S Co-Doped

Subscriber access provided by The University of Manchester Library

Article

In-situ Generated Dual-template Method for Fe/N/S Co-doped Hierarchically Porous Honeycomb Carbon for High Performance Oxygen Reduction Hongju Zeng, Wang Wang, Jun Li, Jin Luo, and Shengli Chen ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b19645 • Publication Date (Web): 26 Feb 2018 Downloaded from http://pubs.acs.org on February 27, 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.

ACS Applied Materials & Interfaces 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 15 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

In-situ Generated Dual-template Method for Fe/N/S Co-doped Hierarchically Porous Honeycomb Carbon for High Performance Oxygen Reduction Hongju Zeng, Wang Wang, Jun Li, Jin Luo, and Shengli Chen* Hubei Key Laboratory of Electrochemical Power Sources, Key Laboratory of Analytical Chemistry for Biology and Medicine (Ministry of Education), Department of Chemistry, Wuhan University, Wuhan 430072, China.

ABSTRACT: Heteroatoms doping is able to produce catalytic sites in carbon materials for oxygen reduction reaction (ORR); while hierarchically porous structure is necessary for efficient exposure and accessibility of the usually limited catalytic sites in such activated carbon catalysts. This work reports an in-situ generated dual-template method to synthesis the Fe/N/S co-doped hierarchically porous carbon (FeNS/HPC), with NaCl crystallites formed during the precursor lyophilization process as the primary template to generate ~500 nm macropores with ultrathin graphene-like carbon-layer walls, and Fe3O4 nanoparticles formed during the hightemperature carbonization process as the secondary template to produce mesopores on the walls of macropores. As well as the coexistence of graphitic-N, pyridinic-N and thiophene-S which are beneficial to ORR, the as prepared FeNS/HPC possesses a highly graphitized and interconnected hierarchical porous structure, giving a specific surface area as high as 938 m2 g-1. As a consequence, it exhibits excellent four-electron oxygen reduction performance in both alkaline and acid electrolytes. The in-situ generation and facile solution removal make the present template method a promising way for scale-up preparation of active porous carbon materials for various applications.

KEYWORDS: oxygen reduction reaction, hierarchically porous carbon, nitrogen doping, sulfur doping, dual template method, NaCl crystal template

INTRODUCTION The global energy crisis has motivated a worldwide search for clean and renewable energy sources in recent years, including fuel cells1-2 and metal-air batteries3-4. However, the cathodic oxygen reduction reaction (ORR), the crucial step in these devices, is kinetically sluggish5 and requires far more Platinum (Pt) catalyst than anodic hydrogen oxidation on electrodes. Nowadays, Pt and Pt-based catalysts are regarded as the most active electrocatalysts for ORR but they are low in reserves, overpriced, and the stability is poor.6 Therefore, great efforts and progress have been made to develop low cost and highly active nonprecious electrocatalysts (NPMCs).3, 7-10 In recent years, Fe and heteroatom (B, N, F, P, or S)-doped carbon materials have become more and more attractive to ORR catalysts,6, 11-15 especially Fe/N/S co-doping ones which have shown promising ORR activity enhancement. 5, 8, 16 As for N atom, it

1 Environment ACS Paragon Plus

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 15

has the similar radius with carbon and higher electronegativity than carbon, which guarantees that the carbon frame will remain integrated after N doping, while at the same time activating the neighboring carbon atoms for reducing O2 due to electronic interaction and/or through the formation of pyridinic and other active N species. Regarding sulfur, it has been found that the sulfur species such as thiophene-S at the zigzag and armchair edges of the carbon framework can induce “electron spin” redistribution, contributing to the enhancement of ORR activity.17 The catalytic electrodes based on doped carbon materials generally suffer from a mass transfer problem because of the high catalyst loading. This problem will naturally reduce the effective utilization of the active sites 18-24, sharply reducing their performance. Hierarchically porous carbon materials (HPCs), with well-distributed micro-, meso-, and macropores would possibly solve this problem.16, 25-27 Macropores and large mesopores can provide transport pathways for oxygen and electrolyte, while enormous micropores as well as small mesopores allow efficient exposure of active sites. 28-29 Usually, HPCs are attained by using various hard or soft templates such as silica15, montmorillonite15, 30 and polystyrene, etc. These templates inevitably involve complicated synthesis and removal procedures, which usually need to use toxic chemicals like organic solvents and HF30, limiting the scale-up production of catalysts. Thus, there is a great need to develop a much more facile template method to synthesize HPCs. In this work, we present a facile template synthesis of Fe/N/S co-doped hierarchically porous carbon (FeNS/HPC) through an in-situ generated dual-template method. In which sucrose, thiourea and ferric chloride were pyrolyzed in the existence of NaCl submicro-crystallites template generated in the process of precursor lyophilization, and homogenoesly dispersed Fe 3O4 nanoparticles template generated during this pyrolysis process. Following the pyrolysis, a H2SO4-leaching process was used to remove the templates, which resulted in FeNS/HPC catalyst, having abundant macropores with the average size below 500 nm and having mesoscale pores on the walls of these macropores (Scheme 1). The very limited amounts of precursor substances squeezed within the packed NaCl crystals allow the formation of graphene-like ultrathin carbon walls and abundant micropores during hightemperature pyrolysis. NaCl, as confining agents can also reduce the weight loss of precursors and active intermediates during hightemperature pyrolysis process and increase the graphitization degree of the resulted carbon materials. 18, 31-33 The coordination between thiourea and ferric ions should make Fe homogeneously distribute in the precursor, causing the formation of highly dispersed fine Fe oxide particles in the high-temperature process. The prepared FeNS/HPC shows a hierarchically porous structure, large graphitization degree, the co-existence of active N species including graphitic-N (54%), pyridinic-N (32%), and thiophene-S (83%). With the synergistic effect of the above aspects, the FeNS/HPC presents superior activity and long-term stability for ORR. The insitu generation during the material formation process and the facile removal through dilute sulfuric acid solution leaching make the present template method a relatively straightforward way to prepare active HPCs for various applications.

