Different phases of few-layer MoS₂ and their silver/gold nanocomposites for efficient hydrogen evolution reaction

Abstract

Molybdenum disulfide (MoS₂) is expansively studied as the most promising non-precious metal catalyst for the hydrogen evolution reaction (HER) because of its excellent properties. However, the properties of MoS₂ are quite distinct when the contents of the trigonal (1T) phase and hexagonal (2H) phase constituting MoS₂ are different. Herein, three MoS₂ representatives with different contents of the 1T phase are synthesized by a simple method, denoted as 1T-MoS₂, 1T/2H-MoS₂ and 2H-MoS₂ from high to low contents of the 1T phase. The chemical composition and crystalline phase of the three samples are systematically analyzed and characterized. To improve charge transport between adjacent nanosheets of MoS₂, silver and gold nanoparticles (Ag/AuNPs) are decorated on MoS₂, denoted as 1T-MoS₂@Ag/AuNPs, 1T/2H-MoS₂@Ag/AuNPs and 2H-MoS₂@Ag/AuNPs. The HER catalytic experiments of various MoS₂ samples in 0.5 M H₂SO₄ show that 1T-MoS₂ has the best catalytic performance compared to 1T/2H-MoS₂ and 2H-MoS₂. Owing to the cooperative effect of the 1T phase and Ag/AuNPs, the 1T-MoS₂@Ag/AuNP nanocomposite exhibits better HER electrocatalytic activity than the other MoS₂ samples because it contains a higher-percentage 1T phase and a higher charge transfer rate. This study offers an abundant foundation for designing and improvement of electrocatalysts based on MoS₂ and its nanocomposites.

---

Introduction

Because of the seriously excessive consumption of fossil fuel and related environmental pollution problems, sustainable hydrogen production through electrochemical water splitting has been proposed as an ideal solution for energy crises and global warming.^1,2 Generally, electrochemical water splitting processes require a good electrochemical catalyst which should be cost effective, stable, environment-friendly, and efficient and have a lower overpotential.^3,4 Currently, as the most active and chemically stable electrocatalysts for the hydrogen evolution reaction (HER), Pt group metals are the most effective catalysts for the generation of hydrogen. However, the high cost of the Pt group metal elements limits their large-scale applications.^5–7 Fortunately, molybdenum disulfide (MoS₂) with two-dimensional (2D)-layered crystal structures has been desired as a catalyst for the HER owing to its low cost and high electrocatalytic performance.⁸

As we all know, MoS₂ is composed of covalently bonded S–Mo–S sheets that are held together by weak van der Waals interactions in layer packing structures. Due to the different arrangements of S atoms, MoS₂ has two distinct crystalline phases principally: semiconducting 2H phase and metallic 1T phase.^9–12 In general, they were synthesized through a hydrothermal method by changing the reaction temperatures of ammonium molybdate and thiourea, i.e. thiourea not only acted as a sulfur source but also as a reductant to reduce Mo⁶⁺ to Mo⁴⁺, and excess ammonium ions hydrolyzed from starting materials might play a key role in the formation of intercalated structures.⁹ Due to the presence of NH₄⁺ ions, MoS₂ gradually undergoes an irreversible phase transition from the metallic 1T phase to the semiconductor 2H phase as the temperature increases.¹³ Nowadays, much attention has been devoted to 1T phase MoS₂ which exhibits excellent properties in the fields of photoluminescence,¹⁴ electrocatalysts,^15,16 photoelectrochemical cells,¹⁷ field effect transistors,¹⁸ photothermal cancer therapy¹⁹ and supercapacitors.^20,21 In particular, 1T phase MoS₂ shows superior performance for catalytic hydrogen evolution to that of the 2H phase, because the charge transfer resistance decreased noticeably in the metallic phase.^22–24 Generally, 1T phase MoS₂ can be obtained by chemical lithium intercalation/extraction exfoliation,²⁵ electron-beam irradiation²⁶ and plasmonic hot electron injection.^27,28 However, these methods commonly require tough experimental conditions such as highly hazardous reagents, expensive technical equipment and so on.^29–32 Therefore, it is still highly desirable to synthesize a high percentage of 1T phase MoS₂ by a simple strategy. The previously reported high-content 1T phase MoS₂ nanosheets have a few-layer structure, but these nanosheets are prone to restack and accumulate between layers which will hinder charge transport. Thus, it is necessary to improve the conductivity of MoS₂ catalysts.

In this study, taking ammonium molybdate and thiourea as the precursors, we synthesized 1T-MoS₂ (up to 80.1% 1T phase, Fig. 1a) and 1T/2H-MoS₂ (57.6% 1T phase, Fig. 1b) through a facile one-step hydrothermal method at 180 °C and 200 °C, respectively. Then 2H-MoS₂ was obtained by calcining the as-prepared 1T-MoS₂ at 300 °C (Fig. 1c). The composition and crystal phase of the three MoS₂ samples were systematically characterized and compared. In addition, to decrease the accumulation between the MoS₂ nanosheets and improve the conductivity, 1T-MoS₂@AgNPs or 1T-MoS₂@AuNPs were obtained by ultrasonication of the mixture of AgNPs and 1T-MoS₂, or through the spontaneous redox between HAuCl₄ and 1T-MoS₂. Results showed that the two nanocomposites showed better enhancement in the catalytic performance toward hydrogen evolution. To further increase the current density, we tried to load both gold and silver nanoparticles on MoS₂ nanosheets (1T-MoS₂@Ag/AuNPs) by three different methods. First, silver nanoparticles and HAuCl₄ were simultaneously added into the MoS₂ suspension; second, silver nanoparticles were added into the as-synthesized 1T-MoS₂@AuNP suspension; third, HAuCl₄ was added into the as-prepared 1T-MoS₂@AgNP suspension. The electrochemical performance of three 1T-MoS₂@Ag/AuNP nanocomposites toward the HER in 0.5 M H₂SO₄ showed that the catalytic activity and current density are improved compared to those of 1T-MoS₂@AgNPs or 1T-MoS₂@AuNPs, indicating that the high content of the 1T phase and the synergistic effect of MoS₂, AgNPs and AuNPs played an important role. The 1T-MoS₂@Ag/AuNPs obtained by the third method presents the best performance, hence it was selected for the preparation of 1T-MoS₂@Ag/AuNP, 1T/2H-MoS₂@Ag/AuNP and 2H-MoS₂@Ag/AuNP nanocomposites. The present work may pave the way for designing more superior MoS₂ based electrocatalysts and other applications.

Fig. 1 Illustration of the preparation of (a) 1T-MoS₂, (b) 1T/2H-MoS₂ and (c) 2H-MoS₂.

