1. Introduction

In recent years, MoS₂, used as a typical two-dimensional (2D) layered transition metal dichalcogenide (TMD), has been extensively researched for various applications owing to its electronic and optoelectronic properties, particularly in surface-enhanced Raman scattering (SERS) sensors and photoelectrocatalytic (PEC) water splitting [1–3]. SERS is a powerful analytical spectroscopy technique that can provide high-resolution vibrational information and achieve label-free single-molecule detection. [4–6] The SERS signal can be typically amplified by two distinct mechanisms: charge-transfer complexes (CM with an enhancement factor of approximately 10²–10⁴) and a localized electromagnetic field (EM with an enhancement factor of approximately 10⁶–10⁸) [7–9]. MoS₂-based SERS is largely ascribed to charge transfer and dipole-dipole coupling between the substrate and molecules [10]. PEC water splitting is considered the most effective method for producing clean, affordable, and eco-friendly hydrogen (H₂) [11–13]. It is critical to accelerate the PEC performance of MoS₂ with abundant active sites, efficient light harvesting, and smooth charge transfer [14,15].

Pure MoS₂ can only engender weak SERS activity owing to chemical mechanism enhancement, limiting the ability to achieve low concentration detection. In addition, although several advances have been achieved for MoS₂-based materials in PEC, its low conductivity and high charge recombination rates remain critical concerns, that restrict their catalytic application. Thus, based on MoS₂, heterostructures incorporating noble metal nanoparticles (NPs) have been developed to improve the SERS and PEC performances. For example, Zuo et al. developed a method to induce photogenerated electrons of MoS₂ through in situ metal NPs (Ag or Pt) decorated on MoS₂ nanosheets to form metal-MoS₂ nanohybrids, which exhibited prominent hydrogen evolution reaction (HER) activity and SERS sensitivity [16]. Zheng et al. demonstrated that the coupling effect of the MoS₂/Au heterostructure could result in both enhancement of PEC water splitting and SERS performance [17]. Dong et al. reported an ultrafast electron transfer heterostructure composed of Pt nanoparticles grown in situ on MoS₂ nanosheet edge sites representing high PEC performance and SERS activity [18]. These heterostructures only used unitary metal NPs to decorate MoS₂ material, and the improvement in SERS or PEC was restricted. It is well known that the structure of metal NPs is vital for their plasmonic characteristics and performance. For better PEC properties based on quantum effects, the noble metal particle should be less than ten nanometers in size, whereas metal particles should be tens of nanometers in size with outstanding SERS properties owing to plasmon resonance [19–21]. Therefore, it is still a serious challenge to propose a substrate with both excellent SERS performance and PEC activity.

In this study, Ag-Au bimetal was prepared with different particle sizes and semiconductor MoS₂ was combined as a supporting layer decorated on FTO conducting glass to form a ternary heterostructure (MoS₂/Au/Ag). The smaller-sized AuNPs play a significant role in PEC performance, and larger-sized AgNPs possess excellent SERS sensitivity, solving the contradictory issue above mentioned. Furthermore, the vertical MoS₂ nanosheets prepared by the hydrothermal method can provide a larger surface area and more active sites, creating a higher density of “hot spots” and faster charge transfer. Importantly, the bimetal-semiconductor composite structure can introduce multiple surface plasmonic coupling because of the size effect, which plays a vital role in both enhanced SERS and PEC performance. The enhanced SERS performance and excellent PEC water splitting of the proposed MoS₂/Au/Ag heterostructure were demonstrated by comparison with the individual MoS₂, MoS₂/Au, and MoS₂/Ag substrates, which could serve as a feasible approach for the advancement of renewable energy and sensor research.

2. Methods

2.1. Fabrication of the MoS₂ on a FTO substrate

The FTO conducting glass (1×1 cm²) was cleaned with acetone, alcohol, and deionized water (DI water) for 20 min in sequence with an ultrasonic cleaner prior to use. Vertical MoS₂ nanosheets grew on the surface of the FTO glass using the hydrothermal method. First, thiourea (CH₄N₂S 4.56 g) and sodium molybdate dihydrate (Na₂MoO₄·2H₂O 3.4 g) were dissolved in 30 mL of DI water and 30 mL of ethanol solution under vigorous stirring. Then, the mixture solution was placed in a 100 mL Teflon-lined stainless-steel autoclave, and clear FTO glass was placed into the autoclave with the right side up. Thereafter, to complete the sulfuration reaction, they were heated in a laboratory oven at 180 °C for 7 h. When the autoclave was cooled to room temperature, the sample was removed and continuously washed with DI water to remove excess MoS₂ without attachment to the surface. Finally, MoS₂ on FTO was acquired after drying in a vacuum drying oven.