EXPERIMENT SECTION Materials Preparation. Thiourea (99 wt %) and sucrose (99 wt %) were purchased from Aladdin Industrial Corporation (America) in China. FeCl3·6H2O (analytical grade) and NaCl (guaranteed reagent) were purchased from Shanghai Chemical Reagent Corporation. (China). All the chemicals were used as received.


Page 3 of 15 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 In a typical synthesis of sample FeNS/HPC, it was prepared by dissolving 1 g (13 mmol) Thiourea, 1 g sucrose (Suc for short)

and 36 mg FeCl36H2O (0.13 mmol, with S/Fe molar ratio equals to 100) into 12 mL saturated NaCl solution, followed by 5 minutes of ultrasonic irradiation to obtain a crystal clear solution. The solution was then immediately frozen into a white solid bulk by injecting into enough liquid nitrogen, followed by putting it into a lyophilizer for 12 h. The resulted white powder was transferred to a ceramic crucible to be heated under an argon atmosphere to 900C and kept at this temperature for 1 h. After cooling to room temperature, the black sample was grounded up and ultrasonically rinsed with 0.5 M H2SO4 at 80 C for 5 h to remove the salts and other impurities. The wet sample was then collected by filtration and dried in a vacuum at 80 C for 12 h. Finally, the obtained black powder was annealed at 900 C again for 2 h to get the FeNS/HPC catalyst. The samples with different ratio of S/Fe that are FeNS/HPC-50 and FeNS/HPC-10, without the addition of NaCl that is FeNS/C, without Fe that is NS/HPC and the sample prepared though water leaching that is FeNS/HPC-W, were all synthesized in identical conditions. (see Supporting Information). Structure and Morphology Characterization. The structure and morphology of the prepared materials were characterized by Transmission Electron Microscopy (TEM, JEM-2100F), Powder X-ray diffraction (XRD, D8-Advance Bruker, CuKa radiation source, =0.154178 nm), and Raman spectrometer (Renishaw in Visa, 532 nm excitation wavelength). X-ray photoelectron spectroscopy (XPS) spectra were recorded on an ESCALAB 250Xi, produced by Thermo Fisher Corporation, with a monochromic Al Kα source. The Brunauer–Emmett–Teller (BET) surface area was recorded by N2 adsorption using a Quantachrome NOVA 4200e. The micropores distribution of prepared catalysts was measured by by N2 adsorption on a Quantachrome AutoSorb iQ. SEM images were obtained with a Zeiss Sigma scanning electron microscope. Electrochemical Measurements. All the ORR performance data of the prepared catalysts were collected at room temperature. The electrocatalytic experiments were performed in a standard three-electrode cell with a Pt plate as the counter electrode, saturated calomel electrode (SCE) and Hg/HgO electrode as the reference electrode in 0.1 M HClO4 and 0.1 M KOH respectively. The RDE measurements of all catalysts were achieved on the surface of a glassy carbon rotating disk electrode (GC RDE) and the RRDE performance was carried out with a glassy carbon rotating ring-disk electrode (GC RRDE), both of which had been polished using alumina (0.05 mm) slurry and washed with ultrapure water. The RDE (5 mm in diameter) and RRDE (5 mm in diameter for disk electrode) loading with the catalyst sample served as the working electrode. To prepare the working electrode, 5 mg catalyst was first dispersed in 1 mL mixture of 5 wt% Nafion and isopropanol (volume ratio equaled to 1:49) under sonication for 30 min to get a homogeneous ink. Then, 20 μL ink (loading, 500 μg cm-2) suspension liquid was deposited on the surface of the GC RDE, and dried naturally. Pt/C (Johnson Matthey, 20 wt % Pt/C) was used as reference catalyst. For Pt/C in comparison, 3 μL of catalyst ink was injected onto the GC electrode, corresponding to a Pt loading of 15 μg cm-2 on electrode. All potentials in this work were calibrated to the RHE. The steady-state polarization curves and cyclic voltammograms (CV) were tested to evaluate the electrocatalytic performance of the catalysts for ORR on a CHI400 electrochemical workstation. The steady-state polarization curves were measured using RDE with an electrode rotating speed of 900 rpm and potential scanning rate of 5 mV s-1 in O2-saturated solution in 0.1 M KOH or 0.1 M



Page 4 of 15

HClO4. The CV measurements were performed with a potential scanning rate of 50 mV s-1 in both O2- and Ar-saturated electrolytes. RRDE experiments were carried out on a CHI900 electrochemical workstation and the constant ring potential of the ORR was set at 1.26 V to detect the mediate product of H2O2. In the accelerated durability tests (ADTs), the working electrode was cycled within a potential range of 0.6–1.0 V in O2-saturated 0.1 M HClO4 or 0.1 M KOH at 100 mV s-1 using a graphite rod as a counter electrode. Thecorresponding steady-state polarization curves of the catalyst were tested before and after the cycling. The kinetic current density (jK), which describes the intrinsic catalytic activity of the catalyst at a certain potential, was obtained by the mass transport correction based on the Koutecky–Levich equation: 𝑗K =

𝑗 × 𝑗L 𝑗L − 𝑗

where j and jL are the measured current density and diffusion-limiting current density on the polarization curve respectively. Four-electron selectivity of the catalyst was evaluated on a basis of H2O2 yield, which was calculated from the following equation: H2 O2 (%) = 200 ×

I R ⁄N (IR ⁄N) + ID

The electron transfer number (n) was calculated from the following equation: n= 4×

ID (IR ⁄N) + ID

where ID and IR refer to the disk and ring currents, respectively. N is the ring collection efficiency, which was calibrated using 10 mM K3[Fe(CN)6] in 0.1 M KNO3. The measured value for N is 0.241, which is close to the producer’s value of 0.25.

