Efficient Infrared-Light-Driven CO2 Reduction Over Ultrathin Metallic Ni-doped CoS2 Nanosheets

Article abstract of DOI:10.1002/anie.202017041

Converting CO₂ and H₂O into carbon-based fuel by IR light is a tough task. Herein, compared with other single-component photocatalysts, the most efficient IR-light-driven CO₂ reduction is achieved by an element-doped ultrathin metallic photocatalyst-Ni-doped CoS₂ nanosheets (Ni-CoS₂). The evolution rate of CH₄ over Ni-CoS₂ is up to 101.8 μmol g^?1 h^?1. The metallic and ultrathin nature endow Ni-CoS₂ with excellent IR light absorption ability. The PL spectra and Arrhenius plots indicate that Ni atoms could facilitate the separation of photogenerated carriers and the decrease of the activation energy. Moreover, in situ FTIR, DFT calculations, and CH₄-TPD reveal that the doped Ni atoms in CoS₂ could effectively depress the formation energy of the *COOH, *CHO and desorption energy of CH₄. This work manifests that element doping in atomic level is a powerful way to control the reaction intermediates, providing possibilities to realize high-efficiency IR-light-driven CO₂ reduction.

Full text of DOI:10.1002/anie.202017041

Angewandte
Communications
Chemie
How to cite: Angew. Chem. Int. Ed. 2021, 60, 8705–8709
International Edition: doi.org/10.1002/anie.202017041
German Edition: doi.org/10.1002/ange.202017041
Photocatalysis
Hot Paper
Efficient Infrared-Light-Driven CO2 Reduction Over Ultrathin
Metallic Ni-doped CoS2 Nanosheets
Jiaqi Xu,* Zhengyu Ju, Wei Zhang, Yang Pan, Junfa Zhu, Jiawei Mao, Xueli Zheng, Haiyan Fu,
Maolin Yuan, Hua Chen, and Ruixiang Li*
Abstract: Converting CO2 and H2O into carbon-based fuel by
IR light is a tough task. Herein, compared with other single-
component photocatalysts, the most efficient IR-light-driven
CO2 reduction is achieved by an element-doped ultrathin
metallic photocatalyst-Ni-doped CoS2 nanosheets (Ni-CoS2).
The evolution rate of CH4 over Ni-CoS2 is up to
101.8 mmolgꢀ1 hꢀ1. The metallic and ultrathin nature endow
Ni-CoS2 with excellent IR light absorption ability. The PL
spectra and Arrhenius plots indicate that Ni atoms could
facilitate the separation of photogenerated carriers and the
decrease of the activation energy. Moreover, in situ FTIR, DFT
calculations, and CH4-TPD reveal that the doped Ni atoms in
CoS2 could effectively depress the formation energy of the
*COOH, *CHO and desorption energy of CH4. This work
manifests that element doping in atomic level is a powerful way
to control the reaction intermediates, providing possibilities to
realize high-efficiency IR-light-driven CO2 reduction.
component photocatalyst to achieve IR-light-driven CO2
reduction without sacrificial agent. In this context, to max-
imize the use of solar energy, it is imperative to design novel
photocatalysts that can fulfill the efficient IR-light-driven
CO2 reduction.
Metallic photocatalysts are expected to be promising
candidates for IR-light-driven CO2 reduction. Compared with
semiconductor photocatalysts, the small (or even no) band
gaps of metallic photocatalysts ensure their excellent absorp-
tion of IR light;[5] more importantly, the partially-filled
conduction band of metallic photocatalysts could act as the
intermediate band and allows them to satisfy the redox
potentials for IR-light-driven CO2 reduction.[6] However,
their high carrier concentration often causes severe recombi-
nation of electron-hole pairs;[7] moreover, the CO2 reduction
activity is still very low and CO2 could only be converted into
CO in those systems.[4,8] Thus, it remains a tough task to
efficiently convert CO2 and H2O into carbon-based fuels by
metallic photocatalysts under IR light irradiation.