2.2. Fabrication of MoS₂/Au on a FTO substrate

The AuNPs were synthesized on the MoS₂ nanosheets by an in situ reduction reaction with HAuCl₄, forming a MoS₂/Au hybrid structure. The prepared FTO/MoS₂ substrate was immersed in a 0.6 mM HAuCl₄ solution to generate AuNPs. After 10 min, the solution was transferred three times in a clean DI water to remove the residual HAuCl₄. Finally, the MoS₂/Au substrate on the FTO glass was dried in a vacuum drying oven.

2.3. Fabrication of MoS₂/Au/Ag on a FTO substrate

The AgNPs were fabricated using a chemical reduction method, as described in our previous study [22]. Briefly, 20 ml ethylene glycol was placed in a clear flask and heated in an oil bath under continuous stirring. The PVP powder (Mw = 55000) of 0.25 g was added in the flask at 70 °C, then AgNO₃ power of 0.05 g was added to the mixture at 125 °C. The temperature was maintained at 135 °C for 1 h to incubate uniform AgNPs with the color from brownish to milky green. Then, 30 mL of acetone was added to the cooled colloid for separation. A few minutes later, the solution was centrifuged three times at 12000 rpm for five minutes and dispersed in DI water. To firmly cling AgNPs to the surface of the MoS₂/Au substrate, the substrate was placed in the AgNP solution for 12 h. Finally, the sample was cleaned with DI water to remove excessive AgNPs, forming a MoS₂/Au/Ag substrate on the FTO glass.

2.4. Apparatus and characterization

Scanning electron microscope (SEM) images (5.0 kV) were obtained with a ZEISS Sigma500 equipped with an energy dispersive spectrometer (EDS) (20 kV). A more detailed morphology and composition were characterized using a high-resolution transmission electron microscope (HRTEM, JEM-2100F) at 200 kV. X-ray diffraction (XRD, Smartlab9) and X-ray photoelectron spectroscopy (XPS, Thermo Fisher Scientific 250Xi) were performed to analyze and identify the crystal phase, composition, and valence state of the prepared substrate. The UV-Vis adsorption spectra were measured in the range of 200–800 nm using a UV-3600 spectrophotometer.

2.5. SERS spectra measurement

All Raman spectra were measured using a Raman spectrometer with a 532 nm laser (Horiba HR Evolution 800). The laser power was 0.48 mW with an integration time of 4 s and a diffraction grid of 600 g/nm. A 50× objective was used to focus the laser beam on the substrate. The average intensities of the Raman signals at five random positions on the substrate were employed to ensure accuracy.

2.6. Photoelectrocatalysis measurements

All photoelectrochemical measurements were performed at room temperature in a three-electrode system using a CHI 760E electrochemical workstation. A Ag/AgCl (KCl saturated) electrode was used as the reference electrode, and a graphite rod was used as the counter electrode for HER tests in a high-purity Ar-saturated 0.5 M H₂SO₄ electrolyte. The polarization curves were recorded at a scan rate of 5 mV/s. Potentials were referenced to an RHE by adding a value of 0.197 V. The electrochemical impedance spectroscopy (EIS) tests were performed at open-circuit potentials with a frequency range of 0.1–10⁵ Hz with an amplitude of 5 mV. A 300 W Xenon arc lamp equipped with an AM 1.5 filter was used as a light source to irradiate the working electrode.

2.7. Electric field distribution simulation

Finite-difference time-domain (FDTD) simulations were conducted to simulate the electromagnetic field distributions. The linearly polarized monochromatic plane wave was 532 nm with polarization along the X-direction. The absorption boundary condition is the perfect matching layer (PML) in the Z-direction, anti-symmetric in the X-direction, and symmetric in the Y-direction. The geometrical parameters of the structures were obtained from the SEM measurements. The refractive index data of MoS₂ were obtained from Beal et al. [23], and the dielectric dates of Ag and Au were obtained from Johnson et al. [24]. A mesh size of 0.8 nm was used.

