analysis and the calcu-
lated electron density differences revealed that more electrons
transfer from the Fe–N–C to *COOH than Ni–N–C (Fig. 4c and d),
demonstrating that Fe–N–C could promote the electron accumu-
lation in *COOH and facilitate the CO2 reduction.13 Thus, we
conclude that the stronger coordination of *COOH to the Fe
atom is the critical factor that enables the Fe–N–C’s outstanding
catalytic activity. These theoretical results are consistent with
the experiments.
´ `
5 C. Cometto, L. Chen, P.-K. Lo, Z. Guo, K.-C. Lau, E. Anxolabehere-
Mallart, C. Fave, T.-C. Lau and M. Robert, ACS Catal., 2018, 8,
3411–3417.
´
6 S. Ren, D. Joulie, D. Salvatore, K. Torbensen, M. Wang, M. Robert
and C. P. Berlinguette, Science, 2019, 365, 367–369.
7 R. Francke, B. Schille and M. Roemelt, Chem. Rev., 2018, 118,
4631–4701.
8 H. Takeda, C. Cometto, O. Ishitani and M. Robert, ACS Catal., 2016,
7, 70–88.
9 C. S. Diercks, Y. Liu, K. E. Cordova and O. M. Yaghi, Nat. Mater.,
2018, 17, 301–307.
10 C. T. Dinh, F. P. G. de Arquer, D. Sinton and E. H. Sargent,
ACS Energy Lett., 2018, 3, 2835–2840.
11 M. Liu, Y. Pang, B. Zhang, P. De Luna, O. Voznyy, J. Xu, X. Zheng,
C. T. Dinh, F. Fan, C. Cao, F. P. de Arquer, T. S. Safaei, A. Mepham,
A. Klinkova, E. Kumacheva, T. Filleter, D. Sinton, S. O. Kelley and
E. H. Sargent, Nature, 2016, 537, 382–386.
12 J. Gu, C.-S. Hsu, L. Bai, H. M. Chen and X. Hu, Science, 2019, 364,
1091–1094.
13 Y. Li, S. Wang, X. S. Wang, Y. He, Q. Wang, Y. Li, M. Li, G. Yang, J. Yi,
H. Lin, D. Huang, L. Li, H. Chen and J. Ye, J. Am. Chem. Soc., 2020,
142, 19259–19267.
14 Z. Wang, J. Yang, J. Cao, W. Chen, G. Wang, F. Liao, X. Zhou,
F. Zhou, R. Li, Z. Q. Yu, G. Zhang, X. Duan and Y. Wu, ACS Nano,
2020, 14, 6164–6172.
15 T. Zheng, K. Jiang, N. Ta, Y. Hu, J. Zeng, J. Liu and H. Wang, Joule,
2019, 3, 265–278.
16 J.-H. Guo, X.-Y. Zhang, X.-Y. Dao and W.-Y. Sun, ACS Appl. Nano
Mater., 2020, 3, 2625–2635.
17 J.-H. Guo and W.-Y. Sun, Appl. Catal., B, 2020, 275, 119154.
18 F. Pan, H. Zhang, K. Liu, D. Cullen, K. More, M. Wang, Z. Feng,
G. Wang, G. Wu and Y. Li, ACS Catal., 2018, 8, 3116–3122.
19 J. Li, Y. Song, G. Zhang, H. Liu, Y. Wang, S. Sun and X. Guo,
Adv. Funct. Mater., 2017, 27, 1604356.
20 Q. Jia, N. Ramaswamy, H. Hafiz, U. Tylus, K. Strickland, G. Wu,
B. Barbiellini, A. Bansil, E. F. Holby, P. Zelenay and S. Mukerjee,
ACS Nano, 2015, 9, 12496–12505.
21 C. Santoro, R. Gokhale, B. Mecheri, A. D’Epifanio, S. Licoccia, A. Serov,
K. Artyushkova and P. Atanassov, ChemSusChem, 2017, 10, 3243–3251.
22 Q. Li, B. Guo, J. Yu, J. Ran, B. Zhang, H. Yan and J. R. Gong, J. Am.
Chem. Soc., 2011, 133, 10878–10884.
23 M. F. Kuehnel, K. L. Orchard, K. E. Dalle and E. Reisner, J. Am. Chem.
Soc., 2017, 139, 7217–7223.
24 N. Zhang, R. Long, C. Gao and Y. Xiong, Sci. China Mater., 2018, 61,
771–805.
25 Q. Wang, L. Zhang, Y. Guo, M. Shen, M. Wang, B. Li and J. Shi,
Chem. Eng. J., 2020, 396, 125347.
In conclusion, we utilized M–N–C catalysts derived from CB and
metalloporphyrins to construct heterogeneous hybrid catalytic sys-
tems with CdS, delivering highly efficient CO2 photoreduction.
Significantly, the Fe–N–C and CdS hybrid system efficiently catalyzes
the CO2 reduction to CO under visible light irradiation, delivered a
À1
high CO yield of 111 mmol gcat and a large TON of 1.22 Â 103
based on the Fe–N–C catalyst in 8 h with a selectivity of 85% to
outperform most of the current photocatalysis systems. The efficient
performance of the Fe–N–C catalysts results from the strong coordi-
nation between iron and *COOH intermediate that reduces the
energy barrier of the rate-determining step. This effective hybrid
catalytic system provides a new perspective for CO2 photoreduction.
We gratefully acknowledge the National Key Research and
Development Project of China (grant no. 2017YFA0303504) for
financial support of this work. This work was also supported by
a Project Funded by the Priority Academic Program Develop-
ment of Jiangsu Higher Education Institutions. We are grateful
to the High-Performance Computing Center of Nanjing
University for conducting the DFT calculations in this work
on its Flex cluster system.
26 X. F. Xie, X. Y. Dao, F. Guo, X. Y. Zhang, F. M. Wang and W. Y. Sun,
ChemistrySelect, 2020, 5, 4001–4007.
27 Q.-Q. Bi, J.-W. Wang, J.-X. Lv, J. Wang, W. Zhang and T.-B. Lu,
ACS Catal., 2018, 8, 11815–11821.
28 C. Bie, B. Zhu, F. Xu, L. Zhang and J. Yu, Adv. Mater., 2019,
31, 1902868.
Conflicts of interest
There are no conflicts to declare.
2036 | Chem. Commun., 2021, 57, 2033À2036
This journal is The Royal Society of Chemistry 2021