Experimental section

Materials and reagents

Ammonium molybdate [(NH₄)₆Mo₇O₂₄·4H₂O], thiourea [CS(NH₂)₂] and hydrogen tetrachloroaurate hydrate (HAuCl₄·4H₂O) were bought from Sinopharm Chemical Reagent Co., Ltd. (Beijing, China). Silver nitrate (AgNO₃) and sodium borohydride (NaBH₄) were purchased from Beijing Chemical Company (Beijing, China). All the reagents were of analytical grade and used as received without further purification. All solutions were prepared with Milli-Q-purified water (18.2 MΩ cm).

Synthesis of 1T-MoS₂, 1T/2H-MoS₂ and 2H-MoS₂ nanosheets

Hydrothermal reaction was used to synthesize 1T-MoS₂ and 1T/2H-MoS₂ nanosheets in a sealed autoclave system. Briefly, 1 mmol (NH₄)₆Mo₇O₂₄·4H₂O and 30 mmol CS(NH₂)₂ were dissolved in 35 mL distilled water under vigorous stirring to form a homogeneous solution. To investigate the effect of temperature on the ratio of the 1T and 2H phases, the solution was transferred into a 50 mL Teflon-lined stainless steel autoclave, maintained at 160, 170, 180, 190, 200, 210 and 220 °C for 24 h and naturally cooled down to room temperature, respectively. The final black product was washed several times with absolute ethanol and dried at 60 °C under vacuum. The XPS spectra of the Mo element of the MoS₂ products obtained at different temperatures were deconvoluted (Fig. S1a–e†), and the contents of the 1T and 2H phases at each temperature were calculated from the XPS fitting curves (Table S1 and Fig. S1f†). It can be seen that when the temperature increases from 160 °C to 180 °C, the content of the 1T phase gradually increases, and reaches a maximum at 180 °C. As the temperature increases up to 220 °C, the content of the 1T phase gradually decreases and the content of the 2H phase increases, and the content of the 1T and 2H phases of the product at 200 °C is almost identical. Therefore, we chose products obtained at 180 °C and 200 °C denoted as 1T-MoS₂ and 1T/2H-MoS₂, respectively. To obtain 2H-MoS₂, the as-prepared 1T-MoS₂ at 180 °C was then annealed at 300 °C with a heating rate of 10 °C min⁻¹ and maintained at this temperature for 5 h, then allowed to cool down to room temperature naturally. Through the contact angle experiment, it was found that all of the three products showed significant hydrophilicity, and the contact angle was less than 60° (Fig. S2†), indicating that the three products were more easily dispersed in water to form a suspension.

Synthesis of 1T-MoS₂@Ag/AuNP, 1T/2H-MoS₂@Ag/AuNP and 2H-MoS₂@Ag/AuNP nanocomposites

Firstly, AgNPs were prepared by the reduction of AgNO₃ with NaBH₄.³³ Briefly, under vigorous magnetic stirring, 0.005 g NaBH₄ was rapidly added into 50 mL aqueous solution of AgNO₃ (0.1 mM), producing a light yellow colored solution. The solution was filtered with a 0.22 μm Millipore filter and then stored in the dark at 4 °C. Then, three different MoS₂ products (1.0 mg) were mixed with the above AgNO₃ solution (1.0 mL) and ultrasonicated for 30 min. The solution was centrifuged at 8000 rpm for 10 min and the supernatant was removed after centrifugation. Then the precipitate of MoS₂@AgNPs was re-dispersed in water. The precipitation and re-dispersion process was repeated twice. Hereafter, 0.5 mL of 10 mM HAuCl₄ solution was added into the MoS₂@AgNP suspension. The mixture solution was ultrasonicated for 10 min and purified with a method similar to the above process. Finally, three different MoS₂@Ag/AuNP nanocomposites were obtained.

Characterization

The UV-vis absorbance spectra were recorded with a UV-2550 spectrophotometer at ambient temperature (20 ± 2 °C) with quartz cuvettes. X-ray diffraction (XRD) data were collected on an MSALXD2 using Cu Kα (1.5406 Å) radiation. Raman measurements were performed at room temperature using a Raman spectrometer (DXR, USA) with a 514 nm laser excitation source. Chemical compositions and valence band spectra of the samples were analyzed using X-ray photoelectron spectroscopy (XPS, Thermo Escalab 250Xi). Contact angles between water droplets and the samples were measured using a JC2000D1 (POWEREACH, China). Transmission electron microscopy (TEM) images were obtained by using a JEM 1200EX, and high-resolution TEM (HRTEM) characterization was performed on a Tecnai G2 F20 S-TWIN (200 kV). A Micromeritics ASAP2460 instrument was employed to measure the Brunauer–Emmett–Teller (BET) specific surface area by N₂ adsorption at 77 K. The pore size distribution was calculated by using density functional theory (DFT) from the absorption branch. The scanning electron microscopy (SEM) images were obtained on a Hitachi SU8010 scanning electron microscope. An energy dispersive X-ray spectroscopy (EDX) instrument attached to the SEM was used to obtain elemental mapping images. Atomic force microscopy (AFM) images were obtained using an Ntegra Prima, NT-MDT. Electrochemical impedance spectroscopy (EIS) measurements were performed on a Zennium E electrochemical workstation (Zahner, German). pH measurements were made with a PB-10 precision pH meter (Sartorius, Germany) and sonication was carried out with a KuDos ultrasonic cleaner (Shanghai, China).

Electrochemical measurements

Electrochemical tests were performed with a three-electrode system in a N₂-saturated 0.5 M H₂SO₄ aqueous solution on a CHI760d electrochemical workstation (Shanghai Chenhua Instruments). 1 mg of the sample powder was suspended in water (1 mL) and sonicated for 30 min. The suspension was centrifuged to remove the supernatant to a final volume of 250 μL and again sonicated for dispersion. Then 5 μL of catalyst ink was drop-cast onto a glassy carbon electrode (GCE) as the working electrode and dried under an infrared lamp. A saturated calomel electrode (SCE) and graphite rod were used as reference and counter electrodes, respectively. All the potentials were reported against a reversible hydrogen electrode (RHE) by adding a value of (0.241 + 0.059 × pH) V. Linear sweep voltammetry (LSV) measurements were performed from 0.1 to −0.6 V (vs. RHE) with a 5 mV s⁻¹ scan rate and the electrodes were cycled at least 50 cycles prior any measurements to refresh the catalytic surface. Tafel slopes were derived from the polarization curves by fitting the polarization curve data to the Tafel equation η = a + b log j (where η is the overpotential, j is the current density, and b is the Tafel slope). Cyclic voltammetry (CV) measurements were recorded from 0 to 0.3 V (vs. RHE) at a series of scan rates ranging from 20 to 200 mV s⁻¹ for evaluation of the double-layer capacities of the samples. The durability tests of the catalysts were conducted by use of time-dependent current density (i–t curve) and CV. The Nyquist plots were measured with frequencies ranging from 100 KHz to 0.01 Hz at an overpotential of 200 mV with an AC voltage of 5 mV.