3. Results and discussions

The morphology of the as-fabricated substrate was studied using SEM, as shown in Fig. 1(a)–(d). The pristine FTO glass possessed a rugged surface, as shown in Fig. S1(a) in Supplement 1, which was covered with vertical nanosheets with 3D morphology by a hydrothermal process after MoS₂ was grown (Fig. 1(a)). It can be clearly observed that the bare MoS₂ nanosheets exhibit closely aligned flake-like structures with an average wall thickness of 12 nm and length of 200 nm. Figure 1(b) shows the morphology of denser AuNPs decorated on the MoS₂ nanosheets by in situ reaction with HAuCl₄ forming a MoS₂/Au substrate, where the size of the AuNPs was approximately 8 nm and the interparticle gap was 3 nm. The AgNPs on the surface of MoS₂ are presented in Fig. 1(c) (MoS₂/Ag), and the average diameter of the AgNPs was approximately 50 nm. Figure 1(d) illustrates the heterostructure of MoS₂ composed of Au-Ag bimetal (MoS₂/Au/Ag), where the large AgNPs and small AuNPs were evenly attached on the MoS₂, and the vertical nanosheet structure was clearly visible. To further identify and verify the components in the hybrid, HRTEM was performed, as shown in Fig. 1(e), where small AuNPs, large AgNPs, and MoS₂ can be clearly observed, indicating that the MoS₂ nanosheets can act as an efficient substrate for the nucleation of AuNPs and support of AgNPs. The distinct interlayer spacings of 0.235, 0.236, and 0.605 nm can be assigned to the (111) plane of Au, (111) plane of Ag, and (003) plane of MoS₂, respectively. The SAED pattern of MoS₂/Au/Ag is shown in Fig. 1(f), in which the diffraction rings of Au (111) and Ag (111) are well indexed to those observed in HRTEM, and the (004) and (106) planes of MoS₂ can also be identified with lattice spacings of 0.309 nm and 0.164 nm. In addition, the corresponding EDS spectra of the hybrid (Fig. S1(b), Supplement 1) also prove the existence of Mo, S, Au, and Ag in the prepared substrate. The elemental maps are shown in Fig. S1(d), Supplement 1) from the local composition of the sample (Fig. S1(c), Supplement 1), where the uniform distribution of MoS₂, Au, and Ag was further confirmed, that was important for SERS detection and PEC water splitting.

 

Fig. 1. SEM images of (a) MoS₂ nanosheets, (b) MoS₂/Au, (c) MoS₂/Ag, and (d) MoS₂/Au/Ag on FTO glass. (e) HRTEM image of AuNPs and AgNPs decorated on the MoS₂ nanosheets. (f) SAED pattern of MoS₂/Au/Ag. The local electric field distributions for (g) the MoS₂, (h)MoS₂/Au, (i) MoS₂/Ag, and (j) MoS₂/Au/Ag substrate.

Download Full Size | PDF

Based on the above sample geometries shown by the dashed line in Fig. 1(d), the FDTD method with Maxwell’s equations was utilized to understand the near-field enhancement of the MoS₂-based substrate in Fig. 1(g)-(j). As shown in Fig. 1(g), the pristine MoS₂ nanosheets exhibited poor electromagnetic field enhancement. With the decoration of AuNPs (Fig. 1(h)), the “hot spots” were mainly distributed between the AuNPs owing to the local surface plasmon resonance (LSPR) effect of AuNPs. For MoS₂/Ag (Fig. 1(i)), the electric field was greatly enhanced, mainly between the gap of the AgNPs. With respect to the MoS₂/Au/Ag substrate (Fig. 1(j)), the dense “hot spots” were distributed between AuNPs and AgNPs as well as between AgNPs and AuNPs, forming multiple surface plasmonic resonance coupling, where the number and intensity of the “hot spots” were clearly enhanced compared with those of the MoS₂, MoS₂/Au, and MoS₂/Ag substrates.