Herein, we designed an element-doped ultrathin metallic
photocatalyst to achieve high-efficiency IR-light-driven CO2
reduction. The ultrathin morphology could reduce carrier
diffusion length and is beneficial to decrease the electron-hole
recombination rates; meanwhile, it also benefits the imple-
mentation of element doping at atomic level, which is an
effective way to regulate their CO2 reduction activity and
selectivity simultaneously. CoS2 were taken as an example,
and ultrathin Ni-doped CoS2 nanosheets (called Ni-CoS2
nanosheets or Ni-CoS2 for short) were fabricated by a modi-
fied salt-mediated strategy (Supporting Information, Fig-
ure S1).[9] XRD pattern (Supporting Information, Figure S3)
and Raman spectra (Supporting Information, Figure S4)
could be indexed to pure cubic-CoS2,[10] while the XPS
spectra (Supporting Information, Figure S5A,B) clearly veri-
fied the introducing of Ni (3.69 atomic%) into CoS2. More
defects might also be introduced in Ni-CoS2 during the doping
processes (Supporting Information, Figure S5C,D). TEM
image (Figure 1A) manifested their ultrathin morphology,
while the annular dark-field TEM image and the correspond-
ing element mappings (Figure 1B) revealed the homogene-
ous distribution of Co, S, and Ni elements. The lattice plane
spacings of Ni-CoS2 (Figure 1C) were 0.28 nm with a dihedral
angle of 908, implying their [0 0 1] orientation. Furthermore,
AFM image and the height profiles (Figure 1D) showed that
the average thickness of Ni-CoS2 were ca. 1.0 nm, which
corresponds to 2-unit cells (Supporting Information, Fig-
ure S10). Above results verified the successful synthesis of Ni-
CoS2 nanosheets. For comparison, pristine CoS2 nanosheets
I
t is extremely meaningful to convert CO2 and H2O into
carbon-based fuels by solar energy, holding great promise to
solve the energy crisis and environmental problems.[1,2]
Though great efforts have been put into this area, CO2
photoreduction efficiency is still very low. One of the reasons
is the insufficient solar energy utilization. Previously, CO2
photoreduction mainly focused on the ultraviolet and visible
range,[3] while the infrared light (ca. 50% of solar spectrum)
driven CO2 reduction has been rarely reported. Considering
the band edges matching and overpotential issues, the band
gaps of photocatalysts often need to be larger than 1.8 eV,
which theoretically hinders the utilization of infrared light
(hn < 1.55 eV).[4] Therefore, it is difficult for traditional single-
[*] Dr. J. Q. Xu, Dr. W. Zhang, Dr. J. W. Mao, Prof. X. L. Zheng,
Prof. H. Y. Fu, Prof. M. L. Yuan, Prof. H. Chen, Prof. R. X. Li
Key Laboratory of Green Chemistry & Technology
Ministry of Education, College of Chemistry, Sichuan University
Chengdu, Sichuan 610064 (P. R. China)
E-mail: jqxu@scu.edu.cn
liruixiang@scu.edu.cn
Prof. Y. Pan, Prof. J. F. Zhu
National Synchrotron Radiation Laboratory
University of Science and Technology of China
Hefei, Anhui 230026 (P. R. China)
Z. Y. Ju
Materials Science and Engineering Program
Texas Materials Institute, The University of Texas at Austin
Austin, TX 78712 (USA)
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.202017041.
Angew. Chem. Int. Ed. 2021, 60, 8705 –8709
ꢀ 2021 Wiley-VCH GmbH
8705

Angewandte
Communications
Chemie
Figure 1. Characterizations for Ni-CoS2 nanosheets: A) TEM image;
B) annular dark-field TEM image and the corresponding element
mappings, the scale bar is 100 nm; C) HRTEM image; D) AFM image.
Figure 2. A) Temperature dependence of resistivities; B) density of
states (the upper part is spin-up and lower part is spin-down) and
C) calculated band structure of Ni-CoS2 nanosheets (spin-up is red
and spin-down is gray); D) UV/Vis-NIR diffuse reflectance spectra, the
inset denotes the energy gaps; E) secondary electron cutoff acquired
by synchrotron-radiation photoemission spectra; F) diagram of the
electronic band structures for Ni-CoS2 and CoS2 nanosheets. The B1,
CB, Bꢀ1, EF, and Eg represent the lowest unoccupied band, conduction
band, the highest fully occupied band, Fermi level, and energy gap,
respectively.
(called CoS2) were also synthesized (Supporting Information,
Figures S3–S10).