RESULTS AND DISCUSSION Scheme 1. Schematic illustration of the preparation process of FeNS/HPC

In the present synthesis of FeNS/HPC (Scheme 1), sucrose is the carbon precursor, and thiourea is a source for both nitrogen and sulfur. Here, thiourea also acts as coordination agent to ensure the homogeneous distribution of Fe ions in the pyrolytic precursor mixture. NaCl crystal served as the frame template to generate rich macropores. Meanwhile, the Fe oxide nanoparticles formed during the pyrolysis served as the templates of abundant mesopores. The morphology of the prepared FeNS/HPC is firstly examined by using scanning electron microscope (SEM) (Figure 1a, 1b), which reveals a comby, 3D and interconnect porous framework


Page 5 of 15 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


with compact cubic and spherical macropores constructed by ultra-thin carbon layers. The size of these macropores inherited from the NaCl crystal template generated in the process of lyophilization is 500 nm in average. To further investigate the role of NaCl crystals in the formation of the hierarchically porous structure, we prepared catalyst (denoted as FeNS/C) without the addition of NaCl. As compared to the FeNS/HPC, FeNS/C shows a totally different morphology, dominated by bulk structure without obvious porous structure (Figure S1), confirming that the NaCl crystals did serve as templates to form the macropores in FeNS/HPC.

Figure 1. (a, b) SEM and (c, d) TEM images of the FeNS/HPC catalyst prepared through acid solution leaching after pyrolysis; (e) TEM and HRTEM (inset) images of the catalyst prepared through water leaching instead of acid solution leaching (FeNS/HPC-W) after pyrolysis; (f) XRD patterns of FeNS/HPC-W and FeNS/HPC.

Transmission electron microscope (TEM) and X-ray diffraction (XRD) measurements were carried out to further study the structural features and composition of the prepared samples. As seen from the TEM images in Figure 1c and 1d, the FeNS/HPC possesses ultra-thin graphene-like wrinkled pore walls, in which there are abundant mesoscale pores with their size ranging from 5 nm to 50 nm (Figure 1d). Except the peak for (002) plane of graphitic carbon, no characteristic diffraction peak is detected on the



Page 6 of 15

FeNS/HPC (Figure 1f, black curve). This suggests the absence of segregated phase of elementary Fe, or Fe carbides, nitrides or sulfides, which agrees well with the TEM image in Figure S2. Thus, if there is any, Fe should be homogeneously dispersed in the carbon lattice of graphene walls. This could be a result of the coordination interaction between ferric ion and S and N. We hypothesize that the large amounts of mesopores on the graphene walls were produced due to the removal of Fe-based nanoparticles formed during the pyrolysis. To verify this, a comparison sample was prepared by leaching the pyrolytic product with ultrapure water instead of acid solution. As seen from the SEM images in Figure S3a and 3b, the thus-obtained FeNS/HPC-W sample shows similar comby 3D interconnected porous framework to that of FeNS/HPC. No Fe-based nanoparticles are seen in the SEM images, which should be due to that the SEM samples are generally produced as thick and aggregated form, so that the very small nanoparticles are largely wrapped into thick carbon layers and hardly visualized by the SEM. As shown in the TEM image in Figure 1e, however, large amount of nanoparticles are indeed present in the FeNS/HPC-W sample. The insert image in Figure 1e gives the HRTEM image and the corresponding Fourier-transformed reciprocal space mapping (FT-RSM) of a typical nanoparticle, which indicated a crystalline lattice distances of 0.25 nm for (111) plane and 0.48 nm for (311) plane and were consistent with those revealed by the XRD data (Figure 1f, red curve), namely, 0.253 nm for the (111) plane (18.2º) and 0.485 nm for the (311) plane (35.5º) of magnetite Fe3O4 respectively. The other XRD peaks marked with hollow triangles (Figure 1f, red curve), locating at 30.1º, 43.1º, 53.4º, 57.0ºand 62.5º, also correspond to those of magnetite Fe3O4 (JCPDS, No.07-0322). The diffraction patterns associated with Fe3O4 did not appear on the XRD responses of the FeNS/HPC prepared by acid-leaching (Figure 1f, black curve), which should have removed the Fe oxide nanoparticles and produced the mesoscale pores on the carbon walls. We also conducted EDX mapping test with the FeNS/HPC-W sample. The results show that Fe is evenly distributed in the carbon matrix (Figure S3c-f). This should further support the TEM and XRD results that suggested the great existence of Fe 3O4 nanoparticles. It suggests that the nanoparticles are homogeneously distributed in the carbon walls rather than only dispersed on the surface of carbon walls; otherwise the element map of Fe should exhibit some concentrated area of Fe signals. Thus we can draw a conclusion that these in-situ generated Fe3O4 nanoparticles did act as a template to produce abundant mesoscale pores seen in the carbon walls (Figure 1d). Figure S4 compares the N 2 adsorption−desorption isotherms and pore size distribution of FeNS/HPC and FeNS/HPC-W samples. It can be seen that the FeNS/HPC absorbed much more N2 and possessed a larger pore volume in the mesoscale region, which again implies the template effect of Fe3O4 nanoparticles. Figure 2a compares the nitrogen adsorption/desorption isotherms of the FeNS/HPC with that of the NS/HPC prepared without adding Fe salt in the precursors. The two catalyst samples both possess type II isotherms with H4 loops according to the IUPAC classification. The drastic increase on the curve at low pressure (P/P 0 < 0.02) indicates the presence of rich micropores while the sharp increase at high pressure (P/P0 > 0.9) confirms the presence of plenty amounts of macropores. As compared with NS/HPC, FeNS/HPC shows not only higher amount of N2 adsorption but also an much larger extended loop at medium pressure (0.2 < P/P 0 < 0.8), which indicates the existence of abundant mesopores in this catalyst. Moreover, the BET specific surface area (SSA) calculated from the isotherms increases from 727 m2 g−1 for NS/HPC to 938 m2 g−1 for FeNS/HPC. As shown in Table S6, the SSA of the