To analyze the elemental composition and chemical states of the MoS₂/Au/Ag heterostructure quantitatively and qualitatively, XPS characterization was performed, as shown in Fig. 2. The survey spectrum of the MoS₂/Au/Ag composites shown in Fig. S2 demonstrates the presence of Mo, S, Au and Ag elements, and all the elemental spectra were calibrated by a carbon 1s peak (285.0 eV). As shown in Fig. 2(a), the high-resolution Mo 3d XPS scan identified two main peaks at 228.3 and 231.5 eV assigned to Mo 3d_5/2 and Mo 3d_3/2 respectively, from the IV oxidation state of Mo⁴⁺ [25], and the weak peak at 235.6 eV was attributed to Mo-O because of slight oxidation of the MoS₂ surface [26]. Moreover, the peak at 225.4 eV was attributed to S 2s. With respect to the S 2p spectrum of MoS₂ shown in Fig. 2(b), the fitted peaks located at 161.1 eV and 162.2 eV were assigned to the S 2p_3/2 and S 2p_1/2 orbital of S^2- respectively [25]. Two well-defined peaks at 83.7 eV and 87.4 eV in the XPS spectrum of Au 4f (Fig. 2(c)) were attributed to Au 4f_7/2 and Au 4f_5/2 respectively, which were derived from the in situ grown AuNPs [27]. Meanwhile, there were two prominent peaks of Ag 3d at 367.6 eV and 373.6 eV, as shown in Fig. 2(d), supporting the insertion of crystallized Ag [28]. Thus, the above characterizations confirm the final synthesis of the MoS₂/Au/Ag hybrid substrate.

 

Fig. 2. XPS spectra in (a) Mo 3d, (b) S 2p, (c) Au 4f, and (d) Ag 3d regions of MoS₂/Au/Ag substrates. (e) SERS spectra of different samples: MoS₂, MoS₂/Au, MoS₂/Ag, and MoS₂/Au/Ag substrates. (f) UV-Vis absorption spectra of MoS₂, MoS₂/Au, MoS₂/Ag, and MoS₂/Au/Ag substrates.

Download Full Size | PDF

In addition, the formation of the MoS₂/Au/Ag composite crystal structure was confirmed by XRD (Fig. S2(b), Supplement 1). The pure MoS₂ exhibited multi-correlation diffraction peaks at 29.02°, 32.67°, 49.78°, and 58.33°, corresponding to the (004), (100), (105), and (110) planes of MoS₂, respectively (PDF#37-1492). With respect to the MoS₂/Au substrate, the peaks of MoS₂ at 29.02° and 32.67° were clearly observed, and new diffraction peaks were located at 38.18° and 44.39°, which could be assigned to the (111) and (200) diffraction peaks of Au (PDF#04-0784), respectively. For MoS₂/Ag, the new diffraction peak at 38.11° was attributed to the (111) planes of Ag (PDF#04-0783). Meanwhile, the XRD patterns of MoS₂/Au/Ag showed distinct peaks at 29.02°, 38.11°, and 44.39° assigned to MoS₂ (004), Ag (111), and Au (200), respectively. The XRD results further illustrated that the Au-Ag bimetal successfully decorated the surface of MoS₂, which is consistent with the results of SEM and TEM. A comparison of the Raman spectra of the different MoS₂-based substrates is shown in Fig. 2(e). There are two representative vibration peaks of MoS₂ in all spectra attributed to the in-plane vibrations of Mo and S atoms ($E_{2g}^1$) and the out-of-plane lattice vibration of S atoms (A_1g), respectively [29]. The peak at 402 cm⁻¹ (A_1g) of pristine MoS₂ was gradually red-shifted to 405 cm⁻¹, 407 cm⁻¹ and 408 cm⁻¹ by the decoration of AuNPs, AgNPs, and Au-Ag bimetal, respectively, indicating a strong interaction between the metal nanoparticles and the MoS₂ nanosheets [17]. In addition, the intensity of the two peaks was enhanced after the introduction of AuNPs, which was attributed to the LSPR of AuNPs, and MoS₂/Ag had a higher intensity because AgNPs generated a large local electromagnetic field, whereas the MoS₂/Au/Ag sample exhibited the strongest enhancement owing to the multiple surface plasmonic resonance coupling [30]. To understand the optical behavior of the MoS₂-based samples, Fig. 2(f) shows the corresponding UV-Vis spectra. The spectrum of pristine MoS₂ on FTO shows a wide absorption band above 300 nm towing to the multilayer nanostructure and intrinsic narrow band gap [31]. After the introduction of AgNPs, the absorption intensity of MoS₂/Ag was enhanced in both the UV and visible range, and the distinct peak at 480 nm was because of the LSPR mode of AgNPs. Furthermore, the UV-Vis spectra of MoS₂/Au displayed a distinct band located at approximately 550 nm owing to the LSPR of AuNPs. The spectrum of the MoS₂/Au/Ag substrate exhibited the strongest light absorption compared with MoS₂, MoS₂/Ag, and MoS₂/Au, where the two prominent peaks of AgNPs at 480 nm and AuNPs at 550 nm were observed, supporting the formation of multiple metal NPs [32].

