Research Article www.afm-journal.de Site-Specific Axial Oxygen Coordinated FeN4 Active Sites for Highly Selective Electroreduction of Carbon Dioxide Ting Zhang,* Xu Han, Hong Liu, Martí Biset-Peiró, Jian Li,* Xuan Zhang, Pengyi Tang, Bo Yang, Lirong Zheng,* Joan Ramon Morante, and Jordi Arbiol* 1. Introduction Regulating the coordination environment via heteroatoms to break the sym- metrical electronic structure of M-N4 active sites provides a promising route Electrochemical CO2 reduction reaction (CO2 RR) into CO, which represents a to engineer metal-nitrogen-carbon catalysts for electrochemical CO2 reduction crucial raw product in Fischer-Tropsch reaction. However, it remains challenging to realize a site-specific introduc- process for the production of value-added tion of heteroatoms at atomic level due to their energetically unstable nature. chemical feedstocks, is considered as a Here, this paper reports a facile route via using an oxygen- and nitrogen-rich promising route for energy conversion and metal–organic framework (MOF) (IRMOF-3) as the precursor to construct the carbon recycling.[1–6] However, the inert- Fe-O and Fe-N chelation, simultaneously, resulting in an atomically dispersed ness of CO2 molecule and the competitive hydrogen evolution reaction (HER) greatly axial O-coordinated FeN4 active site. Compared to the FeN4 active sites limit the conversion efficiency toward the without O coordination, the formed FeN4-O sites exhibit much better catalytic practical implementation.[7–9] Atomically performance toward CO, reaching a maximum FECO of 95% at −0.50 V versus dispersed transition metals embedded in reversible hydrogen electrode. To the best of the authors’ knowledge, such nitrogen-doped carbon matrices (M-N-C) performance exceeds that of the existing Fe-N-C-based catalysts derived have recently appeared at the forefront of CO2 RR toward CO because of their fea- from sole N-rich MOFs. Density functional theory calculations indicate that sible preparation, optimized atomic utili- the axial O-coordination regulates the binding energy of intermediates in the zation and chemical stability.[10–14] Based reaction pathways, resulting in a smoother desorption of CO and increased on previous studies, the four-coordinated energy for the competitive hydrogen production. M-N4 moieties are considered as the main active centers in common M-N-C.[15,16] T. Zhang, X. Han, J. Arbiol X. Zhang Catalan Institute of Nanoscience and Nanotechnology (ICN2) Department of Materials Engineering CSIC and BIST Campus UAB KU Leuven Bellaterra, 08193 Barcelona, Catalonia, Spain Kasteelpark Arenberg 44, B-3001 Leuven, Belgium E-mail: ting.zhang@icn2.cat; arbiol@icrea.cat P. Y. Tang T. Zhang, M. Biset-Peiró, J. R. Morante State Key Laboratory of Information Functional Materials and 2020 X-Lab Energy Storage Shanghai Institute of Microsystem and Information Harvesting and Catalysis Group Technology, Chinese Academy of Sciences Catalonia Institute for Energy Research (IREC) Shanghai 200050, P. R. China Sant Adrià de Besòs, 08930 Barcelona, Catalonia, Spain L. R. Zheng H. Liu, B. Yang Beijing Synchrotron Radiation Facility School of Physical Science and Technology Institute of High Energy Physics ShanghaiTech University Chinese Academy of Sciences 393 Middle Huaxia Road, Shanghai 201210, P. R. China Beijing 100049, P. R. China J. Li E-mail: zhenglr@ihep.ac.cn Laboratory of Renewable Energy Science and Engineering J. R. Morante Institute of Mechanical Engineering Department of Physics EPFL, Station 9, 1015 Lausanne, Switzerland Universitat de Barcelona E-mail: jian.li@epfl.ch Barcelona, 08028 Catalonia, Spain The ORCID identification number(s) for the author(s) of this article J. Arbiol can be found under https://doi.org/10.1002/adfm.202111446. ICREA Pg. Lluís Companys 23 © 2022 The Authors. Advanced Functional Materials published by Barcelona, 08010 Catalonia, Spain Wiley-VCH GmbH. This is an open access article under the terms of the ­Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. DOI: 10.1002/adfm.202111446 Adv. Funct. Mater. 