Oxygen Species on Nitrogen-Doped Carbon ... - ACS Publications

Mar 23, 2018 - All Publications/Website .... Nanosheets as Efficient Active Sites for Multiple Electrocatalysis .... for Oxygen Electroreduction in Ac...
0 downloads 0 Views 3MB Size
Subscriber access provided by Queen Mary, University of London

Energy, Environmental, and Catalysis Applications

Oxygen Species on Nitrogen Doped Carbon Nanosheets as Efficient Active Sites for Multiple Electrocatalysis Jing-Jing Lv, Yanle Li, Shaojun Wu, Hua Fang, Lingling Li, Rongbin Song, Jing Ma, and Jun-Jie Zhu ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b00240 • Publication Date (Web): 23 Mar 2018 Downloaded from http://pubs.acs.org on March 24, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Oxygen Species on Nitrogen Doped Carbon Nanosheets as Efficient Active Sites for Multiple Electrocatalysis Jing-Jing Lv,†,⊥ Yanle Li,‡,⊥ Shaojun Wu,† Hua Fang,† Ling-Ling Li,† Rong-Bin Song,† Jing Ma,*, ‡ Jun-Jie Zhu*, † †

State Key Laboratory of Analytical Chemistry for Life Science and Collaborative Innovation

Center of Chemistry for Life Sciences, School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210093, China ‡

Key Laboratory of Mesoscopic Chemistry of MOE and Collaborative Innovation Center of

Chemistry for Life Sciences, School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210093, China KEYWORDS: g-C3N4 template, oxygen modified nitrogen doped carbon nanosheets, multifunctional electrocatalysis, Zn-air batteries, water splitting cells ABSTRACT Designing and synthesizing nanomaterials with high coverage of active sites is one of the most pivotal factors in the construction of state-of-the-art electrocatalysts with high performance. Herein, we proposed a facile in-situ templated method for the fabrication of oxygen species modified nitrogen doped carbon nanosheets (O-N-CNs). The epoxy-O, ketene-O combined with graphitic-N defects in O-N-CNs gave more active sites for oxygen 1 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 30

reduction reaction (ORR) and oxygen evolution reaction (OER) as proved via theoretical and experimental results, while the carbonyl-O and epoxy-O species showed more efficient electrocatalytic activity for hydrogen evolution reaction (HER). Hence, the O-N-CNs showed highly active electrocatalytic performance toward ORR, OER and HER. More importantly, the superior multifunctional electrocatalytic activity of O-N-CNs allowed their use in the construction of Zn-air batteries to power the corresponding water splitting cells. This work can offer an understanding of underlying mechanisms of oxygen species on N-doped carbon materials toward multiple electrocatalysis and facilitate the engineering of electrocatalysts for energy storage and conversion devices.

INTRODUCTION With the excessively depleting of fossil fuels and increasing global environmental concerns, immense research interests have been stimulated for developing highly efficient and sustainable alternative energy-storage and conversion technologies.1, 2 Particularly, the fuel cells, metal-air batteries and electrolysis cells are exerting as promising environment friendly devices over the past decades.3-5 Highly efficient electrocatalysts are indispensable in these devices for driving the electrode reactions, such as oxygen reduction reaction (ORR),6 oxygen evolution reaction (OER),7 and hydrogen evolution reaction (HER).8 However, the high cost and scarce reserve of noble-metal catalysts commonly used for these reactions impede their widespread applications. This compels us to spare no effort into the exploration of highly active, affordable and multifunctional electrochemical catalysts. Among masses of alternatives, carbon nanomaterials have been found with great potential to substitute noble-metal catalysts due to their good electrocatalytic activity, low cost, long durability, and environmental friendliness.9,

10

To boost the process of their

practical application, great efforts have been devoted to improving the electrocatalytic performance and expanding the versatility of carbon nanomaterials.11-13 Doping heteroatoms 2 ACS Paragon Plus Environment

Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

(N, P, S, B, F, etc.) or modifying active species to the skeleton of carbon nanomaterials can be an effective strategy for increasing the electrophilicity and active sites.14-17 Furthermore, the electrocatalytic activity of carbon nanomaterials can be promoted by tailoring the structure to accelerate both the electron and mass transfer.9, 18-20 As a result, metal-free N-doped-based carbon nanosheets emerged as one of the most promising electrocatalysts for ORR, OER and HER because of their striking conductivity, large surface area and rich active sites.18, 21, 22 The last decades have witnessed the flourish of N-doped-based carbon nanosheets from the initially sole ORR catalysts,9, 23, 24 to the bi-functional ORR-OER catalysts,15, 18 and the recently tri-functional ORR-OER-HER catalysts.20, 25 However, the unknown multifunctional catalytic mechanisms and low multifunctional performance for N-doped-based carbon nanomaterials stand in the way of their further progress. On the other level, to achieve the Ndoped-based carbon nanosheets, it usually requires the nitrogen sources and porogens, as well as the templates. Generally, the nitrogen containing chemicals can simultaneously serve as nitrogen sources and porogens.18, 22, 23, 26, 27 While some soft or hard extrinsic templates are always employed to direct the formation of nanosheet structure,18,

28-31

which inevitably

involve laborious procedures for pre-preparing and follow-up-removing of the templates. Therefore, facile strategies for designing and in-depth mechanism study of metal-free carbonbased multifunctional electrocatalysts are very important to guide the development of advanced catalytic materials with sufficient active sites. Herein, we proposed a facile in-situ templated strategy for fabrication of oxygen modified nitrogen doped carbon nanosheets (O-N-CNs) using biomass as carbon and oxygen sources. The in-situ formed g-C3N4 during the synthesis simultaneously acted as the nanosheet template, nitrogen source and porogen, successfully simplifying the synthesis procedure of ON-CNs. Strikingly, it was theoretically and experimentally revealed that the formed epoxy-O, ketene-O combined with graphitic-N gave more active sites for ORR and OER, while the carbonyl-O and epoxy-O species showed more efficient electrocatalytic activitity for HER 3 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 30

than that of graphitic-N. As a result, the O-N-CNs showed remarkable multifunctional electrocatalytic performance for ORR, OER, and HER. As a proof-of-concept application, the O-N-CNs could be commendably employed as OER and HER catalysts for oxygen and hydrogen gas production in a water splitting cell as powered by Zn-air batteries, which were assembled with the same electrocatalyst on their air electrode for ORR and OER.

EXPERIMENTAL SECTION Characterization. The morphology and microstructure were characterized by field emission scanning electron microscopy (SEM, LEO153VP, 10 kV) with an energy-dispersive X−ray (EDX) spectrometer, transmission electron microscopy (TEM), high angle annular dark field-scanning transmission electron microscopy (HAADF-STEM), and high resolution TEM (HRTEM) using a JEOL2100 TEM operating at 200 kV. Atomic force microscopy (AFM) images were taken on a Fastscan AFM (Bruker Corporation, America) operated in a fastscan mode. Powder X-ray diffraction (XRD) measurements were conducted on a D8 ADVANCE X-ray diffractometer (Bruker Corporation, America). Fourier transform infrared spectroscopy (FT-IR) spectra were obtained on a Vector 22 spectrometer (Bruker Corporation, America). Raman spectra were measured on a Via-Reflex Laser Confocal Raman Microspectroscopy (Renishaw, England). Ultraviolet visible (UV) spectra were recorded on a UV-3600 spectrophotometer (Shimadzu, Kyoto,

Japan).

Photoluminescence

(PL)

spectra

were

obtained

on

an

F-7000

spectrophotometer (Hitachi High-Technologies Corporation, Japan). The Thermogravimetric analysis (TGA) was performed under a flow of N2 on a NETZCHE STA449F3 analyzer and the samples were heated from 25 to 900 °C with a heat rate of 10 °C min–1. BrunauerEmmett-Teller (BET) surface area was measured with a Micrometrics ASAP2020 analyzer (USA). X-ray photoelectron spectra (XPS) were measured on a PHI 5000 VersaProbe (Ulvac-

4 ACS Paragon Plus Environment

Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Phi, Japan). The elemental analysis was conducted on an elemental analyzer (CHN-O-Rapid, Germany). Synthesis of oxygen modified carbon nanospheres (O-CSs). All chemicals were of analytical grade and used as received without further purification unless otherwise noted. The carbon aerogels (CAs) were fabricated through a literature modified method by employing the strawberry (purchased from local fruit supermarket)