The SERS performance of the four MoS₂-based samples was evaluated by detecting 10⁻⁵ M R6G, as shown in Fig. 3(a). The SERS intensity gradually increased as a function of decoration with AuNPs, AgNPs, and Au-Ag bimetal. This indicated that the best SERS sensing performance was exhibited from MoS₂/Au/Ag substrate compared with pure MoS₂, MoS₂/Au, and MoS₂/Ag, mainly attributed to multiple surface plasmonic coupling effects between the AnNPs, AgNPs, and Au-Ag bimetal. In addition, the MoS₂/Ag exhibited better enhancement than MoS₂/Au in SERS because AgNPs could cause a stronger electromagnetic field prior to AuNPs, and the pristine MoS₂ substrate had the worst SERS effect because of weak CM. To better demonstrate the enhancement effect of the MoS₂-metal samples, Raman spectra of R6G with different concentrations were obtained on the MoS₂/Au, MoS₂/Ag, and MoS₂/Au/Ag substrates. In Supplement 1, Fig. S3(a) shows the Raman spectra of the MoS₂/Au substrate, where the intensity of Raman peaks decreased monotonically with the attenuation of R6G concentration, and the limit of detection (LOD) of R6G was approximately 10⁻⁸ M. For MoS₂/Ag, the LOD could reach 10⁻¹⁰ M (Fig. S3(b), Supplement 1), although the intensity was fairly weak and only the main characteristic peaks of 613 cm⁻¹ and 1360 cm⁻¹ could be observed. The linear relationship of the SERS intensity at 613 cm⁻¹, 774 cm⁻¹, and 1650 cm⁻¹ as a function of the R6G concentration on a log scale is presented in Fig. S3(c) in Supplement 1, which fit the linearity with correlation coefficients (R²) of 0.998, 0.980, and 0.944, respectively. The corresponding SERS spectra of the MoS₂/Au/Ag substrate can be observed in Fig. 3(b), in which the Raman signal diminishes when the R6G solution is diluted from 10⁻⁵ to 10⁻¹² M, and the peak of 613 cm⁻¹ at 10⁻¹² M can still be identified. The LOD of the MoS₂/Au/Ag substrate is two orders of magnitude lower than that of MoS₂/Ag and four orders of magnitude lower than that of MoS₂/Au. In addition, a well-defined linear relationship of the MoS₂/Au/Ag substrate was obtained, as shown in Fig. 3(c), with R² values of 0.997, 0.991, and 0.992 at 613 cm⁻¹, 774 cm⁻¹, and 1650 cm⁻¹ respectively, thus having a great potential for sensitive and quantitative detection of R6G dye. Therefore, these results indicate that the MoS₂/Au/Ag substrate possesses the best sensitivity, whose enhancement ability is evaluated using the following formula [33]:

 

Fig. 3. (a) Raman spectra of R6G (10⁻⁵ M) detected on MoS₂, MoS₂/Au, MoS₂/Ag, and MoS₂/Au/Ag substrates. (b) Raman spectra of R6G (the concentration from 10⁻⁵ M to 10⁻¹² M) on the MoS₂/Au/Ag SERS substrate. (c) Linear relationships: Raman intensities at 613 cm⁻¹, 774 cm⁻¹, and 1650 cm⁻¹ on the MoS₂/Au/Ag as a function of the concentrations of R6G molecules. (d) The SERS spectra of R6G (10⁻⁶ M) detected at 10 random positions on MoS₂/Au/Ag sample.

