One-Step Synthesis of Ultrathin Carbon Nanoribbons from Metal–Organic Framework Nanorods for Oxygen Reduction and Zinc–Air Batteries

Open AccessCCS ChemistryRESEARCH ARTICLE1 Jan 2022One-Step Synthesis of Ultrathin Carbon Nanoribbons from Metal–Organic Framework Nanorods for Oxygen Reduction and Zinc–Air Batteries Lianli Zou, Yong-Sheng Wei, Chun-Chao Hou, Miao Wang, Yu Wang, Hao-Fan Wang, Zheng Liu and Qiang Xu Lianli Zou AIST-Kyoto University Chemical Energy Materials Open Innovation Laboratory (ChEM-OIL), National Institute of Advanced Industrial Science and Technology (AIST), Sakyo-ku, Kyoto 606-8501 Google Scholar More articles by this author , Yong-Sheng Wei AIST-Kyoto University Chemical Energy Materials Open Innovation Laboratory (ChEM-OIL), National Institute of Advanced Industrial Science and Technology (AIST), Sakyo-ku, Kyoto 606-8501 Google Scholar More articles by this author , Chun-Chao Hou AIST-Kyoto University Chemical Energy Materials Open Innovation Laboratory (ChEM-OIL), National Institute of Advanced Industrial Science and Technology (AIST), Sakyo-ku, Kyoto 606-8501 Google Scholar More articles by this author , Miao Wang AIST-Kyoto University Chemical Energy Materials Open Innovation Laboratory (ChEM-OIL), National Institute of Advanced Industrial Science and Technology (AIST), Sakyo-ku, Kyoto 606-8501 Google Scholar More articles by this author , Yu Wang AIST-Kyoto University Chemical Energy Materials Open Innovation Laboratory (ChEM-OIL), National Institute of Advanced Industrial Science and Technology (AIST), Sakyo-ku, Kyoto 606-8501 Google Scholar More articles by this author , Hao-Fan Wang AIST-Kyoto University Chemical Energy Materials Open Innovation Laboratory (ChEM-OIL), National Institute of Advanced Industrial Science and Technology (AIST), Sakyo-ku, Kyoto 606-8501 Google Scholar More articles by this author , Zheng Liu Innovative Functional Materials Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), Shimoshidami, Moriyamaku, Nagoya, Aichi 463-8560 Google Scholar More articles by this author and Qiang Xu *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] AIST-Kyoto University Chemical Energy Materials Open Innovation Laboratory (ChEM-OIL), National Institute of Advanced Industrial Science and Technology (AIST), Sakyo-ku, Kyoto 606-8501 Institute for Integrated Cell-Material Sciences (iCeMS), Kyoto University, Sakyo-ku, Kyoto 606-8501 Department of Materials Science and Engineering, SUSTech Academy for Advanced Interdisciplinary Studies and Guangdong-Hong Kong-Macao Joint Laboratory for Photonic-Thermal-Electrical Energy Materials and Devices, Southern University of Science and Technology (SUSTech), Nanshan, Shenzhen, Guangdong 518055 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.021.202101160 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Two-dimensional (2D) carbon nanostructures play a critical role in energy-related applications, but developing facile and efficient strategies to synthesize these kinds of nanostructures is extremely rare. Herein, ultrathin carbon nanoribbons (CNRibs), with a thickness of 2–6 nm and length over 100 nm, have been strategically fabricated via a one-step pyrolysis of one-dimensional (1D) metal–organic framework nanorods (MOF NRods). Manipulating the diameters of MOF NRods will result in the formation of porous carbon nanostructures in 1D or 2D morphologies. Functional CNRibs with N doping or metal active site immobilization have also been studied. The CNRibs decorated with iron nanoclusters and single atoms have been used as excellent catalysts for the oxygen reduction reaction under both alkaline and acidic conditions, as well as zinc–air batteries. This work gives deep insights into the structural evolution from 1D to 2D morphology, providing an efficient approach to fabricate low-dimensional nanomaterials with controllable morphologies and functionalities for electrochemical applications. Download figure Download PowerPoint Introduction The upsurge on the study of low-dimensional