Ultranarrow Graphene Nanoribbons toward Oxygen Reduction and Evolution Reactions

Abstract Identification of catalytic sites for oxygen reduction and evolution reactions (ORR/OER) is critical to rationally develop highly efficient bifunctional carbon‐based metal‐free electrocatalyst. Here, a unique defect‐rich N‐doped ultranarrow graphene nanoribbon with a high aspect ratio that exhibits excellent ORR/OER bifunctional activities and impressive long‐term cycling stability in Zn–air batteries is successfully fabricated. Density functional theory calculations indicates that the topological defects (e.g., pentagons and heptagons) cooperated with pyridinic‐N dopants on the edges are more favorable to electrocatalytic activity toward ORR and OER. This work provides a new design principle for carbon‐based electrocatalytic nanomaterials.

spectroscopy (XPS) measurements were determined by ESCALab MKII electron spectrometer with an excitation source of Mg Ka radiation (1253.6 eV). N 2 adsorption and desorption isotherms were conducted on Micromeritics ASAP 2020 analyzer at 77 K. Electron paramagnetic resonance (EPR) spectra were obtained using Bruker EMXnano wave spectrometer at room temperature.

Electrochemical Measurements
All the electrochemical measurements were carried out using an electrochemical workstation (CHI 760E) with a typical three-electrode system at the room temperature.
A platinum wire, Ag/AgCl (3 M KCl) and glassy carbon disk (5 mm in diameter) were employed as a counter, reference, and the working electrode, respectively. All potentials in this study were corrected to a reversible hydrogen electrode (RHE). The as-prepared catalysts (5.0 mg) were mixed with ethanol (0.950 mL) and Nafion solution (50 L, 5 wt %, DuPont) to form a homogeneous ink. For comparison, the commercial noble metal catalysts (Pt/C (Johnson Matthey, 20%) or RuO 2 (Macklin)) was also tested. For noble metal catalysts (Pt/C and RuO 2 ), the loading is 0.1 mg cm -2 .
For carbon-based non-noble metal catalyst, the loading is 0.3 mg cm -2 . Here, the carbon black (XC-72, 20 wt%) was added into the RuO 2 ink to improve the electron conductivity of catalyst. The electrocatalytic activity was mainly investigated by cyclic voltammetry (CV) and linear sweep voltammetry (LSV) measurements. CV curves were measured in O 2 saturated 0.1 M KOH solution with a scan rate of 20 mV s -1 . LSV curves were tested in O 2 saturated 0.1 M KOH solution by using the rotating disk electrode (RDE) technique with a sweep speed of 10 mV s -1 and the rotation rate at 1600 rpm. For ORR stability, the difference value of half-wave potential (E 1/2 ) was derived from the LSV plots of the catalyst before and after 3000 potential cycles between 0 V to 1.2 V in oxygen-saturated 0.1 M KOH electrolyte. For OER stability, the difference value of the potential at 10 mA cm-2 (E 10 ) was derived from the LSV plots of the catalyst before and after 3000 potential cycles between 1.1 V to 2.0 V in oxygen-saturated 0.1 M KOH.
For the Zn-air battery test, the air cathode was prepared by evenly covering the catalyst ink onto a carbon gas diffusion layer (Teflon-coated carbon fiber paper). The loading is 1 mg cm -2 for all the materials. A Zn foil was employed as the anode. Zn-air batteries were assembled by using the above two electrodes in 6 M KOH aqueous electrolyte. Measurements were performed on a home-made electrochemical cell at room temperature.

Density Functional Theory Calculations
All the density functional theory (DFT) calculations were performed by using the Vienna Ab initio Simulation Package (VASP) [6] , employing the projected augmented wave (PAW) [7] model. The revised Perdew-Burke-Ernzerhof (RPBE) [8] functional was used to describe the exchange and correlation effect. In all the cases, the cutoff energy was set to be 400 eV. The bulk single-layer graphene was optimized using a 9 × 9 × 1 Monkhorst−Pack [9] k-point mesh, obtaining a lattice constant of graphene 2.47 Å. The periodic surfaces were modeled using 4 × 7 and 3 × 6 supercells for the zigzag and armchair edges, respectively. Of all the surface calculations, a 3 × 1 × 1 mesh was adopted, reciprocally proportional to the surface parameters. When optimizing the surface structures, the edge atoms and the adsorbates were allowed to fully relax until reaching the convergence tolerance; while the other atoms were fixed in their bulk arrangement. The force and energy tolerance was set to be 0.05 eV Å -1 and 10 -5 eV, respectively.
The free energy of OH adsorption on the slabs was defined as where ∆E ads is the electronic adsorption energy, ∆E ZPE is the zero-point energy difference between adsorbed and gaseous species, and T∆S ads is the corresponding entropy difference between these two states.
The binding energy of OH on the slabs was defined as  Table S4.

Equation S1 and S2
The Koutecky-Levich (K-L) equation as given below: where J denotes the measured current density, J K is the kinetic current density, J L is the diffusion-limited current density, is the electrode rotation rate, F is the