Nitrogen, phosphorus co-doped carbon dots/CoS₂ hybrid for enhanced electrocatalytic hydrogen evolution reaction

Abstract

Developing highly efficient and low cost hydrogen evolution reaction (HER) electrocatalysts is still a huge challenge. In this study, we demonstrate a facile method to fabricate a nitrogen, phosphorus co-doped carbon dots/CoS₂ (NPCDs/CoS₂) hybrid as an electrocatalyst for HER with desirable electrocatalytic activities (low overpotential ∼78 mV and small Tafel slope ∼76 mV dec⁻¹) and long-term stability (after 1000 CV cycles with negligible current loss ∼1 mA cm⁻²) in acidic media.

---

Introduction

Hydrogen is considered as one of the most ideal energy sources that can be an alternative to fossil fuels in the future due to its abundant advantages, such as high-efficiency, non-pollution, renewability, etc.¹ Water splitting is used as a major method for hydrogen production with high purity and large quantity. The hydrogen evolution reaction (HER) is a crucial step in the water splitting process, which plays a key role in electrochemical and photoelectrochemical devices.^2,3 The active catalysts are required to increase process efficiency by minimizing the overpotential needed to drive the HER.¹ Pt-related catalysts are the most efficient catalysts for HER.^4,5 However, the large-scale commercial application with Pt-related catalysts for HER is still a distant goal due to its scarcity and high cost. Therefore, searching for highly efficient and low cost electrocatalysts may realize large-scale commercial application for HER.

Co-based complexes have emerged as interesting non-noble metal materials. In the past few decades, great progress has been made in developing Co-based complexes as HER catalysts such as CoP,¹ CoSe,^6,7 and Co–MoS,⁸ and especially for cobalt-sulfide (CoS₂).⁹ However, these Co-based complex catalysts still suffer from high overpotential and instability in acidic media. It was reported that the introduction of conducting nanocarbon materials could act as a benign matrix for uniform growth of CoS₂ to improve the electrochemical activity, which would further increase the surface area to enhance the stability of electrocatalyst for HER.^10,11

Carbon dots (CDs), as a rising star in the nanocarbon family, exhibit strong and tunable photoluminescence, rapid electron transfer and electron reservoir properties.^12–14 The abundant functional groups (–OH, –COOH, etc.) on the surfaces of CDs provide effective sites for construct composites and further enhance the catalytic activities of the original materials.^13,15 More important, it was reported that heteroatom-doping could efficiently improve the catalytic performance. For example, our group systematically reported that heteroatoms (N, P and B) co-doped nanocarbons (NPBC) exhibit good electrocatalytic activity, durability and selectivity for oxygen reduction reaction (ORR) in alkaline media.¹⁶ Dai et al. developed nitrogen, phosphorus co-doped carbon networks as efficient dual-functional catalysts for ORR and HER.³ Qiao and his colleagues demonstrated nitrogen and phosphorus co-doped graphene showed better HER performance than bare graphene.¹⁷ These enhanced catalytic performances are attributed to the heteroatom doping due to their electronegativities are different from that of carbon, which breaks the electroneutrality of carbon and creates more charged sites. In view of above, combining nitrogen, phosphorus co-doped CDs (NPCDs) with CoS₂ may serve as an ideal electrocatalyst for HER in the large-scale commercial application.

Herein, we demonstrate a facile method for the fabrication of nitrogen, phosphorus co-doped CDs/CoS₂ (NPCDs/CoS₂) hybrid, simply by hydrothermal treatment of cobalt salt, sodium thiosulfate and NPCDs. The resultant NPCDs/CoS₂ as electrocatalyst shows excellent HER catalytic activity (low overpotential ∼78 mV and small Tafel slope ∼76 mV dec⁻¹) and long-term stability in acidic media. The high catalytic performance of NPCDs/CoS₂ may attribute to the introduction of NPCDs and the formation of Co–O–C bond in the hybrid.

Experimental

All chemical reagents were of analytical grade and used as-received without further purification.

Synthesis NPCDs

The NPCDs were synthesized by modified-electrochemical etching method reported previously by our group and subsequently hydrothermal treatment.^12–14 Typically, two graphite rods (99.99%, Alfa Aesar Co. Ltd.) were inserted into the mixture of ultrapure water and phosphoric acid (volume ratio 1 : 1) as anode and cathode, respectively. A direct current power (30 V) was applied for the etching method. After three weeks, a brown solution was obtained, which was filtered and centrifuged to remove impurities. After that, the solution was dialyzed in a dialysis tube to remove excess acids until dialysate turned neutral, and P doped CDs (PCDs) were obtained. Next, 30 mL of ammonia water was mixed with 30 mL above dialysate solution to form a homogeneous solution. After that, the mixture was poured into a Teflon-lined stainless steel autoclave at 150 °C for 2 h. After the autoclave cooled down to room temperature, the brown-yellow solution was dialyzed until the pH = 7. Finally, the NPCDs were obtained.