Page 7 of 15 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


FeNS/HPC is much higher than the carbon-based materials prepared though pyrolysis reported in most of other studies, and is comparable with the highest ones.

Figure 2. (a) Nitrogen adsorption−desorption isotherms and (b) Pore size distribution of FeNS/HPC and NS/HPC, The inset is pore size distribution of micropores.

Figure 2b shows the pore size distribution of the two catalysts in the mesoscale region calculated from BJH desorption and the micropore distribution (the insert) derived from the QSDFT analysis. In the size range below 5 nm, both catalysts displays large pore volume, corresponding to abundant micropores (0.5-2 nm) and small mesopores (2-5 nm) which can be commonly generated from the decomposition of organic precursors in the pyrolytic process34. These nanopores would bring about more exposed active sites18. For FeNS/HPC, there is considerable pore volume in the pore size spanning from 5 nm to 50 nm which is in consistence with the size distribution of the mesopores seen in the TEM images (Figure 1d). While NS/HPC exhibits the same macroporous morphology (Figure S10a) as sample FeNS/HPC but negligible pore volume in this mesopores region. This indicates that the Fe salt, leading to the formation of Fe3O4 nanoparticles, has really played a template functional role in the formation of mesoscale pores. Figure 3 displays the XPS characterization results for the FeNS/HPC and NS/HPC. The wide range of XPS survey scan give peaks for S 2s, S 2p, C 1s, N 1s and Fe 2p; while the Fe peak is absent for NS/HPC. Table S1 gives the XPS derived elemental composition information, which suggests that the FeNS/HPC contains a small amount of Fe (0.43%), similar C (90.54% vs 90.32%) but less S (0.52% vs 0.68%) and N (4.94% vs 5.54%) as compared with NS/HPC. As shown in Figure S5, Table S2 and Figure 3b, N exists in three forms, namely, oxidized-N (403.7 eV), graphitic-N (401.2 eV) and pyridinic-N (398.4 eV).35-36 For FeNS/HPC, the contents are 14.15%, 53.83% and 32.02%, respectively. The NS/HPC possesses a little higher content of oxidized-N (17.16%), similar proportion of graphitic-N (54.51%), and less content of pyridnic-N (28.33%). The S 2p responses are given in Figure 3c and Table S3. The responses at binding energy of 163.9 eV and 165.0 eV can be attributed to the thiophene-S (82.63% for FeNS/HPC, 71.82% for NS/HPC) and the one at 169.0 eV should be associated with the oxidized-S species (17.37% for FeNS/HPC, 28.18% for NS/HPC)17, 37. Figure 3d shows the high-resolution scan of Fe 2p, where the deconvolution yields two pairs of peaks for Fe2+ (710.4 and 722.9 eV) and Fe3+ (713.2 and 725.0 eV). The EDX Mapping results (Figure S6) indicates a uniform distribution of the Fe ele-



Page 8 of 15

ment in the FeNS/HPC. Considering the very low total Fe content and no segregated phase of Fe species as indicated through TEM observation (Figure 1d and Figure S2), we may conclude that Fe in oxidized states were homogeneously entrapped in the porous carbon-based matrix of FeNS/HPC38-39,36,8, probably through coordinating with N and S to form well-distributed active sites.

Figure 3. (a) XPS survey spectra for FeNS/HPC and NS/HPC; narrow-scan spectra of (b) N 1s, (c) S 2p, (d) Fe 2p for FeNS/HPC; and (e) atomic percentage of different N and S species in the total content of N and S element in the final catalyst, respectively.

The Raman results (Figure S7) indicates that the FeNS/HPC had higher graphitization degree than NS/HPC, with the intensity ratio of D band (~1360 cm-1, attributing to the defects and disorder in graphitic lattice) to G band (~1600 cm -1, in-plane vibration of the sp2 carbon network) being 0.89 and 0.92 respectively. As reported in literatures, the transition metal components such as Fe often act as catalysts to promote the decomposition of carbon precursors and the formation of the fragments into graphene sheets.40


Page 9 of 15 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 Cyclic voltammetry (CV) measurements were carried out in 0.1 M KOH (Figure 4a) and 0.1 M HClO4 (Figure 4b). Both

FeNS/HPC and NS/HPC shows quasi-rectangular electric-double-layer charging/discharging voltammetry in Ar-saturated electrolyte. In O2-saturated electrolytes, the FeNS/HPC shows much more positive ORR peak potential in both KOH and HClO4 solution, demonstrating higher catalytic activity than NS/HPC. As depicted in Figure 4c and 4d, the FeNS/HPC catalyst exhibits much more positive half-wave potential and onset potential (Table S4) than NS/HPC in acid and alkaline medium. This further demonstrates the better ORR activity of FeNS/HPC. Since that the XPS results have suggest no significant difference in the contents and distribution of N and S species, the catalytic activity difference should be mainly ascribed to the hierarchically porous structure. To be more precise, the mesopores generated through the acid leaching of Fe oxide nanoparticles did contribute much to the better catalytic activity. Besides, the high graphitization degree and the co-existence of pyridinic-N and thiophene-S (Table S2 and S3) should also benefit the ORR activity. As compared with the Pt/C catalyst (20 wt% Pt/C, Johnson-Matthey), the FeNS/HPC exhibited a half-wave potential that was ca. 40 mV more positive in 0.1 M KOH and only ca. 55 mV less positive in 0.1 M HClO4. We optimized the catalyst preparation by changing the amount of thiourea in the precursor while keeping the amounts of other components unchanged. The S/Fe molar ratio (X) was used as a controlling parameter for this optimization. It is found that decreasing the X to values below 100 would cause a decline in ORR activity (Figure S8). This should be mainly due to the decrease in the doping contents of S and N, which should result in decreased density of active sites (Table S5). Interestingly, the variation of X also changed the pore structure and the N2 adsorption isotherm accordingly, of the resulted catalyst. As seen in Figure S9, the N 2 adsorption amount decreases with decreasing the value of X. This should be due to a balanced effect of the macropores and mesopores. Firstly, the decrease of X value corresponded to an increase of Fe content with respect to the organic components in the precursor, which should result in more Fe3O4 nanoparticle templates, and therefore more mesopores in the prepared catalyst. However, the decrease of X also corresponded to an increase of NaCl content with respect to the organic components in the precursor. This could result in larger NaCl crystallites, and therefore larger macrpores in the final catalyst. The decrease in N 2 adsorption with decreasing the value of X below 100 thus seemed to suggest that the increased macropore sizes disfavoured the N2 adsorption. It can be seen that the FeNS/HPC-50, NS/HPC (X=100), and FeNS/HPC-W (X=100) exhibited very similar N2 adsorption. The decreased N2 adsorption of NS/HPC and FeNS/HPC-W with respect to that of FeNS/HPC should mainly due to the decrease in mesopore volumes, because they were all prepared with X=100 but there is no mesopore in NS/HPC and FeNS/HPC-W. The similar N2 adsorption of FeNS/HPC-50 to that of the NS/HPC and FeNS/HPC-W suggested that the increased adsorption due to mesopores in FeNS/HPC50 were nearly completely balanced by the decreased adsorption due to the increase in the sizes of the macropres. Further decrease X value to 10, the disfavoring effect of the macropores might surpass the favoring effect of mesopores, so that FeNS/HPC-10 exhibited even less adsorption than NS/HPC. Figure S10 compares the SEM images of the NS/HPC and FeNS/HPC-10, showing clearly that the latter possesses much larger macropores. The FeNS/HPC catalyst prepared with X=100 and with acid-leaching provided an optimization on the mesopore and macropore effects. We also compared the catalyst in this work with other top nonprecious metal electrocatalysts based on porous carbon materi-



Page 10 of 15

als, showing the prepared FeNS/HPC is among the most active ORR catalysts in an alkaline media and comparable to those of nonprecious catalysts in acid media (Table S6).

Figure 4. CVs of FeNS/HPC and NS/HPC in (a) 0.1 M KOH and (b) 0.1 M HClO4 at 50 mV s-1 in O2-saturated solution (solid line) and Ar-saturated solution (dash line); RDE voltammograms of Pt/C (Johnson-Matthey, 20 wt % Pt/C), FeNS/HPC and NS/HPC in O2-saturated (c) 0.1 M KOH and (d) 0.1 M HClO4 at 5 mV s-1 and 900 rpm, the insert figures show the corresponding tafel plots of the three catalysts in kinetic region; RRDE test for peroxide yield and electron transfer number (n) of Pt/C and FeNS/HPC in (e) 0.1 M KOH and (f) 0.1 M HClO4, respectively. (The curves with same colour present the same catalyst.)


Page 11 of 15 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 The mechanistic and kinetic properties of the catalysts toward ORR were further evaluated using diffusion-corrected Tafel

plots (the insert figures in Figure 4c and 4d).The Tafel slopes of FeNS/HPC and NS/HPC are calculated to be 63 mV/decade and 58 mV/decade in 0.1 M KOH, much smaller than those of Pt/C calculated to be 93 mV/decade. In 0.1 M HClO4, the corresponding tafel slopes are 72 mV/decade and 77 mV/decade, slightly higher than the Pt/C (68 mV/decade). Based on the Tafel's law h= a + b*log j, over potential increases as Tafel slope become lager in linear region. The tafel slope of FeNS/HPC again displays its better ORR ability than Pt/C in an alkaline medium and comparable performance in an acid medium. To further clarify the high activity of the prepared FeNS/HPC catalyst, we calculated the electron-transfer number (n) of ORR for our catalyst and Pt/C. Generally, 4e- process is more efficient than 2e- process in which H2O2 will produce. As shown in Figure 4e and 4f, the RRDE experiments present that the peroxide yield on our as synthesized catalyst remains below 4% all though and less than 2% in the high polar potential range in 01 M HClO4, and remains below 2.5 % in 0.1 M KOH in the whole potential region. Based on the RRDE data, the calculated electron‐transfer number (n) on FeNS/HPC is over 3.9 in both acid and alkaline medium, which means that our catalyst follows a direct 4e- pathway where the O2 is transformed straightforward into water (in acid medium) or OH- (in alkaline medium), close to the ideal reaction pathway. Apparently, the ORR is dominated by a 4e - process on Pt/C.

Figure 5. Rotating disk electrode (RDE) voltammograms of Pt/C and FeNS/HPC in O2-saturated (a) 0.1 M KOH and (b) 0.1 M HClO4 at 5 mV s-1 and 900 rpm, before and after the ADTs within a potential range of 0.6-1.0 V in O2-saturated electrolyte at 100 mV s-1. The stability of the electrocatalyst for ORR also plays an important role in its practical use. To evaluate the stability of the prepared FeNS/HPC, the ORR performance change before and after the accelerated durability tests (ADTs) were measured by cycling the working electrode in a potential range of 0.6–1.0 V in O2-saturated 0.1 M HClO4 and 0.1 M KOH. As shown in Figure 5a, the prepared FeNS/HPC basically remains all the initial limiting current and only undergoes a 25 mV negative shift of half-wave potential in 0.1 M HClO4, slightly larger than Pt/C after 2500 cycles. While figure 5b presents that the E1/2 of FeNS/HPC exhibits only 21



Page 12 of 15

mV negative shift after 8000 cycles in 0.1 M KOH, which is smaller than the twice of the Pt/C, respectively. As seen in Figure S11, the interconnected hierarchically porous structure of the FeNS/HPC changed negligibly after the ADTs both in acid or alkaline medium. Table S7 provides the XPS estimation of the corresponding composition change, which shows that the contents of C and O changed in alkaline medium, while the content of N also changed in acid medium. These results indicates that the slightly activity decline in alkaline can be ascribed to the corrosion of carbon 41. The relatively lower stability in acid media was due to the loss of the nitrogen-based active sites.

CONCLUSION Interconnected hierarchically porous carbon co-doped with Fe/N/S elements has been synthesized through a facile dualtemplate method in which NaCl crystals result in large amount macropores (~500 nm) and in-situ generated Fe3O4 nanoparticles lead to abundant mesopores. The hierarchically porous structure offers a specific surface area as high as 938 m2 g-1, making the catalyst exhibit superior electrocatalytic performance as compared with the state-of-the-art Pt/C in alkaline medium, as indicated by the higher onset potential and half-wave potential. The in-situ generation and aqueous removal make the present template method a promising way to prepare active HPCs for various applications.

ASSOCIATED CONTENT Supporting Information The supplementary information, including SEM and TEM images, XRD patterns, Raman, XPS survey spectras and CV corresponding to the prepared catalysts, the half wave potential and on set potential of the prepared catalysts, is available free of charge via the internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Authors * Shengli Chen: [email protected].

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT This work was supported by the Natural Science Foundation of China (Grant Nos. 21633008 and 21673163).

REFERENCES (1) Dai, Y.; Chen, S. Oxygen Reduction Electrocatalyst of Pt on Au Nanoparticles Through Spontaneous Deposition. ACS Appl. Mater. Interfaces 2015, 7, 823-829. (2) Sasaki, K.; Naohara, H.; Choi, Y.; Cai, Y.; Chen, W. F.; Liu, P.; Adzic, R. R. Highly Stable Pt Monolayer on PdAu Nanoparticle Electrocatalysts for the Oxygen Reduction Reaction. Nat. Commun. 2012, 3, 1-9. (3) Zhang, J.; Zhao, Z.; Xia, Z.; Dai, L. A Metal-free Bifunctional Electrocatalyst for Oxygen Reduction and Oxygen Evolution Reactions. Nat. Nanotechnol. 2015, 10, 444-453. (4) Dai, Y.; Ou, L.; Liang, W.; Yang, F.; Liu, Y.; Chen, S. Efficient and Superiorly Durable Pt-Lean Electrocatalysts of Pt−W Alloys for the Oxygen Reduction Reaction. J. Phys. Chem. C 2011, 115, 2162-2168. (5) Wang, Y. C.; Lai, Y. J.; Song, L.; Zhou, Z. Y.; Liu, J. G.; Wang, Q.; Yang, X. D.; Chen, C.; Shi, W.; Zheng, Y. P.; Rauf, M.; Sun, S. G. SDoping of an Fe/N/C ORR Catalyst for Polymer Electrolyte Membrane Fuel Cells with High Power Density. Angew. Chem. 2015, 54, 1004510048.


Page 13 of 15 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


(6) Kim, J. H.; Sa, Y. J.; Jeong, H. Y.; Joo, S. H. Roles of Fe-Nx and Fe-Fe3C@C Species in Fe-N/C Electrocatalysts for Oxygen Reduction Reaction. ACS Appl. Mater. Interfaces 2017, 9, 9567-9576. (7) Cao, C.; Wei, L.; Wang, G.; Shen, J. Superiority of Boron, Nitrogen and Iron Ternary Doped Carbonized Graphene Oxide-based Catalysts for Oxygen Reduction in Microbial Fuel Cells. Nanoscale 2017, 9, 3537-3546. (8) Hu, K.; Tao, L.; Liu, D.; Huo, J.; Wang, S. Sulfur-Doped Fe/N/C Nanosheets as Highly Efficient Electrocatalysts for Oxygen Reduction Reaction. ACS Appl. Mater. Interfaces 2016, 8, 19379-19385. (9) Qiao, X.; Peng, H.; You, C.; Liu, F.; Zheng, R.; Xu, D.; Li, X.; Liao, S. Nitrogen, Phosphorus and Iron Doped Carbon Nanospheres with High Surface Area and Hierarchical Porous Structure for Oxygen Reduction. J. Power Sources 2015, 288, 253-260. (10) Gao, S.; Liu, H.; Geng, K.; Wei, X. Honeysuckles-derived Porous Nitrogen, Sulfur, Dual-doped Carbon as High-performance Metal-free Oxygen Electroreduction Catalyst. Nano Energy 2015, 12, 785-793. (11) Zhu, H.; Sun, Z.; Chen, M.; Cao, H.; Li, K.; Cai, Y.; Wang, F. Highly Porous Composite Based on Tungsten Carbide and N-doped Carbon Aerogels for Electrocatalyzing Oxygen Reduction Reaction in Acidic and Alkaline Media. Electrochim. Acta 2017, 236, 154-160. (12) Wang, B.; Wang, X.; Zou, J.; Yan, Y.; Xie, S.; Hu, G.; Li, Y.; Dong, A. Simple-Cubic Carbon Frameworks with Atomically Dispersed Iron Dopants toward High-Efficiency Oxygen Reduction. Nano Lett. 2017, 17, 2003-2009. (13) Tan, H.; Li, Y.; Jiang, X.; Tang, J.; Wang, Z.; Qian, H.; Mei, P.; Malgras, V.; Bando, Y.; Yamauchi, Y. Perfectly Ordered Mesoporous IronNitrogen Doped Carbon as Highly Efficient Catalyst for Oxygen Reduction Reaction in Both Alkaline and Acidic Electrolytes. Nano Energy 2017, 36, 286-294. (14) Wu, K. H.; Shi, W.; Wang, D.; Xu, J.; Ding, Y.; Lin, Y.; Qi, W.; Zhang, B.; Su, D. In situ Electrostatic Modulation of Path Selectivity for the Oxygen Reduction Reaction on Fe-N Doped Carbon Catalyst. Chem. Mater. 2017, 29, 4649-4653. (15) Liu, X. J.; Zou, S. Z.; Chen, S. W. Ordered Mesoporous Carbons Codoped with Nitrogen and Iron as Effective Catalysts for Oxygen Reduction Reaction. Nanoscale 2016, 8, 19249-19255. (16) Kone, I.; Xie, A.; Tang, Y.; Chen, Y.; Liu, J.; Chen, Y.; Sun, Y.; Yang, X.; Wan, P. Hierarchical Porous Carbon Doped with Iron/Nitrogen/Sulfur for Efficient Oxygen Reduction Reaction. ACS Appl. Mater. Interfaces 2017, 9, 20963-20973. (17) Jeon, I.-Y.; Zhang, S.; Zhang, L.; Choi, H.-J.; Seo, J.-M.; Xia, Z.; Dai, L.; Baek, J.-B. Edge-Selectively Sulfurized Graphene Nanoplatelets as Efficient Metal-Free Electrocatalysts for Oxygen Reduction Reaction: The Electron Spin Effect. Adv. Mater. 2013, 25, 6138-6145. (18) Zhang, Y.; Huang, L. B.; Jiang, W. J.; Zhang, X.; Chen, Y. Y.; Wei, Z.; Wan, L.-J.; Hu, J. S. Sodium Chloride‐Assisted Green Synthesis of 3D Fe‐N‐C Hybrid as a Highly Active Electrocatalyst for Oxygen Reduction Reaction. J. Mater. Chem. A 2016, 4, 7781-7789. (19) Strickland, K.; Miner, E.; Jia, Q.; Tylus, U.; Ramaswamy, N.; Liang, W.; Sougrati, M.-T.; Jaouen, F.; Mukerjee, S. Highly Active Oxygen Reduction Non-platinum Group Metal Electrocatalyst without Direct Metal–nitrogen Coordination. Nat. Commun. 2015, 6, 67343-67351. (20) Sa, Y. J.; Seo, D. J.; Woo, J.; Lim, J. T.; Cheon, J. Y.; Yang, S. Y.; Lee, J. M.; Kang, D.; Shin, T. J.; Shin, H. S.; Jeong, H. Y.; Kim, C. S.; Kim, M. G.; Kim, T. Y.; Joo, S. H. A General Approach to Preferential Formation of Active Fe-Nx Sites in Fe-N/C Electrocatalysts for Efficient Oxygen Reduction Reaction. J. Am. Chem. Soc. 2016, 138, 15046-15056. (21) Kramm, U. I.; Herrmann-Geppert, I.; Behrends, J.; Lips, K.; Fiechter, S.; Bogdanoff, P. On an Easy Way to Prepare Metal–nitrogen Doped Carbon with Exclusive Presence of MeN4-type Sites Active for the ORR. J. Am. Chem. Soc. 2016, 138, 635-640. (22) Sahraie, N. R.; Kramm, U. I.; Steinberg, J.; Zhang, Y.; Thomas, A.; Reier, T.; Paraknowitsch, J.-P.; Strasser, P. Quantifying the Density and Utilization of Active Sites in Non-precious Metal Oxygen Electroreduction Catalysts. Nat. Commun. 2015, 6, 1-9. (23) Kramm, U. I.; Lefèvre, M.; Larouche, N.; Schmeisser, D.; Dodelet, J.-P. Correlations between Mass Activity and Physicochemical Properties of Fe/N/C Catalysts for the ORR in PEM Fuel Cell via 57Fe Mössbauer Spectroscopy and Other Techniques. J. Am. Chem. Soc. 2014, 136, 978985. (24) Kramm, U. I.; Herranz, J.; Larouche, N.; Arruda, T. M.; Lefèvre, M.; Jaouen, F.; Bogdanoff, P.; Fiechter, S.; Abs-Wurmbach, I.; Mukerjee, S. Structure of the Catalytic Sites in Fe/N/C-catalysts for O2-reduction in PEM Fuel Cells. Phys. Chem. Chem. Phys. 2012, 14, 11673-11688. (25) He, W.; Jiang, C.; Wang, J.; Lu, L. High‐Rate Oxygen Electroreduction over Graphitic‐N Species Exposed on 3D Hierarchically Porous Nitrogen‐Doped Carbons. Angew. Chem. 2014, 53, 9503-9507. (26) Wu, M.; Qiao, J.; Li, K.; Zhou, X.; Liu, Y.; Zhang, J. A Large-scale Synthesis of Heteroatom (N and S) Co-doped Hierarchically Porous Carbon (HPC) Derived from Polyquaternium for Superior Oxygen Reduction Reactivity. Green Chem. 2016, 18, 2699-2709. (27) Eisenberg, D.; Stroek, W.; Geels, N. J.; Tanase, S.; Ferbinteanu, M.; Teat, S. J.; Mettraux, P.; Yan, N.; Rothenberg, G. A Rational Synthesis of Hierarchically Porous, N-doped Carbon from Mg-based MOFs: Understanding the Link between Nitrogen Content and Oxygen Reduction Electrocatalysis. Phys. Chem. Chem. Phys. 2016, 18, 20778-20783.



Page 14 of 15

(28) Li, Y.; Zhang, H.; Wang, Y.; Liu, P.; Yang, H.; Yao, X.; Wang, D.; Tang, Z.; Zhao, H. A Self-sponsored Doping Approach for Controllable Synthesis of S and N Co-doped Trimodal-porous Structured Graphitic Carbon ELectrocatalysts. Energy Environ. Sci. 2014, 7, 3720-3726. (29) Zamani, P.; Higgins, D. C.; Hassan, F. M.; Fu, X.; Choi, J.-Y.; Hoque, M. A.; Jiang, G.; Chen, Z. Highly Active and Porous Graphene Encapsulating Carbon Nanotubes as a Non-precious Oxygen Reduction Electrocatalyst for Hydrogen-air Fuel Cells. Nano Energy 2016, 26, 267275. (30) Liang, H. W.; Wei, W.; Wu, Z. S.; Feng, X.; Mullen, K. Mesoporous Metal-nitrogen-doped Carbon Electrocatalysts for Highly Efficient Oxygen Reduction Reaction. J. Am. Chem. Soc. 2013, 135, 16002-16005. (31) Wang, W.; Luo, J.; Chen, W.; Li, J.; Xing, W.; Chen, S. Synthesis of Mesoporous Fe/N/C Oxygen Reduction Catalysts Through NaCl Crystallite-confined Pyrolysis of Polyvinylpyrrolidone. J. Mater. Chem. A 2016, 4, 12768-12773. (32) Liu, X.; Zou, S.; Chen, S. Ordered Mesoporous Carbons Codoped with Nitrogen and Iron as Effective Catalysts for Oxygen Reduction Reaction. Nanoscale 2016, 8, 19249-19255. (33) Wang, W.; Chen, W.; Miao, P.; Luo, J.; Wei, Z.; Chen, S. NaCl Crystallites as Dual-Functional and Water-Removable Templates to Synthesize a Three-Dimensional Graphene-like Macroporous Fe-N-C Catalyst. ACS Cata. 2017, 7, 6144-6149. (34) Pampel, J.; Fellinger, T.-P. Opening of Bottleneck Pores for the Improvement of Nitrogen Doped Carbon Electrocatalysts. Adv. Energy Mater. 2016, 6, 150238-150249. (35) Jingjing, X.; Fang, W.; Riguo, M.; Zhijie, G.; Xianping, F.; Hui, Y.; Liang, A.; Qixing, W.; Zhongkuan, L. Catalytic Performance of Pyrolyzed Graphene Supported Fe-N-C Composite and its Application for Acid Direct Methanol Fuel Cells. RSC Adv. 2016, 6, 90797-90805. (36) Wu, G.; Zelenay, P. Nanostructured Nonprecious Metal Catalysts for Oxygen Reduction Reaction. Acc. Chem. Res. 2013, 46, 1878-1880. (37) Jeon, I.-Y.; Zhang, S.; Zhang, L.; Choi, H.-J.; Seo, J.-M.; Xia, Z.; Dai, L.; Baek, J.-B. Edge-selectively Sulfurized Graphene Nanoplatelets as Efficient Metal-free Electrocatalysts for Oxygen Reduction Reaction: the Electron Spin Effect. Adv. Mater. 2013, 25, 6138-6145. (38) Cui, X.; Yang, S.; Yan, X.; Leng, J.; Shuang, S.; Ajayan, P. M.; Zhang, Z. Pyridinic‐Nitrogen‐Dominated Graphene Aerogels with Fe–N–C Coordination for Highly Efficient Oxygen Reduction Reaction. Adv. Funct. Mater. 2016, 26, 5708-5717. (39) Velázquez-Palenzuela, A.; Zhang, L.; Wang, L.; Cabot, P. L.; Brillas, E.; Tsay, K.; Zhang, J. Fe–Nx/C Electrocatalysts Synthesized by Pyrolysis of Fe(II)–2,3,5,6-tetra(2-pyridyl)pyrazine Complex for PEM Fuel Cell Oxygen Reduction Reaction. Electrochim. Acta 2011, 56, 47444752. (40) Wu, G.; More, K. L.; Johnston, C. M.; Zelenay, P. High-performance Electrocatalysts for Oxygen Reduction Derived from Polyaniline, Iron, and Cobalt. Science 2011, 332, 443-450. (41) Luo, Y.; Alonso-Vante, N. The Effect of Support on Advanced Pt-based Cathodes towards the Oxygen Reduction Reaction. State of the Art. Electrochim. Acta 2015, 179, 108-118.


Page 15 of 15 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



S Co-Doped

Recommend Documents