Experimental characterizations and theoretical calcula-
tions were utilized to disclose the electrical properties and
band structures of Ni-CoS2 and CoS2. Both Ni-CoS2 and CoS2
possessed superp electrical transport ability (Figure 2A);
more importantly, the resistivities of Ni- CoS2 and CoS2
increased along with the elevating temperature verified
their metallic nature. Meanwhile, their metallic nature was
further confirmed by the valence-band spectra (Supporting
Information, Figure S11). Hence, the partially occupied band
is considered as the conduction band (CB) for the metallic
conductor.[6] Theoretical calculations were applied to further
illustrate their electronic structures (Figure 2B,C; Supporting
Information, Figure S12A,B). It was found that their Fermi
level located within the conduction band and their metallic
nature was validated once again. Notably, metallic materials
could generate electrons and holes through interband and
intraband transitions.[4,6] For instance, photon absorption
could lead to the electrons transition from EF to lowest
unoccupied band (B1) in metallic materials. Herein, UV/Vis-
NIR diffuse reflectance spectra and synchrotron-radiation
photoelectron spectra were used to further analyze their band
structures. The average absorbance for Ni-CoS2 and CoS2
were around 1.63 and 1.78 ranging from UV to near infrared
region (Figure 2D), which meant 97.7% and 98.3% of
incident light was absorbed. And the energy gaps (Eg) of
Ni-CoS2 and CoS2 (Figure 2D) were 1.15 and 1.05 eV. Mean-
while, secondary electron cutoff (Figure 2E) revealed the EF
of Ni-CoS2 and CoS2 were ꢀ5.04 and ꢀ4.81 eV (vs. vacuum),
respectively. Ultimately, the electronic band structures of
metallic Ni-CoS2 and CoS2 were obtained (Figure 2F; Sup-
porting Information, Figure S13). The intraband transitions in
the highest fully occupied band (Bꢀ1, transitions in Bꢀ1 from
the lower level to the top), and interband transitions from Bꢀ1
to EF, EF to B1 for both Ni-CoS2 and CoS2 could be realized by
IR light, where the generated holes in the Bꢀ1 could oxidize
H2O to O2 and the photoinduced electrons in the B1 could
reduce CO2 to CH4 and CO (Figure 2F; Supporting Informa-
tion, Figure S13).
Infrared-light-driven CO2 reduction experiments were
conducted to investigate their performance (Supporting
Information, Figure S15). The temperature of samples will
increase under irradiation, and the increased temperature
might enhance the CO2 reduction rate. However, in many
cases, photocatalysts are dispersed in liquid solvent (such as
water, acetonitrile), in which photoinduced heat is easily
wasted.[11,12] To make better use of photoinduced heat,
a quartz tray was used to separate water and photocatalyst,
where photoinduced heat could be efficiently retained
(Supporting Information, Figure S16A). As expected, both
Ni-CoS2 and CoS2 could enable CO2 reduction under IR light
to form CH4 and CO as the main reduction products
(Figure 3A; Supporting Information, Figures S26, S27). The
evolution rates of CH4 and CO over Ni-CoS2 were up to101.8
and 37.5 mmolgꢀ1 hꢀ1, which were roughly 3.3 and 2.8 times
8706 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 8705 –8709

Angewandte
Communications
Chemie
could rise from 17.28C to 43.28C under IR light irradiation
(Supporting Information, Figure S21A,B). The control
experiments depicted that there was negligible amount of
gas product at 458C in the dark, and the reaction temperature
needed to be around 1808C in pure thermocatalysis to obtain
comparable CH4 formation rates (Supporting Information,
Figure S22A, Table S2). When photoinduced heat was delib-
erately excluded (Supporting Information, Figure S23B, S24),
the performance of Ni-CoS2 declined sharply. The result
suggested the photoinduced heat was also significant for CO2
reduction.[21] These similar phenomena can also be observed
when full-spectrum light was applied (Supporting Informa-
tion, Figures S23C, S25). Therefore, it is a good strategy to
make full use of the IR-light-induced heat and a landmark
activity for IR-light-driven CO2 reduction was achieved over
metallic Ni-CoS2 nanosheets.
To unveil the origin of the boosted IR-light-driven CO2
reduction performances, fundamental steps were analyzed.
The BET surface area of Ni-CoS2 was 17.75 m2 gꢀ1 (Support-
ing Information, Figure S28), which was almost double that of
CoS2 (8.89 m2 gꢀ1). This might be attributed to Ni-CoS2ꢀs
higher degree of disorder (Supporting Information, Fig-
ure S3) or more defects (introduced by doping processes
simultaneously; Supporting Information, Figures S5, S28).
The enlarged surface area could facilitate the CO2 adsorption
in return. As expected, Ni-CoS2 possessed a CO2 adsorption
capacity of 5.16 cm3 gꢀ1, roughly 1.85 times larger than that of
CoS2 (Figure 3E). Then, photoinduced charge carriers were
needed to reduce the adsorbed CO2. Due to the strong
photoabsorption ability, almost all the incident light could be
absorbed by Ni-CoS2, which benefited the generation of
charge carriers. Meanwhile, the photoluminescence intensity
of Ni-CoS2 was weaker than that of CoS2 (Figure 3F),
suggesting the introducing of Ni atoms facilitated the
separation of photogenerated carriers. More importantly,
the activation energy to reduce CO2 into CH4 and CO over
Ni-CoS2, measured by Arrhenius plots, was lower than that
over CoS2 [Supporting Information, Figure S22D, Eq. (S4)].
It indicted that the introducing of Ni atoms could enhanced
their reactive activity.[22] Furthermore, both the higher
desorption temperature of CO and lower desorption temper-
ature of CH4 over Ni-CoS2 than these over CoS2 (Figure 4A;
Supporting Information, Figure S29) were beneficial to
further reduce the adsorbed CO into CH4. In situ FTIR
measurements were conducted to detect the reaction inter-
mediates. As for Ni-CoS2 (Figure 4B), the new peaks at 1643
and 1528 cmꢀ1 could be indexed to *COOH, while those
located at 1157 cmꢀ1 and 1080 cmꢀ1 could be assigned to
*CH3O and *CHO, respectively.[4,14,23] Thus, the IR-light-
driven CO2 reduction processes might be suggested as
follows:
Figure 3. A) Formation rates of CH4 and CO under infrared light
irradiation as well as at 458C in the dark; synchrotron-based vacuum
UV photoionization mass spectrometry of the gas products for the
B) 13C and C) 18O isotope labeling experiments; D) cycling measure-
ments of IR-light-driven CO2 reduction; E) CO2 adsorption isotherms;
F) photoluminescence spectra. The error bars in A represent the
standard deviations of three independent measurements.
higher than those of the CoS2. Notably, the formation rate of
CH4 over Ni-CoS2 is so far the highest records among the
reported single-component photocatalysts (Supporting Infor-
mation, Table S1).[13–20] Meanwhile, the electron selectivity
(carbon-based products) of CH4 for Ni-CoS2 (90.6%) was
also higher than that of CoS2 (88.2%). The control experi-
ments showed that no CH4 or CO was detected when
photocatalysts, IR light or CO2 was removed (Supporting
Information, Figure S4), verifying the crucial role of both the
photocatalysts and IR light in transforming CO2 to CH4 and
CO. 13CH4 and 13CO were generated in the 13CO2 labeling
experiment (Figure 3B; Supporting Information, Figure S18),
which ambiguously proved that CH4 and CO originated from
CO2. Meanwhile, the presence of 16O18O in the 18O labeling
experiment verifies that O2 originated from H2O (Figure 3C;
Supporting Information, Figure S18). Interestingly, the CO2
photoreduction performance of Ni-CoS2 could be further
enhanced under full-spectrum irradiation, and the formation
rates of CH4 and CO could reach 243.2 and 63.5 mmolgꢀ1 hꢀ1
(Supporting Information, Figure S19), respectively. Cycling
measurements showed that both Ni-CoS2 and CoS2 were of
excellent stability and their activities did not show obvious
drop even after 24 hours (Figure 3D; Supporting Informa-
tion, Figure S20).
þ CO2 þ Hþ þ eꢀ
!
ð1Þ
ð2Þ
ð3Þ
ð4Þ
*
*
*
*
*
COOH
COOH þ Hþ þ eꢀ ! H2O þ CO
*
CO þ Hþ þ eꢀ
!
CHO or CO ! þ CO "
*
*
*
To evaluate the photothermal effect, the result of thermo-
graphic photographs showed that the temperature of Ni-CoS2
CHO þ Hþ þ eꢀ
!
CH2O
*
Angew. Chem. Int. Ed. 2021, 60, 8705 –8709
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 8707

Angewandte
Communications
Chemie
reduction and H2O oxidation simultaneously by IR light with
the partially occupied conduction band acting as the media-
tor. In situ FTIR and DFT calculations showed that the
formation of *COOH intermediate was the rate-determining
step, and Ni-CoS2 nanosheets possessed a lower *COOH
formation energy due to the stronger bonding between Ni
atoms and *COOH. Furthermore, the introducing of Ni
atoms could depress the formation energy of *CHO and
desorption energy of CH4, so that their selectivity and
catalytic activity were improved simultaneously.
Acknowledgements
This work was financially supported by China Postdoctoral
Science Foundation (BX20190222, 2019M663490), Postdoc-
toral Interdisciplinary Innovation Fund of Sichuan Univer-
sity, and the Fundamental Research Funds for the Central
Universities (20826041D4117). We thank Comprehensive
Training Platform of the Specialized Laboratory in the
College of Chemistry at Sichuan University and Analytical
& Testing Center of Sichuan University for the support.
Supercomputing of USTC and National Supercomputing
Center in Shenzhen are acknowledged for computational
support.
Figure 4. A) CH4-TPD measurements; B) in situ FTIR for Ni-CoS2
nanosheets; free energy diagrams of IR-light-driven CO2 reduction to
CO and CH4 for C) Ni-CoS2 nanosheets and D) CoS2 nanosheets.
Conflict of interest
CH2O þ Hþ þ eꢀ
CH3O þ Hþ þ eꢀ
!
!
CH3O
ð5Þ
ð6Þ
ð7Þ
ð8Þ
*
*
*
*
*
*
The authors declare no conflict of interest.
O þ CH4 "
Keywords: CO2 conversion · infrared photocatalysis ·
metallic photocatalyst · ultrathin 2D materials
O þ Hþ þ eꢀ
!
OH
*
OH þ Hþ þ eꢀ ! H2O
[1] J. L. White, M. F. Baruch, J. E. Pander III, Y. Hu, I. C. For-
tmeyer, J. E. Park, T. Zhang, K. Liao, J. Gu, Y. Yan, T. W. Shaw,
E. Abelev, A. B. Bocarsly, Chem. Rev. 2015, 115, 12888 – 12935.
[2] M. Aresta, A. Dibenedetto, A. Angelini, Chem. Rev. 2014, 114,
1709 – 1742.
where the * denotes active sites.
Some important information during the CO2 reduction
processes was obtained with DFT calculations. As displayed
in Figure 4C,D and the Supporting Information, Tables S3–5
and Equations (S16)–(S24), the formation of *COOH was the
rate-determining step for two photocatalysts, while Ni-CoS2
possessed a lower *COOH formation energy. The fact that
Ni-COOH bonds (1.95 ꢁ) on Ni-CoS2 were shorter than Co-
COOH bonds (1.97 ꢁ) on CoS2, suggesting the stronger
bonding between Ni and *COOH benefited the stabilization
of *COOH. In addition, the formation energy of *CHO was
much more negative than the desorption energy of CO over
Ni-CoS2 (Figure 4C), implying Ni-CoS2 were apt to further
reduce CO to form CH4. Moreover, the lower desorption
energy of CH4 over Ni-CoS2 indicated CH4 could be easier
desorbed from their surface.
[3] X. Chang, T. Wang, J. Gong, Energy Environ. Sci. 2016, 9, 2177 –
2196.
[4] X. Li, L. Liang, Y. Sun, J. Xu, X. Jiao, X. Xu, H. Ju, Y. Pan, J.
Zhu, Y. Xie, J. Am. Chem. Soc. 2019, 141, 423 – 430.
[5] Y. L. Wang, T. Nie, Y. H. Li, X. L. Wang, L. R. Zheng, A. P.
Chen, X. Q. Gong, H. G. Yang, Angew. Chem. Int. Ed. 2017, 56,
7430 – 7434; Angew. Chem. 2017, 129, 7538 – 7542.
[6] X. Xu, C. Randorn, P. Efstathiou, J. T. Irvine, Nat. Mater. 2012,
11, 595 – 598.
[7] Z. Hu, G. Liu, X. Chen, Z. Shen, J. C. Yu, Adv. Funct. Mater.
2016, 26, 4445 – 4455.
[8] L. Liang, X. Li, J. Zhang, P. Ling, Y. Sun, C. Wang, Q. Zhang, Y.
Pan, Q. Xu, J. Zhu, Y. Luo, Y. Xie, Nano Energy 2020, 69,
104421.
[9] X. Xiao, H. Song, S. Lin, Y. Zhou, X. Zhan, Z. Hu, Q. Zhang, J.
Sun, B. Yang, T. Li, L. Jiao, J. Zhou, J. Tang, Y. Gogotsi, Nat.
Commun. 2016, 7, 11296.
[10] S. G. Lyapin, A. N. Utyuzh, A. E. Petrova, A. P. Novikov, T. A.
Lograsso, S. M. Stishov, J. Phys. Condens. Matter 2014, 26,
396001.
In conclusion, high-efficiency IR-light-driven CO2 reduc-
tion has been achieved by Ni-CoS2 nanosheets. The CH4
evolution rates were 101.8 mmolgꢀ1 hꢀ1, which is the highest
record for IR-light-driven CO2 reduction among all single-
component photocatalysts. The metallic and ultrathin nature
endowed Ni-CoS2 nanosheets with excellent IR light absorp-
tion ability, while their band structures were able to fulfill CO2
[11] S. Sun, M. Watanabe, J. Wu, Q. An, T. Ishihara, J. Am. Chem.
Soc. 2018, 140, 6474 – 6482.
8708 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 8705 –8709

Angewandte
Communications
Chemie
[12] X. K. Wang, J. Liu, L. Zhang, L. Z. Dong, S. L. Li, Y. H. Kan,
[19] Q. Liu, Y. Zhou, J. Kou, X. Chen, Z. Tian, J. Gao, S. Yan, Z. Zou,
D. S. Li, Y. Q. Lan, ACS Catal. 2019, 9, 1726 – 1732.
[13] Z. B. Fang, T. T. Liu, J. Liu, S. Jin, X. P. Wu, X. Q. Gong, K.
Wang, Q. Yin, T. F. Liu, R. Cao, H. C. Zhou, J. Am. Chem. Soc.
2020, 142, 12515 – 12523.
[14] X. Li, Y. Sun, J. Xu, Y. Shao, J. Wu, X. Xu, Y. Pan, H. Ju, J. Zhu,
Y. Xie, Nat. Energy 2019, 4, 690 – 699.
J. Am. Chem. Soc. 2010, 132, 14385 – 14387.
[20] S. Sorcar, Y. Hwang, C. A. Grimes, S. I. In, Mater. Today 2017,
20, 507 – 515.
[21] Y. X. Li, M. M. Wen, Y. Wang, G. Tian, C. Y. Wang, J. C. Zhao,
Angew. Chem. Int. Ed. 2021, 60, 910 – 916; Angew. Chem. 2021,
133, 923 – 929.
[15] S. Zhu, S. Liang, J. Bi, M. Liu, L. Zhou, L. Wu, X. Wang, Green
Chem. 2016, 18, 1355 – 1363.
[16] Y. Bai, J. Zhao, S. Feng, X. Liang, C. Wang, Chem. Commun.
2019, 55, 4651 – 4654.
[22] L. Wang, W. Zhang, X. Zheng, Y. Chen, W. Wu, J. Qiu, X. Zhao,
X. Zhao, Y. Dai, J. Zeng, Nat. Energy 2017, 2, 869 – 876.
[23] G. E. Ewing, W. E. Thompson, G. C. Pimentel, J. Chem. Phys.
1960, 32, 927 – 932.
[17] J. Li, Y. Ye, L. Ye, F. Su, Z. Ma, J. Huang, H. Xie, D. E. Doronkin,
A. Zimina, J. D. Grunwaldt, Y. Zhou, J. Mater. Chem. A 2019, 7,
2821 – 2830.
[18] J. Yu, J. Low, W. Xiao, P. Zhou, M. Jaroniec, J. Am. Chem. Soc.
2014, 136, 8839 – 8842.
Manuscript received: December 23, 2020
Accepted manuscript online: January 20, 2021
Version of record online: March 5, 2021
Angew. Chem. Int. Ed. 2021, 60, 8705 –8709
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 8709