For comparison, N doped carbon dots (NCDs), P doped carbon dots (PCDs) were prepared by similar procedure with NPCDs. The differences are the electrochemical etching process in the absence of phosphoric acid or lack of ammonia in the hydrothermal process.

Synthesis of NPCDs/CoS₂, NCDs/CoS₂, PCDs/CoS₂ and bare CoS₂

The NPCDs/CoS₂ hybrid was synthesized by simple hydrothermal method. In general, 1.19 g CoCl₂ and 0.62 g sodium thiosulfate were dissolved in a solution including 40 mL distilled water and 20 mL as-prepared NPCDs. Then, the mixture was kept at 60 °C and stirred for 30 min to form a homogeneous solution. After that, the mixture was transferred into a Teflon-lined autoclave and maintained at 150 °C for 12 h. After the autoclave cooled down to room temperature, the NPCDs/CoS₂ sample was collected by centrifugation (12 000 rpm, 10 min) and washed with deionized water for several times, then drying at 80 °C for 6 h. Finally, the NPCDs/CoS₂ hybrid was obtained.

For comparison, NCDs/CoS₂, PCDs/CoS₂ and bare CoS₂ were prepared by the similar method of NPCDs/CoS₂, in which the NPCDs were replaced with NCDs, PCDs and deionized water, respectively.

Electrochemical measurements

Electrochemical measurements were performed in a standard three-electrode system at room temperature. A modified rotating disk electrode (RDE, d = 3 mm), Pt wire and a saturated calomel reference electrode (SCE) acted as the working, counter and reference electrodes, respectively. All measurements were carried out using electrochemical workstation (CHI 660, CH Instrument) in a 0.5 M H₂SO₄ aqueous electrolyte. All potentials were referenced to a reversible hydrogen electrode (RHE) by adding a value of (0.242 + 0.059pH) V and current densities were the ratios of currents and geometric areas of working electrodes.

The RDE was used for all electrochemical experiments. Prior to be modified, it was polished carefully with 0.05 mm alumina powders and sonicated for several minutes. 5 mg sample was dispersed in the mixture of 190 μL deionized water and 10 μL 0.5 wt% Nafion solution. After that, the mixture was treated with ultrasonication to form a homogenous solution. Then, 8 μL above solution was dropped onto the RDE (loading ∼ 0.25 mg cm⁻²) and the treated RDE was dried at room temperature.

Controlled potential coulometry for H₂ production

In order to quantitatively detect faradaic yield of H₂ production, a two-compartment H-cell was connected by a proton exchange membrane. One compartment was equipped with a carbon paper modified with the activated NPCDs/CoS₂ used as the working electrode and a SCE used as the reference electrode. Another compartment used a carbon paper with a large surface area as the counter electrode. For the preparation of a modified carbon paper working electrode, the catalyst ink of NPCDs/CoS₂ was fixed onto one side of the carbon paper with an area of about 0.7 cm² (the loading of catalyst was 0.5 mg cm⁻²), followed by dropping 10.0 μL 0.5 wt% Nafion solution. Controlled potential coulometry measurement was performed after the reduction potential was set. The evolved H₂ was analyzed by an online GC-7890T gas chromatograph (GC, TCD detector, 5 Å molecular sieve columns and N₂ carrier).

Materials characterization

Scanning electron microscope (SEM) images were recorded on FEI quanta 200 scanning electron microscope with acceleration voltage of 20 kV. Transmission electron microscopy (TEM), high-resolution TEM (HRTEM) images and EDX mapping were obtained on a FEI-Tecnai F20 (200 kV) transmission electron microscope. The TEM samples were prepared by dropping sample solution onto a copper grid with polyvinyl formal support film and drying in air. Raman spectra were conducted on an HR 800 Raman spectroscope (JY, France) equipped with a synapse CCD detector and a confocal Olympus microscope. X-ray photoelectron spectrum (XPS) was recorded on an ESCA SSX-100 (Shimadzu) using a nonmonochromatized Mg Kα X-ray as the excitation source. Ultraviolet-visible (UV-Vis) spectra were collected with a PerkinElmer Lambda 750 UV-Vis spectrophotometer. The Fourier transform infrared (FT-IR) spectra were obtained by using a Nicolet 360 spectrometer. The structure of the samples was characterized by X-ray powder diffraction (XRD) by using an X'Pert-ProMPD (Holand) D/max-γA X-ray diffractometer with Cu Kα radiation (λ = 0.154178 nm). Fluorescence spectra were conducted on a Horiba Jobin Yvon (Fluoro Max-4) Luminescence spectrometer.