nanomaterials can be attributed to their great potential in various applications.1–4 Two-dimensional (2D) carbon nanostructures, including graphene, carbon nanosheets, and carbon nanoribbons (CNRibs), have emerged as a group of new-generation functional materials, and have been extensively studied in the fields of optics, catalysis, energy storage, and so on.5–7 Among these fascinating carbon materials, ultrathin CNRibs constructed by a few layers of carbon atoms not only acquire the exceptional advantages of graphene but also present the anisotropic features of one-dimensional (1D) nanostructures, rendering them unconventional physical and chemical properties.8–10 To date, several methods, including mechanical or chemical exfoliation process,11,12 bottoms-up organic synthesis,13 and electro deposition,14 have been developed to synthesize 2D CNRibs, while the morphologies and compositions of the resulting materials are difficult to be modulated, especially the engineering pores or grafted metal active species on the surface. Additionally, the complex, time-consuming, and low-yield features of these approaches make them unsuitable for industrial production. Thus, establishing cheap and easy-to-achieve methods with controllable morphologies and functionalities is both highly desired and challenging. In recent years, metal–organic frameworks (MOFs)15,16 have been widely used as precursors or templates for fabricating porous carbons,17–21 metal oxides,22,23 and their composites.24–26 General methods to fabricate 2D carbon nanomaterials from MOF precursors are based on a morphology-preserving strategy, which, to a large extent, requires designing MOFs into desired shapes, and subsequently transforms them into 2D carbons via pyrolysis process.27–29 The sonochemical treatment of MOF-derived carbon nanorods (CNRods) with subsequent chemical activation is a promising method to fabricate CNRibs, while multiple steps are still needed.30 Seeking a reliable approach to directly synthesize 2D CNRibs from bulk or 1D MOFs is of immense value, which opens new avenues in developing low-dimensional nanostructures but poses significant challenges. Owing to the decomposition of ligands and aggregation of metals during pyrolysis, thermal transformation of MOFs is accompanied by the partial or complete collapse of their original morphologies,31,32 where the falling fragments might reorganize into a new profile under high pyrolysis temperatures and provide a rare opportunity to tailor the structure of the resultant carbons, such as from 1D MOF nanorods to porous 2D carbon nanostructures. In this work, we have successfully converted 1D MOF NRods into ultrathin CNRibs through a one-step pyrolysis process, based on a facile size-mediated strategy (Scheme 1). The 1D porous CNRods or 2D ultrathin CNRibs can be easily achieved by the carbonization of MOF NRods with controlled diameters. As a result, large-sized zinc-MOF NRods (MOF-NRod-L, 100–300 nm in diameter) would just retain the 1D morphology after pyrolysis, while small-sized MOF NRods (MOF-NRod-S, 20–40 nm in diameter) could be directly transformed into 2D CNRibs with a thickness of only 2–6 nm. To improve the electrochemical application of this kind of nanostructures, N-doped CNRibs decorated with ultrafine metal nanoclusters (<2 nm) and single metal atoms have been developed that show excellent catalytic performance for the oxygen reduction reaction (ORR)33 in both alkaline and acidic conditions, as well as outstanding capabilities for Zn–air batteries.34,35 Scheme 1 | Synthetic strategies for fabricating functional 2D CNRibs from small-sized 1D MOF NRods. Download figure Download PowerPoint Experimental Methods Syntheses MOF-NRod-S Fifty milligrams of 2,5-dihydroxyterephthalic acid and 50 mg zinc acetate dihydrate were dissolved in 15 mL ethanol and 2 mL methanol to form homogeneous solutions, respectively. Then, the zinc-containing solution was poured into the ligand solution and ultrasonicated for about 5 min. Next, the mixture was transferred