2022, 32, 2111446 2111446 (1 of 8) © 2022 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH www.advancedsciencenews.com www.afm-journal.de However, the high structure/electron symmetry of the M-N4 we propose a facile strategy, which uses an O- and N-rich isore- moiety, which results from the symmetrical planar structure, ticular metal–organic framework-3 (IRMOF-3) for constructing makes it chemically inert to a certain extent.[17] Recently, the an axial O coordination to the FeN4 active sites. In comparison performance of M-N4 moiety for CO2 RR has been significantly with FeN4 structure, the as-prepared axial O-coordinated FeN4 improved through the introduction of an additional axial N (denoted as O-Fe-N-C) active sites showed an excellent CO2 RR ligand.[18–20] Such coordination can break the symmetrical elec- activity in 0.5 m NaHCO3 solution, including a high Faradaic tronic structure of M-N4 and subsequently affect the binding efficiency of CO (95%) at −0.50 V versus reversible hydrogen energies for CO2 RR intermediates (e.g., *COOH and *CO), electrode (RHE), and a robust stability for 30 h. Such perfor- which confirmed the key role of axial coordination.[21] Inspired by mance is superior to those of the previously reported Fe-N-C- this, it is anticipated that the catalytic activity of M-N4 can be fur- based materials and comparable to the reported state-of-the- ther boosted by introducing high-electronegativity atoms, such art single atom catalysts (SACs) even at a lower overpotential. as O, in the axial direction. The high electronegativity of oxygen Density functional theory (DFT) calculations revealed that the will not only break the electron symmetry, but also improve the formed O-Fe-N-C not only facilitates CO desorption, but also strength of the electronic metal-support interaction (EMSI) in limits the undesired HER. This work opens a new way of uti- M-N4 sites.[21–23] Unfortunately, the coordination configuration lizing heteroatom-rich MOFs as an enabler to the rational of oxygen atoms is difficult to be controlled accurately via tradi- design and development of high-efficient catalysts. tional calcination methods. The type of oxygen will change and eventually disappear when the calcination temperature is above 500 °C, which as a consequence hinders the successful coordina- 2. Results and Discussions tion of O atoms with the metal site.[22] Therefore, developing an effective method to realize the control on O atoms coordination, 2.1. Characterization of O-Fe-N-C especially along the axial direction is highly desirable. Built by metal ions and organic ligands, metal–organic The synthesis steps for O-Fe-N-C were schematically displayed frameworks (MOFs) are recognized as one of the most prom- in Figure 1A. In brief, the Fe-IRMOF-3 was firstly prepared ising templates for the targeted creation of MN4 sites.[24–28] by adding FeCl3 into IRMOF-3. Afterward, the O-Fe-N-C was More importantly, the coordination of heteroatoms (e.g., O, S, obtained through pyrolysis of the Fe-IRMOF-3 under argon P, etc.) can be easily achieved after pyrolysis via the rational atmosphere. A reference Fe-N-C sample without O coordination design of organic linkers with functional groups.[25] Herein, was also synthesized through pyrolysis of an Fe-based zeolitic Figure 1. A) Schematic illustration of synthesis of O-Fe-N-C. B) XRD patterns of O-Fe-N-C and Fe-N-C samples. C) Raman spectra of O-Fe-N-C. D,E) High-resolution XPS N 1s spectrum, as well as F,G) O 1s spectrum of O-Fe-N-C and Fe-N-C samples. Adv. Funct. Mater. 