Results and discussion

The crystalline phase of different products was investigated using XRD, Raman spectroscopy, UV-vis and XPS. Fig. 2a shows the XRD patterns of 1T-MoS₂, 1T/2H-MoS₂ and 2H-MoS₂, respectively. For the three samples, the XRD patterns show the characteristic peaks at 2θ = 32.7°, 49° and 58°, which correspond to the (100), (105) and (110) planes of MoS₂. Due to the metallic 1T structure of MoS₂, the new diffraction peak at 2θ = 43° in 1T-MoS₂ or 1T/2H-MoS₂ appears which corresponds to the (006) planes of MoS₂.³⁴ Besides, in contrast to the 14.2° (002) peak of the 2H-MoS₂ crystal phase, it is clearly observed that the (002)_new peak of 1T-MoS₂ and 1T/2H-MoS₂ shifts to a lower degree of 9.3°, and a second-order diffraction peak with 18.5° simultaneously appears which indicates the formation of a new lamellar structure with an increasing interplanar spacing compared to that of hexagonal MoS₂.⁹ Therefore, three diffraction peaks at 2θ = 9.3°, 14.2° and 18.5° corresponding to the (002), (002)_new and (004) planes are observed simultaneously in 1T/2H-MoS₂, respectively. The Raman spectra of 1T-MoS₂, 1T/2H-MoS₂ and 2H-MoS₂ are shown in Fig. 2b. For 2H-MoS₂, only two characteristic peaks are observed at 375 and 403 cm⁻¹, corresponding to in-plane E_2g and out-of-plane A_1g, respectively.^23,35 The distance (28 cm⁻¹) between the E_2g and A_1g modes reveals a few-layer feature of MoS₂ according to a recent report.³⁶ However, in 1T-MoS₂, it is worth noting that four strong peaks at 156 (J₁), 226 (J₂), 281 (E_1g) and 333 (J₃) cm⁻¹ appear which is consistent with those of 1T-MoS₂ prepared via n-butyllithium.³⁷ In addition, all of the characteristic peaks of both 1T-MoS₂ and 2H-MoS₂ appear concurrently in 1T/2H-MoS₂ and all the peaks are relatively strong. Further, UV-vis absorption spectroscopy was also used to characterize the difference between the three samples (Fig. 2c). After being diluted to 0.075 mg mL⁻¹, the 1T-MoS₂ sample appears as gray, whereas the 2H-MoS₂ sample is yellow-green (Fig. 2c, inset). More precisely, compared to the semiconducting 2H-MoS₂ sample which has two weak broad absorption peaks from 600 to 700 nm (blue line), the 1T-MoS₂ dispersion efficiently absorbed light in the full spectral range from 200 to 800 nm, and the absorbance decreases monotonously with the increase of wavelength (black line), indicating the characteristics of a non-plasmonic metallic nanostructure.^38–41 Additionally, 1T/2H-MoS₂ (0.075 mg mL⁻¹) with the same concentration exhibits very low absorption and the color of the solution is light gray which may be caused by the coexistence of phases 1T and 2H. Considering the results of XRD, Raman and UV-vis analysis, it is preliminarily proven that the product obtained at 180 °C was mainly the 1T phase, while the 1T and 2H phases were simultaneously exhibited in the sample synthesized at 200 °C. In addition, the product attained at 180 °C exhibited the 2H phase after calcination at 300 °C.

Fig. 2 The characterization of 1T-MoS₂, 1T/2H-MoS₂ and 2H-MoS₂: (a) XRD spectra, (b) Raman spectra, (c) UV-vis spectra (inset shows the corresponding photographs. The concentration of the three solutions is 0.075 mg mL⁻¹) and (d) XPS survey spectra.

The crystalline compositions and valence states of the three resulting MoS₂ samples were further confirmed by performing XPS analysis which is an efficient method to distinguish 2H from 1T polymorphs of MoS₂. The XPS survey spectra confirm the presence of S and Mo elements (Fig. 2d). The high resolution XPS spectra of Mo 3d and S 2p of the three products are shown in Fig. 3a and b respectively. For 2H-MoS₂, the Mo 3d spectra consist of peaks at around 229.5 eV and 232.7 eV corresponding to the Mo 3d_5/2 and Mo 3d_3/2 components, respectively. Notably, an obvious red shift for both the Mo 3d_5/2 and Mo 3d_3/2 peaks to the lower binding energies was observed in 1T-MoS₂ and 1T/2H-MoS₂.¹⁶ Similarly, with respect to the S 2p peak, an obvious shift toward lower binding energies due to the presence of the 1T phase was also observed in 1T-MoS₂ and 1T/2H-MoS₂ compared to 2H-MoS₂.⁴² The parallel shift of these peaks is attributed to the formation of more content of 1T-MoS₂.^10,14 The Mo 3d spectrum of 1T/2H-MoS₂ nanosheets was deconvoluted into four peaks. Two weak peaks located at 232.9 eV and 229.7 eV corresponding to the characteristic 3d_5/2 and 3d_3/2 of Mo⁴⁺ species from the 2H phase, besides a predominant doublet observed at 232.1 eV and 228.9 eV, revealed the presence of the 1T phase (Fig. 3c). Similarly, in the S 2p region of 1T/2H-MoS₂, additional S²⁻ 2p_3/2 and S²⁻ 2p_1/2 peaks for the 1T phase at 161.9 eV and 163.2 eV are also found except the known doublet peaks for the 2H phase (Fig. 3d). The further deconvolution of the other two Mo spectra reveals that all of them were deconvoluted into four peaks with different intensities, indicating that there is also a small amount of another phase in the individual 1T-MoS₂ or 2H-MoS₂, as shown in Fig. 3e–h. The concentrations of the metallic 1T phase obtained from 1T-MoS₂, 1T/2H-MoS₂ and 2H-MoS₂ XPS fitting curves are estimated to be ≈80.1%, 57.6% and 13.6%, respectively.

Fig. 3 XPS spectra of (a) Mo 3d and (b) S 2p of 1T-MoS₂, 1T/2H-MoS₂ and 2H-MoS₂. High-resolution XPS of (c, e and g) Mo 3d and (d, f and h) S 2p spectra of 1T/2H-MoS₂, 1T-MoS₂ and 2H-MoS₂, respectively.