into a Teflon-lined autoclave, heated at 150 °C for 18 h and cooled down naturally to room temperature (RT). The resulting yellow precipitate (MOF-NRod-S) was collected and washed with ethanol several times and dried in a vacuum oven for further use. Fe-doped MOF-NRod-S Similar to the synthesis of MOF-NRod-S, 1.5 mL methanol containing 40 mg zinc acetate dihydrate was poured into 12 mL ethanol solution with 40 mg 2,5-dihydroxyterephthalic acid. After ultrasonication for about 5 min, 0.5 mg FeCl3•6H2O dissolved in 0.1 mL ethanol was added and mixed homogeneously under stirring. The mixture was transferred into a Teflon-lined autoclave, heated at 150 °C for 18 h, and cooled naturally to RT. The obtained precipitate (Fe/MOF-NRod-S) was washed with ethanol several times and dried in a vacuum oven for further use. For comparison, samples with different amounts of Fe were prepared by adding 0.2, 2.0, and 4.0 mg FeCl3•6H2O. CNRib With a heating rate of 5 °C/min, 0.6 g MOF-NRod-S was placed in a tube furnace and heated at different temperatures for 2 h in argon, and resulting products were denoted as CNRib-T, where T is 400, 500, 600, 700, 800, 900, and 1000 °C. The weights of samples obtained at 1000 °C were about 84 mg, corresponding to a yield of 14 wt %. Fe–N/CNRib To adjust the composition of the carbon nanostructures, N-doped CNRibs were prepared by the pyrolysis of MOF NRods with additional nitrogen sources. The as-prepared Fe/MOF-NRod-S (0.3 g) and melamine (0.9 g) were placed in a tube furnace with melamine upstream of Ar flow. With a ramp rate of 5 °C/min, the temperature was increased from RT to 1000 °C and kept at this temperature for 2 h. The resulting samples (55 mg, corresponding to the yields of 18 wt %) were denoted as Fe–N/CNRib. The inductively coupled plasma (ICP) results confirmed that the Fe content of the samples containing 0.5 mg FeCl3•6H2O was about 1 wt %. Characterizations Powder X-ray diffraction (PXRD) measurements were performed on a Rigaku Ultima IV X-ray diffractometer (Rigaku Corp., Tokyo, Japan) with a Cu Kα source (40 kV, 40 mA). Raman scattering spectra were recorded on a laser Raman microscope system (Nanophoton RAMANtouch, Nanophoton Corp., Osaka, Japan) with an excitation wavelength of 532 nm. Fourier-transform infrared (FTIR) spectroscopy analyses were carried out on a Shimadzu IRTracer-100 (Shimadzu Corp., Kyoto, Japan) in air mode. The specific surface areas and pore structures were analyzed by N2 adsorption/desorption isotherms at 77 K after dehydration under vacuum at 120 °C for 12 h using automatic volumetric adsorption equipment (Belsorp-max, MicrotracBEL Corp., Osaka, Japan). Pore volumes were calculated using the nonlocalized density functional theory (NLDFT) method. Thermogravimetric analysis (TGA) was recorded on a Rigaku Thermo plus EVO2/TG-DTA equipment (Rigaku Corp., Tokyo, Japan). X-ray photoelectron spectroscopy (XPS) analyses were conducted on a KRATOS ULTRA2 instrument (Shimadzu Corp., Kyoto, Japan). Scanning electron microscopy (SEM) analyses were carried out with a scanning electron microscope (JEOL JSMIT100, JEOL Ltd., Tokyo, Japan; Hitachi S-5000, Hitachi Corp., Tokyo, Japan). All transmission electron microscopy (TEM) and two of the high-angle annular dark-field scanning TEM (HAADF-STEM) images were taken using an FEI Tecnai G2 F20 at 200 kV (Thermo Fisher Scientific, OR, USA); other HAADF-STEM images and annular bright-field (ABF)-STEM images were recorded on a JEM-ARM200FC equipped with Corrected Electron Optical Systems (CEOS) Cs correctors at 120 kV. Energy dispersive X-ray spectrometry (EDS) mapping was collected from the JED-2300 attached onto a JEM-ARM200FC (JEOL Ltd., Tokyo, Japan). Atomic force microscopy (AFM) images were recorded on the FastScan Dimension XR (Bruker Corp., MA, USA). Metal contents in samples were detected by ICP-optical emission spectroscopy (ICP-OES) on a Thermo Scientific iCAP6300 (Thermo Fisher Scientific, OR, USA). Electrochemical measurements Electrocatalytic measurements were carried out in a three-electrode cell using a rotating ring-disk electrode (RRDE) constant rotation system (RRDE-3A) with a CHI7088E electrochemical workstation at ambient conditions. A platinum wire and an Ag/AgCl electrode in saturated aqueous KCl solution were used as the counter and reference electrodes, respectively. A catalyst-loaded glassy carbon (GC) rotating disk electrode (RDE, 5 mm in diameter, 0.196 cm2 geometric surface areas) or RRDE (GC disk: 4 mm in diameter; Pt ring: 5 mm ID/7 mm OD) was used as the working electrode. All potentials in this study refer to the reversible hydrogen electrode (RHE; ERHE = EAg/AgCl + 0.059 pH + 0.198 V). Catalyst inks were prepared by dispersing catalysts (2.5 mg) in a 1.0 mL mixture of ethanol–water solution (0.49 mL ethanol, 0.49 mL water, and 20 μL 5%-Nafion solution). After sonication for about 30 min, a certain volume of catalyst ink was dropped onto the GC surface to give a loading of 0.4 mg cm−2 for all samples, except for commercial 20% Pt/C (0.2 mg cm−2). The working electrode was dried at RT naturally and then tested in the Ar- or O2-saturated electrolyte (0.1 M KOH or 0.5 M H2SO4). The linear sweep voltammogram (LSV) data presented in this work deducted the value obtained in Ar-saturated conditions to eliminate the effects of double-layer capacitance (Cdl). Zn–air batteries Zinc–air batteries were assembled with a Zn anode, an aqueous solution with 6 M KOH and 0.2 M zinc acetate, and an air cathode comprised of a gas-diffusion layer, catalyst layer, and a separator to prevent electrolyte leakage. The catalyst layer was prepared by dropping a certain volume of catalyst ink on a nickel foam, forming a circular plane (0.7 cm in diameter) with a loading of 0.5 mg cm−2. The battery was placed at ambient conditions overnight to ensure the electrodes were well-immersed in the electrolyte, and then tested at RT. Results and Discussion Morphology control of MOF NRods and their derivatives For the synthesis of 1D MOF NRods in different diameters, the mass ratio of metals to ligands (RM/L) has been adjusted. Generally, when RM/L was 1∶1, small-sized Zn-MOF NRods with a diameter less than 40 nm could be obtained, while the Zn-MOF NRods with larger sizes (100–300 nm in diameters and lengths of several micrometers) were prepared by adjusting RM/L up to 2∶1 ( Supporting Information Figure S1). PXRD characterization was used to analyze the crystal structure of these MOF NRods, and their diffraction profiles match well with the diffraction curve of Zn-MOF-74 (Figures 1a and 1b).36 After carbonization in argon flow at 1000 °C, PXRD curves of products only show two broad peaks at about 25 and 44°, corresponding to the characteristic peaks of graphitic carbon (002) and (101), respectively.17 Figure 1 | (a) Crystal structure of MOF-74 with Zn (green), O (blue), and C (grey) atomic distribution and the photos (right) of MOF-NRod-L (up) and MOF-NRod-S (below); (b) PXRD, (c) N2 sorption curves of MOF NRods and their corresponding carbon products; (d) SEM images of MOF-NRod-L; (e) SEM images of CNRods; (f) SEM and (g) ABF-STEM images of MOF-NRod-S; (h) SEM and (i) TEM images of CNRibs; (j and k) AFM image of CNRibs with the corresponding height profile along the line scan. Download figure Download PowerPoint Compared with MOF NRods and CNRods, CNRibs exhibit a sharp increase of N2 adsorption at the high relative pressure section (0.9–1.0) in the isotherms (Figure 1c). The Brunauer–Emmett–Teller (BET) surface area of CNRibs is about 1201 m2 g−1, comparable to CNRods with a value of 1480 m2 g−1 ( Supporting Information Table S1). Attributable to the presence of large pores or abundant slits caused by the stacked CNRibs, CNRibs (2.94 cm3 g−1) show a much higher pore volume than that of CNRods (1.33 cm3 g−1).32 It has been demonstrated that the large porosity of electrode materials facilitates mass transport and electrolyte diffusion during catalytic reactions.37,38 The SEM images display that the MOF-NRod-L can retain the original 1D morphology after carbonization, giving porous CNRods with diameters of about 100–200 nm and lengths of 2–3 μm (Figures 1d and 1e and Supporting Information Figure S2). Unexpectedly, the MOF-NRod-S (20–40 nm in diameter) will transform into 2D ultrathin CNRibs with a width distribution between 20 and 50 nm after pyrolysis at the same conditions (Figures 1f–1i and Supporting Information Figure S3). HAADF-STEM and TEM images clearly show the ribbon-like structure of samples with folds along their lengths (200–500 nm), suggesting the excellent flexibility as well as few-layer thickness of CNRibs. The high purity of CNRibs demonstrates the usefulness of this synthesis methodology, which shows potential in scalable production on industrial levels. As shown in AFM images and height profiles (Figures 1j and 1k), CNRibs display an average height between 2 and 6 nm, further demonstrating the ultrathin feature of these CNRibs.30 It is obvious that the ultrathin CNRibs can be directly achieved by using small-sized MOF NRods as precursors, avoiding complex exfoliation or template-removing processes. Thermal evolution of CNRibs from MOF NRods To trace the thermal transformation process of MOF-NRod-S to CNRibs, PXRD analyses were initially performed to monitor the decomposition process of MOF NRods (Figure 2a). Diffraction curves of samples calcined at 300 and 400 °C are different from that of MOF NRods, indicating the change of crystal structure at these temperatures ( Supporting Information Figures S4–S6). By increasing the carbonization temperature, diffraction peaks corresponding to ZnO were observed in samples obtained at 500–800 °C, suggesting the complete decomposition of MOFs to carbons and metal oxides. At 900 °C, the Zn-related diffraction peaks disappeared, demonstrating the reduction and removal of Zn species at high pyrolysis temperatures. Two peaks located at about 25 and 44° illustrate the existence of graphitic carbons, which can be further confirmed from Raman spectra with two bands centered at about 1347 cm−1 (D band) and 1610 cm−1 (G band) ( Supporting Information Figure S7).39 Then, SEM and TEM analyses were used to observe the morphology of intermediates during the carbonization process (Figure 2 and Supporting Information Figures S8–S12). Samples carbonized at 300 °C show many ZnO nuclei, which are evenly distributed in the whole MOF NRods (Figure 2b and Supporting Information Figure S9). With increasing pyrolysis temperature, these nuclei begin to grow and aggregate into large-sized nanoparticles (NPs), which finally move to the surface of MOF-NRods and cause the collapse of 1D MOF NRods, inducing the transformation of rod-like structures into a ribbon-like morphology surrounded by many ZnO NPs (Figure 2c and Supporting Information Figures S8 and S10). Increasing the temperature to 700 °C will further accelerate the growth of ZnO NPs and repair the carbon framework. Note that some ZnO NPs are reduced to metallic Zn species, which are directly evaporated due to the relatively low evaporation temperature. Therefore, most NPs disappear at this temperature and the well-defined CNRibs with a few immobilized NPs are observed (Figure 2d and Supporting Information Figure S11). Uniform CNRibs with nearly no NPs are obtained at 900 °C, demonstrating the successful formation of pure CNRibs (Figure 2e and Supporting Information Figure S12). The above observation suggests that the in situ-formed Zn-related NPs show great contributions to the transformation of MOF NRods to CNRibs. The aggregation and movement of Zn or ZnO NPs are beneficial for morphology tailoring, which collapses or bursts the rod-like morphology into ribbon-like structures along the Z-axis, resulting in the formation of plate-like carbon intermediates wrapped with many ZnO NPs. Afterwards, the Zn-related species will be reduced and evaporated with increasing calcination temperature, which will consume a great quantity of carbon and reduce the thickness of carbon plates. Thus, small-sized MOF NRods are finally transformed into