2022, 32, 2111446 2111446 (2 of 8) © 2022 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH www.advancedsciencenews.com www.afm-journal.de imidazolate framework (Fe-ZIF-8) precursor prepared by adding could be observed (Figure S2, Supporting Information). Further- FeCl3 into N-rich ZIF-8. Two main peaks at around 25° and 43° more, AC HAADF STEM was used for directly detecting the (Figure 1B), which belong to the (002) and (100) planes of gra- single Fe atoms distribution thanks to the different Z contrast phitic carbon, were observed in O-Fe-N-C samples, indicating among Fe, N, O, and C elements. The representative HAADF the successful conversion of MOFs into carbon-based materials STEM images showed isolated starry spots densely planted in without the metal, metal oxide, or metal carbide impurity struc- the graphitic structure, which confirmed that Fe was atomically tures.[29,30] Similar result was also observed on the X-ray dif- dispersed in the O-Fe-N-C sample (Figure 2A,B and Figure S3, fraction (XRD) patterns of Fe-N-C sample. The Raman spectra Supporting Information). Notice that the graphitic structure of O-Fe-N-C samples, featuring peaks at around 1365 cm−1 (D showed the typical (001) basal planes, which bend forming lay- band) and 1590 cm−1 (G band), further confirmed the struc- ered structures. In addition, the Fe atoms were mainly placed ture of graphitic carbon (Figure 1C).[31,32] X-ray photoelectron along these basal planes in a perfect single metal atom distribu- spectroscopy (XPS) analysis was performed to investigate the tion (see in Figure 2B the red arrows pointing to the C-graphitic chemical composition and elemental states of different sam- (001) planes, while the orange circles highlight the presence of ples. The high-resolution N 1s spectra obtained on both the single Fe atoms). Energy dispersive X-rays spectroscopy (EDS) O-Fe-N-C and Fe-N-C samples demonstrate the presence of pyri- analyses revealed that Fe, N, O, and C were homogeneously dis- dinic (398.6 eV), pyrrolic (401 eV), graphitic (402 eV), and Fe-Nx persed on the O-Fe-N-C sample (Figure 2C and middle panel in (399.4 eV) species (Figure 1D,E).[23,30] In contrast to Fe-N-C Figure 2, as well as Figure S4, Supporting Information). As a (Figure 1G), an additional peak at 530 eV, corresponding to the comparison, the morphology, elemental distribution and struc- FeO bond, was observed on sample O-Fe-N-C (Figures 1F), tural HAADF STEM characterization of Fe-N-C and N-C sam- which indicated that the Fe-O chelation in O-Fe-N-C sample was ples is shown in Figures S5 and S6 (Supporting Information). retained after calcination.[22,23,33] The detailed local chemical environment and electronic The O-Fe-N-C sample was further characterized by aberration states of the Fe atoms in the catalysts was further disclosed via corrected (AC) high-angle annular dark-field scanning transmis- X-ray absorption spectroscopy analyses. Fe K-edge X-ray absorp- sion electron microscopy (HAADF STEM). Multiple areas of the tion near edge structure (XANES) spectra of O-Fe-N-C sample O-Fe-N-C sample were examined and only few Fe nanoparticles shifts toward higher binding energy compared to that of a Figure 2. A) Atomic resolution aberration-corrected HAADF STEM image obtained on a representative region of O-Fe-N-C catalyst. B) Magnified detail from (A) where the graphitic (001) planes can be observed (red arrows). Notice that dispersed Fe atoms (some of the highlighted with orange circles) are placed along the graphitic structures. C) HAADF STEM image and (middle panel) representative EDS chemical composition maps in a selected region of the O-Fe-N-C sample. D) Fe K-edge XANES spectra, E) Fourier transformation of the EXAFS spectra at R space. F) Corresponding EXAFS fitting curves for the O-Fe-N-C sample (Fe, O, N, and C atoms are represented in red, blue, orange, and black, respectively). Adv. Funct. Mater. 