These structural features of the three representative products were further investigated by HRTEM (Fig. 4). As shown in Fig. 4a, the interlayer distances were measured to be 0.78 nm (±0.03 nm), which is attributed to the d spacing of the (002) planes of 1T phase MoS₂. Noticeably, such an interlayer distance of 0.78 nm is slightly larger than the layer-to-layer distance of 0.61 nm in bulk MoS₂,^43,44 which could be ascribed to the intercalation of in situ generated NH₃/NH₄⁺ species between S–Mo–S layers during hydrothermal synthesis.⁴⁵ Besides, an interplanar spacing of 0.27 nm (±0.03 nm) can be observed from Fig. 4b, typically ascribed to the (101) facet of 1T-MoS₂.⁴⁶ However, there are two 0.27 nm (±0.03 nm) lattice fringes at 60° angles to each other, which are indexed to the (100) and (010) planes of 2H-MoS₂ ⁴⁷ (Fig. 4c). In addition, the HRTEM images of 1T/2H-MoS₂ (Fig. 4d) clearly show the coexistence of both 1T and 2H phases. Meanwhile, some defects and crystalline deformation are visible, as marked by red circles in Fig. 4d, which can potentially provide HER active sites.⁴⁸ All of the above results entirely demonstrate the chemical composition and crystalline phase of the three products.

Fig. 4 Typical HRTEM images of (a and b) 1T-MoS₂, (c) 2H-MoS₂ and (d) 1T/2H-MoS₂.

The morphology and hierarchical architecture of MoS₂ with three different crystal compositions are shown in Fig. S3.† As can be seen from the TEM and SEM images, these three samples primarily were composed of curly 2D nanosheets. As for 1T-MoS₂ (Fig. S3a†), most nanosheets are upright arranged and uniformly dispersed but some wrinkles are formed between the edges of layers. As for 1T/2H-MoS₂ (Fig. S3c†), the nanosheets are relatively flat, so some of the nanosheets are accumulated together. For 2H-MoS₂, the nanosheets are densely packed together, and the aggregation and stacking of the nanosheets can be clearly seen (Fig. S3e†). Comparing the TEM images of the three samples (Fig. S3b, d and f†), it can be seen that the degree of fragmentation of the 1T-MoS₂ nanosheets is more obvious, hence there may be more edge structures to provide more active catalytic sites.³⁰ The lateral size of the 1T-MoS₂ nanosheets is smaller than those of 1T/2H-MoS₂ and 2H-MoS₂, but the edges of the three nanosheets have partial overlap. For the above three samples, the crowded structures would not only hinder the charge transport but also limit the transfer of protons at the catalytically active sites of MoS₂. These problems can be addressed by the addition of conductive metal nanoparticles (for example Au or Ag), which could enhance the charge transport along interplanar directions and to some extent act as spacers to inhibit accumulation.⁴⁹ Therefore, we adopted two different processes to modify gold or silver nanoparticles on MoS₂ nanosheets to further improve their conductivity and catalytic activity.

We chose 1T-MoS₂ as a representative to prepare a nanocomposite of MoS₂ and gold or silver nanoparticles. When the gold precursor (HAuCl₄) was added to the 1T-MoS₂ dispersion without any other reductants at room temperature, the color of 1T-MoS₂ dispersion obviously changed from gray to light blue after ultrasonication (Fig. 5a, inset), indicating that gold nanoparticles are formed in situ on 1T-MoS₂ and the 1T-MoS₂@AuNP nanocomposite was obtained. The change in the UV-vis spectrum also confirmed the formation of gold nanoparticles. As shown in Fig. 5a (curve x), after the addition of the HAuCl₄ solution to the 1T-MoS₂ dispersion, a new absorption peak at ∼550 nm emerged, which further confirmed the formation of gold nanoparticles.⁴⁹ However, silver nanoparticles cannot be modified on MoS₂ nanosheets by spontaneous redox between Ag⁺ and 1T-MoS₂ like the preparation of 1T-MoS₂@AuNPs. As shown in Fig. S4,† after the addition of the AgNO₃ solution to the 1T-MoS₂ dispersion, no characteristic peak of silver nanoparticles appeared at ∼400 nm, indicating that Ag⁺ and 1T-MoS₂ could not spontaneously react to obtain silver nanoparticles. The reason is as follows: the Fermi level of MoS₂ is higher than the reduction potential of AuCl₄⁻ ions,^50,51 MoS₂/AuCl₄⁻ can form a redox pair, allowing spontaneous electron transfer from MoS₂ to Au³⁺ and leading to the formation of gold nanoparticles. Thus, 1T-MoS₂@AuNPs were prepared by a spontaneous redox reaction between HAuCl₄ and 1T-MoS₂ without adding any other reducing reagents, while the 1T-MoS₂@AgNP nanocomposite is achieved by mixing 1T-MoS₂ suspension with silver nanoparticles obtained by reduction of NaBH₄. Compared to 1T-MoS₂, when the silver nanoparticles were modified on the 1T-MoS₂ nanosheets, the shape of the absorption curve was not significantly changed, but the absorption intensity was slightly decreased (Fig. 5a, curve y). When HAuCl₄ was added into the 1T-MoS₂@AgNP suspension, a new absorption peak corresponding to the Au plasmon band at around 550 nm emerged, suggesting the formation of gold nanoparticles, and the color of the solution turned from gray to blue, which means that the redox reaction still occurs between 1T-MoS₂ and HAuCl₄ to obtain gold nanoparticles (Fig. 5a, curve z).

Fig. 5 (a) UV-vis spectra of 1T-MoS₂ (0.075 mg mL⁻¹), the mixture of 1T-MoS₂ (0.075 mg mL⁻¹) and 0.5 mM HAuCl₄ (curve x), 1T-MoS₂@AgNPs (0.075 mg mL⁻¹, curve y), the mixture of 1T-MoS₂@AgNPs (0.075 mg mL⁻¹) and 0.5 mM HAuCl₄ (curve z); the inset shows the photographs of the four solutions; polarization curves of (b) 1T-MoS₂, 1T-MoS₂@AgNPs, and 1T-MoS₂@AuNPs and (c) three 1T-MoS₂@Ag/AuNP nanocomposites from different approaches (black, red and blue lines) toward the HER at a scan rate of 5 mV s⁻¹ in 0.5 M H₂SO₄.

Next, we measured the electrochemical characteristics of the 1T-MoS₂@AgNP and 1T-MoS₂@AuNP nanocomposites on glassy carbon electrodes toward the HER catalytic performance in 0.5 M H₂SO₄ and compared them with single 1T-MoS₂ (Fig. 5b). As expected, the two nanocomposites showed better enhancement in catalytic performance, but the current density needed to be further improved. Fig. 5c shows the catalytic properties of the 1T-MoS₂@Ag/AuNP nanocomposite obtained by three different methods. First, silver nanoparticles and HAuCl₄ were simultaneously added into the MoS₂ suspension to obtain a 1T-MoS₂@Ag/AuNP nanocomposite (Fig. 5c, black line); second, silver nanoparticles were further added into the 1T-MoS₂@AuNP nanocomposite to obtain the 1T-MoS₂@Ag/AuNPs (Fig. 5c, red line); third, the 1T-MoS₂@Ag/AuNP nanocomposite was formed by a spontaneous redox reaction between the 1T-MoS₂@AgNPs and HAuCl₄ (Fig. 5c, blue line). By comparing the LSV curves, it can be seen that the catalytic activity and current density of all the nanocomposites are improved, but the nanocomposite obtained by the third method has the highest current density. It might because AuNPs are modified on the nanosheets by in situ redox reaction between HAuCl₄ and 1T-MoS₂, which is easier and more preferential than silver nanoparticle modification by ultrasonication on the nanosheets. Furthermore, a relatively strong Au–S bond was formed between the gold nanoparticles with the sulfur atom on the surface of 1T-MoS₂.⁵² Therefore, the catalytic activity of the 1T-MoS₂@Ag/AuNPs obtained by the prior two methods is weaker than that of the third. Thus, we chose the third preparation process for subsequent experiments and characterized the 1T-MoS₂@Ag/AuNP nanocomposite prepared by the third method.