Results and discussion

Fig. 1a shows the TEM image of as-prepared NPCDs, in which the NPCDs are well dispersed with diameter of about ∼5 nm. The inset in Fig. 1a is the high-resolution TEM (HRTEM) image of a single NPCD. The lattice spacing of 0.21 nm is consistent with the (100) plane of the graphite.^13,14 The red trace in Fig. 1b shows the FT-IR spectrum of NPCDs. The broad peaks at about 3192 cm⁻¹ and 3440 cm⁻¹ are attributed to the stretching vibration of N–H and O–H bonds,^18,19 respectively. Two sharp peaks located at about 1440 cm⁻¹ and 1600 cm⁻¹ belong to the C=C stretching of polycyclic aromatic hydrocarbons and stretching vibration of C=O bond,^20,21 respectively. The peak at 950 cm⁻¹ is assigned to P–O stretching vibrations,^21,22 while the character peak of the C–N bending mode appears at 1121 cm⁻¹.^21–23 The black trace in Fig. 1b displays the UV-Vis absorption spectrum of NPCDs, in which a broad absorption band around 300 nm is attributed to π–π* transitions of C=C groups.²⁴

Fig. 1 (a) TEM image (inset: HRTEM image) of NPCDs. (b) FT-IR and UV-Vis absorption spectra of NPCDs. (c) Full-scan XPS spectrum of NPCDs. High-resolution XPS spectra of (d) C 1s, (e) N 1s, and (f) P 2p of NPCDs.

XPS measurements were employed to understanding the composition and chemical states of as-prepared NPCDs. Fig. 1c shows the full XPS survey spectrum of the NPCDs, which reveals the existence of C, O, N, and P elements without other impurities. The high resolution XPS spectra of C 1s for NPCDs display in Fig. 1d, which can be deconvoluted into four peaks, corresponding to C–C (284.60 eV), C–N (285.00 eV), C–O (286.60 eV), O=C–O (288.92 eV) bonds.^22,24 The high resolution XPS spectra of N 1s (Fig. 1e) can be resolved into two types of N species, corresponding to pyrrolic-N (399.40 eV) and graphitic-N (401.41 eV),²² respectively. The high-resolution XPS spectra of P 2p (Fig. 1f) reveal the existence of P–O bonding,²² indicating the presence of phosphorus in the NPCDs. These results demonstrate the active species of heteroatoms (N, P) are mainly located on the edge of CDs.

In the following experiments, a series comparisons among different elements doped CDs were discussed in the ESI.† Fig. S1a† shows the typical TEM image of CDs, in which the CDs are well dispersed. The HRTEM image of CDs (inset in Fig. S1a†) exhibits the lattice spacing of 0.21 nm, which is in well agreement with (100) plane of graphite. The typical size distribution of CDs is also inserted in Fig. S1a,† which suggests that the diameter is in the range of 2–7 nm. Fig. S1b and c† show the TEM images of NCDs, and PCDs. The images reveal that the morphology of different elements doped CDs are similar to that of CDs with diameter in the range of 4–10 nm. No significant change is observed. Fig. S1d† shows the FT-IR spectra of CDs, NCDs, PCDs and NPCDs (black trace, red trace, blue trace, and pink trace, respectively). As shown, the carbon framework is nearly invariable and the characteristic stretching vibration peaks of C=C (about 1440 cm⁻¹), C=O (approximately 1630 cm⁻¹) and O–H (around 3460 cm⁻¹) bonds are all appeared in the FT-IR spectra.¹⁵ Compared to CDs (3410 cm⁻¹), the characteristic peaks of O–H bond in doped CDs shift slightly (NCDs 3472 cm⁻¹, PCDs 3455 cm⁻¹ and NPCDs 3440 cm⁻¹). The characteristic stretching vibration peak of P–O bond in PCDs was observed locating at 1075 cm⁻¹, while the characteristic stretching vibration peak of C–N bond in NCDs situates at 1120 cm⁻¹. These results further prove the nitrogen and phosphorus have been successfully doped into CDs. Fig. S1e† describes the typical UV-Vis absorption spectra of CDs, NCDs, PCDs and NPCDs (black, red, blue, and pink lines, respectively). What can be seen is the NPCDs can absorb significantly more light in the 400–800 nm, indicating the NPCDs can effectively absorb visible light. It is very important for the water splitting with sunlight. Fig. S1f† shows Raman spectra of CDs, NCDs, PCDs and NPCDs (black, red, blue, and pink lines, respectively), two characteristic peaks located at about 1330 and 1595 cm⁻¹, corresponding to the D-band and G-band of carbon, respectively,¹⁴ which are typical characteristic peaks of carbon.