2022, 32, 2111446 2111446 (3 of 8) © 2022 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH www.advancedsciencenews.com www.afm-journal.de standard Fe foil (Figure 2D), suggesting a positive charge state The intrinsic activity of the catalysts was further disclosed by the of Fe atoms in the O-Fe-N-C catalyst.[34,35] Furthermore, the mass activities of N-C, Fe-N-C and O-Fe-N-C at −0.50 V versus intrinsic structure of the reactive sites was corroborated by the RHE. Compared to N-C (0.68 A g−1) and Fe-N-C (2.2 A g−1), Fourier transformed (FT) k3-weighted χ(k)-function of the Fe a much higher mass activity of 4.4 A g−1 was obtained for O-Fe- K-edge extended X-ray absorption fine structure (EXAFS) in R N-C, revealing the axial O coordination plays a key role toward space (Figure 2E). The dominant peak centered at around 1.5 Å CO2 RR. The smaller Tafel slope, of 29 mV dec−1, for O-Fe-N-C for O-Fe-N-C sample was attributed to the light backscattering catalyst further indicates its favorable kinetics for the formation induced by light atoms (N, O, or C) situated in the first coor- of CO than those of Fe-N-C (35 mV dec−1) (Figure S8, Supporting dination shell of the absorbing metal. The slight presence of a Information).[44–45] peak at ≈2.27 Å, corresponding to the FeFe bond, confirmed To prove that the excellent CO2 RR performance of the a co-existence of the atomically isolated Fe dispersion with a O-Fe-N-C catalyst was attributed to the highly dispersed modicum presence of Fe nanoparticles, in accordance with the single active sites rather than Fe nanoparticles or agglomer- HAADF STEM results (Figure S2, Supporting Information).[14,31] ates, the O-Fe-N-C catalyst was treated with acid to remove The Fe K-edge EXAFS spectra were then fitted with the model the Fe nanoparticles. Inductively coupled plasma atomic structures depicted in the insets of Figure 2F. The structural emission spectro­ scopy was performed to evaluate the con- parameters obtained from the fittings are shown in Table S1 tent of Fe in the O-Fe-N-C catalyst with and without the acid (Supporting Information), including the coordination number treatment (Table S2, Supporting Information). Compared (CN) and different bond distances. The optimized fitting results to the total Fe content in the O-Fe-N-C catalyst without acid for O-Fe-N-C sample showed a CN-value of 5.1 and a mean bond treatment (0.73 %), the content of Fe single atoms embedded length of 2.02 Å. Therefore, there are five coordination atoms in the O-Fe-N-C sample after acid treatment, decreased to around the Fe atoms in O-Fe-N-C sample, which differs from the 0.41 % (wt %), suggesting that the Fe nanoparticles were suc- previous reported M-N4 sites derived from ZIF-8.[8,36–40] Experi- cessfully removed. In contrast to the sharp decrease of the mentally and theoretically, metal-nitrogen bonds are more likely amount of Fe, a neglectful depression of the catalytic activity than metal-carbon or metal-oxygen ones to form in-plane FeN4 for O-Fe-N-C after the acid treatment process was observed, as sites in the first coordination sphere.[17,41] The higher average CN- shown in Figure S9 (Supporting Information), revealing the value of 5 for Fe-N4 catalysts strongly suggests that one axial O dominant impact of the exposed Fe single atoms on the high atom is coordinated in the axial direction of the FeN4 moieties, activity and selectivity. In addition, no Fe reduction/oxidation resulting in coordinatively saturated iron cations, in line with the redox peaks for Fe appeared in the CV curve for the O-Fe-N-C high oxophilicity of Fe.