Controlled Growth of Multidimensional Interface for High-Selectivity Ammonia Production

Open AccessCCS ChemistryRESEARCH ARTICLES12 Aug 2022Controlled Growth of Multidimensional Interface for High-Selectivity Ammonia Production Xuchen Zheng, Yurui Xue, Chao Zhang and Yuliang Li Xuchen Zheng CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190 School of Chemical Sciences, University of Chinese Academy of Sciences, Beijing 100190 , Yurui Xue *Corresponding authors: E-mail Address: email protected E-mail Address: email protected CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190 Science Center for Material Creation and Energy Conversion, School of Chemistry and Chemical Engineering, Institute of Frontier and Interdisciplinary Science, Jinan 250100 , Chao Zhang CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190 School of Chemical Sciences, University of Chinese Academy of Sciences, Beijing 100190 and Yuliang Li *Corresponding authors: E-mail Address: email protected E-mail Address: email protected CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190 School of Chemical Sciences, University of Chinese Academy of Sciences, Beijing 100190 https://doi.org/10.31635/ccschem.022.202202189 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail The efficient production of ammonia by reducing nitrates at room temperature and ambient pressure is a promising alternative to the Haber–Bosch process and can effectively overcome the attendant water pollution issues. Herein, a new idea has been realized for rational and selective construction of the sp-carbon–metal–carbon interface, comprised of electronic-donating triple bonds in graphdiyne and electron-withdrawing iron carbides, for a highly efficient nitrate reduction reaction. The as-prepared sp-carbon–metal–carbon interfacial structures greatly increase the charge transfer ability and electrical conductivity of the system. The proposed concept of incomplete charge transfer has demonstrated significantly high selectivity, activity, and stability in catalytic system. The catalyst exhibits high Faradaic efficiency of over >95% and a NH3 yield rate up to 205.5 μmolNH3 cm−2 h−1 in dilute nitrate conditions without any contaminant. Download figure Download PowerPoint Introduction Ammonia (NH3), a fundamental chemical in the production of nitrogen fertilizer, pharmaceuticals, and green carbon-free fuels, is in great demand for the global market and modern industries.1 However, NH3 still suffers from the high economic cost and intensive energy consumption of the traditional Haber–Bosch process, which requires high temperatures (300–500 °C) and high pressures (150–200 atm). Electrochemical nitrogen reduction reaction (NRR), which utilizes water as a proton source and N2 as nitrogen source, provides a promising alternative to synthesize NH3 at ambient temperatures and pressures.2–5 Unfortunately, the efficient production of NH3 through NRR still faces huge obstacles such as how to activate inert nitrogen to break the extremely stable N≡N triple bond and the large kinetic barrier and poor solubility of N2 in water, which result in Faradaic efficiencies (FEs) at high current densities and ammonia yield rates (YNH3) much lower than those of the Haber–Bosch method.6,7 It is urgent and necessary to propose alternative strategies to achieve efficient and large-scale NH3 production without pollution at room temperature and ambient pressure.8 Comparing nitrogenous sources, we can choose target raw materials that can bring us a convenient way to produce NH3. Nitrate (NO3−) is one of the most widespread groundwater contaminants, which gradually endangers ecological environments and causes serious health problems. The electrocatalytic conversion of nitrate to high-value added NH3 through the nitrate reduction reaction (NtRR) not only provides a new route for NH3 production but also overcomes the environmental pollution issues, leading to a win-win scenario. NtRR for NH3 production has many other advantages, such as lower N–O bond dissociation energy (204 kJ mol−1) than N≡N bond (941 kJ mol−1), higher solubility in electrolytes than N2, and abundance in surrounding environments.9 To date, the fundamental and applied research of NtRR electrocatalysis has developed rapidly.10 As the core of research and development in this field is novel catalysts, the emergence of some important catalysts, such as GCC-CoDIM,11 email protected/CNTs,12 and Ni3N/N-C,13 has propelled research in this field. However, the efficient and scalable production of NH3 has not seen significant progress. NtRR has to