Download Full Size | PDF

It should be noted that uniformity and reproducibility play important roles in SERS detection. Thus, the Raman spectra of 10⁻⁶ M R6G was measured from ten random positions on MoS₂/Au/Ag substrates (Fig. 3(d)), and the Raman spectra curves displayed almost the same intensities for all characteristic peaks, proving favorable uniformity. Furthermore, Fig. S3(d) in Supplement 1 reveals the intensities of the three main characteristic peaks at 613 cm⁻¹, 774 cm⁻¹, and 1650 cm⁻¹ of 10⁻⁵ M R6G detected from 10 batches of MoS₂/Au/Ag SERS substrate, which had slight fluctuations with RSDs of 10.32%, 15.67%, and 15.01%, respectively. In general, the outstanding uniformity and reproducibility unambiguously reveal that the proposed MoS₂/Au/Ag substrate has great potential for SERS applications.

Except for the advantages of SERS property, the PEC behavior of MoS₂-based samples was explored investigated in a typical three-electrode system in 0.5 M Ar-saturated H₂SO₄ electrolyte using an Ag/AgCl electrode as the reference electrode and a graphite rod as the counter electrode. Figure 4(a) shows the polarization curves of HER on a series of catalysts in the dark and under illumination, where the pristine MoS₂ nanosheets have the worst catalysis and the catalytic performance gradually increases with the load of AgNP, AuNPs, and Au-Ag bimetal. Moreover, the same sample possessed better catalysis under visible-light illumination than in the dark. For comparison, Fig. 4(c) shows the corresponding overpotential (10 mA/cm²) and onset potential (1 mA/cm²) of the four samples in the dark and light. It is fairly self-explanatory that MoS₂/Au/Ag on FTO glass under illumination has the highest catalytic performance with an overpotential of 124 mV and onset potential of 32 mV, which is not only lower than that of MoS₂/Au, MoS₂/Ag, and MoS₂, but also lower than that in the dark. This is probably because of the more efficient electron-hole separation under illumination resulting from the multiple surface plasmonic coupling of the tripartite structure. To confirm the electron transfer rate during the HER process, EIS analysis is presented in Fig. 4(b). It was clearly observed that the semicircle of MoS₂/Au/Ag is much smaller than MoS₂/Au, MoS₂/Ag, and MoS₂, indicating the smaller charge-transfer resistance and higher intrinsic conductivity [17]. Furthermore, the smaller semicircle under light illumination manifests a more effective charge transfer and superior PEC activity compared with that in the dark, which is consistent with the polarization curve results. In addition, the photocurrent responses of the four substrates were measured to further explore their PEC behavior, as shown in Fig. 4(d). Clearly, the MoS₂/Au/Ag heterostructure exhibits the highest photocurrent density of 43.9 µA/cm², which is approximately 9.1, 2.2, and 1.4 times higher than those of MoS₂, MoS₂/Ag, and MoS₂/Au, respectively, and compared with other reported MoS₂-based samples, such as TiO₂/Ag/MoS₂ with a photocurrent density of 3 µA/cm² and g-C₃N₄/Ag/MoS₂ of 0.375 µA/cm² [34,35], the MoS₂/Au/Ag heterostructure exhibited the best enhancement of photocurrent, indicating the superior light utilization and charge transport efficiency attributed to the synergistic effect of the Au-Ag bimetal and MoS₂ nanosheets. To evaluate the stability of the MoS₂/Au/Ag catalyst, long-term electrochemical measurements were performed, as shown in Fig. 4(e). A photocurrent of approximately 10 mA/cm² at a potential of 150 mV could be maintained for at least 8 h, and the SEM morphology of the heterostructure (inset of Fig. 4(e)) after 8 h shows almost no change compared with Fig. 1(d), demonstrating the outstanding stability of MoS₂/Au/Ag for catalysis.