The NPCDs/CoS₂ hybrid was synthesized by simple hydrothermal method. Fig. 2a exhibits the SEM image of NPCDs/CoS₂, revealing an analogous tetragonum structure of NPCDs/CoS₂ with the diameter of approximately 200 nm. For further observed in Fig. 2a, the analogous tetragonum particles are consist of many agglomerated nanoflakes. Plentiful intermediate pores can be easily observed from the agglomerated NPCDs/CoS₂ nanoflakes. These intermediate pores effectively enlarge the specific surface area and increase the accessibility of the electrolyte, which is benefit to improve the catalytic performance. Fig. 2b shows the TEM image of NPCDs/CoS₂, in which the porous structure can further be observed. The inset in Fig. 2b shows the energy-dispersive X-ray spectroscopy (EDX) spectrum of NPCDs/CoS₂, revealing the presence of C, Co, S, N and P elements in this nanohybrid. Fig. 2c indicates the HRTEM image of NPCDs/CoS₂, in which the lattice fringes of 0.25 and 0.28 nm correspond to the (210) and (200) planes of the cubic cattierite-type CoS₂, respectively.^2,25 The lattice spacing of 0.21 nm for NPCDs is attributed to the (100) lattice spacing of graphite. The X-ray diffraction (XRD) analysis was performed for further characterize the structure of NPCDs/CoS₂, which is shown in Fig. 2d. The identified peaks at 2θ = 27.90, 32.30, 36.30, 39.80, 46.45, 54.97 and 60.26° are assigned to the (111), (200), (210), (211), (220), (311) and (023) planes of the cubic cattierite-type CoS₂ phase, matching well with the standard pattern (JCPDS card no. 41-1471).^2,26 The small peak at around 26° is assigned to the (002) diffraction of graphitic carbon. These results prove the NPCDs/CoS₂ hybrid was obtained.

Fig. 2 (a) SEM image of NPCDs/CoS₂. (b) The typical TEM image of NPCDs/CoS₂. The inset is the EDX spectrum of NPCDs/CoS₂. (c) HRTEM image of NPCDs/CoS₂. (d) XRD patterns of NPCDs/CoS₂ and the standard patterns of cubic CoS₂ (JCPDF 41-1471).

XPS spectra were further recorded to characterize the composition and chemical state of NPCDs/CoS₂ hybrid. The full XPS survey spectrum of NPCDs/CoS₂ is shown in Fig. 3a, showing the hybrids are composed of C, N, P, Co, and S. High resolution XPS spectra of Co 2p for NPCDs/CoS₂ and CoS₂ are shown in Fig. S2.† As shown, only several satellite peaks for cobalt oxide with low intensity are observed in CoS₂. This phenomenon might be caused by CoS₂ is susceptible to be oxidized to cobalt oxide in air which is consistent with that of others reported literatures.^1,27,28 For NPCDs/CoS₂, two new peaks at the binding energies of 778.59 eV, 793.69 eV for Co 2p^3/2 and Co 2p^1/2,^1,28 indicating the NPCDs have successfully composited with CoS₂. As shown in Fig. 3b, the high resolution spectra of C 1s can be deconvoluted into five peaks, corresponding to O=C–O (288.90 eV), C=O (287.80 eV), C–O (286.60 eV), C–N (285.70 eV) and C–C (284.60 eV) bonds.^18,19,24 The high resolution XPS spectra of O 1s displayed in Fig. 3c reveal the presence of Co–O–C (531.00 eV), O–H (531.20 eV), C–O (532.21 eV), and C=O (533.51 eV) in the hybrid.^{14,22,27–30} The high-resolution XPS spectra of N 1s is shown in Fig. 3d. Two peaks for pyridinic N (398.71 eV), pyrrolic N (399.40 eV) are observed. The P 2p XPS spectra in Fig. 3e show the P–O bonding in NPCDs/CoS₂.²² In the high-resolution XPS spectra of S 2p (Fig. 3f), the wide range (160–166 eV) can be deconvoluted into two peaks, corresponding to S 2p^1/2 and S 2p^3/2, respectively. XPS analysis demonstrates that the Co–O–C bond was formed in NPCDs/CoS₂ hybrid by the interfacial interaction between CoS₂ and NPCDs with surface active sites, surface defects, or oxygen-containing functional groups.^30–33 This new interactions would modulate the electron density between O and Co site, which benefit to electron transfer between CoS₂ and NPCDs.^26,27,29

Fig. 3 (a) Full XPS spectrum of NPCDs/CoS₂. High-resolution XPS spectra of C 1s (b), O 1s (c), N 1s (d), P 2p (e), S 2p (f) of NPCDs/CoS₂.

The catalytic abilities of CDs, NCDs, PCDs and NPCDs deposited on rotating disk electrodes (RDE) for HER were first investigated in 0.5 M H₂SO₄ using a standard three-electrode system (the detailed information shown in the Experimental section). The linear-sweep voltammetry (LSV) were performed to evaluate the catalytic activities of different elements doped CDs. As shown in Fig. 4a, the electrode modified with CDs is observed to exhibit poor catalytic activity, while the NCDs and PCDs modified electrodes also show low catalytic activities for HER. However, the catalytic ability of doped CDs are superior than that of CDs. In sharp contrast, the dual-doped CDs (NPCDs) exhibit the best catalytic activity for HER, and show a higher current density at a lower overpotential than that of single-doped CDs (NCDs and PCDs) or CDs. These results prove that through co-doping, the surface properties of CDs could be optimized to more efficiently catalyze HER by facilitating charge transfer and promoting proton adsorption.

Fig. 4 (a) Polarization curves for HER on the RDE modified with CDs (green line), NCDs (pink line), PCDs (black line) and NPCDs (red line) in 0.5 M H₂SO₄, respectively. (b) Polarization curves for HER on the RDE modified with bare CoS₂ (blue line), CDs/CoS₂ (green line), NCDs/CoS₂ (pink line), PCDs/CoS₂ (black line), NPCDs/CoS₂ (red line) and physical mixing of NPCDs and CoS₂ (wine line) in 0.5 M H₂SO₄, respectively. (c) Tafel plots for bare CoS₂, CDs/CoS₂ and NPCDs/CoS₂. (d) The electrochemical impedance spectroscopy (EIS) Nyquist plots for bare CoS₂, CDs/CoS₂ and NPCDs/CoS₂. (e) The polarization curves of NPCDs/CoS₂ and Pt mesh in 0.5 M H₂SO₄ with scan rate of 2 mV s⁻¹. (f) The Polarization curves of the NPCDs/CoS₂ modified RDE before and after 1000 cycles with scan rate of 100 mV s⁻¹.

In the following experiments, the catalytic activities of the bare CoS₂, CDs/CoS₂, NCDs/CoS₂, PCDs/CoS₂, NPCDs/CoS₂ and physical mixing of NPCDs and CoS₂ were carried out in 0.5 M H₂SO₄ with a three-electrode system. The catalyst mass loading on a glassy carbon electrode was 0.25 mg cm⁻². The HER polarization curves are displayed in Fig. 4b. From Fig. 4b, the NPCDs/CoS₂ exhibits better catalytic ability for HER than that of pure or single-doped catalysts with a more positive onset potential and large current density. Specifically, the onset potential of NPCDs/CoS₂ is about −66 mV, which is much lower than that of bare CoS₂ (about −150 mV). Moreover, the overpotential of NPCDs/CoS₂ for achieving the current density of 10 mA cm⁻² (an important metric for comparing the catalytic ability of an electrocatalyst in water electrolysis for solar hydrogen production) is 78 mV, while the CDs/CoS₂, NCDs/CoS₂ and PCDs/CoS₂ (green, pink and black lines, respectively) exhibit overpotentials of 150 mV, 124 mV or 109 mV at a current density of 10 mA cm⁻². Based on the above results, we can infer the excellent HER activity of NPCDs/CoS₂ may due to the formation of Co–O–C binding in the hybrid. To further confirm this result, we compared the HER performance of NPCDs/CoS₂ and the physical mixing of CoS₂ and NPCDs. By contrast, the NPCDs/CoS₂ modified electrode shows better HER activity with extremely high current density at low overpotential than that of the physical mixing electrode. The electrochemical measurements definitely demonstrate that NPCDs/CoS₂ exhibits high catalytic activity for HER in acid media.

The Tafel slopes (a very important factor for HER) are also discussed to predict the reaction mechanism. In theory, a much lower Tafel slope means that generating an equivalent current only need apply a lower overpotential. Fig. 4c describes the Tafel plots for bare CoS₂, CDs/CoS₂ and NPCDs/CoS₂ (blue, green and red lines, respectively). As shown in Fig. 4c, the Tafel slope of NPCDs/CoS₂ was calculated to be 76 mV dec⁻¹, which is much smaller than that of other two samples (134 mV dec⁻¹ for bare CoS₂ and 105 mV dec⁻¹ for CDs/CoS₂). The experimental results displayed a quite large exchange current density and a relatively small Tafel slope at such low overpotential. The results further confirm the catalytic activity of NPCDs/CoS₂ is better than bare CoS₂ for HER, indicating favorable reaction kinetics of the NPCDs/CoS₂ modified electrode.

To investigate the electrode kinetics under the catalytic operating conditions for HER, we preformed electrochemical impedance spectroscopy (EIS) analysis on bare CoS₂, CDs/CoS₂ and NPCDs/CoS₂. The obtained Nyquist plots are shown in Fig. 4d, which reveal that the charge transfer resistance (R_ct) value of NPCDs/CoS₂ (75 Ω) is much lower than those of CDs/CoS₂ (150 Ω) and bare CoS₂ (175 Ω). As known to all, the semicircle in the high-frequency range of the Nyquist plot attributed to the charge-transfer resistance is related to the electrocatalytic kinetics and a faster reaction rate.² The calculated results reveal that the coupling of NPCDs and CoS₂ indeed brings about enhanced electron transport abilities. The low charge transfer resistance is beneficial to obtain highly efficient charge transport, which might be attributed to the excellent interfacial contact between NPCDs and CoS₂. Moreover, the N, P co-doping could efficiently facilitate the electron charge transfer for HER on NPCDs/CoS₂ as evidenced by the decreased charge transfer resistance, and thus enhance the HER activity. For further confirm the electron charge transfer between NPCDs and CoS₂, the photoluminescence (PL) spectra of NPCDs, CoS₂, and the NPCDs/CoS₂ were performed shown in Fig. S3.† The remarkable decay of PL intensity of NPCDs/CoS₂ can be observed, suggesting there have electron transfer between NPCDs and CoS₂.

The catalytic activity of NPCDs/CDs for HER was further evaluated by comparing with the commercial platinum mesh (Pt). Fig. 4e shows the polarization curves of NPCDs/CoS₂ and Pt in 0.5 M H₂SO₄ with scan rate of 2 mV s⁻¹, which shows the NPCDs/CoS₂ is almost comparable to that of Pt. The durability is another key factor in evaluating the catalytic performance. A long-term potential cycle test was conducted to evaluate the electrochemical stability of NPCDs/CoS₂ in 0.5 M H₂SO₄. After that, the NPCDs/CoS₂ catalyst electrode was cycled continuously for 1000 cycles. The polarization plots are recorded in Fig. 4f. As shown, no obvious decay of the catalytic activity is observed, which indicates the high stability of NPCDs/CoS₂ in acid solution. The considerable durability is probably attributed to the tight binding between the active material and the substrate by the in situ fabrication approach.

The controlled potential coulometry was applied to collect the produced H₂ during several hours and analyzed by gas chromatography, thus calculating faradaic yields of H₂ production. The obtained results are shown in Table S1 and Fig. S5.† The experiment confirms that hydrogen was produced with relatively quantitative faradaic yield. For modified electrode with 0.5 mg cm⁻² of NPCDs/CoS₂, 18.32 C was passed after 3600 s at η = 180 mV and 0.09 mmol of hydrogen were detected by gas chromatography corresponding to a faradaic yield of ca. 94.60%.

For further testify the high catalytic activity of obtained NPCDs/CoS₂ for HER, a detailed comparison with previous researches is shown in Table 1. Compared the bare CoS₂ and metal complexes (CoS₂@MoS₂) with NPCDs/CoS₂, the synthesized hybrid exhibits lower onset potential (or overpotential) and Tafel slope with relatively small loading mass. For the carbon-based electrocatalyst (N,S-graphene and N,P-graphene), the NPCDs/CoS₂ also displays lower onset potential and Tafel slope. While for the analogous hybrids (CoS₂/CFP, CoS/CP/CT and CoS₂/RGO–CNT), the NPCDs/CoS₂ shows lower potential to achieve a current density of 10 mA cm⁻² with lesser loading mass. These results of the comparison further indicate the NPCDs/CoS₂ possess high catalytic activity for HER in acid media.

Table 1 The comparison of HER performance with different electrocatalysts in acid media

Catalyst	Loading (mg cm^−2)	η [mV]		Tafel slope [mV dec^−1]	Ref.
		Onset	10 mA cm^−2
Bare CoS2	0.65	190	200	134	30
CoS2@MoS2	—	44	110	57	31
N,S-Graphene	0.50	180	290	81	5
N,P-Graphene	0.20	290	420	91	17
CoS2/CFP	—	61	98	75	32
CoS/CP/CT	0.32	98	180	72	33
CoS2/RGO–CNT	1.15	110	142	51	2
NPCDs/CoS2	0.25	66	78	76	Our work

---

In the following experiments, the influence of different yields of NPCDs/CoS₂ on the HER performance was studied. The digital photos of different yields are shown in Fig. S4.† Table 2 records the statistical data obtained in the analysis HER performance with different yields. As can been seen from Table 2, the electrochemical measurements obtained from different yield matched well with what mentioned above, suggesting that this method can be realized large-scale production.