32

as

the carbon and oxygen precursor. Generally, the appropriate volume of cleaned fresh strawberry fleshes were placed into a Teflon-lined stainless-steel autoclave (100 mL, RNK Science and Tech Company, Nanjing, China) and mashed into the homogeneous slurry. The autoclave then was put into an oven and heated at 180 °C for 12 h. The obtained black carbonaceous hydrogel were subsequently treated with a lyophilization process for 48 h to get the CAs. Then, the O-CSs were prepared by the calcining treatment of CAs in N2 atmosphere at 900 oC for 2 h with a heating rate of 10 oC min–1. Synthesis of oxygen modified nitrogen-doping carbon nanosheets (O-N-CNs). To obtain the O-N-CNs, the CAs (50 mg) were mixed with melamine (500 mg), grinded together and annealed in N2 atmosphere at 900 oC for 2 h with a heating rate of 10 oC min–1. In a typical experiment, the yield respected to CAs is ca. 42 wt%. For comparison, the oxygen modified nitrogen-doping carbon namomaterials (O-N-CMs-X, X = the usage mass ratio of CAs to melamine) with different content of N were acquired by adding various usage of melamine under the same conditions. Besides, the g-C3N4, g-C3N4-O-CSs-600 oC, and OCSs-600 oC samples were fabricated by calcining melamine, mixture of CAs (50 mg) and melamine (500 mg), and CAs in N2 atmosphere at 600 oC for 2 h with a heating rate of 10 oC min–1, respectively. Electrochemical measurements. Electrochemical studies were conducted in a standard three electrode system controlled by a CHI 760E electrochemistry workstation (CH Instruments Inc., Shanghai, China). Taking O5 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 30

N-CNs modified electrode as an example, 4 mg of catalyst was ultrasonically dispersed in 1.0 mL of a mixture with 250 µL of distilled water (Milli-Q), 700 µL of ethanol and 50 µL of 0.5 wt% Nafion. An aqueous dispersion of the catalyst (30 µL, 4 mg/mL) was then carefully transferred onto the glassy carbon rotating ring disk electrode (RRDE, 0.247 cm2, Pine Research Instrumentation, Durham, NC) to achieve a catalyst loading of 0.49 mg cm−2, which was served as a working electrode. All electrodes were prepared by depositing the same loading mass of active materials on RDE using the same method. A platinum wire was used as the counter electrode and Ag/AgCl (saturated KCl) as the reference electrode. All potential measurements were converted to the RHE based on the following formula Evs RHE = Evs Ag/AgCl + Eθ Ag/AgCl + 0.059 pH (in volts). For the ORR measurements, the cyclic voltammetry curves (CVs) were performed with a scan rate of 50 mV s−1 in N2-saturated and O2-saturated 0.1 M KOH. The liner scan voltammetric curves (LSVs) were obtained in O2-saturated 0.1 M KOH at various rotating speeds from 900 to 2500 rpm at a rate of 10 mV s−1, and the ring current curve was held on a constant potential at 1.5 V vs RHE. The percent of H2O2 and the number of electron transfer (n) were determined based on LSVs by the equations (1) and (2):23

%(H 2O 2 ) = 200 ⋅

n=

Ir/N Id + Ir/N

4Id (Id + Ir/N)

(1)

(2)

Id: The disk current; Ir: The ring current; N: The current collection efficiency of the Pt ring (0.37). The n can also be calculated based on LSVs by removing the capacitive current at various rotating speeds by the K–L Equations (3) and (4):33

1 1 1 1 1 = + = + J J L J K Bω1 / 2 J K

(3)

6 ACS Paragon Plus Environment

Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

B = 0.2nF(DO 2 ) 2/3 ν −1/6CO 2

(4)