overcome the following issues during the production process: (1) multistep electron and proton transfer in order to increase the reaction rate and speed up NH3 synthesis; (2) selective stabilization of intermediate *NO2; (3) low concentration of nitrate ions limiting the reaction kinetics and reduces inevitably the overall catalytic efficiency. Based on the systematic analysis of the above scientific issues, the core issue determining the efficient conversion of the reaction is how to produce a transformative catalyst to realize the production of ammonia on a large scale with high selectivity and high yield of NH3. Considering the conversion of the mechanism and process from NO3− to NH3, we suggest that such an active center should be composed of basic units with a strong electron-donor and electron-receptor to form a perfect donor–acceptor (D–A) interface with a semiconductor property and a strong charge transfer ability.14–18 Such a specially structured interface has the potential to generate new functions, especially in the selectivity of reactions of the system. As we develop a rational design, we expect our catalysts to meet the demanding requirements of advanced catalysts for high selectivity, activity, and robustness. The sp and sp2 hybridized graphdiyne (GDY) is a novel two-dimensional porous all-carbon network with large pores, a large π conjugate system, natural band gaps, high intrinsic activity, and stability, and it has shown great advantages and transformative properties in various fields from catalysis to energy conversion and storage and many others.19–40 One of the advantages of GDY is that it can grow reasonably well on any material surface, which means that we can construct D–A structures of active centers with target constituents, valence states, and catalytic properties. The incorporation of GDY not only can effectively chelate metal atoms, resulting in specific incomplete and fast charge transfer between metal atoms and GDY, but it also can hybridize with heteroatoms (such as N and S) causing a more uneven distribution of surface charges, creating more active sites, enhancing the charge transfer ability, and finally forming ideal interface structures with high intrinsic catalytic activity and reaction selectivity. By building such a D–A interface, we can control multiple functions at the same time to achieve comprehensive control of the performance of the catalytic system. Here, we report the construction of the highly selective and active interface of sp-carbon–metal–carbon with special composition and properties to achieve incomplete charge transfer between D and A, which is composed of electron donors with triple bonds in GDY and electron withdrawing of iron carbide. We find that the GDY can effectively modulate the coordination environments of Fe atoms, improve the adsorption/desorption ability of the reactants and reaction intermediates, and therefore significantly enhance the activity of the catalyst. In the reaction, the FE is over 95% and the NH3 yield rate up to 205.5 μmolNH3 cm−2 h−1 at dilute nitrate conditions without any contaminant. The coordination of sp-hybridized C∼Fe characterized by X-ray absorption fine structure (XAFS) probes the structure of carbon–metal–carbon. The incomplete charge transfer of D–A results demonstrates the superior activity and selectivity of nitrate reduction. Surface interrogation mode of scanning electrochemical microscopy (SI-SECM) results further prove that Fe3email protected has faster electron transfer ability and higher reactivity as compared to pure iron carbide. The incorporation of GDY endows the electrocatalysts with excellent long-term stability. Experimental Methods Synthesis of Prussian blue Prussian blue (PB) samples were synthesized through a hydrothermal method. Typically, 168.9 mg potassium ferricyanide and 64 mg sodium citrate were added to 40 mL HCl (0.1 M) solution under magnetic stirring for 10 min. The mixed solution and a piece of carbon cloth (CC; 2 cm × 3 cm) were added to a Teflon-lined stainless steel autoclave and maintained at 120 °C for 3 h. The products were subsequently thoroughly washed by water and ethanol and dried at 80 °C in the vacuum oven. Synthesis of Fe3C The as-synthesized PB samples were placed at the center of the tube furnace and kept at 800 °C (heating rate: 5 °C min−1) under Ar protection for 2 h, followed by being cooled down to room temperature. The produced samples were subsequently cleaned by water and ethanol three times, and dried in an Ar flow. Synthesis of Fe3email protected The freshly prepared Fe3C was added to 