[33,41,42] In this way, all the results obtained sample (Figure S10, Supporting Information). In combination on the O-Fe-N-C sample revealed the presence of four in-plane with the HAADF STEM, electron energy loss spectro­scopy ele- nitrogen atoms and one oxygen atom as an axial coordination. mental mapping, and high-resolution TEM (HRTEM) results, it could be corroborated that the Fe clusters were rigorously encapsulated by a few layers of carbon, which would encumber 2.2. CO2 RR in H-Type Cell the interaction between the Fe nanoparticles and the electrolyte, resulting in an inactive performance of these Fe nanoparticles The catalytic performance of the different samples was studied (Figures S11 and S12, Supporting Information). A long-term systematically in 0.5 m NaHCO3. First, the cyclic voltammetry stability test for the O-Fe-N-C electrocatalyst was performed (CV) curves were carried out to roughly evaluate the electro- for 30 h at a constant −0.50 V versus RHE cathode potential. catalytic CO2 RR performance. As shown in Figure S7 (Sup- The outlet gases were analyzed every 1 h by GC for calculating porting Information), upon saturating the solution with CO2, the corresponding FE of CO. The current density of O-Fe-N-C the increased reduction current on all samples suggested an effi- maintained a steady value of ≈−4.5 mA cm−2 with no significant cient catalytic performance toward CO2 RR. Compared to N-C decay (Figure 3F) during the 30 h test. The corresponding FE without Fe atoms, the total current densities of O-Fe-N-C and of CO decreased slightly to 88% after 30 h stability test and no Fe-N-C increased significantly at a wide range of applied poten- liquid product was detected by 1H nuclear magnetic resonance tials, indicating the crucial role of the Fe metal center in CO2 (1H-NMR) spectroscopy (Figure S13, Supporting Information). RR (Figure 3A).[43] As expected, the best faradaic efficiency (FE) toward CO in a potential range from −0.40 to −0.70 V versus RHE was obtained for O-Fe-N-C, reaching the maximum FE 2.3. Computational Studies (CO) of 95% at −0.50 V versus RHE (Figure 3B), which is supe- rior to those of previously reported Fe-N-C-based catalysts and To further understand the intrinsic activity of the axial O group comparable to the state-of-the-art SACs even at lower overpoten- for CO2 RR, DFT calculations were performed to calculate the tial (Table S4, Supporting Information). The decreasing trend of free energies of possible intermediates in the reaction pathways FE (CO) for all samples as the potential shifted to more negative from CO2 to CO by using the computational hydrogen electrode values mainly stems from the dominance of the H2 evolution model and parameters reported in the literature.[46–49] As the over the CO2 RR (Figure 3C). The sums of FE for CO and H2 are counterpart, we created a simulation model with an Fe atom ≈100% for all samples, which indicates that no other products coordinated with 4 N atoms (tetra-nitrogen) by replacing six C were produced. The potential-dependent CO/H2 partial current atoms in a graphene surface to represent the reported normal densities were also calculated based on the total current densi- FeN4 catalysts.[50] For the O-Fe-N-C catalyst, an axial OH ties and the corresponding FE (CO) and FE (H2) (Figure 3D,E). ligand was added to coordinate with the Fe single atom in the Adv. Funct. Mater. 2022, 32, 2111446 2111446 (4 of 8) © 2022 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH www.advancedsciencenews.com www.afm-journal.de Figure 3. A) Total current density, B) FE of CO at various potentials, C) current density for CO production, D) FE of H2 at various potentials, E) current density for H2 production on N-C, O-Fe-N-C, and Fe-N-C. F) Stability test of O-Fe-N-C at −0.50 V versus RHE. simulated FeN4 catalyst model. The optimized structures and smaller ΔG of the O-Fe-N-C catalyst (0.47 eV) than the normal the optimal adsorption configurations of reaction intermediates FeN4 (0.89 eV) indicated a superior catalytic activity for CO con- are shown in Figure 4A,B. There are three elementary steps and version from a thermodynamic perspective. In addition, HER as two important intermediates (COOH* and CO*) involved in the a side reaction was