Table 2 The comparison of HER performance with different electrocatalyst yields in acid media

Yield	Loading [mg cm^−2]	η [mV]		Tafel slope [mV dec^−1]
		Onset	10 mV cm^−2
0.50 g	0.25	66	78	76
1.00 g	0.25	66	79	76
10.00 g	0.25	68	82	78

---

Based on the above discussion, the high catalytic activity of NPCDs/CoS₂ for HER may be attributed to the synergetic effects of NPCDs and CoS₂: (1) both Co and S could be the active sites of CoS₂ for HER, and more exposed sites and defects may be formed after combination with NPCDs by hydrothermal treatment. (2) The introduction of N, P into carbon-based materials may create more defect sites that can modulate the physical and chemical properties of nanocarbon. This is thought to be due to N serves for electron acceptors for the adjacent C, while P acted as electron donors. Moreover, N and P co-doped CDs have the favorable H* adsorption–desorption property that show a much lower HER overpotential than those of other pure and single-doped CDs samples.^22,31 (3) The formed Co–O–C bond in NPCDs/CoS₂ may facilitate electron transfer as evidenced by the decreased charge transfer resistance, and thus enhance the HER activity.

Conclusion

We demonstrate a facile method for the fabrication of NPCDs/CoS₂ hybrid, simply by hydrothermal treatment of cobalt salt, sodium thiosulfate and NPCDs. The obtained NPCDs/CoS₂ hybrid exhibits excellent HER electrocatalytic activity with a more positive onset potential, small Tafel slope (∼76 mV dec⁻¹) and high stability in 0.5 M H₂SO₄. The enhanced electrocatalytic activity of NPCDs/CoS₂ may be attributed to the synergetic effects of NPCDs and CoS₂: (1) the hydrothermal method may provide more active sites and exposed sites and defects in CoS₂. (2) The introduction of N, P into carbon-based materials may create more defect sites and N, P co-doped CDs have the favorable H* adsorption–desorption property. (3) The formation of chemical bond (Co–O–C) and resultant interface between CoS₂ and NPCDs could facilitate electron transfer as evidenced by the decreased charge transfer resistance. Thus, considering the above results, the HER activity of NPCDs/CoS₂ significantly enhanced. Our work may provide a way for the designing electrocatalyst by using dual-doped carbon matrices with metal disulfide with excellent HER performance.

Acknowledgements

This work is supported by the Collaborative Innovation Center of Suzhou Nano Science and Technology, the National Basic Research Program of China (973 Program) (2012CB825803, 2013CB932702), the National Natural Science Foundation of China (51422207, 51132006, 51572179, 21471106, 21501126), the Specialized Research Fund for the Doctoral Program of Higher Education (20123201110018), the Natural Science Foundation of Jiangsu Province of China (BK20140310), China Postdoctoral Science Foundation (2014M560445, 2015T80581) and a project funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).