30 mL hexaethynylbenzene pyridine solution (0.25 mg mL−1) in the Teflon-lined stainless steel autoclave at 120 °C for 12 h. The produced Fe3email protected sample was thoroughly washed by dimethyl formamide, acetone and water. The as-prepared sample was used directly for electrochemical measurements. Results and Discussion The routes to the controlled growth of the highly selective and active sp-carbon–metal–carbon interface is shown in Figure 1, including (1) the growth of PB on the CC surface41,42; (2) the pyrolysis of PB in an Ar atmosphere to obtain Fe3C; and (3) the construction of the D– A interface by growing GDY on the surface of Fe3C. The scanning electron microscopy (SEM) images in Figure 2a,b show that the PB nanocubes feature smooth surfaces ( Supporting Information Figures S1 and S2) were obtained and uniformly distributed on the CC surface. The Fe3C nanoparticles show wrinkled surfaces and smaller sizes (Figure 2c–f) as compared to those of pristine PB (Figure 2a). Transmission electron microscopy (TEM, Figure 2g) and high-resolution TEM (HRTEM; Figure 2h) images show that the Fe3C is crystalline with lattice spacings of 0.26 and 0.32 nm, corresponding to the (200) and (020) planes of the Fe3C phase, respectively, which was also visible from the corresponding nano-area electron diffraction images ( Supporting Information Figure S3). SEM (Figure 2i,j), TEM (Figure 2k), and HRTEM (Figure 2l,m) images of Fe3email protected confirm the successful growth of GDY on the surface of Fe3C, resulting in a D–A interface between GDY and Fe3C nanoparticles. The interlayer spacing of GDY film is 0.36 nm. Raman results show that Fe3email protected has the characteristic peaks of both Fe3C (around 633 cm−1 reflecting the vibration of Fe-C) and GDY (1968 and 2175 cm−1 corresponding to the vibration of the conjugated diyne linking ( Supporting Information Figure S4), further demonstrating the successful synthesis of Fe3email protected materials (Figure 2n). As the result of the extra sp-hybridized C–Fe coordination, the vibrational peak position of Fe–C bond changes to 633 cm−1. Compared with Fe3C, the Fe 2p X-ray photoelectron spectroscopy (XPS) spectra of Fe3email protected show the negative shift in binding energies (by 0.42 eV), revealing an obvious incomplete charge transfer of electron donor-GDY to electron acceptor-iron carbide (Figure 2o). Note that these results indicate the successful in situ growth of a new D–A interface (Figure 2p), which features incomplete charge transfer between Fe3C and GDY. Figure 1 | Synthesis routes to the Fe3email protected electrodes through a two-step procedure including the controlled synthesis of Fe3C and in situ growth of GDY. Download figure Download PowerPoint Figure 2 | Morphological characterization. (a) SEM image and (b) the model of PB. (c and d) Structural models, (e and f) SEM, (g) TEM, and (h) HRTEM images of Fe3C. (i and j) SEM, (k) TEM, and (l) HRTEM images of images Fe3email protected (m) The interface between GDY film and Fe3C. (n) Raman spectroscopy of Fe3C and Fe3email protected (o) Fe 2p XPS spectra of Fe3C and Fe3email protected (p) Schematic representation of the incomplete charge transfer at the interface between GDY and Fe3C. Download figure Download PowerPoint Thermogravimetric analysis (TGA) results show that initially there is a substantial weight loss (100–650 °C) due to the decomposition of PB.43 The platform observed could be attributed to the phase transition of the sample during the pyrolysis (Figure 3a). X-ray diffraction (XRD) patterns of catalysts prepared under the corresponding pyrolysis temperature were measured to elaborate the change of crystal structure (Figure 3b). The XRD pattern shows the characteristic diffraction peaks of Fe3C (PDF card no.85-1317; Figure 3b, Supporting Information Figures S5 and S6), and no peak could be observed at 20–30 deg, indicating the complete transition of PB to Fe3C. XPS measurements were performed to determine the elemental composition and electronic states of catalysts ( Supporting Information Figure S7).44,45 The calculated area ratio of the sp2-C and sp-C peaks in C 1s XPS spectra of Fe3email protected maintains 0.5 (Figure 3c,d), revealing the structural robustness of GDY. Note that the peak corresponding to the C–Fe is obtained at 283.54 eV in C 1s XPS spectra of Fe3email protected Similarly, Fe–C double peaks at 708.17 and 720.97 eV were also observed from the Fe 2p XPS spectra of Fe3email protected (Figure 3e). The interaction between the sp-carbon and the neighboring iron atoms leads to a higher ratio of the Fe–C