Hydrogen, a clean and renewable energy, is one of the most promising alternatives for traditional fossil fuels [1]. Water electrolysis is an ideal approach for hydrogen production, and the key to achieving its practical applications is to develope cost-effective electrocatalyst to accelerate the hydrogen evolution reaction (HER) [2-5]. Pt-based materials possess the best performance in HER, yet their expensiveness and scarcity have hampered the large-scale applications [6-9]. To this end, enormous efforts have been devoted to the non-noble and earth-abundant HER catalysts in the past years [10-15]. Molybdenum disulfide (MoS₂) has drawn great attention due to its low cost and high abundance [16, 17]. However, the bulk MoS₂ is not an active HER catalyst because its exposed basal planes are generally inert and the electron/mass transportation is sluggish [18]. Fortunately, both theoretical and experimental studies have verified that the edge sites and defects of MoS₂ are catalytically active owing to the unsaturated sulfur [19-21]. Hence, nanostructured MoS₂ HER catalysts with smaller sheet size and more edge sites exposed are highly desirable in order to realize the superior performance of electrocatalysts. Using ultrasonic exfoliation to cleave bulk MoS₂ into a few nanolayers can increase the exposure of active edge sites; nevertheless, the number of layers and the lateral size of resultant nanosheets are hard to control [22]. Chemical vapor deposition is beneficial for the growth of high-quality MoS₂ nanosheets with controllable sizes, shapes, and structures, but rigorous experimental conditions, such as high vacuum, high temperature, and specific substrates, restricting its wide practical applications [23]. Benefiting from the low-cost precursors (metal salts) and simple manipulation to meet the practical requirements that obtain abundant edge defects [23, 24], hydrothermal synthesis of MoS₂ nanosheets has drawn great attention in synthesizing nanostructured MoS₂ electrocatalysts.

Favorable HER performance has been successfully achieved by preparing the defect-rich nanosized MoS₂ flakes with promoted intrinsic catalytic activity [24]. Although the as-prepared MoS₂ nanosheets in the solution can be conveniently stored, separated and easily transferred onto any substrates for electrode construction [23], those MoS₂ nanosheets with exposed edges are easily restacked in the solution because of their high surface energy and interlayer van der Waals attractions. Besides, the aggregate MoS₂ nanoparticles, which greatly impede the ion accessibility of the exposed active sites, reducing the electrocatalytic performance in the electrode level. Therefore, various controllable nanostructures such as hybrid nanoflowers [25, 26], hybrid core-shell structure [27, 28], nanoconfined structure [29] and vertically aligned nanolayers [30-32] have been reported to design the transport channels with exposed edges by introducing supporters. However, because of the restacking problem in nanostructured MoS₂ with abundant defects, it remains a great challenge that how to balance the intrinsic catalytic activity (mainly determined by the unsaturated sulfur on the sheet edges) and the active edge accessibility (influenced by the assembly structure) of MoS₂ nanosheets without other dopants to maximize the catalytic performance during electrochemical processes.

Herein, we report a hydrothermal method for the fast synthesis of size-controllable MoS₂ nanocatalysts by pH regulation. We investigated the effects of the acid concentrations on the synthesizing process and the resulting structure of the MoS₂ nanosheets. Results suggested that increasing the proton concentrations can downsize MoS₂ nanosheets with better intrinsic catalytic activity, but lead to serious aggregation of the nanosheets: The increase of proton concentrations may provide more nucleation sites during pretreatment and thus result in the size decrease of the as-formed MoS₂ nanosheets; however, the small size renders the nanosheets freedom to well restack. In other words, our results show that varying the size of the nanosheets, there is a trade-off between the intrinsic catalytic activity and active site accessibility. By the facile pH regulation, the MoS₂ nanosheets with the moderate intrinsic activity and the restacking degree deliver the best catalytic property of ~200 mV at the current density of 10 mA/cm², a Tafel slope of 46.3 mV/dec, and long-run stability in our work. Furthermore, MoS₂ nanocatalysts were prepared within 3 h via the proton-induced hydrothermal preparation process, leading to significant time savings (the reaction time of recent MoS₂ catalysts prepared by hydrothermal synthesis are shown in Table S1 in Supporting information), which was a benefit for its practical use. Our success in fast synthesis and insights into the edge-exposed MoS₂ catalyst properties provide a simple method to adjust catalytic performance without other dopant and a pathway for the practical applications of non-noble catalysts for water electrolysis.

Sodium molybdate dihydrate (Na₂MoO₄·2H₂O), sulfuric acid (H₂SO₄), and hydrochloric acid (HCl) were purchased from Sinopharm Chemical Reagent Co., Ltd. Thiourea (SC(NH₂)₂) and Nafion solution (5 wt%) were purchased from Sigma-Aldrich.

Typically, 2.5×10^-3 mol Na₂MoO₄·2H₂O and 10×10^-3 mol SC(NH₂)₂ were dissolved in 70 mL HCl solution (pH 1.00, 0.69, 0.50, 0.37 and 0.24). After being stirred to form a homogeneous mixture, the solution was transferred into a 100 mL Teflon-lined stainless-steel autoclave. The autoclave was heated at 200 ℃ for 3 h and then cooled to room temperature naturally. Thereafter, the products were washed with deionized water assisted by centrifugation for several times to remove redundant HCl and unreacted reagents. The resulting products were collected and maintained in deionized water. The MoS₂ catalyst synthesized with pH 1.00, 0.67, 0.50, 0.37, 0.24 are denoted as MoS₂-1.00, MoS₂-0.67, MoS₂-0.50, MoS₂-0.37, MoS₂-0.24, respectively.

Powder X-ray diffraction (XRD) profiles were recorded on an X-ray diffract meter (D2 PHASER, BRUKER). The scanning electron microscope (SEM) was performed in FEI Nova NanoSEM 450. A JEM-2100 electron microscopy with an accelerating voltage of 200kV was performed for transmission electron microscope (TEM) measurements. X-ray photoelectron spectroscopy (XPS) measurement was carried out on a Thermo Scientific Escalab 250Xi X-ray photoelectron spectrometer using Al as the exciting source. The dynamic light scattering (DLS) was carried out on Malvern Zetasizer Nano ZS90.

Electrochemical measurements were performed in a three-electrode electrochemical cell with Bio-Logic potentiostat (VMP3). Typically, 1 mL of water-isopropanol solution (volume ratio, 4:1) containing 3 mg MoS₂ catalyst was mixed with 60 μL Nafion by sonicating for 30 min to form a homogeneous ink. Then 5 μL of the ink was dropped onto the glassy carbon electrode with a 3 mm diameter (catalyst loading: 0.21 mg/cm²). All measurements were performed in the electrolyte of N₂-saturated 0.5 mol/L H₂SO₄ aqueous solution with a graphite rod as the counter electrode, saturated calomel electrode (SCE) as the reference electrode and the glassy carbon electrode loaded with MoS₂ as the working electrode.

The linear sweep voltammetry (LSV) with a scan rate of 5 mV/s and Tafel analysis were conducted for the characterization of HER activity. The double-layer capacitance was measured by cyclic voltammetry (CV) scan with various scan rates (25, 50, 100, 200, 400 mV/s) in the range of 0.25–0.35 V vs. SCE. The electrochemical impedance spectroscopy (EIS) measurements were carried out at an overpotential of 250 mV in the frequency range from 10⁵ Hz to 0.01Hz. To investigate the electrochemical stability, chronopotentiometry scan (at 10 mA/cm²) and CV (-0.3–0.1 V vs. RHE at 50 mV/s) were conducted. All data in this work were corrected with iR (removing the effect of solution resistance). All potentials were calibrated to a reversible hydrogen electrode (RHE), E (vs. RHE)=E (vs. SCE)+0.241 V.

Proton-induced MoS₂ nanocatalysts were fast prepared by the hydrothermal synthesis from the pre-treatment of precursors (Na₂MoO₄·2H₂O and SC(NH₂)₂) with the addition of aqueous HCl solution (Fig. S1 in Supporting information). The products MoS₂ powder was characterized by TEM, XRD, and XPS as shown in Fig. S2 (Supporting information), which demonstrate that our fast-prepared material was predominately composed of low-crystalline 2H-MoS₂ nanosheets [31, 33]. To investigate the effect of HCl pre-treatment on resultant products and their formed electrode architectures, XRD of the MoS₂ products prepared with various pH was further carried out to analyze the crystalline structure of the synthesized MoS₂ in the assembled electrode. Note that samples for XRD shown in Fig. 1A were prepared from the dried ink for drop-casting to simulate the state of the real electrode, while that for XRD in Fig. S2 was prepared by freeze drying which has an obvious crystal structure. A typical 2H-MoS₂ profile of JCPDS card No. 75-1539 of these samples can be observed [34]. The lower intensity of (100) and (002) peaks indicates a shorter basal plane and less stacking of MoS₂ layers [35]. As seen inFig. 1B, the intensity of (100) peak decreases along with the decrease of pH. When the pH value decreases to 0.24, the (100) peak even becomes difficult to be observed, which suggests that the lateral size of resultant MoS₂ products was decreased by lowering pH and more edge sites and defects were exposed. Notably, in Fig. 1C, with the decrease of pH, the intensity of (002) peak becomes weaker at first (from pH 1.00 to 0.50) but then becomes stronger (from pH 0.50 to 0.24). The weakening of intensity of the (002) peak is probably due to the proton-induced downsizing of MoS₂ nanosheets at c axis. Since the smaller nanosheets with higher surface energy are easier to restack, we could assume that the subsequent increase in (002) peak is mainly attributed to the restacking of further size-decreased MoS₂ nanosheets which decreases the edge-exposing degree [16, 36].

To further investigate the pH effect on the size control, SEM was carried out to further investigate the effect of pH on the size of MoS₂ nanosheets as shown in Fig. 1D, which was inconsistent with the results of XRD. Three SEM samples prepared from the ethanol solution of MoS₂ for drop-casting merely shown that the size of resultant MoS₂ products was decreased by lowering pH, which is consistent with the results of DLS (Fig. S3 in Supporting information). Here, we have a clear understanding of the effect of HCl pre-treatment on resultant MoS₂ by summarizing those characterizations. Decreasing the pH value is propitious to decrease the lateral size and thickness of resultant MoS₂ nanosheets. However, the too small MoS₂ nanosheets would tend to restack and aggregate after being assembled into the electrode. Meanwhile, the size decrease is beneficial for the enhancement of intrinsic activity, but the resultant restacking with narrow d-spacing between neighboring sheets is harmful for the exposition of active sites (Fig. 1E). Therefore, the size of MoS₂ nanosheets should be optimized.

After having demonstrated the particle size was determined by various pH values from characterization in Fig. 1, we tried to address the synthesis process of proton-induced MoS₂ that is schematically shown in Fig. 2A. When dissolved in HCl solution, the MoO₄^2- can be condensed into polymolybdate groups by protonation as following reactions, simultaneously [37]:

(1)

(2)

Compared with the MoO₄^2-, the as-formed polymolybdate groups are more reactive and can be reduced by thiourea even under normal temperature and pressure, indicated by a vivid color change, in sharp contrast with that without proton pre-treatment (Fig. 2A and Fig. S1). These polymolybdate groups may serve as nucleation sites for the subsequent fast hydrothermal reaction. Consequently, the reaction time was greatly shortened from 20 h to 3 h assisted by the pre-formation of polymolybdate groups. As schematically shown in Fig. 2B, we further proposed the following mechanism for this pH regulation effect. At high pH (e.g., pH 1.00), less polymolybdate groups are formed to play the role of nucleation active centers, so more Mo sources participate in the crystal growth process, which results in the formation of large-sized MoS₂ sheets; at low pH (e.g., pH 0.24), more polymolybdate groups are formed, leading to the formation of the small-sized MoS₂ nanosheets. Since the smaller nanosheets suffer from a restacking problem, which is unfavorable to edges exposure and the catalytic performance on the electrode level.

We further investigated the electrocatalytic performance of these drop-casted size-controlled MoS₂ via a standard three-electrode setup with 0.5 mol/L H₂SO₄ as the electrolyte. The electrochemical measurements were carried out with a constant mass loading of 0.21 mg/cm² of MoS₂ catalysts on the glassy carbon electrode and carbon paper (Fig. S4 in Supporting information), and the effects of substrate on the final activity of MoS₂ catalysts can be ignored. As shown in Fig. 3A, among the three MoS₂ species, the one synthesized under the medium pH of 0.50 exhibits the best performance, delivering the lowest overpotential (η ~200 mV) at the current density of 10 mA/cm². Corresponding Tafel plots of the three MoS₂ samples also show the same tendency (Fig. 3B). The Tafel slope decreases from 83.3 mV/dec (MoS₂-1.00) to 46.3 mV/dec (MoS₂-0.50) and then increase to 77 mV/dec (MoS₂-0.24). The Tafel slope of MoS₂-0.50 suggests that they match the Volmer-Heyrovsky mechanism. The electrocatalytic performance of MoS₂-0.50 is compatible with recent notable works (Table S2 in Supporting information).

Notably, the catalytic performances of the three kinds of MoS₂ are well consistent with their intensity of (002) peaks in the XRD profiles (Fig. 1C). These demonstrate that MoS₂-1.00 with larger and thicker nanosheets and MoS₂-0.24 with serious restacking show a compromised HER performance. Furthermore, the long-term electrochemical stability of MoS₂-0.50 is also examined by CV and chronopotentiometry scan (at 10 mA/cm²). As shown in Fig. 3C, MoS₂-0.50 affords a similar CV curve after 1000 cycles, and a negligible increase of overpotential can be observed under a constant current density of -10 mA/cm² for 12 h (Fig. 3D).

To further demonstrate the possible origins of such different performances of these size-controlled MoS₂, exchange current density (j₀), double layer capacitance (C_dl), and EIS were carried out. j₀ represents the inherent activity for HER, which is obtained by extrapolating E-I data to the Tafel equation [20]. It can be seen that the activity of the MoS₂ catalyst is improved along with the decrease of pH (from pH 1.00 to 0.24), demonstrated by the increase of j₀ from 2.9 μA/cm² (MoS₂-1.00) to 7.8 μA/cm² (MoS₂-0.50) and 26.3 μA/cm² (MoS₂-0.24) as shown in Fig. 4A. These results can be explained by the size of the three kinds of MoS₂ nanocatalysts. C_dl was carried out to evaluate the electrochemically active surface area (ECSA) of each sample [38, 39]. As shown in Fig. 4B, the capacitive current corresponding to the measured CV curves (Fig. S5 in Supporting information) is plotted as the function of the scan rates, from which the C_dl is extracted to evaluate the ECSA. As can be seen, the C_dl decreases from 1.09 mF (MoS₂-1.00) to 0.09 mF (MoS₂-0.24), suggesting adsorbable sites decreases along with the decrease of pH during the synthesis process. Note that not all adsorbable sites can be used for hydrogen evolution. EIS was performed to examine the electrode kinetics. As shown in Fig. 4C, the two semicircles at the low and the high frequencies represent the charge transfer resistance (R_ct) and the non-negligible electronic resistance of MoS₂ (R_i), respectively [40]. The smaller R_ct and R_i can afford faster HER kinetics. Detailed data of R_ct and R_i are exhibited in Table S3 (Supporting information). The R_ct decreases from 54.1 Ω (MoS₂-1.00) to 22.3 Ω (MoS₂-0.50), and then increases to 131.6 Ω (MoS₂-0.24). The R_ct corresponds to their LSV performance and Tafel slopes. Notably, the R_i keeps increasing along with the decrease of pH, from 5.3 Ω (MoS₂-1.00) to 5.9 Ω (MoS₂-0.50) and 40.4 Ω (MoS₂-0.24) (Fig. 4D). This result can be attributed to the downsizing of MoS₂ nanosheets, which impairs the electron migration in the electrode.

From the above electrochemical characterizations, we summary the mechanism of the pH regulation as follows. The increase of proton concentrations can effectively decrease the size of MoS₂ nanosheets to produce more unsaturated sulfur, endowing an enhanced intrinsic catalytic activity of each nanosheet. However, the serious aggregation occurs in these highly-active small-sized nanosheets, which will impair the system conductivity and electrolyte-accessibility. That may be the reason for MoS₂-0.50, which holds moderate intrinsic activity and the restacking degree and delivers the best electrode performance among the three.

Importantly, we have shown that the small difference in the size of these MoS₂ products will lead to such diverse electrode-level performances. Much attention should be paid to catalytic electrode construction apart from the chemical structure of the catalysts.

In this work, we put forward a strategy to realize the fast synthesis method of size-controllable MoS₂ nanocatalysts. Through an additional pre-treatment by HCl, nanosized MoS₂ catalysts can be synthesized simply and controllably by the hydrothermal reaction that achieves a superior performance of electrocatalysts without dopant. We demonstrate that lowing pH value can effectively decrease the size of resultant MoS₂ nanosheets, which will increase the intrinsic catalytic activity but lead to the restacking problem. By adjusting the size of MoS₂ nanoparticle, an excellent HER performance can be exhibited with a small overpotential of ~200 mV at 10 mA/cm², a low Tafel slope of 46.3 mV/dec and long-time stability. These findings will encourage further studies on the engineering of the catalytic electrode construction apart from the intrinsic activity of catalyst to improve the catalytic performance. Moreover, the proton-induced fast hydrothermal reaction in 3 h provides a new pathway for the large-scale application of highly active MoS₂ catalysts.

Declaration of competing interest

The authors report no declarations of interest.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (No. 21905206) and Shanghai Sail Program (No. 19YF1450800).

Appendix A. Supplementary data

Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.cclet.2020.08.005.