Fig. 6a and b show the SEM images of 1T-MoS₂@AgNPs and 1T-MoS₂@AuNPs. After the decorating process, silver or gold nanoparticles can be evidently observed and uniformly dispersed on the edge of 1T-MoS₂ nanosheets. By comparing Fig. 6a and b, it can be seen that the gold nanoparticles on the surface of the nanosheets are denser than the silver nanoparticles since the content of HAuCl₄ is relatively higher. In addition, the SEM and TEM images of the 1T-MoS₂@Ag/AuNP nanocomposite (Fig. 6c and d) show that a small amount of gold and silver nanoparticles are aggregated, resulting in the larger size of the nanoparticles. Fig. 6e shows the AFM image of the 1T-MoS₂@Ag/AuNP nanocomposite. It can be seen that it has a similar thickness (∼2.9 nm) to the 1T-MoS₂ (Fig. 6f) nanosheets, suggesting that 1T-MoS₂ has a nearly uniform thickness of 3–4 layers. In addition, since the gold and silver nanoparticles are supported on the surface of the MoS₂ nanosheets, the height of 1T-MoS₂@Ag/AuNPs increases to ∼9.6 nm. Further, the EDS mapping results in Fig. S5† confirm the existence of Mo, S, Au and a small amount of Ag in the sample, indicating that Au and Ag are successfully introduced in the prepared 1T-MoS₂@Ag/AuNPs. It can be seen from Fig. S6† that the adsorption–desorption isotherms of the respective samples have no obvious saturated adsorption platform, indicating that the pore structure is irregular, and it can be seen from the Barrett–Joyner–Halenda (BJH) pore-size distributions (Fig. S6,† inset) that there are primarily irregular mesoporous structures. The decoration of gold or silver nanoparticles makes the specific surface area of 1T-MoS₂ larger (Fig. S6a–c†), but when both the gold and silver nanoparticles are simultaneously loaded on the surface of the 1T-MoS₂ nanosheets, the surface area is slightly reduced (Fig. S6d†), which may be due to the accumulation of nanoparticles, as confirmed by the TEM image. The XRD patterns of the 1T-MoS₂@AgNP and 1T-MoS₂@AuNP nanocomposites further proved that the silver and gold nanoparticles respectively are successfully decorated on the surface of the MoS₂ nanosheets (Fig. 7a). In comparison with 1T-MoS₂, new obvious diffraction peaks at 38.17° and 38.24° are observed, which could be assigned to the (111) diffraction peaks of cubic Ag (JCPDS no. 87-0720) and Au (JCPDS no. 89-3697), respectively.⁵³ The XRD results further illustrate that Ag and Au have been successfully decorated on the surface of the MoS₂ nanosheets, consistent with SEM observations. Then, the XRD spectra of 1T-MoS₂@Ag/AuNPs obtained by three different methods illustrated that the gold and silver nanoparticles can be simultaneously modified on the 1T-MoS₂ nanosheets by the three methods (Fig. S7†). Furthermore, a comparison of the XPS spectra of 1T-MoS₂ and 1T-MoS₂@Ag/AuNPs also confirmed that the gold and silver nanoparticles were simultaneously modified on the nanosheets. As depicted in Fig. 7b, there is no significant deviation observed between them, except for the appearance of two new significant peaks of Ag 3d doublet and Au 4f in the 1T-MoS₂@Ag/AuNP nanocomposite. In the deconvolution of the Ag 3d spectrum (Fig. 7c), the peaks appear at 368.2 eV and 374.2 eV, corresponding to Ag 3d_5/2 and Ag 3d_3/2, respectively, further proving the existence of Ag.^53,54 Similarly, the peaks located at 87.8 eV and 84.3 eV are assigned to the binding energy of Au 4f_5/2 and Au 4f_7/2, respectively, of the zero valence Au (Fig. 7d).^53,55 All of the above analyses indicate that the gold and silver nanoparticles together were successfully modified on the surface of the MoS₂ nanosheet. Subsequently, we prepared three nanocomposites (1T-MoS₂@Ag/AuNPs, 1T/2H-MoS₂@Ag/AuNPs and 2H-MoS₂@Ag/AuNPs) by modifying gold and silver nanoparticles on different kinds of MoS₂ using the third approach. The XRD patterns of the three MoS₂@Ag/AuNPs (Fig. 7e) show a new diffraction peak at 38°, which is attributed to the (111) crystal planes of Ag and Au. From the XRD patterns and Raman spectra of the three MoS₂ nanocomposites (Fig. 7e and f), it can be seen that, for the three MoS₂ nanocomposites, the XRD peak and Raman spectral position of MoS₂ in the nanocomposites are similar to those of the original MoS₂ samples, demonstrating that the crystal structure of MoS₂ was not obviously affected after surface modification.

Fig. 6 SEM images of (a) 1T-MoS₂@AgNPs, (b) 1T-MoS₂@AuNPs and (c) 1T-MoS₂@Ag/AuNPs; (d) TEM image of 1T-MoS₂@Ag/AuNPs; AFM images of (e) 1T-MoS₂ and (f) 1T-MoS₂@Ag/AuNP nanosheets.

Fig. 7 (a) XRD spectra of 1T-MoS₂ and its Ag and Au nanocomposites; (b) XPS spectra of 1T-MoS₂ and 1T-MoS₂@Ag/AuNP composite; high-resolution XPS (c) Ag 3d and (d) Au 4f spectra of 1T-MoS₂@Ag/AuNPs; XRD patterns (e) and Raman spectra (f) for the three MoS₂ nanocomposite products (1T-MoS₂@Ag/AuNPs, 1T/2H-MoS₂@Ag/AuNPs and 2H-MoS₂@Ag/AuNPs).