also considered, and its corresponding free CO2 RR process. Their corresponding free energy profiles at a energy profiles were shown in Figure 4E. It can be concluded potential of 0 V versus RHE are shown in Figure 4C. The ΔG that the HER is less active on O-Fe-N-C than that on normal for the formation of COOH* over O-Fe-N-C and normal FeN4 FeN4. All these results are in good agreement with the experi- were calculated to be 0.48 and 0.15 eV, respectively. The ΔG for ments. The projected density of states (PDOS) of O-Fe-N-C and the dissociation of COOH* assisted by proton-electron transfer normal FeN4 surfaces were calculated to investigate the origin of to produce CO* and H2O is downhill on both catalyst models. the difference between adsorption of the reaction intermediates As for the final step of CO desorption, the ΔG over O-Fe-N-C is over these two catalysts, and the results are shown in Figure 4F. 0.47 V, which is significantly lower than that over normal FeN4 Compared to normal FeN4, the d-band center of Fe-3d orbits (0.89 eV). Under these circumstances, it is obvious that the shifted to a more negative value with introduction of an axial O potential determining step (PDS) is COOH* formation (ΔG = group, which leads to weaker CO adsorption over the O-Fe-N-C, 0.48 eV) on O-Fe-N-C, while the CO desorption step is more dif- and thus improve the selectivity toward CO.[51] ficult on normal FeN4 (ΔG = 0.89 eV). As the potential became more negative (−0.48 V vs RHE), the ΔG for the formation of COOH* decreased, whilst the ΔG for the non-electrochemical 3. Conclusions step of CO desorption remained unchanged on both models (Figure 4D). Consequently, CO desorption became the most dif- In summary, the axial O-coordinated FeN4 active site was pre- ficult step on both catalysts’ surfaces at −0.48 V versus RHE. The cisely constructed by using a self-sacrificing O- and N-rich Adv. Funct. Mater. 2022, 32, 2111446 2111446 (5 of 8) © 2022 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH www.advancedsciencenews.com www.afm-journal.de Figure 4. A,B) Top view and side view optimized adsorption configuration on simulated FeN4 and O-Fe-N-C (Fe, O, N, and C atoms are represented in purple, red, blue, and gray, respectively). C) Free energy profiles for the CO2 RR to CO at 0 V (vs RHE) and D) Free energy profiles for the CO2 RR to CO at −0.48 V (vs RHE). E) Free energy profiles for the HER at 0 V (vs RHE) on simulated FeN4 and O-Fe-N-C. F) Projected d-density of states (PDOS) of Fe over O-Fe-N-C and FeN4 surfaces. IRMOF-3. The obtained FeN4-O sites were experimentally Acknowledgements proved to be much more active than the FeN4 active sites, T.Z, X.H., and H.L contributed equally to this work. ICN2 acknowledges showing an excellent FE (CO) of 95% at a low applied potential funding from Generalitat de Catalunya 2017 SGR 327. IREC acknowledges of −0.50 V versus RHE. Such performance outperforms previ- Generalitat de Catalunya for financial support M2E 2017 SGR 1246. T.Z., ously reported Fe-N-C-based catalysts and most of the reported J.R.M., and J.A. thank support from the project NANOGEN (PID2020- single atom catalysts at lower overpotential. Meanwhile, DFT 116093RB-C43), funded by MCIN/AEI/10.13039/501100011033/. ICN2 was supported by the Severo Ochoa program from Spanish MINECO simulations revealed that the axial O coordination could not (Grant No. SEV-2017-0706). ICN2 and IREC were funded by the CERCA only weaken the binding energy of CO but also inhibit the com- Programme/Generalitat de Catalunya. This project received funding from petitive hydrogen evolution. This work opens a new way for the European Union’s Horizon research and innovation programme engineering the coordination of FeN4 sites via heteroatom-rich under Grant Agreement No. 823717-ESTEEM3. J.L acknowledges funding MOFs precursors. from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodovska-Curie Grant Agreement No 101030637. X.Z. acknowledges funding from FWO project (12ZV320N). Part of the present work was performed in the framework of Universitat Supporting Information Autònoma de Barcelona Materials Science Ph.D. program. T.Z. and X.H. thank China Scholarship Council for scholarship support (201706180028, Supporting Information is available from the Wiley Online Library or 201804910551). The authors thank Guillaume Sauthier, Javier Saiz, and from the author. Yunhui Yang for the XPS spectrum, FTIR, and 1H-NMR tests. Adv. Funct. Mater. 