Notes and references

1. M. Li, X. T. Liu, Y. P. Xiong, X. J. Bo, Y. F. Zhang, C. Han and L. P. Guo, J. Mater. Chem. A, 2015, 3, 4255.
2. S. J Peng, L. L. Li, X. P. Han, W. P. Sun, M. Srinivasan, S. G. Mhaisalkar, F. Y. Cheng, Q. Y. Yan, J. Chen and S. Ramakrishna, Angew. Chem., Int. Ed., 2014, 53, 12594.
3. J. T. Zhang, L. T. Qu, G. Q. Shi, J. Y. Liu, J. F. Chen and L. M. Dai, Angew. Chem., Int. Ed., 2015, 54, 1.
4. C. G. M. Guio, L. A. Stern and X. L. Hu, Chem. Soc. Rev., 2014, 43, 6555.
5. Y. Ito, W. T. Cong, T. Fujita, Z. Tang and M. W. Chen, Angew. Chem., Int. Ed., 2015, 54, 2131.
6. J. Y. Li, X. M. Zhou, Z. M. Xia, Z. Y. Zhang, J. Li, Y. Y. Ma and Y. Q. Qu, J. Mater. Chem. A, 2015, 3, 13066.
7. D. Kong, J. J. Cha, H. Wang, H. Lee and Y. Cui, Energy Environ. Sci., 2013, 6, 3553.
8. D. Merki, H. Vrubel, L. Rovelli, S. Fierro and X. Hu, Chem. Sci., 2012, 3, 2515.
9. H. Zhang, Y. Li, G. Zhang, P. Wan, T. Xu, X. Wu and X. Sun, Electrochim. Acta, 2014, 148, 170.
10. Z. Y. Lu, W. Zhu, X. Y. Yu, H. C. Zhang, Y. J. Li, X. M. Sun, X. W. Wang, H. Wang, J. M. Wang, J. Luo, X. D. Lei and L. Jiang, Adv. Mater., 2014, 26, 2683.
11. Z. Chen, D. Cummins, B. N. Reinecke, E. Clark, M. K. Sunkara and T. F. Jaramillo, Nano Lett., 2011, 11, 4168.
12. J. Liu, Y. Liu, N. Y. Liu, Y. Z. Han, X. Zhang, H. Huang, Y. Lifshitz, S. T. Lee, J. Zhong and Z. H. Kang, Science, 2015, 347, 970.
13. H. Ming, Z. Ma, Y. Liu, K. M. Pan, H. Yu, F. Wang and Z. H. Kang, Dalton Trans., 2012, 41, 9526.
14. Y. M. Yang, J. Liu, S. J. Guo, Y. Liu and Z. H. Kang, J. Mater. Chem. A, 2015, 3, 18598.
15. H. T. Li, Z. H. Kang, Y. Liu and S. T. Lee, J. Mater. Chem., 2012, 22, 24230.
16. S. Y. Zhao, J. Liu, C. X. Li, W. B. Ji, M. M. Yang, H. Huang, Y. Liu and Z. H. Kang, ACS Appl. Mater. Interfaces, 2014, 6, 22297.
17. Y. Zheng, Y. Jiao, L. H. Li, T. Xing, Y. Chen, M. Jaroniec and S. Z. Qiao, ACS Nano, 2014, 8, 5290.
18. X. X. Chen, Q. Q. Jin, L. Z. Wu, C. H. Tung and X. J. Tang, Angew. Chem., 2014, 126, 12750.
19. X. J. Gong, Q. Hu, M. C. Paau, Y. Zhang, S. M. Shuang, C. Dong and M. M. F. Choi, Nanoscale, 2014, 6, 8162.
20. H. T. Li, X. D. He, Z. H. Kang, H. Huang, Y. Liu, J. L. Liu, S. Y. Lian, C. H. A. Tsang, X. B. Yang and S. T. Lee, Angew. Chem., Int. Ed., 2010, 49, 4430.
21. Y. M. Yang, J. Liu, Y. Z. Han, H. Huang, N. Y. Liu, Y. Liu and Z. H. Kang, Phys. Chem. Chem. Phys., 2014, 16, 25350.
22. J. Liu, S. Y. Zhao, C. X. Li, M. M. Yang, Y. M. Yang, Y. Liu, Y. Lifshitz, S. T. Lee and Z. H. Kang, Adv. Energy Mater., 2016, 1502039.
23. J. Wu, X. J. Zheng, C. Jin, J. H. Tian and R. Z. Yang, Carbon, 2015, 92, 327.
24. H. Li, W. Q. Kong, J. Liu, N. Y. Liu, H. Huang, Y. Liu and Z. H. Kang, Carbon, 2015, 91, 66.
25. B. Wang, J. Park, D. W. Su, C. Y. Wang, H. J. Ahn and G. X. Wang, J. Mater. Chem., 2012, 22, 15750.
26. Y. Y. Liang, Y. G. Li, H. L. Wang, J. G. Zhou, J. Wang, T. Regier and H. J. Dai, Nat. Mater., 2011, 10, 780.
27. R. F. Nie, J. J. Shi, W. C. Du, W. S. Ning, Z. Y. Hou and F. S. Xiao, J. Mater. Chem. A, 2013, 1, 9037.
28. J. Yang, H. Liu, W. N. Martens and R. L. Frost, J. Phys. Chem. C, 2010, 114, 111.
29. D. Y. Wang, M. Gong, H. L. Chou, C. J. Pan, H. A. Chen, Y. P. Wu, M. C. Lin, M. Y. Guan, J. Yang, C. W. Chen, Y. L. Wang, B. J. Hwang, C. C. Chen and H. J. Dai, J. Am. Chem. Soc., 2015, 137, 1587.
30. J. X. Xu, Q. M. Yu, C. X. Wu and L. H. Guan, J. Mater. Chem. A, 2015, 3, 21647.
31. H. C. Zhang, Y. J. Li, T. H. Xu, J. B. Wang, Z. Y. Huo, P. B. Wan and X. M. Sun, J. Mater. Chem. A, 2015, 3, 15020.
32. H. C. Zhang, Y. J. Li, G. X. Zhang, T. H. Xu, P. B. Wan and X. M. Sun, J. Mater. Chem. A, 2015, 3, 6306.
33. J. Wang, H. X. Zhong, Z. L. Wang, J. Wang, H. X. Zhong, Z. L. Wang, F. L. Meng and X. B. Zhang, ACS Nano, 2016, 10, 2342.

---

Footnote

† Electronic supplementary information (ESI) available: TEM images, FTIR spectra, UV-Vis absorption spectra and Raman spectra of CDs, NCDs, PCDs and NPCDs; high resolution XPS spectra of Co 2p in CoS₂ and NPCDs/CoS₂; PL spectra of NPCDs, CoS₂, and NPCDs–CoS₂; the digital photos of different electrocatalyst yields; faradaic yields of H₂ production with NPCDs/CoS₂ at different overpotentials during 3600 s; faradaic yields (both experimentally measured and theoretically calculated) of H₂ production versus time with NPCDs/CoS₂ at overpotential (η = 180 mV) for 2.5 h in 0.5 M H₂SO₄. See DOI: 10.1039/c6ra11396f

---