ultrathin CNRibs. With the removal of zinc species at high temperature, pure CNRibs with well-defined morphologies are obtained (Figure 2f). In comparison with large-sized MOF crystals, it was proved that small-sized MOF nanocrystals facilitate the formation of layered carbon structures.32 Large crystals show a superior structure stability, which can bear the aggregation and transport processes of NPs during pyrolysis. Therefore, only 1D porous CNRods are obtained when using large-sized MOF NRods (hundred nanometers) as precursors. Figure 2 | (a) PXRD patterns of MOF-NRod-S carbonized at different temperatures; (b–e) TEM images showing the morphology of MOF-NRod-S-derived samples obtained at 300, 500, 700, and 900 °C, respectively; (f) schematic illustration of the transformation process from MOF NRods to CNRibs. Download figure Download PowerPoint Metal nanoclusters immobilized on CNRibs Benefiting from the extraordinary electrical, mechanical, and thermal properties, 2D ultrathin carbon nanostructures are excellent supports to immobilize metal nanoclusters,40 which have been widely investigated in various catalysis reactions. To achieve the uniform dispersion of metal nanoclusters on ultrathin CNRibs, metal ions (Fe, Co, and Ni) are initially encapsulated in the pores or adsorbed on the surface of MOF precursors through a one-pot synthesis strategy. After a simple carbonization process, catalysts with metal nanoclusters immobilized on ultrathin CNRibs were obtained ( Supporting Information Figures S13–S15). It is of vital importance that the chemical composition of the resulting catalysts can be further adjusted by the pyrolysis conditions. The use of additional precursors such as N, P, and S sources helps to design excellent catalysts with various functionalities.41 The carbonization of Fe-doped MOF-NRods with melamine as a nitrogen source leads to the formation of N-doped CNRibs immobilized with ultrafine Fe nanoclusters and single metal atoms (Fe–N/CNRib). Considering that too much Fe doping might result in the formation of large Fe NPs or other related phases ( Supporting Information Figures S16 and S17), samples with Fe content of about 1.0 wt % have been investigated. No Fe-related diffraction peaks on PXRD curves suggest the tiny size of Fe nanoclusters (Figure 3a). The obtained nanocatalysts not only inherit the exceptional features of CNRibs including high surface area (518 cm2 g−1) and large pore volume (1.76 cm3 g−1) (Figure 3b) but also merge the ultrathin 2D morphology and ultrafine metal nanoclusters into a solid framework, which vastly improves the utilization of metals and accelerates the mass transfer ability during the catalytic process.37 XPS analyses show the chemical status of the Fe–N/CNRib, in which the C, N, and Fe have been detected. The high-resolution C 1s spectrum confirms the presence of C–C (284.8 eV), C–N (285.6 eV), and C–O (289.2 eV) species (Figure 3c).6 The N 1s spectrum can be divided into three peaks located at 398.4, 400.1, and 401.2 eV, corresponding to the pyridinic, pyrrolic, and graphitic N species, respectively (Figure 3d).42,43 As shown in the Fe 2p spectrum, the large area of Fe(0) arises from the existence of Fe NPs, while the Fe2+ and Fe3+ might be attributed to the formation of FeCx or FeNx in the matrix (Figure 3e).32 As expected, Fe–N/CNRib exhibited a typical 2D belt-like structure, which is different from the Fe–N/CNRod with 1D morphology, suggesting the advantage of this synthetic approach (Figure 3f and Supporting Information Figure S18). Considering the ultrafine size of metal nanoclusters that are hardly observed from SEM and TEM imaging ( Supporting Information Figure S19), HAADF-STEM and EDS elemental mappings were performed for further investigation. As shown in Figure 3g, HAADF-STEM images clearly show the uniform distribution of Fe nanoclusters on N-doped CNRibs, with an average diameter of 1.5 nm. The very small bright dots suggest the existence of atomically distributed Fe single atoms (Figures 3h and 3i and