bond compared with Fe3C. The double peaks of 712.49 and 723.90 eV could be assigned to Fe2+ while peaks of 710.55 and 725.65 eV correspond to Fe3+. The density functional theory calculations were conducted to study the electron distribution at the interface between Fe3C and GDY. The electron distribution patterns show the obvious charge transfer from the carbon atoms on GDY to neighboring Fe atoms (Figure 3f), resulting in electron accumulation around Fe atoms.46–48 Figure 3g shows the XAFS of the samples. Compared with Fe3C, the adsorption edge and the main energy of Fe3email protected shift to lower energy. Also, the E0 peak in the first-derivative curve of X-ray absorption near edge structure (XANES) clearly indicates that the iron valence of Fe3email protected is higher than Fe foil but lower than Fe3C (Figure 3h). These all confirm the charge transfer from GDY to Fe3C, which is consistent with the XPS results and demonstrate the successful construction of the sp-carbon–metal–carbon interface with a strong electron-donor and electron-acceptor to form a D–A interface. The coordination environment of Fe is further characterized by extended X-ray absorption fine structure (EXAFS). Fe3C shows a predominant peak at ∼1.6 Å corresponding to Fe–C (Figure 3i). The intensity of Fe–C for Fe3email protected is higher than that of Fe3C due to the formation of the sp-carbon–metal–carbon. These results solidly demonstrate the incomplete charge transfer on the interface of the as-prepared D–A interface. Figure 3 | Structural characterizations. (a) TGA of the samples. (b) XRD patterns of Fe3C and Fe3email protected (c and d) C 1s XPS spectra of pristine GDY, Fe3C, and Fe3email protected (e) Fe 2p XPS spectra of Fe3C and Fe3email protected (f) Side and top views of the charge density difference maps of Fe3email protected (g) XANES and (h) the first-derivative curves of the samples. (i) EXAFS spectra of the samples. Download figure Download PowerPoint To gain deeper insights into the relationship of the catalytic performance with the coordination environment of the active sites, we explore the kinetic behavior of nitrate reduction on the catalyst surface by SI-SECM (Figure 4a,b and Supporting Information Figure S8).49,50 First, we prove that the nitrate can be reduced on catalyst substrate and the ammonia can be oxidized on the tip by the substrate generation/tip collection (SG/TC) mode of SECM (Figure 4c). The positive feedback, as we anticipated, is the reduction of current increases with more negative substrate potential. And the feedback on the tip is about the same (Figure 4d). Next, mediator ferrocenemethanol (FcMeOH) is added into the electrolyte to titrate residual charge on the substrate of the SI-SECM test. The active sites on the substrate are activated with a given transient test potential, then turn to the open circuit, and are rapidly titrated with FcMeOH+ generated at the tip. The charge plateaus can be observed in the charge-potential curve in the potential range of −0.1 to −0.5 V (Figure 4e). At potentials of −0.1 to −0.5 V, the active sites on the surface were charged. As the potential decreases to −0.55 V, the inflection point was observed, which indicated that the charge originated from the reaction between the FcMeOH+ and the adsorbed NO3−.51,52 The residual charge was also consumed by the on-site competition between water and NO3− before FcMeOH was generated.53 Thus, we set a time delay before the titration to explore the influence factors of the consumption rate. As shown in Figure 4f, the adsorption rate of nitrate is significantly larger than water on Fe3email protected, which is proposed to be the high FE for ammonia production. The titration curve is fitted by assuming the nitrate adsorption as the first-order kinetic reaction (Figure 4g). The kinetic constants in the testing nitrate concentration ranges (95% and NH3 yield rate up to 205.5 μmolNH3 cm−2 h−1 at dilute nitrate without any contaminant. research has a platform of new structural catalytic materials with high selectivity, high stability, and activity, an for the development of Supporting Information Supporting Information is and and of There is no of to Information This research was as a result of a from the Science of and the Key and of and the Key of the Chinese Academy of Li N2 for Ammonia Synthesis on an with under a and of N2 to Zhang Zhang Synthesis of Ammonia from by under and in the of of to of to Zhang of of to Ammonia under Energy of but Zheng Zhang of Ammonia from of a to Li Zhang for Nitrate Zhang on for Nitrate Zhang for Li of to the in Energy