 

Fig. 4. (a) Polarization curves of HER for MoS₂, MoS₂/Au, MoS₂/Ag, and MoS₂/Au/Ag on FTO glass in the dark and under illumination. (b) The corresponding electrochemical impedance spectroscopy (EIS) Nyquist plots. (c) The corresponding overpotential (10 mA/cm²) and onset potential (1 mA/cm²) in the dark and under illumination. (d) Photoresponse of MoS₂, MoS₂/Au, MoS₂/Ag, and MoS₂/Au/Ag on FTO glass. (e) Photocurrent stability of the MoS₂/Au/Ag on FTO glass in 8 h at an overpotential of 150 mV.

Download Full Size | PDF

To illustrate the possible derivation for the superior catalysis activity of MoS₂/Au/Ag, the electrochemical surface area (ESCA) of each sample was evaluated. The potential range where no faradic current was selected and the differences between the positive and negative current densities at the center of the scanning potential ranges were plotted versus the different scan rates, where the slopes represent the electrochemical double-layer capacitances (Cdl). The CV curves at 0–0.2 V vs. RHE with different scan rates of MoS₂, MoS₂/Ag, MoS₂/Au, and MoS₂/Au/Ag on FTO are respectively shown in Fig. 5(a)-(d). Clearly, the MoS₂/Au/Ag catalyst shows the maximum slope compared to MoS₂, MoS₂/Ag, and MoS₂/Au (Fig. 5(e)), which represents the largest Cdl value of MoS₂/Au/Ag, suggesting that it has more exposed catalytic active sites during PEC.

 

Fig. 5. Cyclic voltammetry (CV) curves at different scan rates of (a) MoS₂, (b) MoS₂/Ag, (c) MoS₂/Au, and (d) MoS₂/Au/Ag. (e) Plot of current density vs. scan rate for the four samples. (f) Illustration of the mechanism of PEC with MoS₂/Au/Ag.

Download Full Size | PDF

Based on the aforementioned results, a proposed mechanism under visible-light irradiation in the PEC process is described, as shown in Fig. 5(f). Apparently, PEC reactions occurred between the Au-Ag hybrid and MoS₂ nanostructures owing to the LSPR and electron transfer. For the Au-Ag bimetal, near-field enhancements existed at the interface of AgNPs and AuNPs, inducing a direct interfacial electronic transition [36,37]. Under visible-light irradiation, the surface plasmons excited in the Au-Ag bimetal owing to the LSPR coupling of Au and Ag, where photogenerated electrons were generated and the reaction rate significantly increased, resulting in a distinct improvement in the PEC activity [38]. Moreover, the photogenerated electrons induced by plasmons with sufficiently high energies could overcome the Schottky barrier between MoS₂ and nanoparticles and transfer to the MoS₂ conduction band (CB). In addition, the Schottky barrier could efficiently prevent the injected photogenerated electrons from combining holes during the PEC process. Thus, the results of this study demonstrate that the superior light-driven PEC performance of MoS₂/Au/Ag distinctly improves its activity by combining MoS₂ nanosheets with Au-Ag bimetal.

4. Conclusion

In summary, a ternary heterojunction consisting of vertical MoS₂ nanosheets decorated with small-sized AuNPs and large-sized AgNPs as substrates with good synergistic enhancement of SERS and PEC performance utilizing the size effect was developed. For SERS, the MoS₂/Ag hybrid structure exhibited better performance than the MoS₂/Au substrate, whereas it was in contrast during PEC. The proposed heterostructure of MoS₂/Au/Ag manifests outstanding SERS enhancement with an EF of 1.1×10⁹ compared with MoS₂, MoS₂/Au, and MoS₂/Ag samples; the performance of the MoS₂/Au/Ag hybrid catalyst based on PEC remains the most efficient with the lowest overpotential and onset potential and the highest photocurrent response, which are all because of the introduction of multiple surface plasmonic coupling effects of the Au-Ag bimetal and generate more plasmonic hot spots with the size effect. Furthermore, the MoS₂/Au/Ag heterostructure also possesses excellent homogeneity and reproducibility for SERS and outstanding stability for catalysis. The strategy proposed in this study can pave the way toward the fabrication of a series substrate of a multi-size bimetal composited semiconductor with perfect synergistic enhancement of SERS and PEC performance.