Construction of Synergistic Co and Cu Diatomic Sites for Enhanced Higher Alcohol Synthesis

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE24 May 2022Construction of Synergistic Co and Cu Diatomic Sites for Enhanced Higher Alcohol Synthesis Gaofeng Chen, Olga A. Syzgantseva, Maria A. Syzgantseva, Shuliang Yang, Guihua Yan, Li Peng, Changyan Cao, Wenxing Chen, Zhiwei Wang, Fengjuan Qin, Tingzhou Lei, Xianhai Zeng, Lu Lin, Weiguo Song and Buxing Han Gaofeng Chen College of Energy, Xiamen University, Xiamen 361102 Google Scholar More articles by this author , Olga A. Syzgantseva Department of Chemistry, Lomonosov Moscow State University, Moscow 119991 Google Scholar More articles by this author , Maria A. Syzgantseva Department of Chemistry, Lomonosov Moscow State University, Moscow 119991 Google Scholar More articles by this author , Shuliang Yang *Corresponding author: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] College of Energy, Xiamen University, Xiamen 361102 Fujian Engineering and Research Center of Clean and High-Valued Technologies for Biomass; Xiamen Key Laboratory of Clean and High-Valued Utilization for Biomass, Xiamen 361102 Google Scholar More articles by this author , Guihua Yan College of Energy, Xiamen University, Xiamen 361102 Google Scholar More articles by this author , Li Peng College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005 Google Scholar More articles by this author , Changyan Cao Beijing National Laboratory for Molecular Science, CAS Research/Education Center for Excellence in Molecular Sciences, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190 University of Chinese Academy of Sciences, Beijing 100049 Google Scholar More articles by this author , Wenxing Chen *Corresponding author: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] Energy & Catalysis Center, School of Materials Science and Engineering, Beijing Institute of Technology, Beijing 100081 Google Scholar More articles by this author , Zhiwei Wang School of Environmental Engineering, Henan University of Technology, Zhengzhou 450001 Google Scholar More articles by this author , Fengjuan Qin Energy & Catalysis Center, School of Materials Science and Engineering, Beijing Institute of Technology, Beijing 100081 Google Scholar More articles by this author , Tingzhou Lei Institute of Urban and Rural Mining, Changzhou University, Changzhou 213164 Google Scholar More articles by this author , Xianhai Zeng *Corresponding author: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] College of Energy, Xiamen University, Xiamen 361102 Fujian Engineering and Research Center of Clean and High-Valued Technologies for Biomass; Xiamen Key Laboratory of Clean and High-Valued Utilization for Biomass, Xiamen 361102 Google Scholar More articles by this author , Lu Lin College of Energy, Xiamen University, Xiamen 361102 Fujian Engineering and Research Center of Clean and High-Valued Technologies for Biomass; Xiamen Key Laboratory of Clean and High-Valued Utilization for Biomass, Xiamen 361102 Google Scholar More articles by this author , Weiguo Song Beijing National Laboratory for Molecular Science, CAS Research/Education Center for Excellence in Molecular Sciences, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190 University of Chinese Academy of Sciences, Beijing 100049 Google Scholar More articles by this author and Buxing Han Beijing National Laboratory for Molecular Science, CAS Research/Education Center for Excellence in Molecular Sciences, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190 University of Chinese Academy of Sciences, Beijing 100049 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.022.202201930 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Higher alcohol synthesis (HAS) from syngas could efficiently alleviate the dependence on the traditional fossil resources. However, it is still challenging to construct high-performance HAS catalysts with satisfying selectivity, space–time yield (STY), and stability. Herein, we