Finally, we measured the electrochemical characteristics of various MoS₂ samples toward the HER catalytic performance in 0.5 M H₂SO₄. As can be seen from the LSV curves (Fig. 8a), as the amount of 1T phase increases, the overpotential (η) decreases, while the current density (j) increases, indicating that the 1T phase plays a significant role in the enhancement of catalytic activity. In addition, the overpotentials of the three kinds of MoS₂@Ag/AuNP nanocomposites toward the HER were obviously smaller than those of the individual MoS₂. From the Tafel equation, we obtained Tafel slopes of 54, 59, 88, 53, 57 and 67 mV dec⁻¹ for 1T-MoS₂, 1T/2H-MoS₂, 2H-MoS₂, 1T-MoS₂@Ag/AuNPs, 1T/2H-MoS₂@Ag/AuNPs and 2H-MoS₂@Ag/AuNPs, respectively (Fig. 8b). As far as we know, the Tafel slope is an inherent property of the catalyst that is determined by the rate-limiting step of the HER process, and is used to evaluate the type of mechanism: Volmer–Tafel or Volmer–Heyrovsky.³⁷ There are possible reaction steps in acidic aqueous medium for the generation of hydrogen by water splitting. The first step is a discharge step (Volmer reaction) with a Tafel slope of ∼120 mV dec⁻¹. The following step involves a Tafel recombination step (Tafel reaction) with a Tafel slope of ∼30 mV dec⁻¹, or an electrochemical desorption step (Heyrovsky reaction) with a Tafel slope of ∼40 mV dec⁻¹.³⁴

Fig. 8 Electrochemical characterization of various MoS₂ samples for HER catalysis. (a) Polarization curves show the catalytic performance of various MoS₂ samples at a scan rate of 5 mV s⁻¹ in 0.5 M H₂SO₄; (b) Tafel plots for the data presented in (a); (c) the current density variation (ΔJ = J_a − J_c) at an overpotential of 0.15 V plotted against a scan rate fitted to a linear regression enables the estimation of C_dl; (d) electrochemical impedance spectroscopy Nyquist plots for various MoS₂ samples at an overpotential of 200 mV in 0.5 M H₂SO₄; (e) LSV curves of 1T-MoS₂@AgNPs before and after 5000 CV cycles at a scan rate of 5 mV s⁻¹ in 0.5 M H₂SO₄ and (f) i–t curve at an overpotential of 210 mV versus RHE in 0.5 M H₂SO₄.

With a Tafel slope of about ∼60 mV dec⁻¹, the HER process of the MoS₂ catalyst is ascribed to follow the Volmer–Heyrovsky reaction mechanism with electrochemical desorption as the rate-determining step.⁴⁶ To further investigate the difference in HER performance by the various MoS₂ samples, we measured the double-layer capacitance (C_dl) derived from the results of the cyclic voltammetry measurement at different scan rates (Fig. S8†), which is proportional to the effective electrochemically active surface area (ECSA).³⁰ The ECSA is calculated from the equation ECSA = C_dl/C_s, where C_s is 0.035 mF cm⁻².⁵⁶ The calculated C_dl values were 24.7, 17.8, 1.2, 36.7, 21.2 and 4.9 mF cm⁻² for 1T-MoS₂, 1T/2H-MoS₂, 2H-MoS₂, 1T-MoS₂@Ag/AuNPs, 1T/2H-MoS₂@Ag/AuNPs and 2H-MoS₂@Ag/AuNPs, respectively (Fig. 8c). All of the above results indicate that 1T phase MoS₂ displays higher exposure of catalytically active sites, and the HER properties of the MoS₂ nanosheets are closely related to the quantity of the 1T phase. The introduction of gold and silver nanoparticles leads to an increase in current density and a decrease in overpotential at a 10 mA cm⁻² current density.

We also applied EIS to provide further insight into the electrode kinetics during the HER. The Nyquist plots were fitted using an equivalent circuit as shown in the inset in Fig. 8d, which revealed that the charge transfer resistance (R_ct) values of 1T-MoS₂, 1T/2H-MoS₂, 2H-MoS₂, 1T-MoS₂@Ag/AuNPs, 1T/2H-MoS₂@Ag/AuNPs and 2H-MoS₂@Ag/AuNPs are approximately 75, 123, 807, 36, 71 and 398 Ω, respectively. The R_ct of all the 1T-MoS₂ samples is at least one order of magnitude smaller than that of 2H-MoS₂, which suggests a faster electron transfer process due to the presence of 1T phase MoS₂, and the existence of gold and silver nanoparticles further enhances conductivity and catalytic activity. Furthermore, the electrochemical stability of the 1T-MoS₂@Ag/AuNPs was evaluated by conducting LSV scanning before and after repeating cyclic voltammetry tests in the working voltage range. As depicted in Fig. 8e, the curve of 1T-MoS₂@Ag/AuNPs after 5000 cycles is nearly coincident with the initial one. The i–t curve (Fig. 8f) indicates that even after a long period of 10 h, the current density of the modified electrode displays only slight degradation. These results suggest that the 1T-MoS₂@Ag/AuNP electrode has excellent long-term stability for the HER.

The comprehensive comparison of the electrochemical performances from these representative MoS₂ samples above reveals several important points, such as overpotentials (η₁₀) at a current density of 10 mA cm⁻², Tafel slope, C_dl, R_ct and ECSA. From Table 1, it can be seen that with the increase of the 1T phase content and the introduction of gold and silver nanoparticles, the overpotential (η₁₀), Tafel slope and R_ct of MoS₂ and their gold/silver nanocomposites decrease, indicating that 1T-MoS₂ or 1T-MoS₂@Ag/AuNPs are more efficient in the kinetics of H₂ evolution and faster faradaic processes lead to superior HER performance.⁵⁷ This phenomenon is consistent with previous reports.^30,49 In addition, the increase of C_dl and ECSA caused by the increase of 1T phase content far exceeds the increase caused by the combination of the same type of MoS₂ with gold/silver nanoparticles, which means that MoS₂ with a high 1T phase content has more catalytic active sites. The contribution of the 1T phase content toward the HER activity process is greater than that of the gold and silver nanoparticles. Therefore, the significantly enhanced electrochemical HER activity can be ascribed to the following reasons: first, the hydrophilic properties of MoS₂ are favorable for the easy access of electrolytes to catalytic active sites. Second, they are attributed mostly to the high-percentage (≈80%) of 1T phase MoS₂ which exhibited more active sites and promoted electronic conductivity. Third, the introduction of gold and silver nanoparticles further decreases the charge transfer resistance and facilitates the fast electron transfer for the HER. Hence, the 1T-MoS₂@Ag/AuNP nanocomposite exhibits good electrocatalytic activity for the HER.