2022, 32, 2111446 2111446 (6 of 8) © 2022 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH www.advancedsciencenews.com www.afm-journal.de Conflict of Interest [19] Y. Pan, R. Lin, Y. Chen, S. Liu, W. Zhu, X. Cao, W. Chen, K. Wu, W.-C. Cheong, Y. Wang, L. Zheng, J. Luo, Y. Lin, Y. Liu, C. Liu, J. Li, The authors declare no conflict of interest. Q. Lu, X. Chen, D. Wang, Q. Peng, C. Chen, Y. Li, J. Am. Chem. Soc. 2018, 140, 4218. [20] H. Chen, X. Guo, X. Kong, Y. Xing, Y. Liu, B. Yu, Q.-X. Li, Z. Geng, R. Si, J. Zeng, Green Chem. 2020, 22, 7529. Data Availability Statement [21] Qu, S. J.i , Y. Chen, D. Wang, Y. Li, Chem. Sci. 2021, 12, 4201. The data that support the findings of this study are available from the [22] Z. Chen, A. Huang, K. Yu, T. Cui, Z. Zhuang, S. Liu, J. Li, R. Tu, corresponding author upon reasonable request. K. Sun, X. Tan, J. Zhang, D. Liu, Y. Zhang, P. Jiang, Y. Pan, C. Chen, Q. Peng, Y. Li, Energy Environ. Sci. 2021, 14, 3430. [23] X. Wang, Y. Pan, H. Ning, H. Wang, D. Guo, W. Wang, Z. Yang, Q. Zhao, B. Zhang, L. Zheng, J. Zhang, Appl. Catal., B 2020, 266, Keywords 118630. CO generation, CO2 electroreduction, FeN4-O active sites, metal–organic [24] S. Ji, Y. Chen, X. Wang, Z. Zhang, D. Wang, Y. Li, Chem. Rev. 2020, frameworks, single atom catalysts 120, 11900. [25] Y.-S. Wei, M. Zhang, R. Zou, Q. Xu, Chem. Rev. 2020, 120, Received: November 10, 2021 12089. Revised: December 16, 2021 [26] C.-C. Hou, H.-F. Wang, C. Li, Q. Xu, Energy Environ. Sci. 2020, 13, Published online: January 22, 2022 1658. [27] L. Jiao, H.-L. Jiang, Chem 2019, 5, 786. [28] L. Zou, Y.-S. Wei, C.-C. Hou, C. Li, Q. Xu, Small 2021, 17, 2004809. [1] Q. Guo, F. Liang, X.-B. Li, Y.-J. Gao, M.-Y. Huang, Y. Wang, S.-G. Xia, [29] L. Ye, Y. Ying, D. Sun, Z. Zhang, L. Fei, Z. Wen, J. Qiao, H. Huang, X.-Y. Gao, Q.-C. Gan, Z.-S. Lin, C.-H. Tung, L.-Z. Wu, Chem 2019, 5, Angew. Chem., Int. Ed. 2020, 59, 3244. 2605. [30] W. Ren, X. Tan, W. Yang, C. Jia, S. Xu, K. Wang, S. C. Smith, C. Zhao, [2] H. B. Yang, S.-F. Hung, S. Liu, K. Yuan, S. Miao, L. Zhang, Angew. Chem., Int. Ed. 2019, 58, 6972. X. Huang, H.-Y. Wang, W. Cai, R. Chen, J. Gao, X. Yang, W. Chen, [31] X. Ao, W. Zhang, Z. Li, J.-G. Li, L. Soule, X. Huang, W.-H. Chiang, Y. Huang, H. M. Chen, C. M. Li, T. Zhang, B. Liu, Nat. Energy 2018, H. M. Chen, C. Wang, M. Liu, X. C. Zeng, ACS Nano 2019, 13, 3, 140. 11853. [3] D. U. Nielsen, X.-M. Hu, K. Daasbjerg, T. Skrydstrup, Nat. Catal. [32] C. Zhao, Y. Wang, Z. Li, W. Chen, Q. Xu, D. He, D. Xi, Q. Zhang, 2018, 1, 244. T. Yuan, Y. Qu, J. Yang, F. Zhou, Z. Yang, X. Wang, J. Wang, J. Luo, [4] B. You, Y. Sun, Acc. Chem. Res. 2018, 51, 1571. Y. Li, H. Duan, Y. Wu, Y. Li, Joule 2019, 3, 584. [5] Q. Cheng, K. Mao, L. Ma, L. Yang, L. Zou, Z. Zou, Z. Hu, H. Yang, [33] L. Gong, H. Zhang, Y. Wang, E. Luo, K. Li, L. Gao, Y. Wang, Z. Wu, ACS Energy Lett. 2018, 3, 1205. Z. Jin, J. Ge, Z. Jiang, C. Liu, W. Xing, Angew. Chem., Int. Ed. 2020, [6] J. Huang, R. Buonsanti, Chem. Mater. 2019, 31, 13. 59, 13923. [7] J. Pei, T. Wang, R. Sui, X. Zhang, D. Zhou, F. Qin, X. Zhao, [34] Y. Lou, J. Liu, M. Liu, F. Wang, ACS Catal. 2020, 10, 2443. Q. Liu, W. Yan, J. Dong, L. Zheng, A. Li, J. Mao, W. Zhu, W. Chen, [35] J. Wang, R. You, C. Zhao, W. Zhang, W. Liu, X.