designed a diatomic catalyst by anchoring Co and Cu sites onto a hierarchical porous N-doped carbon matrix (Co/Cu–N–C). The Co/Cu–N–C is efficient for HAS and is among the best catalysts reported. With a CO conversion of 81.7%, C2+OH selectivity could reach 58.5% with an outstanding C2+OH STY of 851.8 mg/g·h. We found that the N4–Co1 and Cu1–N4 showed an excellent synergistic effect. The adsorption of CO occurred on the Co site, and the surrounding nitrogen sites served as a hydrogen reservoir for the CO reduction reactions to form CHxCo. Meanwhile, the Cu sites stabilized a CHOCu species to interact with CHxCo, facilitating a barrier-free formation of C2 species, which is responsible for the high selectivity of higher alcohols. Download figure Download PowerPoint Introduction Higher alcohol (C2+OH) synthesis from direct thermocatalytic conversion of syngas (CO and H2) is a promising strategy to alleviate the intensified energy crisis and global environmental issues because the fossil-based syngas route could be replaced by alternative green approaches such as biomass gasification and the electrochemical CO2 reduction reaction.1–4 In our daily life, higher alcohols (HAs) have been applied widely in different fields such as pharmaceuticals, solvents, fuel additives, hydrogen carriers, and so on.1,3,5 In the past decades, significant advances have been made in higher alcohol synthesis (HAS) from syngas with Rh-based,6,7 Mo-based,8,9 modified methanol/Fischer–Tropsch (F–T) synthesis catalysts, and so on.4,10–12 Among them, modified F–T Co–Cu-based catalysts have been considered one of the most attractive candidates owing to its low feedstock cost, abundant reserves, benign reaction condition, and high CO hydrogenation activity.13 The HAS reaction requires “dual-active sites” in close proximity for the Co–Cu catalyst, where Co active sites are active for CO dissociation, alkyl group (CxHy*) formation, and carbon chain propagation, and the Cu species are responsible for the insertion of non-dissociated CO molecule, further facilitating alcohol formation.14,15 Thus, it is crucial to enhance the intimacy of dual-active sites to make full use of the synergistic catalytic effect to improve the yield of HAs. Even though great efforts have been dedicated to fabricate adjacent Co–Cu dual-active-center catalysts derived from oxalate co-precipitation oxides,16 perovskite structure oxides,17 and layered double hydroxides,18 catalyst stability issues, such as the agglomeration and phase separation of active particles, are often inevitable due to the collapse of the catalyst skeleton under the relatively harsh reaction conditions.16,17 For example, recently in 2022, CoFe alloy or α-Fe phases in the CoFe-300 catalyst carburized into ε-carbide and χ-carbide during the HAS reaction process.19 At the same time, the HAS catalysts often suffer decreased selectivity at a relative higher conversion. Thus, it is very hard to obtain high selectivity for alcohol (ROH, such as methanol, ethanol, etc.) in HAS at high conversion,10,16,20–25 which results in low efficiency for HA production with a relatively low space-time yield (STY) of HAs. Therefore, developing novel strategies to construct stable and highly efficient catalysts for HAS is highly demanded. Compared with the bulk catalysts and catalysts with big particles, the atomic-scale single-atom catalysts (SACs) could in theory endow the catalysts with ultrahigh atom utilization efficiency and abundant synergistic sites.26–34 However, it is still problematic for SACs, with only one kind of active site, to catalyze the complicated reactions involving multiple intermediates.35 In contrast to SACs, the diatomic site catalysts (DASCs) possess more flexible and sophisticated active sites, a substantially different coordination environment, and quantum size effects, thereby offering more chances to obtain multiple active sites for a multistep catalytical process.36–39 Particularly, the electronic interactions between adjacent heteroatomic metal species could be tailored to make the catalytic environment suitable for the reaction route involving multiple intermediates.40–43 Moreover, DASCs have enormous potential to improve the intrinsic activity as a result of the multi-functional activity centers with more than one kind of adsorption site. Therefore, designing DASCs with proper metal sites could offer new opportunities for the development of HAS catalysts considering that the interfacial synergy of different active metal sites is vital for HAS to obtain high alcohol selectivity. Herein, we have successfully designed a dual-metal single-atom Co/Cu–N–C catalyst consisting of high-density Co and Cu atoms anchored onto a N-doped porous carbon matrix. The preparation process of Co/Cu–N–C is simple and cost-efficient without using expensive metal precursors. Gratifyingly, Co/Cu–N–C displayed extraordinary thermocatalytic performance for syngas-to-higher alcohols (CO conversion of 81.7%, C2+OH selectivity of 58.5%, C2+OH STY of 851.8 mg/g·h) and excellent stability over 200 h, greatly outperforming most of the reported catalysts for HAS ( Supporting Information Table S8). The in-depth electronic structure analysis revealed that the Co/Cu–N–C possesses high density of states (DOS) centered on both Co and Cu metals near the Fermi level. This enabled an easy electron transfer between these sites, a strong interaction with the adsorbates, a facile accommodation of reduction electrons from adsorbed hydrogen, and a series of barrier-free HAS reaction steps, including a C–C coupling state, thus resulting in good catalytic activity and selectivity of Co/Cu–N–C. This work provides critical insights into the catalyst rational design, structure–activity relationship, and application of hetero-DASCs. To the best of our knowledge, there is no previous report on the application of Co/Cu–N–C DASCs for HAS. We anticipate that the results reported here could move the industrialization process of the HAS reaction forward a step. Experimental Methods Preparation of Co/Cu–N–C In a typical process, 4.0 g sucrose and 40.0 g urea were dissolved in 270 mL 70% (v:v) ethanol–aqueous solution. Cobalt(II) acetate tetrahydrate (0.17 g) and 0.23 g copper(II) acetate monohydrate were dissolved in 30 mL 70% (v:v) ethanol–aqueous solution and added to the above solution, followed by stirring for 2 h. The as-obtained solution was rotary evaporated at 65 °C until the solvent was completely removed. The as-prepared sol-gel mixture was further vacuum dried at room temperature for 24 h. After drying, the sample was thoroughly ground. Subsequently, the powder was transferred into a tube furnace and heated to 300 °C for 1 h, and then pyrolyzed at 700 °C for 2 h with a heating rate of 2 °C/min under a flowing argon atmosphere. The collected black powders were denoted as Co/Cu–N–C and used directly without further treatment. Catalytic reaction experiments The dual-metal single-atom Co/Cu–N–C catalyst was investigated for HAS from syngas in a Microactivity Effi (Micromeritics Instrument Ltd., Norcross, GA, United States) with a tubular fixed-bed reactor. The catalyst was loaded into the middle of the reaction tube. Then, the catalyst was evaluated under 260 °C, 3.0 MPa syngas (60 vol % H2, 30 vol % CO, and 10 vol % N2) with gas hourly space velocity (GHSV) of 6000 h−1. Here, N2 was used as the internal standard. The tail gas containing CO, H2, N2, CO2, CH4, and C2+H was analyzed using an Agilent GC 7890B gas chromatograph equipped with a thermal conductivity detector, a hydrogen flame ionization detector (FID), and a TDX-01 column and HayeSep Q packed column using Ar as the carrier gas. The liquid products collected from a high-precision gas-liquid separator ( Supporting Information Figure S1) were analyzed offline over an Agilent GC 7890B with an HP-5 capillary column and FID using 2-butyl alcohol as the internal standard with Ar as the carrier gas. CO conversion (XCO), carbon-based product selectivity (Si), STY, and Anderson–Schulz–Flory (ASF) distribution of HAs were calculated using the following equations. The carbon/mass balance obtained throughout the performed studies was within 95–99%. X CO = n CO , in − n CO , out n CO , in × 100 % (1) S CO 2 = n CO 2 n CO , in − n CO , out × 100 % (2) S i = N i × n i ∑ ( N i × n i ) × 100 % (3) STY = weight of alcohols product ( g ) weight of catalysts ( g ) × h (4) W n n = ( 1 − α ) 2 a n − 1 → ln V n n = n ln α + c (5) C = M R × N R A + ∑ ( N i × n i ) M R × N R B × 100 % (6)where n CO , in and n CO , out were the moles of CO at the inlet and outlet, respectively. MR was the number of carbon atoms in the reactant; NRB and NRA were the moles of the reactant before and after the reaction. The N i and n i denote the moles and carbon number of products, respectively. Computational details All calculations were performed with density functional theory (DFT) applying the Perdew–Burke–Ernzerhof density functional,44 plane waves with 40 Ry and 320 Py kinetic energy cutoffs for wave function and charge density, respectively, as a basis set and ultrasoft pseudopotentials45,46 for the description of the core region, as implemented in the Quantum ESPRESSO software package.47 Spin polarization was accounted for during the geometry optimizations and electronic structure calculations. A 4 × 4 × 1 Monkhorst–Pack k-point grid was used to sample the reciprocal space. The Co/Cu–N–C atomic structures were constructed from a pristine graphene 6 × 6 unit cell by corresponding substitutions followed by a full relaxation of both cell parameters and atomic positions. A vacuum thickness of at least 10 Å was used in all calculations. The structures were visualized using Visual Molecular Dynamics (VMD)48 software. Results and Discussion Synthesis and structural characterization of Co/Cu–N–C The one-pot complexation annealing strategy for preparation of Co/Cu–N–C DASCs is schematically depicted in Figure 1. The Co/Cu–N–C catalyst was obtained by homogeneously mixing precursors of copper(II) acetate monohydrate, cobalt(II) acetate tetrahydrate, sucrose, and urea followed by calcination under Ar atmosphere at 700 °C for 2 h. The Co2+/Cu2+ ions coordinated with the abundant –OH and –NH2 groups of sucrose and urea through a simultaneous complexation process,49 which was beneficial for the formation of highly dispersed Co and Cu and will further be helpful for the construction of dual atomic catalyst. During the preparation process, sucrose and urea not only acted as the sources of carbon and nitrogen, but also formed the hierarchical porous nanosheet to support the atomically isolated Co and Cu atoms. Figure 1 | Schematic illustration of the synthesis of Co/Cu–N–C dual atomic catalyst. Download figure Download PowerPoint The powder X-ray diffraction (PXRD) patterns of the Co/Cu–N–C exhibit two broad peaks located at ∼26° and ∼44°, which were assigned to the (002) and (101) diffractions of graphitic carbon, respectively ( Supporting Information Figure S2).50 These peaks indicate the presence of amorphous carbon matrix. Moreover, no X-ray diffraction (XRD) peaks of Co and Cu related to metal clusters, nanoparticles, alloys, oxides, or carbides were observed, indicating that the metal species were highly dispersed in the carbon matrix.41,42,51 As revealed by field emission scanning electron microscopy (SEM) (Figure 2a and Supporting Information Figure S3) and transmission electron microscopy (TEM) (Figure 2b), a hierarchical porous nanosheet was obtained. The detailed morphology feature was further identified by scanning TEM (STEM) (Figure 2c). It revealed that the Co/Cu–N–C was composed of N-doped porous carbon without detectable nanoparticles, indicating that the Co and Cu existed in an atomic dispersion state. These results were consistent with the above PXRD analysis result. High-resolution TEM images of Co/Cu–N–C in Figure 2d show an expanded interlayer space of 0.39 nm compared to 0.34 nm for typical graphite, which is due to the doping of metal sites and carbon defects.29 The metal species (single atom or nanoparticle) were further uncovered by aberration-corrected high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) (Figure 2e). The atomically dispersed Co and Cu on the N-doped porous carbon was confirmed by the appearance of bright spots in Co/Cu–N–C based on the different Z-contrast; no clusters or nanocrystals were observed (Figure 2e). The elemental distributions in Co/Cu–N–C were further determined by HAADF-STEM imaging and the corresponding energy dispersive X-ray spectroscopy (EDS) maps. As shown in Figures 2f–2i, the Co, Cu, C, and N elements coexisted on the carbon matrix, and Co and Cu single atoms distributed uniformly on the N-doped porous carbon nanosheets. Inductively coupled plasma optical emission spectroscopy (ICP-OES) analysis revealed that the Co and Cu contents in Co/Cu–N–C were 1.39 and 2.60 wt %, respectively ( Supporting Information Table S1). The Raman spectrum of the Co/Cu–N–C showed the characteristic D band (∼1340 cm−1) and G band (∼1584 cm−1) resulting from carbon lattice defects and sp2-hybridized carbon, respectively ( Supporting Information Figure S4). The ratio values of ID/IG for Co/Cu–N–C was 0.96, indicating high structural defect density introduced by dual-metal single-atom Co/Cu on the N-doped porous carbon nanosheet. Additionally, the N2 adsorption–desorption test at 77 K showed that the Co/Cu–N–C had a Brunauer–Emmet–Teller surface area of 246 m2/g ( Supporting Information Figure S5a). The elemental composition and surface chemical states were further characterized by X-ray photoelectron spectroscopy (XPS). The survey XPS spectra of the Co/Cu–N–C confirmed the presence of Co, Cu, C, N, and O ( Supporting Information Figure S6 and Table S2). The high-resolution C 1s spectrum of the Co/Cu–N–C was deconvoluted into C=C (284.4 eV), C=N (285.5 eV), C–N/C–O (287.3 eV), and C=O (290.2 eV), respectively ( Supporting Information Figure S7a). The N 1s spectra of the Co/Cu–N–C showed the existence of pyridinic N (398.1 eV, 41.3%), Co–N/Cu–N (399.1 eV, 19.7%), pyrrolic N (400.2 eV, 19.7%), graphitic N (401.4 eV, 14.2%), and oxidized N (404.1 eV, 5.1%) species, respectively (Figure 3h, Supporting Information Figure S8a and Table S3).52 Accordingly, the high-resolution Co 2p spectrum of Co/Cu–N–C in Supporting Information Figure S9a exhibited two peaks at 780.5 eV (2p3/2) and 795.9 eV (2p1/2), which could be attributed to the oxidation state of a Co single atom. The peaks of Cu 2p3/2 at 931.6 eV and Cu 2p1/2 at 951.4 eV in the Cu 2p spectrum corresponded to the oxidation state of a Cu single atom ( Supporting Information Figure S9b). Figure 2 | Structural characterizations. (a) SEM, (b) dark-field TEM, (c) STEM and (d) (d) HRTEM images of Co/Cu–N–C (inset is the lattice distance of carbon layer in the selected area). (e) HAADF-STEM image of Co/Cu–N–C (isolated bright dots marked with yellow circles are Co or Cu single atoms). (f) HAADF-STEM image of Co/Cu–N–C and (g) the corresponding EDS elemental mapping. (h) HAADF-STEM images of Co/Cu–N–C and (i) its corresponding EDS elemental mapping with higher magnification at 5 nm scale. Download figure Download PowerPoint For comparison, single metal Co–N–C and Cu–N–C were also prepared following a similar method, but with only cobalt(II) acetate tetrahydrate or copper(II) acetate monohydrate as the metal precursor, respectively. Detailed synthetic procedures are shown in the Supporting Information. HAADF-STEM and EDS elemental mapping analysis of both samples confirmed the similar atomically dispersed features with that of Co/Cu–N–C ( Supporting Information Figures S10 and S11). Inductively coupled plasma optical emission spectroscopy (ICP-OES) analysis revealed the Co–N–C had 1.33 wt % Co loading and Cu–N–C had 2.54 wt % ( Supporting Information Table S1). In contrast, the ratio values of ID/IG for Co–N–C and Cu–N–C are 0.91 and 0.87, indicating lower structural defect density when compared with that of Co/Cu–N–C ( Supporting Information Figure S4). In addition, the N2 adsorption–desorption test at 77 K reported the Co–N–C and Cu–N–C with surface areas of 184 and 168 m2/g, which are lower than that of the Co/Cu–N–C surface area of 246 m2/g ( Supporting Information Figure S5). Figure 3 | Detailed coordinated structure analysis of Co/Cu–N–C, Co–N–C, and Cu–N–C catalysts. (a) Co K-edge XANES spectra of Co/Cu–N–C, Co–N–C, Co foil, CoO, and Co3O4. (b) FT k3-weighted EXAFS spectra for Co K-edge of Co/Cu–N–C, Co–N–C, Co foil, CoO, and Co3O4. (c) WT k3-weighted EXAFS spectra of Co–N–C, Cu–N–C, and Co/Cu–N–C. (d) Cu K-edge XANES spectra of Co/Cu–N–C, Cu–N–C, Cu foil, Cu2O, and CuO. (e) FT k3-weighted EXAFS spectra for Cu K-edge of Co/Cu–N–C, Cu–N–C, Cu foil, Cu2O, and CuO. (f) Corresponding Co K-edge EXAFS fitting curves of Co/Cu–N–C in R space (inset: model of N4–Co–Cu–N4 sites). (g) Corresponding Cu K-edge EXAFS fitting curve of Co/Cu–N–C in R space (inset: model of N4–Co–Cu–N4 sites). (h) High-resolution N 1s XPS spectra of Co/Cu–N–C. Download figure Download PowerPoint Compared with the metal-N content in Co–N–C (13.4%) and Cu–N–C (11.9%), the content of Co-N/Cu-N species of Co/Cu–N–C are higher (19.7%), which is consistent with the ICP-OES result of Co/Cu–N–C with a higher total metal content of 3.99 wt % ( Supporting Information Figure S8 and Table S3). However, the binding energy of Co 3d in Co/Cu–N–C shifted toward a higher valence state than that of the Co–N–C. In contrast, the binding energy of Cu 2p in Co/Cu–N–C shifted to a lower binding energy compared to that of Cu–N–C. These results firmly indicate the presence of a strong electronic interaction between the Co and Cu atoms assisted by N in Co/Cu–N–C. Synchrotron-radiation-based X-ray adsorption fine structure (XAFS) analysis, including X-ray adsorption near-edge spectroscopy (XANES) and extended X-ray adsorption fine structure (EXAFS) analyses, were conducted to interpret the detailed electronic structure and coordination environment of Co/Cu–N–C.40 In the Co K-edge XANES spectra (Figure 3a), the Co/Cu–N–C, Co–N–C, and standard Co species (e.g., Co foil, CoO, and Co3O4) were employed as the control samples for X-ray adsorption spectroscopy (XAS) investigation. It is well known that the adsorption threshold can be applied to monitor the electronic state of the Co. The near-edge adsorption energy of Co sites in Co/Cu–N–C is located between that of the Co foil and CoO. These tendencies disclose that the electronic state of Co in Co/Cu–N–C is between Co0 and Co2+, suggesting that the electronic state of Co atoms in Co/Cu–N–C is close to +1.53 The energy level of the pre-edge peak for Co/Cu–N–C decreased compared with Co–N–C (inset of Figure 3a), implying a small distortion of D4h symmetry for Co atoms in Co/Cu–N–C.54 As shown in the Fourier transform (FT) k3-weighted EXAFS spectra, the prominent peak of Co/Cu–N–C is located at about 1.50 Å stemming from the first shell scattering of the Co–N1 path, without the appearance of the Co–Co scattering at 2.21 Å, confirming the atomically dispersed Co in nitrogen coordination environments (Figure 3b). Interestingly, a shoulder peak located at ∼2.15 Å was observed, which is ascribed to the Co–N2 scattering. Also, the Co–N–C displayed a peak at 1.65 Å (attributed to the Co–N), confirming the existence of the atomically dispersed Co in Co–N–C (Figure 3b). The wavelet transform (WT) plots (Figure 3c and Supporting Information Figure S12) were conducted to further analyze the Co K-edge EXAFS oscillation. The only intensity maximum at about 2.4 Å−1 is ascribed to the Co–N bonds in the Co WT contour plots of Co/Cu–N–C, a little more positive than that of Co–N–C (1.8 Å−1), and the Co–Co signals at about 6.2 Å−1 are not observed (compared to the Co foil, CoO, and Co3O4) in Supporting Information Figure S12, implying the absence of Co–Co bonds. By fitting the EXAFS spectra of Co/Cu–N–C in R space, the local structure of Co was identified. The EXAFS fitting result showed that coordination number is ca. 3.8 and the average bond length of the Co–N bond is ∼2.00 Å in Co/Cu–N–C ( Supporting Information Table S4), indicating that each Co atom in Co/Cu–N–C is coordinated with four N atoms (Figure 3f). Moreover, the Co in Co–N–C is coordinated with four N atoms, as shown in Supporting Information Figure S13a. The corresponding Co K-edge EXAFS k space fitting curves of Co/Cu–N–C and Co–N–C are shown in Supporting Information Figure S14. XAFS measurements were also performed to investigate the coordination environments of Cu in Co/Cu–N–C and Cu–N–C. Several standard Cu species (e.g., Cu foil, Cu2O, and CuO) were employed as references. The Cu K-edge of Co/Cu–N–C is situated at Cu2O and CuO, revealing the electronic state of Cu species is between Cu1+ and Cu2+ (Figure 3d). However, the value is slightly more negative than that obtained for Cu in Cu–N–C (inset of Figure 3d), which is consistent with the XPS results. As shown in the FT k3-weighted EXAFS spectra, one main peak at about 1.50 Å ascribed to Cu–N shell (Cu–N1) scattering could be observed in Co/Cu–N–C, and there is no peak corresponding to the Cu–Cu scattering pathway, indicating a single Cu atomic coordination structure. Additionally, a significant signal at about 2.09 Å just like that of the Cu–N2 scattering was monitored at the Cu K-edge EXAFS (Figure 3e). Only the peak at ∼1.50 Å could be observed in the Cu–N–C, which was ascribed to the Cu-N scattering pathway (Figure 3e). The WT was also conducted to analyze the Cu (Figure 3c and Supporting Information Figure S15) K-edge EXAFS oscillation. The only intensity maximum at about 2.2 Å−1 was assigned to the Cu–N pair in Co/Cu–N–C and Cu–N–C. The Cu–Cu signal at ∼3.4 Å−1, as referenced by the WT of Cu foil, Cu2O, and CuO, was not detected ( Supporting Information Figure S15), suggesting a negligible number of Cu–Cu bonds. The EXAFS fitting result showed that coordination number is ca. 4.0 and the average bond length of Cu–N is ca. 2.00 Å in Co/Cu–N–C and 1.93 Å in Cu–N–C ( Supporting Information Table S5), indicating that each Cu atom in Co/Cu–N–C and Cu–N–C is coordinated with four N atoms (Figure 3g and Supporting Information Figure S13b). The corresponding Cu K-edge EXAFS k space fitting curves of Co/Cu–N–C and Cu–N–C are shown in Supporting Information Figure S16. Based on these results, we conclude that the Co and Cu in Co/Cu–N–C dominantly coexist as hetero-DASCs sites (N4–Co–Cu–N4), separately anchored on the N-doped porous carbon support. Catalytic performance in HAS The catalytic performance of the Co/Cu–N–C DASCs catalyst for CO hydrogenation was evaluated on a Microactivity Effi with a tubular fixed-bed reactor. The parameters including temperature, GHSV, and pressure were screened to explore the optimum reaction conditions. First, the reaction temperature was optimized from 240 to 270 °C with a step of 10 °C. As CO conversion increased with the increasing reaction temperature, the product selectivity of CH4 and CO2 improved concurrently, accompanied by a decreased ROH selectivity (Figure 4a). The yield of ROH reached the highest value at 260 °C. The lower alcohol selectivity at higher temperature may be because of the exothermic reaction of the alcohol production.20 Regarding the variation of the total reaction pressure (Figure 4b), CO conversion increased with increasing pressure, and this was also the case for the alcohol fraction. However, the selectivity for C2+H as well as CH4 and CO2 decreased within the range of total reaction pressure from 1 to 4 MPa, which should be due to the volumetric contraction for HAs synthesis under increasing reaction pressure based on the Le Chatelier’s principle.55 Increasing the GHSV within the range from 4000 h−1 to 10,000 h−1 in steps of 2000 h−1 led to a decrease in CO conversion (Figure 4c). This is ascribed to the shorter residence reaction time with the higher GHSV. As for the product distribution, an increase in the GHSV resulted in a decrease in C1 by-product (CH4 and CO2),