Table 1 Summary of the electrocatalytic parameters for various MoS₂ samples

Samples	η 10 (mV vs. RHE)	Tafel slope (mV dec^−1)	C dl (mF cm^−2)	R ct (Ω)	ECSA (cm^2)
1T-MoS2	214	54	24.7	75	705.8
1T/2H-MoS2	247	59	17.8	123	508.6
2H-MoS2	317	88	1.2	807	34.3
1T-MoS2@Ag/AuNPs	201	53	36.7	36	1048.6
1T/2H-MoS2@Ag/AuNPs	225	57	21.2	71	605.8
2H-MoS2@Ag/AuNPs	263	67	4.9	398	140.0

---

Conclusions

In conclusion, we successfully synthesized three representative MoS₂ products with different percentages of the 1T phase through a simple process, including 1T-MoS₂ (80.1% 1T phase), 1T/2H-MoS₂ (57.6% 1T phase), and 2H-MoS₂ (13.6% 1T phase) nanosheets. We systematically analyzed and compared the difference of the three samples using UV-vis, Raman, XRD and XPS. Then, three kinds of nanocomposites were obtained by modifying gold and silver nanoparticles on MoS₂ nanosheets through different ways. Simultaneously, we investigated the catalytic hydrogen evolution activity for these different MoS₂ samples. The results showed that the crystal phase served as the key role in determining the HER catalytic activity, that is to say, 1T-MoS₂ exhibited higher electrocatalytic hydrogen activity than 2H-MoS₂ and the higher the 1T phase content of the sample, the better the catalytic performance. In addition, with the same 1T phase content, the enhancement of the electrocatalytic activity of the MoS₂@Ag/AuNP nanocomposite compared to that of MoS₂ is due to the greatly improved charge transfer, as indicated by the smaller charge-transfer impedance. This study performed comprehensive systematic characterization and comparison of the three representative MoS₂ samples, and analyzed the reasons for affecting the catalytic activity of MoS₂, which provides a new idea for the design and improvement of electrocatalysts based on MoS₂ and its nanocomposites with further enhanced catalytic performance.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by a grant from the National Natural Science Foundation of China (Grant no. 21838010, 21890762 and 21676270), the National Natural Science Foundation of China-Key Projects of Shanxi Coal Based Low Carbon Joint Foundation (Grant no. U1610222) and the National Key Research and Development Plan of China (Grant no. 2016YFF0203700).

