Templated Synthesis of Exfoliated Porous Carbon with Dominant Graphitic Nitrogen

We present here a new approach for the synthesis of nitrogen-doped porous graphitic carbon (g-NC) with a stoichiometry of C6.3H3.6N1.0O1.2, using layered silicate as a hard sacrificial template. Autogenous exfoliation is achieved due to the heterostacking of 2D silicate and nitrogen-doped carbon layers. Micro- and meso-porosity is induced by melamine and cetyltrimethylammonium (C16TMA). Our density functional calculations and X-ray photoelectron spectroscopy (XPS) observations confirm that the most dominant nitrogen configuration in g-CN is graphitic, while pyridinic and pyrrolic nitrogens are thermodynamically less favored. Our large-scale lattice dynamics calculations show that surface termination with H and OH groups at pores accounts for the observed H and O in the composition of the synthesized g-NC. We further evaluate the electrocatalytic and the supercapacitance activities of g-NC. Interestingly, this material exhibits a specific capacitance of ca. 202 F g–1 at 1 A g–1, retaining 90% of its initial capacitance after 10,000 cycles.

The electrochemical performance of the prepared electrodes was investigated through a threeelectrode system.A homogeneous ink was prepared as follows: first, as-prepared samples were ground, and then, 5.0 mg of the ground sample was dispersed into 950 µL of ethanol: water mixed solution (volume ratio of 1 : 3) and 50 µL of 5.0 wt% Nafion.After 60 min sonication, 5 µL of the suspension was dropped onto a glassy carbon electrode (GCE) with an area of 0.1256 cm 2 and dried slowly.The mass loading was 0.2 mg cm −2 .For GCE activation, the sample was pre-polished with 1 µm diamond, then 0.05 µm alumina powder, and washed with water.All electrochemical S2 measurements were carried out using CHI 842B instrument, the cyclic voltammetry (CV), linear sweep voltammetry (LSV) at 10 mV s −1 , and chronoamperometry (i -t) measurements were carried out in O 2 saturated 0.1 M KOH at ≈ 0.5 V vs. RHE.The electrolyte solution was purged with ultra-pure O 2 or N 2 for at least 1 h before starting the measurements.A platinum wire and silver/silver chloride (Ag/AgCl) were used as the counter and reference electrode, respectively.
The potentials were expressed with regard to the reversible hydrogen potential electrode (RHE).
The overall electron transfer numbers per oxygen molecule were calculated from the slope of the Koutecky-Levich plots according to the following equation: where J is the current density, J k is the kinetic current density, J L is the diffusion-limited current density, n is the transferred electron number, F is Faraday constant (F = 96485 C mol −1 ), C 0 is the ν is the kinematic viscosity of the electrolyte (0.01 cm 2 s −1 ), and ω is the electrode rotating rate.

In situ Raman spectroscopy
Raman spectra were measured with a Renishaw inVia Raman system.A 50× long working distance (8 mm) objective was used.The wavelength of the excitation laser was 633 nm from a He Ne laser.The laser power was 6 mW.Raman frequencies were calibrated using Si wafer spectra.All Raman spectra were measured over an acquisition time of 10 s and 1 accumulation.
The Raman cell used for this measurement is shown in Figure S4.This cell configuration was employed to perform Raman spectroscopy at a biased electrode immersed in a very thin layer of electrolyte (0.1 M KOH).The drop-casted electrode was characterized at the following applied potentials (using chronoamperometry) in an O 2 -saturated electrolyte.
B. Results

Electrocatalytic oxygen reduction reaction (ORR)
In addition to supercapacitance, we investigated the ORR activity for the oxygen reduction reaction in g NC. Figure S4a  As shown in Figure S5b, there was a linear relationship between the inverse of the square root of rotating speed (ω −1/2 ) and the inverse of measured current density (J −1 ) irrespective of potential.The number of electrons in ORR was extracted from the slopes of each straight line and found to be between 2.4-2.8e, regardless of potential, indicating that a two-electron reduction reaction proceeded on these electrodes (Figure S5c).The stability measurement of the g NC sample was also carried out under constant potential seen in Figure S5d.
It was found that after 100 minutes, the initial current value decreased by 27%.By evaluating the outputs of the results, it can be claimed that g NC can be an alternative candidate as a supporting carbon material for metal-based electrocatalysts instead of commercial carbon black.This way, it may be possible to produce more active electrocatalysts that show better onset and higher current due to the synergistic effect obtained between the metal and g NC-based supporting material.

In situ Raman spectroscopy of g NC in electrocatalysis
In the next step, we studied the stability of the material through in situ Raman spectroscopy in an electrochemical process using a cell running where the Raman spectra measurement is monitorable.The setup details are presented in Figure S3.We performed this experiment to S4 investigate any functional group alteration and stability of g NC during the electrocatalysis.As shown in Figure S6, two prominent bands appeared in 1352 cm −1 and 1590 cm −1 , respectively, attributable to D and G bands.The intensity ratio of the D band (I D ) to the G band (I G ) intensity before starting the ORR reaction was 1.7.However, after applying voltage, a minor change occurs in the ID/IG ratio, representing that the material is stable during the electrocatalytic reaction while applying the voltage.Therefore, we did not see a remarkable change in the overall shape and the bands' intensities.A new shoulder at ∼ 1440 cm −1 appeared immediately after adding the electrolyte because of the electrolyte and g NC interaction.This can be recognised by comparing the Raman spectra before and after adding the electrolyte, shown in Figure S7.
represents LSV curves taken before and after silicate etching at 1600 rpm.For comparison, the LSV curve of commercial carbon black (Vulcan XC-72) measured as a reference commercial electrocatalyst was shown in the same Figure.Onset potential, half-wave potential, and limiting current density at 0 V vs. RHE of these samples were extracted from LSV curves shown in FigureS4aand summarised in FigureS4b.Compared to the carbon black and SiO 2 NC, we obtained a better activity for g NC in all three terms, indicating g NC has a better catalytic activity for the ORR than the other samples.In detail, onset potential was enhanced from 0.72 V to 0.8 V vs. RHE, most probably due to the less or no activity of the silicate part.A similar trend was also observed for the half-wave potential.Moreover, limiting current density has also increased almost twice compared to SiO 2 NC.While SiO 2 NC shows nearly the same onset and half-wave potential as the carbon black (Vulcan XC-72), the limiting current value is much lower since the presence of SiO 2 in the structure raises the charge transfer resistance.Hence, the best limiting current value was reached after eliminating SiO 2 , as expected for the g NC sample.For electrocatalytic ORR, we have two already-accepted mechanisms in alkaline solutions: reduction via four electrons from O 2 to OH -and the reduction of O 2 to H 2 O -via two electrons.To clarify the mechanism, we calculated the electron numbers using the Koutecky-Levich formula to find the relevant mechanism.FigureS5aexhibits LSV curves at various rotating speeds via a rotating disk electrode (RDE) and the Koutecky-Levich plots at multiple potentials for the g NC in O 2 saturated 0.1 M KOH aqueous solution.

FigureFigure S4 .
Figure S1.Schematic representation of the in situ Raman spectroscopy measurement during the oxygen reduction reaction.

TABLE S1 .
CHNS analysis results of g NC and SiO 2 NC materials.