S,N co-doped reduced graphene oxide sheets with cobalt hydroxide nanocrystals for highly active and stable bifunctional oxygen catalysts

Sung-Woo Park, Hyun Jung Shin and Dong-Wan Kim*
School of Civil, Environmental and Architectural Engineering, Korea University, 145, Anam-Ro, Seongbuk-Gu, Seoul 136-713, South Korea. E-mail: dwkim1@korea.ac.kr

Received 30th August 2019 , Accepted 13th October 2019

First published on 15th October 2019


It is imperative to develop cost-effective and stable cathode materials with satisfactory activities for both oxygen evolution and oxygen reduction reactions (OER and ORR) in order to build next-generation rechargeable metal–air battery systems. Herein, we developed an effective strategy for fabricating highly active and stable bifunctional oxygen catalysts using a highly active ORR catalyst and hybridizing it with an OER catalyst. The effects of single (S)- and multi (S and N)-doping on the ORR catalytic efficiency of reduced graphene oxide (rGO) sheets were investigated. Cobalt hydroxide (Co(OH)2) nanocrystals with diameters of less than 5 nm were directly grown on SN-rGO via a simple precipitation method at room temperature to produce a bifunctional oxygen catalyst (Co(OH)2@SN-rGO). The Co(OH)2@SN-rGO bifunctional oxygen catalyst exhibited outstanding catalytic activities for both OER and ORR and showed excellent stability up to 40 h. The results demonstrated that the bifunctional oxygen electrode activities of the non-noble metal-based catalyst prepared in this study were as high as those of commercial noble metal-based oxygen electrodes.


1. Introduction

With a rapid increase in the demand for clean and cost-effective energy storage and conversion devices with high energy density, metal–air batteries have evolved as a promising alternative to lithium-ion batteries.1–6 To realize efficient metal–air batteries, the development of highly active cathode materials for oxygen reduction reaction (ORR) and oxygen evolution reaction (OER) is essential. This is because the reaction kinetics of cathode materials significantly affect the performance of batteries.3–5,7 Noble metals such as Pt, Ru, and Ir and their alloys are used as cathode materials owing to their outstanding oxygen catalysis activities. However, the use of noble metals hinders the large-scale commercial applications of metal–air batteries because of their high cost and rare availability. Furthermore, bare noble metal-based electrocatalysts suffer from dissolution and aggregation of active sites during the electrocatalytic reactions in acidic or alkaline media resulting in insufficient durability.7,8 Thus, in order to fabricate next-generation rechargeable metal–air battery systems, it is imperative to develop cost-effective and stable cathode materials with satisfactory activities for both ORR and OER.

Over the past few years, various efforts have been made to develop noble metal-free electrocatalysts for highly efficient and stable oxygen catalysis using carbon nanomaterials such as graphene and carbon nanotubes. It has been reported that nanoarchitecturing, surface modification, and heteroatom doping can induce numerous catalytic active sites and increase the electron transfer number and electrical conductivity of electrocatalysts, thus enhancing their oxygen catalytic efficiency.9–17 Heteroatom (B, N, S and P) single- and multi-doping of carbon nanomaterials, which show unique intrinsic properties, have been extensively investigated.13–17 The dopants in the carbon network create electron-deficient (p-type) and electron-donating (n-type) regions, which in turn reduce the work function of carbon by re-distributing the spin and charge densities on carbon nanomaterials.12,13 In terms of catalysis kinetics, the dopants affect the adsorption behavior of H and O atoms on the catalytic active sites of carbon nanomaterials. Li et al. reported that n-type dopants act as more efficient active sites for ORR and OER as compared to p-type dopants. This is because n-type doping favors the adsorption of O atoms, while p-type doping favors the adsorption of H atoms.14 Although heteroatom-doped carbon nanomaterials show ORR catalytic activities comparable to those of noble metal-based electrocatalysts, their applications are limited owing to their intrinsic high OER overpotentials. Therefore, it is challenging to develop cost-effective highly efficient oxygen catalysts.

In order to improve the OER catalytic activities of heteroatom-doped carbon nanomaterials, earth abundant transition metal-based oxides and/or hydroxides have been used. Conventional approaches for preparing OER/ORR bifunctional catalysts involve the physical mixing of heteroatom-doped carbon nanomaterials with transition metal-based oxides/hydroxides to form composite materials.18,19 Physically mixed bifunctional electrocatalysts provide active sites for OER and ORR. However, they show high contact and charge transfer resistances at the interface between the OER and ORR catalysts owing to their weak interaction. This deteriorates the catalytic activity of such catalysts.19,20 Recently, Yang et al. grew NiCo2O4 nanoparticles on N-doped mesoporous graphene via pulsed-laser irradiation followed by a hydrothermal treatment. They found that the NiCo2O4/NLG hybrids showed outstanding reversible oxygen electrocatalytic activities owing to the chemical bonds between Co and pyridinic-N.20 As a result, the hybridization of OER catalysts with ORR catalysts via bottom-up approaches provides strong interaction between the catalysts, and hence is considered as a promising approach to develop bifunctional electrocatalysts.21

Herein, we report a facile strategy to develop S and N co-doped reduced graphene oxide-Cobalt hydroxide (Co(OH)2) nanocrystals (Co(OH)2@SN-rGO) as a highly active and stable bifunctional oxygen catalyst. The effects of single (S)- and multi (S and N)-doping on the ORR catalytic efficiency of the rGO sheets were investigated. Co(OH)2 nanocrystals with diameters less than 5 nm were directly grown on the S and N co-doped rGO via a simple precipitation method at room temperature. The Co(OH)2@SN-rGO bifunctional oxygen catalyst so-obtained exhibited excellent OER and ORR electrocatalytic activities. Furthermore, the chronopotentiometric (CP) (for OER) and chronoamperometric (CA) (for ORR) measurements revealed that the catalysts showed excellent catalytic stability over 40 h. The bifunctional oxygen electrode activities of the catalyst were compared with those of commercial noble metal-based electrodes (Pt/C for ORR and RuO2 for OER). The bifunctional oxygen activity of the Co(OH)2@SN-rGO electrode was comparable to those of the commercial noble metal-based oxygen electrodes. The results demonstrated the potential of the Co(OH)2@SN-rGO oxygen catalyst prepared in this study to replace high-cost noble metal-based oxygen catalysts.

2. Experimental procedure

2.1 Synthesis of heteroatom-doped rGO

Partially S-doped rGO (PS-rGO) was prepared via freeze drying a graphene oxide (GO) gel (N002-PS, pH = ∼2, Angstron materials) followed its thermal reduction at 700 °C for 2 h. High S-doped rGO (HS-rGO) was synthesized using the same procedure as that used for preparing PS-rGO with additional elemental sulfur (reagent grade, Sigma-Aldrich). The sulfur[thin space (1/6-em)]:[thin space (1/6-em)]GO weight ratio was 10[thin space (1/6-em)]:[thin space (1/6-em)]1. Sulfur was placed 10 cm downstream from GO in the tube furnace. S and N co-doped rGO (SN-rGO) was prepared by mixing freeze-dried GO with urea (99.3+%, Alfa Aesar) using a mortar and pestle followed by thermal reduction at 700 °C for 2 h. The urea[thin space (1/6-em)]:[thin space (1/6-em)]GO weight ratio was 3[thin space (1/6-em)]:[thin space (1/6-em)]1. All the samples were immersed in a 1 M HCl aqueous solution to remove any residual species and were then washed with deionized water until a pH value of 7 was obtained. The cleaned samples were dried by lyophilization.

2.2 Synthesis of Co(OH)2 nanocrystals anchored on SN-rGO (Co(OH)2@SN-rGO)

Co(OH)2@SN-rGO was prepared via a precipitation method. Briefly, 30 mg of SN-rGO and 30 mg of CoCl2·6H2O (98.0%, Alfa Aesar) were dispersed in 300 mL of a solution consisting of 200 mL of ethanol and 100 mL of isopropyl alcohol using an ultrasonic processor (500 W, 20 kHz, VC 505, Sonics & Materials) for 30 min in an ice bath. Then, 30 mg of NH4HCO3 (95.0%, Samchun) was added to the resulting homogeneous suspension under vigorous stirring. Stirring was continued for 10 h. The Co(OH)2@SN-rGO powder was collected by centrifugation and washed with ethanol and deionized water several times. The cleaned samples were dried by lyophilization.

2.3 Characterization

The mophological and microstructural features of all the samples were observed using a field emission scanning electron microscope (FESEM, Hitachi, S-4700) and a transmission electron microscope equipped with an energy dispersive spectroscope (TEM and EDS, JEOL, JEM-F200). The X-ray diffraction patterns of the samples were recorded on an X-ray diffractometer (XRD, Rigaku, Ultima III) with Cu Kα radiation (λ = 1.5406 Å). The surface compositions of the samples were analyzed by X-ray photoelectron spectroscopy (XPS, Thermo Fisher Scientific, K-alpha+) and their Raman spectra were obtained using a Raman spectrometer with a laser wavelength of 532 nm (NRS-3100, JASCO).

2.4 Electrochemical characterization

The electrocatalytic performances of the samples for ORR and OER were evaluated using IVIUMnSTAT (IVIUM Technologies, Netherlands) with a rotating ring disk electrode apparatus (RRDE-3A, ALS Co., Ltd) in a three-electrode system. The three-electrode system was composed of a glassy carbon electrode (GCE) as the working electrode, a Pt foil as the counter electrode, and an Ag/AgCl (saturated KCl) electrode as the reference electrode. The catalyst ink was prepared by ultrasonically dispersing 5 mg of the catalyst and 50 μL of 5 wt% Nafion in 450 μL of ethanol. The catalyst ink (10 μL) was loaded onto the GCE with a diameter of 5 mm. The GCE was then dried in an oven for a few minutes. The catalyst loading of the CGE was about 0.5 mg cm−2. The ORR and OER activities of the samples were evaluated by carrying out their iR-compensation linear sweep voltammetry (LSV) measurements at a scan rate of 5 mV s−1 in O2-saturated 0.1 M KOH at a rotating speed of 1600 rpm. The electron transfer numbers of the catalysts were determined using the Koutecky–Levich (K–L) equation by carrying out their LSV measurements at various rotating speeds (400–2025 rpm). The durability of the catalysts was evaluated by carrying out their CP measurements at 10 mA cm−2 for OER and CP measurements at 0.7 V (vs. reversible hydrogen electrode (RHE)) for ORR. Electrode impedance spectroscopy (EIS) measurements were carried out at 0.8 V (vs. RHE) at a rotating speed of 1600 rpm. The frequency was varied from 100 kHz to 10 MHz.

All the potentials measured in this study were calibrated to potentials vs. RHE (E vs. RHE) using the following equation:

image file: c9qi01108k-t1.tif

3. Results and discussion

3.1 Heteroatom-doped rGO for ORR catalysts

The elemental compositions of PS-rGO, HS-rGO, and SN-rGO were analyzed by XPS (Fig. 1 and S1). The XPS spectra of all the samples showed C 1s and O 1s peaks. HS-rGO showed S 2s and S 2p peaks, while SN-rGO showed an N 1s peak. High-resolution XPS spectra were obtained to examine the binding states of the samples. Fig. 1 shows the deconvoluted high-resolution C 1s, S 2p, and N 1s XPS spectra of PS-rGO, HS-rGO, and SN-rGO. The high-resolution C 1s XPS spectra of the samples revealed that the carbon species of the samples were composed primarily of C–C/C–H bonds. This indicates that GO was successfully reduced by the thermal reduction process (Fig. 1a, d, and g).22 Interestingly, the S 2p spectra of all the samples (Fig. 1b, e, and h) revealed the existence of S species with mainly C[double bond, length as m-dash]S/C–S–C bonding. This indicates that S was doped in the carbon lattice of all the samples.23 The N contents of the samples (Fig. 1c, f, and i) were determined from their high-resolution N 1s spectra. SN-rGO was composed of pyridinic N (N-6, 47.1%), pyrrolic N (N-5, 23.7%), and quaternary N (N-Q, 21.5%).24
image file: c9qi01108k-f1.tif
Fig. 1 High-resolution C 1s (a, d, and g), S 2p (b, e, and h), and N 1s (c, f, and i) XPS spectra of PS-rGO, HS-rGO, and SN-rGO.

Table S1 lists the S and N concentrations of all the samples. The S dopant concentrations of PS-rGO, HS-rGO, and SN-rGO were 0.1, 5.9, and 0.1 at%, respectively. The N dopant concentration of SN-rGO was 8.9 at%. The low S dopant concentrations of PS-rGO and SN-rGO can be attributed to the presence of organosulfate groups in the GO gel (N002-PS, pH = ∼2, Angstron Materials). Hirsch et al. reported that organosulfate groups are covalently bonded at the edges of GO flakes exfoliated by the Hummers’ method using sulfuric acid. These groups are not affected by extensive aqueous work-up, but can be reduced by base-treatment.25 Hence, it can be stated that the initial GO platelets prepared by freeze drying of the non-pH-titrated GO gel contained some organosulfate groups, and the organosulfate groups bound to GO acted as S dopant sources during the thermal reduction process.

The XRD patterns of PS-rGO, HS-rGO, and SN-rGO were recorded over the 2θ range of 10–80° (Fig. 2a). All the samples showed a peak at around 2θ = 26.4° corresponding to the (002) plane of hexagonal graphite. This indicates that no impurity was introduced during the thermal reduction of GO with different dopant sources (elemental S and urea). To examine the graphitic degrees of the samples, their Raman spectra were obtained over the 1000–1950 cm−1 range (Fig. 2b). All the spectra exhibited two sharp peaks at ∼1350 and ∼1590 cm−1, corresponding to the D- and G-bands, respectively. The D-band is attributed to the disorders in the carbon lattice, whereas the G-band corresponds to the graphitic carbon lattice. The D-band[thin space (1/6-em)]:[thin space (1/6-em)]G-band peak intensity ratios (ID/IG values) of PS-rGO, HS-rGO, and SN-rGO were found to be 0.90, 1.07, and 1.08, respectively. This indicates that high dopant concentrations induced lattice distortions and defects in the carbon network of rGO.26 The SEM images of the samples (Fig. S2) revealed that they showed a shape similar to that of two-dimensional (2D) graphene nanosheets.27 The samples did not exhibit morphological differences. The images revealed that the PS-rGO, HS-rGO, and SN-rGO catalysts were successfully prepared via thermal reduction using the different dopant sources (elemental S and urea).


image file: c9qi01108k-f2.tif
Fig. 2 (a) XRD patterns and (b) Raman spectra of PS-rGO, HS-rGO, and SN-rGO.

The ORR catalytic activities of the heteroatom-doped rGO catalysts were evaluated by LSV using an RRDE, as shown in Fig. 3a. Pristine GO showed poor electrocatalytic activity for ORR owing to its low conductivity. The ORR activity of PS-rGO was significantly better than that of pristine GO because of its higher electric conductivity and the presence of a large number of defect sites generated by both the thermal reduction of GO and partial S doping.28,29 The LSV curve and ORR activity of HS-rGO were similar to those of PS-rGO. The increase in the S dopant concentration resulted in a slight increase in the limiting current density (JL) of the catalysts. This indicates that the S species doped in rGO induced defects in the carbon lattice, however the defect sites did not affect the catalytic activity of rGO.28,29 The SN-rGO catalyst showed the highest ORR activity. The onset potentials of the samples were determined by enlarging their LSV curves over the current density range of −0.1–0 mA cm−2 (Fig. 3b). The onset potentials of the samples shifted positively in the order: GO (0.77 V), PS-rGO (0.82 V), HS-rGO (0.83 V), and SN-rGO (0.88 V). The predominant N species in SN-rGO contributed to the improvement in its catalytic activity, while the partially doped S species generated defect sites, thus increasing the electrocatalytic surface area of SN-rGO. These factors enhanced the ORR activity of SN-rGO.30


image file: c9qi01108k-f3.tif
Fig. 3 (a) LSV curves of GO, PS-rGO, HS-rGO, and SN-rGO obtained at 1600 rpm. (b) Enlarged LSV curve to determine the onset potential. (c) CV curves of SN-rGO in N2- and O2-saturated 0.1 M KOH solutions. (d) LSV curves of SN-rGO at different rotating speeds ranging from 400 to 2025 rpm (inset: the corresponding K–L plot obtained at 0.2 V vs. RHE).

The ORR activity of SN-rGO was evaluated by carrying out its cyclic voltammetry (CV) measurements in N2- and O2-saturated 0.1 M KOH solutions (Fig. 3c). The CV curve in the N2-saturated medium revealed that the SN-rGO catalyst exhibited a pseudocapacitive behavior. Well-defined oxygen reduction peaks were observed at potentials lower than 0.9 V (vs. RHE), indicating that the SN-rGO catalyst showed excellent ORR activity without any redox reactions attributing to metal impurities.31 The LSV curves of SN-rGO were obtained at different rotating speeds ranging from 400 to 2025 rpm to determine its electron transfer number (per O2) (Fig. 3d). The onset potential of SN-rGO remained almost constant, while the JL increased with an increase in the rotating speed. The electron transfer number of the catalyst was calculated using the K–L equation at 0.2 V (vs. RHE).21,32 The K–L plot of the SN-rGO catalyst (inset of Fig. 3d) showed a linear relationship between J−1 and ω−0.5, and the corresponding electron transfer number was calculated to be 3.2. In alkaline solutions, ORR occurs via either the four-electron reduction pathway (direct reduction: O2 → OH) or the two-electron reduction pathway (indirect reduction: O2 → HO2− → OH).33,34 In general, the direct reduction pathway is preferred owing to its faster oxygen reduction rate. The electron transfer number of 3.2 indicates that the ORR in SN-rGO occurred predominantly via the four-electron reduction pathway involving the two-electron reduction pathway. Hence, we selected SN-rGO as the highly active ORR catalyst.

3.2 Co(OH)2 nanocrystals attached on SN-rGO for bifunctional electrocatalyst

Co(OH)2 nanocrystals were attached on SN-rGO (Co(OH)2@SN-rGO) to fabricate a bifunctional electrocatalyst via a simple precipitation method. The SEM images of Co(OH)2@SN-rGO are shown in Fig. S3. The surface morphology of Co(OH)2@SN-rGO was similar to that of 2D graphene nanosheets. There was no significant difference between the morphologies of SN-rGO and Co(OH)2@SN-rGO. The microstructural features of SN-rGO and Co(OH)2@SN-rGO were observed by obtaining their TEM images (Fig. 4). Fig. 4a and the inset show the low- and high-resolution TEM images of SN-rGO, respectively. SN-rGO exhibited partially wrinkled and crumpled 2D graphene sheets without any impurities. Like SN-rGO, Co(OH)2@SN-rGO showed wrinkled 2D graphene sheets, as revealed by its low-magnification TEM image (Fig. 4b). It can be clearly observed from Fig. 4c that nanocrystals with diameters less than 5 nm were uniformly distributed on the 2D graphene sheets. The lattice parameter (d) of the nanocrystals was calculated to be 0.235 nm, which corresponds to the (102) plane of the Co(OH)2 phase, as shown in Fig. 4d. The distribution of the Co(OH)2 nanocrystals on the graphene sheets was further investigated by obtaining their scanning TEM (STEM) images and the corresponding EDS mapping results (Fig. 4e). The EDS mapping results revealed that Co(OH)2@SN-rGO consisted of C (red), N (orange), S (yellow), Co (cyan), and O (green). All these regions could be seen overlapping in the STEM image of Co(OH)2@SN-rGO, indicating the successful N and S doping of rGO and the uniform attachment of the Co(OH)2 nanocrystals on SN-rGO.
image file: c9qi01108k-f4.tif
Fig. 4 TEM images of (a) SN-rGO (inset: high-resolution TEM image) and (b) Co(OH)2@SN-rGO. c and (d) High-resolution TEM image, (e) STEM image and the corresponding EDS mapping of Co(OH)2@SN-rGO.

Table S2 lists the element ratios of Co(OH)2@SN-rGO, as obtained from the EDS mapping analysis. The analyzed elements contents in Co(OH)2@SN-rGO catalyst are 81.7 at% for carbon, 9.3 at% for oxygen, 0.1 at% for sulfur, 7.5 at% for nitrogen, and 1.4 at% for cobalt. Consequently, the S and N doping contents in SN-rGO are well maintained during the process for hybridization of SN-rGO with Co(OH)2 nanocrystals. The increased O ratio of Co(OH)2@SN-rGO can be attributed to the oxygen species derived from the Co(OH)2 nanocrystals. The Co content of Co(OH)2@SN-rGO was as small as 1.4 at%. Fig. S4a shows the XRD pattern of Co(OH)2@SN-rGO. Like SN-rGO, Co(OH)2@SN-rGO showed an XRD peak at around 2θ = 26.4°, corresponding to the (002) plane of hexagonal graphite. It also showed three additional peaks corresponding to the Co(OH)2 phase (JCPDS #51-1731). The high-resolution Co 2p and O 1s XPS spectra of Co(OH)2@SN-rGO revealed that the nanocrystals uniformly attached on SN-rGO were Co(OH)2 nanocrystals (Fig. S4b and S4c).35,36 Hence, it can be stated that Co(OH)2 nanocrystals with diameters less than 5 nm were successfully grown on SN-rGO via a facile and cost-effective precipitation method at room temperature without involving any harsh and complex process.

To investigate the insights of hybridization effects, Co(OH)2 powders were prepared via same precipitation method without SN-rGO. Fig. S5a represents the XRD pattern of as-prepared Co(OH)2 powders. Like Co(OH)2@SN-rGO, Co(OH)2 powder exhibited the peaks corresponding to the Co(OH)2 phase (JCPDS #51-1731). There were no peaks corresponding to the (002) plane of hexagonal graphite, which indicates the formation of pure Co(OH)2 without SN-rGO. The morphological features of as-prepared Co(OH)2 powders were observed by SEM images as shown in Fig. S5b and c. Interestingly, the Co(OH)2 powders exhibited bulk shape particles with diameters of over 30 μm, despite prepared via same procedure with synthesis of Co(OH)2@SN-rGO. Moreover, irregular and aggregated nanosheets structure could be observed on the surface of the bulk Co(OH)2 particles as shown in inset of Fig. S5c. The results indicate that SN-rGO is contributed to develop the Co(OH)2 nanocrystals with diameters less than 5 nm during the precipitation method for hybridization of SN-rGO with Co(OH)2.