J: The measured current density; JK: The kinetic current densities; JL: The diffusion-limiting current densities; ω: The rotation speed in rpm; F: The Faraday constant (F = 96 485 C moL−1); DO2: The diffusion coefficient of O2 (1.9 × 10−5 cm2 s−1); ν: The kinetic viscosity (0.01 cm2 s−1); CO2: The bulk concentration of oxygen (1.2×10−6 mol cm−1). The OER LSVs were obtained in O2-saturated 0.1 M KOH at a rotating speed of 1600 rpm with a scan rate of 10 mV s−1. For HER measurements, the LSVs were carried out in N2saturated 0.1 M KOH with the rotating speed of 1600 rpm and the scan rate of 10 mV s−1. The Tafel slopes were calculated according to the Tafel equation η = b log(j/j0) based on the LSV curves, where η is the overpotential, b is the Tafel slope, j is the current density, and j0 is the exchange current density.18 All LSVs for OER and HER were corrected with 95 % iRcompensation. The stability measurements for ORR, OER, and HER were tested by the chronoamperometry method for 20 000 s at corresponding given potentials. The CVs of the samples measured in 0.1 M KOH solution by employing the scan rates (20, 40, 60, 80 and 100 mV s-1) were acquired to compare the effective electrode surface area (ECSA) of the relevant samples. The electrochemical impedance spectroscopy (EIS) was tested in potentiostatic mode (0.7 V vs. RHE for ORR, 1.65 V vs. RHE for OER and –0.5 V vs. RHE for HER) from 10 kHz to 1 Hz at the amplitude of the applied voltage of 5 mV. For the Zn-air battery tests, 4.0 mg of O-N-CNs or 20 wt% Pt-C was mixed with 950 µL of Milli-Q water and 50 µL of 5.0 wt% Nafion solution, and with the assistance of at least 30 7 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 30

min sonication to form a homogeneous ink. This catalyst ink (250 µL) was uniformly dropcast onto 1.0 cm2 of carbon paper electrode to obtain a catalyst loading of 1.0 mg cm−2. The air cathode covered with a laminating film for circulating air was then paired with a Zn foil anode and assembled in a customized electrochemical cell filled with 6.0 M KOH. The LSVs were collected at a scan rate of 10 mV s−1. Discharge–charge cycling were performed at room temperature using the double-pulse method, where one cycle consisted of a discharging step (10 mA cm−2 for 10 min) followed by a charging step with the same current and duration time. To measure the overall water splitting performance in the two-electrode alkaline electrolyzer, the O-N-CNs or 20 wt% Pt-C on the carbon paper with a catalyst loading of 1.0 mg cm−2 simultaneously behaved as an anode electrode for the OER and a cathode electrode for the HER in 1.0 M KOH solution. LSVs were obtained at a scan rate of 10 mV s−1 and chronoamperometry data were tested for 20 000 s at a given potential of 1.65 and 1.75 V vs. RHE for O-N-CNs and 20 wt% Pt-C, respectively.

Computational methods. Density functional theory (DFT) calculations were carried out using the Generalized Gradient Approximation (GGA) exchange-correlation functional in the Perdew-Burke-Ernzerhof (PBE),34 and the projector augmented wave (PAW) method,35 as implemented in Vienna ab

initio simulation package (VASP).36, 37 The cut-off energies for plane waves was 400 eV, the convergence tolerance was set as 10-5 eV in energy and 0.02 eV/Å in force. The 14×14×1 Monkhorst-Pack grid k-points were applied for all the geometric optimizations. The theoretical models were constructed from 7×7 graphene unit cell with a vacuum layer as large as 20 Å along the c direction. We focused on on the role of graphitic nitrogen (graphitic-N), hole, epoxy oxygen (epoxy-O), carbonyl oxygen (carbonyl-O), and ketene oxygen (ketene-O) played in the eletrocatalysis reactions. The detail simulation processes and models are shown in Supporting Information. 8 ACS Paragon Plus Environment

Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

RESULTS AND DISCUSSION

A Melamine

Strawberries Strawberry

g-C 3N43-CNs CNs-C N4

CAs

B

O- N-CNs

D

C

In-plane holes 50 nm

100 nm

500 nm

E

100 nm

F1

F 1

C-K

N-K

0.335 nm

F3

O-K

0.5 nm

F2

0.278 nm

2 2 nm

3

0.5 nm

0.5 nm

Figure 1. (A) Schematic illustration of the O-N-CNs synthesis by using the in-situ formed g-C3N4 as template. (B) SEM and (C, D) TEM images of O-N-CNs. (E) HAADF-STEM image of O-N-CNs and the corresponding C, N and O element mapping images. (F) HRTEM and (F1-F3) its corresponding selected domain’s HRTEM images of the O-N-CNs.