Supporting Information Tables S2 and S3), which also play a crucial role in improving catalytic performance.44,45 It is believed that the highly distributed N and Fe atoms and unique 2D ultrathin structure endow the Fe–N/CNRibs with more accessible active sites for the reaction to occur. Figure 3 | (a) PXRD and (b) N2 sorption and pore size distribution curves of Fe–N/CNRib and Fe–N/CNRod; (c–e) XPS results of Fe–N/CNRib showing the C 1s, N 1s, and Fe 2P spectra; (f) SEM, (g and h) HAADF-STEM, and (i) EDS elemental mapping images of Fe–N/CNRib. Download figure Download PowerPoint ORR and Zn–air battery ORR, a critical reaction in fuel cells, has been widely studied in recent years, but challenges still exist in the preparation of high-performance catalysts with high activity, fast kinetics, and low cost.46–48 Taking advantage of the ultrathin morphology shortening the pathway of electron transfer and mass diffusion and the uniform distribution of ultrafine metal nanoclusters providing abundant active sites, 2D Fe–N/CNRibs exhibit excellent performance in both alkaline and acidic conditions. In 0.1 M KOH aqueous solution, the Fe–N/CNRib shows outstanding ORR activity with an onset potential of 1.00 V and half-wave potential (E1/2) as high as 0.89 V, much better than those of commercial Pt/C (0.86 V) and Fe–N/CNRod (0.84 V), and comparable with the best ORR catalysts recently reported (Figure 4a and Supporting Information Figures S20–S23 and Table S4). The limiting current density (JL) of Fe–N/CNRib at 1600 rpm is about 6.0 mA cm−2, which is up to 7.8 mA cm−2 at 2500 rpm. Tafel curves extracted from LSV curves are shown in Figure 4b. The Fe–N/CNRib shows the smallest Tafel slope (71 mV dec−1) with the best ORR kinetics, superior to that of Pt/C (87 mV dec−1), Fe–N/CNRod (78 mV dec−1), Fe/CNRib (81 mV dec−1), and CNRib (118 mV dec−1). This result indicates that the Fe and N species are important to the enhancement of ORR kinetics, as the Tafel slope of Fe–N/CNRib is superior to Fe/CNRib, and much better than CNRib. Of course, 2D CNRibs with atomically thin and high aspect ratioscan effectively shorten the diffusion length of electrolytes and accelerate the mass transport, and thus improve their electrochemical performances.29 The RRDE measurement was used for monitoring the yield of H2O2 intermediates and surveying the ORR pathway of Fe–N/CNRib (Figure 4c). The H2O2 yield is below 4% when the potential window is 0.2–1.0 V, corresponding to the electron transfer number larger than 3.9 ( Supporting Information Figure S21b). Figure 4 | (a) Comparison of LSV curves at 1600 rpm, (b) the corresponding Tafel curves in (b and c) show the RRDE measurements providing the electron transfer number and H2O2 yield of Fe–N/CNRib. (d) Chronoamperometric response of Fe–N/CNRib and Pt/C at 0.5 V (the inset shows the corresponding methanol tolerance curves). (e) LSV curves of Fe–N/CNRib tested in 0.5 M H2SO4 aqueous solution. (f) Polarization curves, (g) corresponding power density plots, and (h) discharge–charge cycling curves at a current density of 25 mA cm–2 of Zn–air batteries using Fe–N/CNRib and Pt/C + IrO2 as the cathode. Download figure Download PowerPoint The electrochemical surface areas (ECSAs) of Fe–N/CNRib and Fe–N/CNRod were investigated by measuring the Cdl. The Cdl of Fe–N/CNRib catalyst (32.1 mF·cm–2) is larger than that of the Fe–N/CNRod (26.8 mF cm–2), suggesting that the Fe–N/CNRib can provide more accessible active sites for reactions ( Supporting Information Figure S24). It is worth noting that the ORR activity of Fe–N/CNRibs highly depends on the Fe and N contents in CNRibs. The N atoms in carbon frameworks will facilitate the anchoring and distribution of Fe functional species, such as Fe–N–C groups, the main active centers in the ORR process.45,49 However, too much Fe doping on CNRibs will decrease their ORR activities because of the aggregation of Fe NPs ( Supporting Information Figures S25 and S26). Taking advantage of the ultrathin 2D morphology along with highly di