The ORR and OER activities of the Co(OH)2@SN-rGO catalyst were investigated using an RRDE in an O2-saturated 0.1 M KOH medium, as shown in Fig. 5. Fig. 5a compares the ORR LSV curves of SN-rGO, Co(OH)2, Co(OH)2@SN-rGO, and Pt/C (20 wt%) obtained at the rotating speed of 1600 rpm. It could be observed that the ORR activity of the SN-rGO and Co(OH)2 increased significantly by the hybridization of SN-rGO with the Co(OH)2. The onset potential of SN-rGO showed a positive shift from 0.88 to 0.96 V with the attachment of the Co(OH)2 nanocrystals. Furthermore, the JL value of SN-rGO increased by about 1.6 times (from −2.9 mA to −4.5 mA cm−2) with the attachment of the Co(OH)2 nanocrystals. The enhanced ORR activity of Co(OH)2@SN-rGO can be attributed to the synergistic effect between the Co(OH)2 nanocrystals and SN-rGO. Although the Pt/C catalyst exhibited the highest ORR activity with high onset potential and JL among all the investigated catalysts, it should be noted that the ORR activity of the Co(OH)2@SN-rGO catalyst with a Co(OH)2 nanocrystal content of only 1.4 at% was comparable to that of Pt/C.


image file: c9qi01108k-f5.tif
Fig. 5 (a) ORR LSV curves of SN-rGO, Co(OH)2, Co(OH)2@SN-rGO, and Pt/C (20%), and (b) the corresponding Tafel plots of Co(OH)2@SN-rGO and Pt/C (20%). (c) ORR LSV curves of Co(OH)2@SN-rGO at different rotating speeds ranging from 400 to 2025 rpm (inset: the corresponding K–L plot obtained at 0.2 V vs. RHE). (d) OER LSV curves of SN-rGO, Co(OH)2, Co(OH)2@SN-rGO, and RuO2, and (e) the corresponding Tafel plots of Co(OH)2@SN-rGO and RuO2. (f) Combined ORR/OER LSV curves of Co(OH)2@SN-rGO and Pt/C//RuO2.

The ORR kinetics of the Co(OH)2@SN-rGO and Pt/C catalysts were investigated by obtaining their Tafel plots (Fig. 5b). At low current densities (J ≤ 0.3 mA cm−2), the Tafel slopes of the Pt/C and Co(OH)2@SN-rGO catalysts showed low values of 70 and 73 mV dec−1, respectively. This indicates that the last electron-transfer step (OOH* → O* + OH*) in the ORR process was the rate-determining step.37,38 The higher Tafel slope value of Co(OH)2@SN-rGO can be attributed to the weaker bonding between its N-doping sites and OOH*.37 After the initial region, the Tafel slope of Co(OH)2@SN-rGO increased rapidly from 73 to 52 mV dec−1, whereas that of Pt/C decreased from 70 to 100 mV dec−1. This indicates that the Co(OH)2@SN-rGO catalyst showed more favorable kinetics for the O2 → OOH* process in the presence of the OH-abundant electrolyte.37,39 To investigate the electron transfer kinetics of Co(OH)2@SN-rGO, its LSV curves were obtained at different rotating speeds (400–2025 rpm), as shown in Fig. 5c. The onset potential of Co(OH)2@SN-rGO remained constant, while the JL increased with an increase in the rotating speed. The electron transfer number of Co(OH)2@SN-rGO was calculated using the K–L equation at 0.2 V (vs. RHE).21,32 The K–L plot of Co(OH)2@SN-rGO (inset of Fig. 5c) showed a linear relationship between J−1 and ω−0.5, and the electron transfer number was calculated to be ∼4. The low Tafel slope and high electron transfer number of Co(OH)2@SN-rGO indicate that it showed facile oxygen adsorption and that the oxygen reduction occurred via the four-electron reduction pathway. These results demonstrate the excellent ORR activity of Co(OH)2@SN-rGO.

The mechanism of the excellent ORR activity of Co(OH)2@SN-rGO hybrid catalysts can be explained by enhanced electron-transfer number from 3.2 (for SN-rGO) to ∼4 (for Co(OH)2@SN-rGO). There are possible three reactions can be occurred during ORR in alkaline electrolytes: (i) O2 + H2O + 2e → HO2 + OH; (ii) 2HO2 → O2 + 2OH; and (iii) HO2 + H2O + 2e → 3OH. Four-electron pathway involves reactions (i) and (ii), while two-electron pathway involves reactions (i) and (iii).40,41 Therefore, the ORR on the surface of SN-rGO with low electron-transfer number involves reaction (i), (ii) and (iii) resulting in slow kinetics. As for Co(OH)2@SN-rGO with high electron-transfer number of ∼4, Co(OH)2 nanocrystals strongly interconnected to SN-rGO contribute to instantaneous disproportionation of OH2 derived from reaction (i) to O2 and 2OH (reaction (ii)) before the electro-reduction reaction of HO2 (reaction (ii)), which facilitates fast kinetics.

To evaluate the OER activity of the Co(OH)2@SN-rGO catalyst, the OER LSV curves of SN-rGO, Co(OH)2, Co(OH)2@SN-rGO, and RuO2 were obtained by using an RRDE in a three-electrode system filled with 0.1 M KOH, as shown in Fig. 5d. The SN-rGO catalyst showed the lowest OER activity with a current density of less than 10 mA cm−2 (with an increase in the potential up to 1.8 V). The Co(OH)2 catalyst represented the relatively high activity with overpotential of 420 mV at a current density of 10 mA cm−2 which similar with RuO2 catalyst (422 mV). The Co(OH)2@SN-rGO catalyst exhibited excellent OER activity with lower overpotential (370 mV) than that of RuO2 at the current density of 10 cm−2. At the high current density of 30 mA cm−2, the overpotential of the Co(OH)2@SN-rGO increased slightly to 398 mV. The Tafel plots of Co(OH)2@SN-rGO and RuO2 are shown in Fig. 5e. As expected, the Co(OH)2@SN-rGO catalyst showed lower Tafel slope (66 mV dec−1) than RuO2 (143 mV dec−1). The bifunctional oxygen electrode activity of the Co(OH)2@SN-rGO catalyst was compared with those of commercial noble metal-based electrodes (Pt/C for ORR and RuO2 for OER), as shown in Fig. 5f. The oxygen activities (ΔV) of the electrodes were calculated by subtracting the potential at −3 mA cm−2 for ORR from that at 10 mA cm−2 for OER. The current density of 10 mA cm−2 for OER is the equivalent current density for 10% efficiency of an ideal solar cell device, and the current density of −3 mA cm−2 is equivalent to the current density at half-wave potential of state of art Pt-based catalyst.42 The ΔV value of the Co(OH)2@SN-rGO oxygen catalyst was 0.82 V, which is close to those of the Pt/C//RuO2 oxygen catalysts (0.79 V). The ΔV decreased when apply the their half-wave potentials (ΔV = 0.80 V for Co(OH)2@SN-rGO and ΔV = 0.77 V for Pt/C//RuO2). These results demonstrate the potential of the cost-effective Co(OH)2@SN-rGO oxygen catalyst prepared in this study to replace high-cost noble metal-based oxygen catalysts.

The electrocatalytic stability of a catalyst significantly affects its electrocatalytic activity. Therefore, we carried out the CP (for OER) and CA (for ORR) measurements of the Co(OH)2@SN-rGO catalyst for 40 h to evaluate its stability (Fig. 6a). During the CP measurement (for OER stability), the initial potential at −10 mA cm−2 remained constant. This indicates that the Co(OH)2@SN-rGO catalyst showed excellent OER stability. During the CA measurement (for ORR stability), the initial current density at 0.7 V (vs. RHE) decreased slightly. However, the normalized current density after 40 h was as high as 73.3%. The higher stability of Co(OH)2@SN-rGO catalyst for OER than ORR suggests that the surface chemistry of Co(OH)2, especially their surface oxygen species, might be more stable on the electro-oxidative catalytic response (OER: 4OH → 2H2O + 4e + O2) than on electro-reductive catalytic response (ORR: 2H2O + 4e + O2 → 4OH) in alkaline medium.43–45


image file: c9qi01108k-f6.tif
Fig. 6 (a) CP and CA responses of Co(OH)2@SN-rGO for OER and ORR, respectively at 900 rpm. (b) Nyquist plots of Co(OH)2, PS-rGO, SN-rGO, and Co(OH)2@SN-rGO. (c) Schematic illustration of the bifunctional Co(OH)2@SN-rGO electrocatalyst.

To investigate the excellent catalytic stability of the Co(OH)2@SN-rGO catalysts, further investigations were proceeded by CA measurement for OER and subsequent high-resolution XPS spectra of the catalysts after CA measurement for 40 hours. Fig. S6 shows the CA result for 40 hours, which indicates excellent OER stability of Co(OH)2@SN-rGO catalyst, in common with CP measurement. Fig. S7 represents high-resolution XPS spectrum. The Co 2p spectrum of Co(OH)2@SN-rGO was deconvoluted into Co2+ and Co3+ electronic states. The peaks at 779 eV and 795 eV correspond to Co3+ and those at 780 and 796 eV correspond to Co2+, which agree with previous reports.46,47 Furthermore, the O 1s spectrum was deconvoluted into M–O–M, M–OH and C–O bonding at 529, 531 and 533 eV, respectively. The high-resolution XPS results represented that the surface of Co(OH)2 (OER active sites) were partially oxidized from Co(OH)2 to CoOOH after CA measurement for 40 hours.47,48 The irreversible oxidation of Co(OH)2 to CoOOH during the OER induces degradation of catalytic activity of Co(OH)2 by reducing the ratio of Co2+/Co3+ because the Co2+ sites are more favorable to catalytic activity. Although the Co(OH)2 phase in the Co(OH)2@SN-rGO was undergo inevitable oxidation of Co(OH)2 to CoOOH during the OER. The ratio of Co2+/Co3+ was as high as approximate 2 after CA measurement for 40 hours, which demonstrate the excellent catalytic stability of Co(OH)2@SN-rGO catalyst.

The EIS spectra of Co(OH)2, PS-rGO, SN-rGO, and Co(OH)2@SN-rGO were obtained to determine their electron transfer resistances, as shown in Fig. 6b. The Nyquist plots of the catalysts showed semi-circles with different diameters, which were related to their charge-transfer resistances (Rct) resulting from electron diffusion at the interface between the electrolyte and electrode.49 The diameter of the semi-circles decreased in the order: Co(OH)2 (Rct: 4231.1 Ω) > SN-rGO (Rct: 597.2 Ω) > SN-rGO (Rct: 474.5 Ω), >Co(OH)2@SN-rGO (Rct: 115.1 Ω). This indicates that the predominant N species in SN-rGO and the hybridization of SN-rGO with the Co(OH)2 nanocrystals significantly contributed to the increased charge transfer at the electrolyte/electrode interface of the Co(OH)2@SN-rGO catalyst.

The results of this study suggest that the outstanding bifunctional oxygen electrocatalytic activity of the Co(OH)2@SN-rGO catalyst can be attributed to the heteroatom dual-doping of rGO (for ORR catalysis), hybridization of the Co(OH)2 nanocrystals (as an OER catalyst) with SN-rGO, and unique nano-structural features of Co(OH)2 and SN-rGO (Fig. 6c). SN-rGO was prepared by thermally reducing GO with urea as the N source. The initial organosulfate groups covalently bonded at the edges of the GO flakes acted as the S dopants.25 The predominant N species enhanced the intrinsic catalytic activity of SN-rGO, causing a positive shift in its ORR onset potential. The S species acted as defect sites, which increased the JL of the catalyst.

The hybridization of the Co(OH)2 nanocrystals (OER catalyst) with SN-rGO occurred via a facile precipitation method at room temperature. The Co(OH)2@SN-rGO bifunctional oxygen catalyst exhibited unique structural features with Co(OH)2 nanocrystals less than 5 nm in diameter uniformly distributed on the partially wrinkled SN-rGO sheets. Surprisingly, even a small Co(OH)2 nanocrystal content of 1.4 at% could improve both the ORR and OER activities of the Co(OH)2@SN-rGO catalyst. The Co(OH)2@SN-rGO catalyst showed high ORR activity with a high onset potential of 0.96 V, low Tafel slope of 52 mV dec−1, high JL of −4.5 mA cm−2, and high electron transfer number of ∼4. Furthermore, the Co(OH)2@SN-rGO catalyst showed outstanding OER activity with a low overpotential of 370 mV at the current density of 10 mA cm−2 and a low Tafel slope of 66 mV dec−1. The Co(OH)2@SN-rGO catalyst could retain its initial OER and ORR catalytic efficiencies for 40 h. This demonstrates the excellent catalytic stability of Co(OH)2@SN-rGO. The calculated oxygen electrode activity (ΔV) of the catalyst was as low as 0.82 V, which is comparable to those of commercially available noble metal-based oxygen electrodes (Pt/C for ORR and RuO2 for OER, 0.79 V). Hence, we prepared a cost-effective, highly active, and stable bifunctional Co(OH)2@SN-rGO oxygen catalyst as an alternative to high-cost noble metal-based oxygen catalysts via a facile strategy.

4. Conclusions

In summary, we developed Co(OH)2@SN-rGO as a bifunctional electrocatalyst for oxygen catalysis (OER and ORR) using an effective strategy: selecting a highly active ORR catalyst and hybridizing it with an OER catalyst. First, SN-rGO was prepared as a highly active ORR catalyst by thermally reducing non-pH-titrated GO (pH = ∼2) with urea as the N source. Then, the hybridization of the Co(OH)2 nanocrystals (as the OER catalyst) with SN-rGO was carried out via a facile precipitation method at room temperature. The Co(OH)2@SN-rGO bifunctional oxygen catalyst exhibited outstanding electrocatalytic activity and stability with high onset potential, low Tafel slope, high ORR electron transfer number, and low OER overpotential. The oxygen electrode activity of the Co(OH)2@SN-rGO catalyst, as estimated by subtracting the potential at −3 mA cm−2 for ORR from the potential at 10 mA cm−2 for OER, was comparable to those of the commercial noble metal-based oxygen electrodes. The outstanding oxygen catalytic activity of the Co(OH)2@SN-rGO catalyst can be attributed to its predominant N doping sites (which enhanced the intrinsic activity of the catalyst), S doping sites (which provided defect sites), and the hybridization of the Co(OH)2 nanocrystals (with diameters less than 5 nm) with SN-rGO. The results showed that the cost-effective Co(OH)2@SN-rGO oxygen catalyst prepared in this study is a promising alternative to high-cost noble metal-based oxygen catalysts.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work is supported by the National Research Foundation of Korea (NRF) Grant funded by the Ministry of Science and ICT, South Korea (2016M3A7B4909318), and by a Korea University Grant.

Notes and references

  1. K. Qiu, G. Chai, C. Jiang, M. Ling, J. Tang and Z. Guo, ACS Catal., 2016, 6, 3558–3568 CrossRef CAS.
  2. C. Hu and L. Dai, Angew. Chem., Int. Ed., 2016, 55, 11736–11758 CrossRef CAS PubMed.
  3. Z. Ma, X. Yuan, L. Li, Z.-F. Ma, D. P. Wilkinson, L. Zhang and J. Zhang, Energy Environ. Sci., 2015, 8, 2144–2198 RSC.
  4. H. Yadegari, Q. Sun and X. Sun, Adv. Mater., 2016, 28, 7065–7093 CrossRef CAS PubMed.
  5. Y. Li and H. Dai, Chem. Soc. Rev., 2014, 43, 5257–5275 RSC.
  6. Y. Jiao, Y. Zheng, M. Jaroniec and S. Z. Qiao, Chem. Soc. Rev., 2015, 44, 2060–2086 RSC.
  7. Z.-F. Huang, J. Wang, Y. Peng, C.-Y. Jung, A. Fisher and X. Wang, Adv. Energy Mater., 2017, 7, 1700544 CrossRef.
  8. S. Rojas-Carbonell, K. Artyushkova, A. Serov, C. Santoro, I. Matanovic and P. Atanassov, ACS Catal., 2018, 8, 3041–3053 CrossRef CAS.
  9. B. Z. Desalegn, H. S. Jadhav and J. G. Seo, ChemCatChem, 2019, 11, 2870–2878 CrossRef CAS.
  10. J. Zhang, Z. Zhao, Z. Xia and L. Dai, Nat. Nanotechnol., 2015, 10, 444–452 CrossRef CAS PubMed.
  11. Y. Ji, H. Dong, C. Liu and Y. Li, J. Mater. Chem. A, 2018, 6, 13489–13508 RSC.
  12. M. Sharma, J.-H. Jang, D. Y. Shin, J. A. Kwon, D.-H. Lim, D. Choi, H. Sung, J. Jang, S.-Y. Lee, K. Y. Lee, H.-Y. Park, N. Jung and S. J. Yoo, Energy Environ. Sci., 2019, 12, 2200–2211 RSC.
  13. Q. Wang, Y. Ji, Y. Lei, Y. Wang, Y. Wang, Y. Li and S. Wang, ACS Energy Lett., 2018, 3, 1183–1191 CrossRef CAS.
  14. H. Jiang, J. Gu, X. Zheng, M. Liu, X. Qiu, L. Wang, W. Li, Z. Chen, X. Ji and J. Li, Energy Environ. Sci., 2019, 12, 322–333 RSC.
  15. V. Mazanek, S. Matejkova, D. Sedmidubsky, M. Pumera and Z. Sofer, Chemistry, 2018, 24, 928–936 CrossRef CAS PubMed.
  16. Q. Lv, W. Si, J. He, L. Sun, C. Zhang, N. Wang, Z. Yang, X. Li, X. Wang, W. Deng, Y. Long, C. Huang and Y. Li, Nat. Commun., 2018, 9, 3376 CrossRef PubMed.
  17. H. L. Tan, A. Du, R. Amal and Y. H. Ng, Chem. Eng. Sci., 2019, 194, 85–93 CrossRef CAS.
  18. X. Li, Y. Fang, S. Zhao, J. Wu, F. Li, M. Tian, X. Long, J. Jin and J. Ma, J. Mater. Chem. A, 2016, 4, 13133–13141 RSC.
  19. X. Li, Y. Fang, X. Lin, M. Tian, X. An, Y. Fu, R. Li, J. Jin and J. Ma, J. Mater. Chem. A, 2015, 3, 17392–17402 RSC.
  20. D. U. Lee, B. J. Kim and Z. Chen, J. Mater. Chem. A, 2013, 1, 4754–4762 RSC.
  21. G.-H. Lee, M.-C. Sung, J.-C. Kim, H. J. Song and D.-W. Kim, Adv. Energy Mater., 2018, 8, 1801930 CrossRef.
  22. P. Wang, J. Wang, X. Wang, H. Yu, J. Yu, M. Lei and Y. Wang, Appl. Catal., B, 2013, 132–133, 452–459 CrossRef CAS.
  23. J. Li, Y. Zhang, X. Zhang, J. Huang, J. Han, Z. Zhang, X. Han, P. Xu and B. Song, ACS Appl. Mater. Interfaces, 2017, 9, 398–405 CrossRef CAS PubMed.
  24. S.-W. Park, J.-C. Kim, M. A. Dar, H.-W. Shim and D.-W. Kim, Chem. Eng. J., 2017, 315, 1–9 CrossRef CAS.
  25. S. Eigler, C. Dotzer, F. Hof, W. Bauer and A. Hirsch, Chemistry, 2013, 19, 9490–9496 CrossRef CAS.
  26. A. A. King, B. R. Davies, N. Noorbehesht, P. Newman, T. L. Church, A. T. Harris, J. M. Razal and A. I. Minett, Sci. Rep., 2016, 6, 19491 CrossRef CAS.
  27. I. Kim, S.-W. Park and D.-W. Kim, Chem. Eng. J., 2019, 375, 122033 CrossRef CAS.
  28. Z. Lu, S. Li, C. Liu, C. He, X. Yang, D. Ma, G. Xu and Z. Yang, RSC Adv., 2017, 7, 20398–20405 RSC.
  29. Z. Yang, Z. Yao, G. Li, G. Fang, H. Nie, Z. Liu, X. Zhou, X. A. Chen and S. Huang, ACS Nano, 2012, 6, 205–211 CrossRef CAS PubMed.
  30. K. Preuss, A. M. Siwoniku, C. I. Bucur and M. M. Titirici, ChemPlusChem, 2019, 84, 457–464 CrossRef CAS.
  31. Y. Cheng, X. Wu, J. P. Veder, L. Thomsen, S. P. Jiang and S. Wang, Energy Environ. Mater., 2019, 2, 5–17 CrossRef CAS.
  32. G. A. Ferrero, A. B. Fuertes, M. Sevilla and M.-M. Titirici, Carbon, 2016, 106, 179–187 CrossRef CAS.
  33. J. Yan, H. Lu, Y. Huang, J. Fu, S. Mo, C. Wei, Y.-E. Miao and T. Liu, J. Mater. Chem. A, 2015, 3, 23299–23306 RSC.
  34. Z.-J. Jiang and Z. Jiang, J. Mater. Chem. A, 2014, 2, 14071 RSC.
  35. H. Zhang, W. Tian, X. Guo, L. Zhou, H. Sun, M. O. Tade and S. Wang, ACS Appl. Mater. Interfaces, 2016, 8, 35203–35212 CrossRef CAS PubMed.
  36. T. Qin, S. Dang, J. Hao, Z. Wang, H. Li, Y. Wen, S. Lu, D. He, G. Cao and S. Peng, Dalton Trans., 2018, 47, 11503–11511 RSC.
  37. G. Zhang, H. Luo, H. Li, L. Wang, B. Han, H. Zhang, Y. Li, Z. Chang, Y. Kuang and X. Sun, Nano Energy, 2016, 26, 241–247 CrossRef CAS.
  38. C. H. Choi, H. K. Lim, M. W. Chung, J. C. Park, H. Shin, H. Kim and S. I. Woo, J. Am. Chem. Soc., 2014, 136, 9070–9077 CrossRef CAS PubMed.
  39. A. Holewinski and S. Linic, J. Electrochem. Soc., 2012, 159, H864–H870 CrossRef CAS.
  40. J. Liu, L. Jiang, B. Zhang, J. Jin, D. S. Su, S. Wang and G. Sun, ACS Catal., 2014, 4, 2998–3001 CrossRef CAS.
  41. J. Liu, L. Jiang, Q. Tang, B. Zhang, D. S. Su, S. Wang and G. Sun, ChemSusChem, 2012, 5, 2315–2318 CrossRef CAS PubMed.
  42. S. Chakrabarty, A. Mukherjee, W.-N. Su and S. Basu, Int. J. Hydrogen Energy, 2019, 44, 1565–1578 CrossRef CAS.
  43. S. Chen, Y. Li, F. Ma, F. Chen and W. Lu, Catal. Sci. Technol., 2015, 5, 1213–1221 RSC.
  44. X. Huang, J. Wang, H. B. Tao, H. Tian and H. Xu, Chem. Sci., 2019, 10, 3340–3345 RSC.
  45. R. Berenguer, C. Quijada and E. Morallon, ChemElectroChem, 2019, 6, 1057–1068 CrossRef CAS.
  46. Y. Gao, L. Mi, W. Wei, S. Cui, Z. Zheng, H. Hou and W. Chen, ACS Appl. Mater. Interfaces, 2015, 7, 4311–4319 CrossRef CAS PubMed.
  47. J. A. Koza, C. M. Hull, Y.-C. Liu and J. A. Switzer, Chem. Mater., 2013, 25, 1922–1926 CrossRef CAS.
  48. Z. Song, X. Han, Y. Deng, N. Zhao, W. Hu and C. Zhong, ACS Appl. Mater. Interfaces, 2017, 9, 22694–22703 CrossRef CAS.
  49. H. J. Qiu, P. Du, K. Hu, J. Gao, H. Li, P. Liu, T. Ina, K. Ohara, Y. Ito and M. Chen, Adv. Mater., 2019, 31, e1900843 CrossRef.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c9qi01108k

This journal is © the Partner Organisations 2019