Figure 1A illustrates the fabrication procedure of the N-CNs. Initially, carbon aerogels (CAs) as carbon sources were prepared by the hydrothermal treatment of the strawberry fleshes. Then, the N-CNs were fabricated by directly calcining the mixture of CAs and melamine, wherein the melamine was in-situ condensed to form g-C3N4 with increasing temperature in the furnace. Since the g-C3N4 would be decomposed at higher temperature, it could then serve as template and porogen to direct the formation of porous carbon nanosheets and as nitrogen source for doping N element into the carbon skeleton. The scanning electron microscopy (SEM, Figure 1B) and transmission electron microscopy (TEM, Figure 1C, D) images show that the O-N-CNs possess a dominantly few-layer porous sheet-like structure 9 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 30

with numerous wrinkles and nanoholes randomly distributing in the plane. The high angle annular dark field-scanning transmission electron microscopy (HAADF-STEM) image confirms the porous sheet-like feature as shown in Figure 1E. And the corresponding energy dispersive X-ray spectroscopy (EDX) elemental mapping in this area shows homogeneous distribution of C, N and O (Figure 1E). The formation of thin sheets with 3-6 single graphitic atomic layers can be further verified by the atomic force microscopy (AFM, Figure S1) with ca. 0.3–3.5 nm in thickness. To confirm the formation mechanism of the O-N-CNs, a series of controlled experiments were performed. As can be observed in the SEM image (Figure S2A) and Fourier transform infrared spectroscopy (FT-IR, Figure S2B) spectrum, the CAs present cross-linked sphere structure and contain numerous hydroxyl, carbonyl, carboxyl, and aromatic groups on the surface.32 Besides, the thermogravimetric analysis (TGA, Figure S2C) curves reveal that the CAs have good thermal stability after 400 oC and can keep their ca. 40% mass after the thermal treatment, while the g-C3N4 fabricated by melamine can be decomposed after 750 oC. Therefore, the CAs and melamine were suitably chosen as the carbon and oxygen sources, and the in-situ formed g-C3N4 precursor, respectively. The product via thermal annealing CAs in the absence of melamine under otherwise identical conditions remains the morphology of CAs consisting of cross-linked oxygen modified carbon nanospheres (O-CSs) as indicated from its SEM and TEM images (Figure S3). And when the pyrolysis temperature for the melamine, mixture of CAs and melamine, and CAs was decreased to 600 oC, the products show the sheet-like, sphere-on-sheet-like and sphere-like structures, respectively, suggesting the formation of g-C3N4 (Figure S4A), hybrid of g-C3N4 and O-CSs (g-C3N4-CSs-600 oC, Figure S4B), and O-CSs (O-CSs -600 oC, Figure S4C). Even though the signal of the g-C3N4 was weakened a lot by the O-CSs-600 oC on the ultraviolet visible (UV, Figure S5B) and photoluminescence (PL, Figure S5C) spectra of gC3N4-O-CSs-600 oC,38 it can still identify the g-C3N4 characteristic peaks on XRD (2Ɵ=13o 10 ACS Paragon Plus Environment