-P. Fu, Y. Li, F. Zhou, Z. Zhuang, Energy Environ. Sci. 2021, 14, 3019. X. Zheng, Q. Xu, T. Yao, C.-J. Jia, Y.-G. Wang, W. Huang, Y. Wu, ACS [8] C. Zhao, X. Dai, T. Yao, W. Chen, X. Wang, J. Wang, J. Yang, S. Wei, Catal. 2020, 10, 2754. Y. Wu, Y. Li, J. Am. Chem. Soc. 2017, 139, 8078. [36] Q. Wang, T. Ina, W.-T. Chen, L. Shang, F. Sun, S. Wei, D. Sun- [9] T. Zhang, X. Han, H. Liu, M. Biset-Peiró, X. Zhang, P. Tan, P. Tang, Waterhouse, S. G. Telfer, T. Zhang, G. I. N. Waterhouse, Sci. Bull. B. Yang, L. Zheng, J. R. Morante, J. Arbiol, Energy Environ. Sci. 2021, 2020, 65, 1743. 14, 4847. [37] P. Yin, T. Yao, Y. Wu, L. Zheng, Y. Lin, W. Liu, H. Ju, J. Zhu, X. Hong, [10] A. S. Varela, W. Ju, A. Bagger, P. Franco, J. Rossmeisl, P. Strasser, Z. Deng, G. Zhou, S. Wei, Y. Li, Angew. Chem., Int. Ed. 2016, 55, ACS Catal. 2019, 9, 7270. 10800. [11] M. Li, H. Wang, W. Luo, P. C. Sherrell, J. Chen, J. Yang, Adv. Mater. [38] X. X. Wang, D. A. Cullen, Y.-T. Pan, S. Hwang, M. Wang, 2020, 32, 2001848. Z. Feng, J. Wang, M. H. Engelhard, H. Zhang, Y. He, Y. Shao, [12] G. Zhang, Y. Jia, C. Zhang, X. Xiong, K. Sun, R. Chen, W. Chen, D. Su, K. L. More, J. S. Spendelow, G. Wu, Adv. Mater. 2018, 30, Y. Kuang, L. Zheng, H. Tang, W. Liu, J. Liu, X. Sun, W.-F. Lin, H. Dai, 1706758. Energy Environ. Sci. 2019, 12, 1317. [39] J. Li, M. Chen, D. A. Cullen, S. Hwang, M. Wang, B. Li, K. Liu, [13] T. N. Nguyen, M. Salehi, Q. V. Le, A. Seifitokaldani, C. T. Dinh, ACS S. Karakalos, M. Lucero, H. Zhang, C. Lei, H. Xu, G. E. Sterbinsky, Catal. 2020, 10, 10068. Z. Feng, D. Su, K. L. More, G. Wang, Z. Wang, G. Wu, Nat. Catal. [14] X.-M. Hu, H. H. Hval, E. T. Bjerglund, K. J. Dalgaard, M. R. Madsen, 2018, 1, 935. M.-M. Pohl, E. Welter, P. Lamagni, K. B. Buhl, M. Bremholm, [40] L. Jiao, G. Wan, R. Zhang, H. Zhou, S.-H. Yu, H.-L. Jiang, Angew. M. Beller, S. U. Pedersen, T. Skrydstrup, K. Daasbjerg, ACS Catal. Chem., Int. Ed. 2018, 57, 8525. 2018, 8, 6255. [41] J. Li, P. Pršlja, T. Shinagawa, A. J. M. Fernández, F. Krumeich, [15] F. Pan, H. Zhang, K. Liu, D. Cullen, K. More, M. Wang, Z. Feng, K. Artyushkova, P. Atanassov, A. Zitolo, Y. Zhou, R. García- G. Wang, G. Wu, Y. Li, ACS Catal. 2018, 8, 3116. Muelas, N. López, J. Pérez-Ramírez, F. Jaouen, ACS Catal. 2019, 9, [16] X. Yang, J. Cheng, X. Xuan, N. Liu, J. Liu, ACS Sustainable Chem. 10426. Eng. 2020, 8, 10536. [42] Y. Chen, S. Ji, Y. Wang, J. Dong, W. Chen, Z. Li, R. Shen, L. Zheng, [17] Y. Wu, C. Chen, X. Yan, X. Sun, Q. Zhu, P. Li, Y. Li, S. Liu, J. Ma, Z. Zhuang, D. Wang, Y. Li, Angew. Chem., Int. Ed. 2017, 56, Y. Huang, B. Han, Angew. Chem., Int. Ed. 2021, 60, 20803. 6937. [18] H. Zhang, J. Li, S. Xi, Y. Du, X. Hai, J. Wang, H. Xu, G. Wu, J. Zhang, [43] O. S. Bushuyev, P. De Luna, C. T. Dinh, L. Tao, G. Saur, J. Lu, J. Wang, Angew. Chem., Int. Ed. 2019, 58, 14871. J. van de Lagemaat, S. O. Kelley, E. H. Sargent, Joule 2018, 2, 825. Adv. Funct. Mater. 2022, 32, 2111446 2111446 (7 of 8) © 2022 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH www.advancedsciencenews.com www.afm-journal.de [44] J. Gu, C.-S. Hsu, L. Bai, H. M. Chen, X. Hu, Science 2019, 364, [48] H. Liu, J. Liu, B. Yang, Phys. Chem. Chem. Phys. 2020, 22, 9600. 1091. [49] J. Li, A. Slassi, X. Han, D. Cornil, M.-H. Ha-Thi, T. Pino, [45] C. W. Li, M. W. Kanan, J. Am. Chem. Soc. 2012, 134, 7231. D. P. Debecker, C. Colbeau-Justin, J. Arbiol, J. Cornil, M. N. Ghazzal, [46] J. K. Nørskov, J. Rossmeisl, A. Logadottir, L. Lindqvist, J. R. Kitchin, Adv. Funct. Mater. 2021, 31, 2100994. T. Bligaard, H. Jónsson, J. Phys. Chem. B 2004, 108, 17886. [50] J. Li, J. Liu, B. Yang, J. Energy Chem. 2021, 53, 20. [47] H. Liu, J. Liu, B. Yang, Phys. Chem. Chem. Phys. 2019, 21, [51] X. Liu, H. Liu, C. Chen, L. Zou, Y. Li, Q. Zhang, B. Yang, Z. Zou, 9876. H. Yang, Nano Res. 2019, 12, 1651. Adv. Funct. Mater. 2022, 32, 2111446 2111446 (8 of 8) © 2022 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH