Hydrogen generation via photoelectrochemical water splitting using chemically exfoliated MoS₂ layers

Hydrogen is one of the promising energy sources beyond fossil fuel era because of its green, storable and high energy density characteristics.¹ Among the renewable energy technologies for hydrogen production, photoelectrochemical water-splitting has been widely studied as solar energy resources without environmental pollution.² Dissociation of water into hydrogen can be achieved via photoelectrochemical process that utilizes both light harvesting and solar fuel production. Recently, transition metal di-chalcogenides (TMDCs) consisting of few-atom-thick layers have emerged as an alternative of traditional materials in various applications due to their unique properties. TMDCs such as MoS₂ and WSe₂ have shown a great potential in immense applications of valleytronics,^3,4 flexible electronics,⁵ high-mobility transistors^6–8 and optoelectronic devices.^9,10 Recently, it has been predicted that TMDCs are useful for photovoltaic applications due to their tunable band gap and strong photo-excitement.¹¹ The exfoliated MoS₂ exhibits high photoluminescence quantum yield and other unusual optical properties.^12,13 In addition, MoS₂ nanoribbons or nanoflakes have excellent catalytic effect at the highly reactant edges, resulting in superior photo-induced catalyzing abilities.^14–16 The strong electrocatalytic activity of the nanostructured 2D MoS₂ is attributed to the highly active edges of the MoS₂. From the first-principle calculations, nanostructured MoS₂ is preferred over bulk MoS₂ for photocatalytic application due to the larger availability of catalytically active edge sites.¹⁷ Therefore, increasing the number of active edges in the 2D MoS₂ is critically important to bring out an efficient hydrogen evolution reaction (HER).^18–20 In this work, we employed thin MoS₂ film with numerous active edges as an efficient photocatalytic material. Photo-catalytic hydrogen generation using 2D materials shows a promising way toward straightforward and cost-effective procedures of hydrogen production.

Similar to graphene, ultrathin MoS₂ flakes can be obtained using mechanical or chemical exfoliation methods. Here MoS₂ layers were prepared using chemical exfoliation method in solvent. For this purpose, a mixture of MoS₂ powder and N-Methyl-2-pyrrolidone (NMP) solvent was ultra-sonicated for 10 hours. The solution was further centrifuged repeatedly three times at 10, 000 RPM for 10 minutes. The final product contains the few layered MoS₂ flakes. In order to produce the thin MoS₂ film for water splitting and other electrochemical experiments, MoS₂ dispersion was sprayed on the chosen substrates. For over-pressure observation experiments, 1μm-thick MoS₂ film was prepared on an alumina template using a vacuum filtration. Current-voltage measurements were conducted to study the photocatalytic hydrogen generation from these layers.

Figure 1 represents the typical transmission electron microscopy (TEM) micrograph, selected area electron diffraction (SAED) pattern, Raman spectrum and photoluminescence spectrum of the chemically exfoliated MoS₂ layers used for photocatalytic hydrogen generation. Figure 1(a) shows the morphology of the exfoliated MoS₂ layers. The rolled-up edges are evidently observed. The inset of Figure 1(a) is the high resolution TEM image of the exfoliated MoS₂, indicating that a typical hexagonal pattern of MoS₂ is maintained without any defects in the grain. The spacing of 2.7 Å corresponds to the inter-atomic spacing of MoS₂ can be seen. The SAED of Figure 1(b) shows that the exfoliated MoS₂ sample has a single crystal structure. In the Raman spectrum of bulk MoS₂, two distinct vibrational modes can be noticed at 383 cm⁻¹ for in-plane mode (E¹_2g) and at 408 cm⁻¹ out-of-plane mode (A_1g).²¹ Of these two vibrational modes, the interlayer vibrational mode is the fingerprint of the material that corresponds to the chemical composition of the material synthesized. Furthermore, their position is directed by the number of atomic layers present in the material. In the present work, the MoS₂ layers synthesized exhibited a peak centered at 388.6 cm⁻¹ and 403.1 cm⁻¹ due to the E¹_2g and A_1g modes respectively. Our experimental values upon comparison with the theoretical values reported in the literature,²¹ with frequency difference of 22.5 cm⁻¹ suggest few (3) layered samples. Photoluminescence (PL) measurement was performed on the selected MoS₂ samples. Figure 1(d) shows two emission peaks around 650 nm and 710 nm, which correspond to direct (1.90eV), and indirect (1.74eV) bandgap transition, respectively suggesting presence of exfoliated layers. Enhancement in bandgap suggests the possible photocatalytic application using chemically exfoliated layers.

First principle calculations using density functional theory suggests direct optical transition at the K point in the Brillouin zone corresponds to the peak marked by a red arrow in the electronic band structure for mono-layered MoS₂ as shown in figure 2. This band gap is found to decrease and the lowest band (green) becomes doubly degenerate at the K point in bulk MoS₂. First principle calculations were performed using density functional theory as implemented in the Vienna Ab-initio Simulation Package (VASP). The details of calculations have been described elsewhere.²² K-points mesh of 16x16x8 and was generated using Monkhorst-Pack scheme to sample the reciprocal space for electronic structures calculations.

In order to measure the amount of the photocatalytically generated hydrogen through the MoS₂ film, we fabricated the simple set-up of figure 3. The thin film of exfoliated MoS₂ coated on the porous alumina template by a vacuum filtration was immersed in a container filled with a deioned water. Top of the container was sealed with a thin rubber baloon, which can be blown up as hydrogen is generated. Upon exposure to visible light (60 W-AM 1.5) at a distance of 10 cm from the MoS₂ film, the baloon starts to expand and. The size of baloon increase with time and formation of bubbles on the MoS₂ film surface was observed. This confirms that hydrogen generation occurs through water spliting initiated by photocatalytic activity of MoS₂ films. The size of the baloon is evantually saturated after a day and maintained unchanged for three weeks. It should be noted that this experiment was conducted only for 3 weeks. To confirm the photocatalytic effect of the exfoliated MoS₂, we also performed the same experiment with bulk MoS₂ (before exfoliation) bare alumina template, or nothing in a water container with the sealed baloon on its top. No bubbles or gas generation was found in these three conditions.

For quantitaive analysis of the photocatalytic effect of exfoliated MoS₂, photo-current generated by the MoS₂ was measured. The MoS₂ suspension was sprayed on Si substrates of 1 × 1 cm². The current-voltage (I-V) characteritics was measrued in a dark condition or under visible light illumination. I-V characteristics from positive to negative bias clearly indicated that the MoS₂ film is highly sensitive to the visble light. In contrats, no such photo-current was measured for bulk MoS₂. With no external voltage applied, the visible light creates enough carriers on the surface of semiconducting MoS₂, corresponding to a current of approximately 0.2μA, which is roughly three magnitude of orders higher than that under no illumination of the light. To measure the photoelectochemical resposne of the exfolidated MoS₂ film, we set up the experiment of figure 4(a). The MoS₂ film deposited on the gold substrate of 1 x 1 cm² was used as anode and platinum plate was employed as cathode. When the MoS₂ film is exposed to a visble light, electron-hole pairs are generated. Then, photo-excited electrons in MoS₂ are transferred to the Pt counter electrode. On the Pt electrode, H⁺ ions are reduced, leading to generation of hydrogen bubbles, as reported in the typical photoelectrochemical experiments.^23–27 During the experiment, bubble formation was clearly detected in a water. The holes in the MoS₂ electrode oxidize the OH⁻ ions so that a current flows through photoelectrochemical reaction. The photoelectrochemical current curves of Figure 4(a) show that the MoS₂ film on Au are more efficient than the bare Au electrode when exposed to the visible light. To verify the dynamic optical response in photo-current, the electrical current was mesured at 0.5 V under the periodic illumination. Figure 4(b) shows the photo-current ratio of I/I_o, where I and I_o is the photo-currents under illumination and in dark, respectively. Under, the photo-current ratio reaches over 10, which indicates the high efficiency of hydrogen generation by photoelectrochemical activity of the MoS₂ film. This result shows that 2D MoS₂ can be used as an active catalyst for hydrogen generation. It is expected that the larger area monolayer film of the exfoliated MoS₂ will have the more enhanced photocatalytic behaviour due to increased area of MoS₂ film, indicative of more ractive MoS₂ edges.

In conclusions, we have developed a straightforward procedure to study hydrogen generation through water-spliting by the photoelectrochemical activity of 2D MoS₂. The chemically exfolaited MoS₂ layers were characterized for their structural quality prior to their use for hydrogen generaion experiment. The over-presure created by formation of hydrogen via water splitting by MoS₂ catalyst was investigated by measuring the blowing-up of the baloon. In the quantitative anlysis of photocatalytic effect, a large amount of photo-current in exfoliated MoS₂ film, which is higher than that in bulk MoS₂, was observed under the illumination of visible light. Our work shows a new way toward hydrogen generation by 2D MoS₂, which is required for our future energy plan.