Page 11 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

and 27o) spectra (Figure S5A), further illustrating the formation and decomposition of g-C3N4 in the pyrolysis process for the fabrication of O-N-CNs. Moreover, the content of the melamine affect the morphology of the obtained oxygen modified N-doped carbon nanomaterials (O-N-CMs). The sheet-like O-N-CNs (Figure 1) and O-N-CMs-1:20 (Figure S6C) were prepared using the mass ratio between CAs and melamine of 1:10 and 1:20, while decreasing the mass ratio to be 1:2 or 1:6.4 achieved the O-N-CMs-1:2 (Figure S6A) or O-NCMs-1:6.4 (Figure S6B) with plate-like enclosed with spheres feature. These results verify the in-situ formation of g-C3N4, and its key role as template to induce the growth of sheet-like ON-CNs, along with its removal under high temperature to dope nitrogen and generate pores. To further probe into the morphology, porosity, and crystalline structure of the O-N-CNs, the high-resolution TEM (HRTEM) images were provided. It shows the O-N-CNs compose of randomly orientated amorphous graphitic texture, porous structure and defective carbon feature with a certain crystallization degree (Figure 1F). The distinct lattice fringes in the crystallization domains can be measured to be 0.335 nm (Figure 1F1) and 0.278 nm (Figure 1F2), corresponding to the (002) and (200) planes of graphite-like carbon (JCPDS Card No. 41-1487) and CNx crystal (JCPDS Card No. 50-0664), respectively, suggesting the O-N-CNs with polycrystalline graphitic structure.39 In addition, the lattice distortions of carbon hexagons with atomic holes can be observed in the crystallization domain (Figure 1F3), which might be due to the variation of the perfect hexagon lattices to accommodate the curvature for the nitrogen doping and pore introducing under high temperature.40 And the nitrogen adsorption–desorption isotherm was measured to verify the porosity of the O-N-CNs (Figure S7). The O-N-CNs exhibit the specific surface area of 507 m2 g −1 and the total pore volume of 0.44 cm3 g−1. The pore size distribution suggests a mesoporous structure (Figure S7 inset), which is favorable for the mass transfer and exposing more active sites in the electrochemical reactions.

11 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

B

D G

2D ID/IG=0.99

ID/IG=1.17 1000 1500 2000 2500 3000 3500 4000 -1 Wavenumber / cm

C

O-CSs O-N-CNs

O

1000

N

800 600 400 200 Binding Energy / eV

E

540

C=O carbonyl

536 532 528 Binding Energy / eV

30

D

0

O-N-CNs O1s 5.44 at% C-O epoxy

C=C=O ketene

15

524

45 60 2θ / degree

75

O-N-CNs N1s quaternary-N 2.57 at%

Intensity / a.u.

Intensity / a.u.

C

O-CSs O-N-CNs

Intensity / a.u.

Raman Intensity / a.u.

O-CSs O-N-CNs

pyridinic-N 1.08at%

408

404 400 396 Binding Energy / eV

F

C-O epoxy Intensity / a.u.

A

Intensity / a.u.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 30

C=C=O ketene

540

392

O-CSs O1s 4.37 at%

C=O carbonyl

536 532 528 Binding Energy / eV

524

Figure 2. (A) Raman spectra and (B) XRD patterns of O-N-CNs and O-CSs; (C) survey, (D) highresolution N1s, and (E, F) O1s XPS spectra of O-N-CNs and O-CSs, O-N-CNs, and O-N-CNs and O-CSs, respectively.

Moreover, the corresponding ID/IG ratio in Raman spectra increasing from 0.99 of O-CSs to 1.17 of O-N-CNs confirms the higher concentration of defects in O-N-CNs (Figure 2A).18 The broader and higher 2D peak of O-N-CNs at 2700 cm–1 than that of the O-CSs further demonstrates the O-N-CNs with few-layer carbon structure of a high degree carbon graphitization, which benefit for promoting the electronic conductivity and corrosion resistance for the electrocatalysts.24 Furthermore, the O-N-CNs show two typical peaks located at 24.6o and 43.7o in the XRD pattern (Figure 2B), which can be indexed to the (002) and (100) lattice planes of turbostratic graphitic carbon, respectively (JCPDS Card No. 411487).24 And the absolute height of peaks at 24.6o and 43.7o of O-CSs is 1.06 and 1.66 times 12 ACS Paragon Plus Environment

Page 13 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

higher than those of the O-N-CNs, respectively, verifying the O-N-CNs with more amorphous defect sites.22 The specific chemical compositions of the samples were determined by X-ray photoelectron spectroscopy (XPS) and elemental analysis (EA). The XPS spectra reveal that the O-CSs contain the atomic content of C and O with 94.72 at% and 4.37 at%, along with the trace of N (0.9 at%, Figure 2C). As to the O-N-CNs, three evident peaks corresponding to C, N and O elements are observed with the atomic content of 90.90 at%, 3.65 at% and 5.44 at% (Figure 2C), respectively, in good accordance to the EA result with the C and N mass content of 81.73 wt% and 4.7 wt%. The curve-fitted N1s high-resolution XPS spectrum of O-N-CNs (Figure 2D) shows two distinguished nitrogen species at 401.2 eV and 398.8 eV, assigning to the quaternary-N (2.57 at%) and pyridinic-N (1.08 at%), respectively.30 Compared to the O-N-CNs, O-N-CMs-1:2, O-N-CMs-1:6.4 and O-N-CMs-1:20 all consist of lower nitrogen content with 2.18 at%, 3.3 at% and 2.88 at%, respectively (Figure S8). The high content of pyridinic and graphitic N is generally beneficial to ORR, OER, and HER.18, 41, 42 Meanwhile, the analysis of O1s XPS spectra deconvolution (Figure 2E, F) provides significant clues on the higher oxidized vacancies concentration of 5.44 at% for O-N-CNs than that of O-CSs (4.37 at%), which can be further evidenced by its more intense peaks of epoxy oxygen (C-O), carbonyl oxygen (C=O), and ketene oxygen (C=C=O).43 A certain content of stable oxidized vacancies on ON-CNs can make it with additional catalytic active sites and better hydrophilicity for more accessible catalytic surfaces.44, 45 The high-resolution C1s XPS spectra of O-N-CNs and OCSs indicate the typical carbon species in nitrogen doped carbons, such as the C-C/C=C and C-O/C=O species (Figure S9).22 Intrinsic (zig-zag, armchair, hole, etc.),46, 47 topological (C5, C7, C5+C7, etc.),18, 40 and component (N, B, S, P, etc.)14,

48

defects of metal-free carbon materials have been

experimentally and theoretically demonstrated to contribute to their electrocatalytic activity. 13 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

For most of the carbon materials, however, the specific roles of the O-containing carbon defects played in ORR, OER or HER processes are not well established.43, 49 Here, to gain insights into the O-containing carbon defects for ORR, OER and HER catalytic mechanisms, the theoretical models containing the epoxy oxygen (epoxy-O), ketene oxygen (ketene-O), and carbonyl oxygen (carbonyl-O) defects in O-N-CNs were built in density functional theory (DFT) calculations, in comparison with the established models of graphitic nitrogen (graphitic-N) and hole defects in O-N-CNs (Figure 3A, Figure S10, 11). graphitic-N

A

B

12

graphitic-N hole epoxy-O ketene-O carbonyl-O

epoxy-O

hole

ketene-O

6 4

carbonyl-O

D

graphitic-N hole epoxy-O ketene-O carbonyl-O

6 3 0 -3 -6

O2

OOH*

2 0

O*

ORR U = 0 V OH* OH

2 graphitic-N

1

∆GH* / eV

C8

Free energy / eV

9

Free energy / eV

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 30

0

+ H +e

carbonyl-O

-1

epoxy-O ketene-O

-2

hole

1/2H2

-2 -4 OH

OH*

O*

OER U = 0 V O2 OOH*

HER U = 0 V Reaction coordinate

Figure 3. (A) A schematic graphitic model with different kinds of carbon defects for O-N-CNs (C: grey ball, N: blue ball, O: red ball and H: white ball). Calculated free energy diagrams of (B) ORR, (C) OER and (D) HER for the corresponding carbon defects.

An ideal ORR or OER catalyst should be able to facilitate overall reaction with lower Gibbs free energy change.48, 50 As for the ORR, the graphitic-N, ketene-O and epoxy-O, are all downhills with similar lower free energy change, while the carbonyl-O and hole defects both involving a uphill step and much higher free energy change, indicating the former species are more active than the later ones (Figure 3B, Table S1), in good agreement to the previous reported results.51 The same scenario is found on the OER processes for those 14 ACS Paragon Plus Environment

Page 15 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

defects (Figure 3C, Table S1). These results reveal that the graphitic-N, ketene-O and epoxyO defects are more responsible for the good ORR and OER performance among these model defects. In terms of HER, the free energy of H*, ∆G(H*), is usually used as an effective index for evaluating HER activity.52 The smaller absolute value of ∆G(H*) indicates a better HER activity of the system.53 As can be seen in the Figure 3D and Table S1, the epoxy-O and carbonyl-O defects exhibit almost the same lowest absolute value of ∆G(H*) among these defects, suggesting they are more active for HER than the other defect models. Therefore, the defects induced by nitrogen doping and oxidized carbon defects in graphite phase can actively regulate electrocatalytic activity. Combining with its large specific surface area and high degree graphitization, the O-N-CNs can be expected with remarkable electrocatalytic performance for ORR, OER, and HER. The electrocatalytic ORR/OER/HER performances for the obtained different carbon catalysts were evaluated in 0.1 M KOH electrolyte. For the ORR, cyclic voltammetric (CV) measurements were firstly performed and the CV curves of O-N-CNs, O-CSs, and Pt-C (20 wt%) were provided in the Figure 4A. Compared to the O-CSs and Pt-C, a much more obvious redox peak was observed for the O-N-CNs in O2 saturated electrolyte, but there is no characteristic oxygen reduction peak for all the three samples in N2 saturated electrolyte, implying the higher sensitivity of O-N-CNs toward ORR. Rotating disk electrode (RDE) tests were then performed for O-N-CNs, O-CSs, and Pt-C to further investigate their ORR activity (Figure 4B). The O-N-CNs gives a limiting current density of 8.0 mA cm–2 at 0.2 V versus reversible hydrogen electrode (vs. RHE, j0.2 V), exceeding the O-CSs (j0.2 V=6.4 mA cm–2), PtC (j0.2 V=6.7 mA cm–2) and some reported state-of-the-art electrocatalysts (Table S2). In addition, the half-wave potential (E1/2) of O-N-CNs is 0.88 V vs. RHE, which is more positive than those of the O-CSs (E1/2=0.86 V vs. RHE), Pt-C (E1/2=0.87 V vs. RHE) and most of the N-doped nanocatalysts in the reported work (Table S2).

15 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

A

O-CSs O-N-CNs Pt-C

B

0 O-CSs O-N-CNs Pt-C

-2 j / mA cm

j / mA cm

-2

-2

5 mA cm-2

-4 -6 -8

0.4

C 0

ir

D

0.4 0.6 0.8 Potential / V vs. RHE

25

4

O-N-CNs 20 O-CSs 15 Pt-C 10

3

-4

2

id

-6

5

1

-8 0.2

E

0.4 0.6 0.8 Potential / V vs. RHE

1.0

O-CSs O-N-CNs Pt-C

j / mA cm

-2

0.95

F

0.90

1.0

30

5

n

-2

0.2

1.0

O-CSs O-N-CNs Pt-C

-2

j / mA cm

0.6 0.8 E / V vs. RHE

H2O2 / %

0.2

Potential / V vs. RHE

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 30

0 0 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Potential / V vs. RHE 0 O-CSs O-N-CNs -2 Pt-C

-4 -6 -8

0.85

-10

0

4

8 12 -2 ik / mA cm

16

0

5000

10000 15000 Time / s

20000

Figure 4. (A) Cyclic voltammograms of ORR at a scan rate of 50 mV s–1 in N2 and O2-saturated 0.1 M KOH aqueous solution; (B) LSVs for ORR and (C) the ring current, disk current at a rotational speed of 1600 rpm at a scan rate of 10 mV s–1 in O2-saturated 0.1 M KOH aqueous solution on O-CSs, O-N-CNs and Pt-C electrodes; (D) the electron transfer number and H2O2 production yields of ORR, (E) specific kinetic current densities (ik) at different potentials and (F) chronoamperometry curves on corresponding electrodes.

To probe into the mechanism of ORR on O-N-CNs, O-CSs, and Pt-C, we carried out the rotating ring-disk electrode (RRDE) liner scan voltammetric curves (LSVs) at 1.5 V vs. RHE under 1600 rpm (Figure 4C) and RDE LSVs at rotation speed ranging from 400 to 2500 rpm (Figure S12A-C). In comparison with the O-CSs and Pt-C, the resulted RRDE LSVs in Figure 4D suggests the O-N-CNs with a similar four electron transfer number for the ORR at almost all measured potentials and a pretty lower H2O2 yield (