Notes and references

1. H. Xu, J. J. Wei, K. Zhang, M. Zhang, C. F. Liu, J. Guo and Y. K. Du, J. Mater. Chem. A, 2018, 6, 22697–22704.
2. H. Lu, C. C. Zhou, X. Zhou, H. Sun, H. T. Bai, Y. X. Wang, F. T. Lv, L. B. Liu, Y. G. Ma and S. Wang, Adv. Electron. Mater., 2017, 3, 1700161.
3. W. X. Yang, X. J. Liu, L. L. Chen, L. Liang and J. B. Jia, Chem. Commun., 2017, 53, 4034–4037.
4. Q. Y. Hao, S. Y. Li, H. Liu, J. Mao, Y. Li, C. Liu, J. Zhang and C. C. Tang, Catal. Sci. Technol., 2019, 9, 3099–3108.
5. M. H. Suliman, A. Adam, M. N. Siddiqui, Z. H. Yamani and M. Qamar, Catal. Sci. Technol., 2019, 9, 1497–1503.
6. W. X. Yang, L. L. Chen, X. J. Liu, X. Y. Yue, C. Y. Liu and J. B. Jia, J. Mater. Chem. A, 2016, 4, 5834–5838.
7. X. Yue, S. L. Huang, Y. S. Jin and P. K. Shen, Catal. Sci. Technol., 2017, 7, 2228–2235.
8. L. L. Chen, W. X. Yang, X. J. Liu and J. B. Jia, Int. J. Hydrogen Energy, 2017, 42, 12246–12253.
9. Q. Liu, X. L. Li, Q. He, A. Khalil, D. B. Liu, T. Xiang, X. J. Wu and L. Song, Small, 2015, 11, 5556–5564.
10. G. Eda, H. Yamaguchi, D. Voiry, T. Fujita, M. W. Chen and M. Chhowalla, Nano Lett., 2011, 11, 5111–5116.
11. F. Wypych, K. Sollmann and R. Schollhorn, Mater. Res. Bull., 1992, 27, 545–553.
12. S. X. Min and G. X. Lu, J. Phys. Chem. C, 2012, 116, 25415–25424.
13. F. Wypych and R. Schollhorn, J. Chem. Soc., Chem. Commun., 1992, 1386–1388.
14. Q. Liu, Q. C. Shang, A. Khalil, Q. Fang, S. M. Chen, Q. He, T. Xiang, D. B. Liu, Q. Zhang, Y. Luo and L. Song, ChemCatChem, 2016, 8, 2614–2619.
15. K. Chang, X. Hai, H. Pang, H. B. Zhang, L. Shi, G. G. Liu, H. M. Liu, G. X. Zhao, M. Li and J. H. Ye, Adv. Mater., 2016, 28, 10033–10041.
16. D. Voiry, M. Salehi, R. Silva, T. Fujita, M. W. Chen, T. Asefa, V. B. Shenoy, G. Eda and M. Chhowalla, Nano Lett., 2013, 13, 6222–6227.
17. Q. Ding, F. Meng, C. R. English, M. Caban-Acevedo, M. J. Shearer, D. Liang, A. S. Daniel, R. J. Hamers and S. Jin, J. Am. Chem. Soc., 2014, 136, 8504–8507.
18. R. Kappera, D. Voiry, S. E. Yalcin, W. Jen, M. Acerce, S. Torrel, B. Branch, S. D. Lei, W. B. Chen, S. Najmaei, J. Lou, P. M. Ajayan, G. Gupta, A. D. Mohite and M. Chhowalla, APL Mater., 2014, 2, 092516.
19. W. Y. Yin, L. Yan, J. Yu, G. Tian, L. J. Zhou, X. P. Zheng, X. Zhang, Y. Yong, J. Li, Z. J. Gu and Y. L. Zhao, ACS Nano, 2014, 8, 6922–6933.
20. N. Joseph, P. M. Shafi and A. C. Bose, New J. Chem., 2018, 42, 12082–12090.
21. M. Acerce, D. Voiry and M. Chhowalla, Nat. Nanotechnol., 2015, 10, 313–318.
22. M. A. Lukowski, A. S. Daniel, F. Meng, A. Forticaux, L. S. Li and S. Jin, J. Am. Chem. Soc., 2013, 135, 10274–10277.
23. Y. F. Yu, G. H. Nam, Q. Y. He, X. J. Wu, K. Zhang, Z. Z. Yang, J. Z. Chen, Q. L. Ma, M. T. Zhao, Z. Q. Liu, F. R. Ran, X. Z. Wang, H. Li, X. Huang, B. Li, Q. H. Xiong, Q. Zhang, Z. Liu, L. Gu, Y. H. Du, W. Huang and H. Zhang, Nat. Chem., 2018, 10, 638–643.
24. C. L. Tan, Z. M. Luo, A. Chaturvedi, Y. Q. Cai, Y. H. Du, Y. Gong, Y. Huang, Z. C. Lai, X. Zhang, L. R. Zheng, X. Y. Qi, M. H. Goh, J. Wang, S. K. Han, X. J. Wu, L. Gu, C. Kloc and H. Zhang, Adv. Mater., 2018, 30, 1705509.
25. M. A. Py and R. R. Haering, Can. J. Phys., 1983, 61, 76–84.
26. B. Radisavljevic and A. Kis, Nat. Mater., 2013, 12, 815–820.
27. T. Hong, B. Chamlagain, S. R. Hu, S. M. Weiss, Z. X. Zhou and Y. Q. Xu, ACS Nano, 2015, 9, 5357–5363.
28. Y. M. Kang, Y. J. Gong, Z. J. Hu, Z. W. Li, Z. W. Qiu, X. Zhu, P. M. Ajayan and Z. Y. Fang, Nanoscale, 2015, 7, 4482–4488.
29. H. Tan, W. Hu, C. Wang, C. Ma, H. L. Duan, W. S. Yan, L. Cai, P. Guo, Z. H. Sun, Q. H. Liu, X. S. Zheng, F. C. Hu and S. Q. Wei, Small, 2017, 13, 1701389.
30. Y. Yin, J. C. Han, Y. M. Zhang, X. H. Zhang, P. Xu, Q. Yuan, L. Samad, X. J. Wang, Y. Wang, Z. H. Zhang, P. Zhang, X. Z. Cao, B. Song and S. Jin, J. Am. Chem. Soc., 2016, 138, 7965–7972.
31. V. Kochat, A. Apte, J. A. Hachtel, H. Kumazoe, A. Krishnamoorthy, S. Susarla, J. C. Idrobo, F. Shimojo, P. Vashishta, R. Kalia, A. Nakano, C. S. Tiwary and P. M. Ajayan, Adv. Mater., 2017, 29, 1703754.
32. X. B. Fan, P. T. Xu, D. K. Zhou, Y. F. Sun, Y. G. C. Li, M. A. T. Nguyen, M. Terrones and T. E. Mallouk, Nano Lett., 2015, 15, 5956–5960.
33. Y. Zhou, H. Zhao, Y. J. He, N. Ding and Q. Cao, Colloids Surf., A, 2011, 391, 179–183.
34. S. Bai, L. M. Wang, X. Y. Chen, J. T. Du and Y. J. Xiong, Nano Res., 2015, 8, 175–183.
35. H. Li, Q. Zhang, C. C. R. Yap, B. K. Tay, T. H. T. Edwin, A. Olivier and D. Baillargeat, Adv. Funct. Mater., 2012, 22, 1385–1390.
36. A. Ambrosi, Z. Sofer and M. Pumera, Chem. Commun., 2015, 51, 8450–8453.
37. J. Yang, K. Wang, J. X. Zhu, C. Zhang and T. X. Liu, ACS Appl. Mater. Interfaces, 2016, 8, 31702–31708.
38. L. H. Yuwen, F. Xu, B. Xue, Z. M. Luo, Q. Zhang, B. Q. Bao, S. Su, L. X. Weng, W. Huang and L. H. Wang, Nanoscale, 2014, 6, 5762–5769.
39. K. G. Zhou, N. N. Mao, H. X. Wang, Y. Peng and H. L. Zhang, Angew. Chem., Int. Ed., 2011, 50, 10839–10842.
40. N. N. Meng, J. Cheng, Y. F. Zhou, W. Y. Nie and P. P. Chen, Appl. Surf. Sci., 2017, 396, 310–318.
41. (a) L. Guardia, J. I. Paredes, J. M. Munuera, S. Villar-Rodil, M. Ayan-Varela, A. Martinez-Alonso and J. M. D. Tascon, ACS Appl. Mater. Interfaces, 2014, 6, 21702–21710.
42. H. W. Wang, P. Skeldon and G. E. Thompson, Surf. Coat. Technol., 1997, 91, 200–207.
43. D. A. Reddy, J. Choi, S. Lee, Y. Kim, S. Hong, D. P. Kumar and T. K. Kim, Catal. Sci. Technol., 2016, 6, 6197–6206.
44. X. H. Zhang, N. Li, J. J. Wu, Y. Z. Zheng and X. Tao, Appl. Catal., B, 2018, 229, 227–236.
45. M. A. R. Anjum, H. Y. Jeong, M. H. Lee, H. S. Shin and J. S. Lee, Adv. Mater., 2018, 30, 1707105.
46. S. Wang, D. Zhang, B. Li, C. Zhang, Z. G. Du, H. M. Yin, X. F. Bi and S. B. Yang, Adv. Energy Mater., 2018, 8, 1801345.
47. Z. P. Liu, Z. C. Gao, Y. H. Liu, M. S. Xia, R. W. Wang and N. Li, ACS Appl. Mater. Interfaces, 2017, 9, 25291–25297.
48. Y. F. Huang, R. J. Nielsen, W. A. Goddard and M. P. Soriaga, J. Am. Chem. Soc., 2015, 137, 6692–6698.
49. J. Kim, S. Byun, A. J. Smith, J. Yu and J. X. Huang, J. Phys. Chem. Lett., 2013, 4, 1227–1232.
50. T. S. Sreeprasad, P. Nguyen, N. Kim and V. Berry, Nano Lett., 2013, 13, 4434–4441.
51. H. L. Zou, B. L. Li, H. Q. Luo and N. B. Li, Sens. Actuators, B, 2017, 253, 352–360.
52. S. Su, C. Zhang, L. H. Yuwen, J. Chao, X. L. Zuo, X. F. Liu, C. Y. Song, C. H. Fan and L. H. Wang, ACS Appl. Mater. Interfaces, 2014, 6, 18735–18741.
53. X. Q. Qiao, Z. W. Zhang, F. Y. Tian, D. F. Hou, Z. F. Tian, D. S. Li and Q. C. Zhang, Cryst. Growth Des., 2017, 17, 3538–3547.
54. A. J. Cheah, W. S. Chiu, P. S. Khiew, H. Nakajima, T. Saisopa, P. Songsiriritthigul, S. Radiman and M. A. A. Hamid, Catal. Sci. Technol., 2015, 5, 4133–4143.
55. S. Su, M. Zou, H. Zhao, C. F. Yuan, Y. A. Xu, C. Zhang, L. H. Wang, C. H. Fan and L. H. Wang, Nanoscale, 2015, 7, 19129–19135.
56. C. C. L. McCrory, S. Jung, I. M. Ferrer, S. M. Chatman, J. C. Peters and T. F. Jaramillo, J. Am. Chem. Soc., 2015, 137, 4347–4357.
57. Y. Z. Xu, L. L. Wang, X. Liu, S. Q. Zhang, C. B. Liu, D. F. Yan, Y. X. Zeng, Y. Pei, Y. T. Liu and S. L. Luo, J. Mater. Chem. A, 2016, 4, 16524–16530.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9cy02158b

---