1. INTRODUCTION

Two-dimensional transition metal dichalcogenides (TMDs) have attracted significant attention as next-generation functional materials owing to their unique electronic structures, high surface-to-volume ratios, and tunable physicochemical properties^[1–6]. Among TMDs, tungsten disulfide (WS₂) is considered one of the most promising candidates for multidisciplinary applications, including photoelectrochemical (PEC) energy conversion, optoelectronics, catalysis, and gas sensing, due to its suitable bandgap, excellent chemical stability, and strong light–matter interaction^[7–10]. The performance of WS₂ for practical device applications, however, is strongly governed not only by its intrinsic electronic structure but also by its nanoscale morphology. In particular, vertical nanosheet architectures have emerged as an effective strategy to enhance functional performance, as they provide enlarged active surface areas, abundant edge sites, and shortened charge transport pathways^[10–12]. These advantages are beneficial for both PEC reactions, which require efficient charge separation and transport, and gas sensing, which relies on surface adsorption and rapid charge transfer between gas molecules and sensing materials. Notwithstanding extensive research on WS₂, most previous studies have focused on optimizing WS₂ for a single application, such as PEC water splitting or gas sensing. Optimization based on a single function limits the expandability of WS₂ across diverse application fields. Moreover, although vertical WS₂ nanosheets are known to be advantageous, systematic strategies to optimize their structural parameters, such as density and thickness, for multifunctional applications have not been sufficiently explored.

Herein, we report a simple yet effective morphology engineering strategy to realize high-performance multifunctional WS₂ using a controllable metal–organic chemical vapor deposition process. By precisely modulating the growth pressure, we are able to systematically tune the vertical nanosheet density and thickness of WS₂ films. The morphology-engineered WS₂ electrodes are then evaluated in both PEC and NO₂ gas sensing systems. We demonstrate that an optimized WS₂ nanosheet architecture simultaneously enhances PEC photocurrent generation and the gas sensing response. Whereas previous studies have largely focused on optimizing vertically aligned WS₂ for a single functional application, such as PEC water splitting or gas sensing, the present work systematically correlates pressure-controlled morphology parameters with performance across two distinct application platforms. By identifying a common optimized nanosheet architecture that simultaneously enhances PEC and gas sensing characteristics, we highlight morphology engineering as a multifunctional design strategy rather than a single-purpose optimization approach. This work highlights morphology control as a universal design principle for extending the applicability of a limited range of TMD materials to diverse technological fields.

2. EXPERIMENTAL

3. RESULTS AND DISCUSSION

Figure 1 presents cross-sectional SEM images of WS₂ grown on FTO substrates under different growth pressures that highlight the evolution of the nanosheet morphology. The insets provide corresponding top view SEM images for comparison. As shown in the cross-sectional and planar SEM images in Figure 1(a), WS₂ grown at 400 °C under a pressure of 1 torr (hereafter denoted as WS₂-1torr) exhibits a sharp vertically oriented nanosheet morphology on the substrate. The average lateral size and thickness of the WS₂ nanosheets are both in a range of approximately 220 to 250 nm. When the growth pressure is increased to 3 torr (referred to as WS₂-3torr in Figure 1(b)), vertically aligned WS₂ nanosheets with uniform thickness and high density are formed. This results in a well interconnected network structure with nanosheets reaching sizes in a range of 400–480nm. With a further increase in growth pressure to 5 torr (referred to as WS₂-5torr in Figure 1(c)), the vertical WS₂ nanosheets continue to grow and become excessively thick and densely packed, with sizes reaching approximately 600 to 750 nm. Upon further increasing the growth pressure to 10 torr (denoted as WS₂-10torr in Figure 1(d)), the WS₂ film exhibits a nanosheet structure with a thickness comparable to that of WS₂-1torr but with a significantly denser arrangement. This is attributed to the competitive effects of vertical growth on the substrate and direct gas phase reactions during WS₂ formation under higher pressure conditions. The morphological evolution of WS₂ is primarily governed by the growth pressure, which modulates the supersaturation level within the MOCVD chamber. At an elevated pressure of 10 torr, the increased partial pressure of precursors leads to high supersaturation, which favors nucleation over individual crystal growth. This results in a high density of initial nuclei that rapidly intersect and promotes vertical stacking and polycrystalline formation rather than lateral expansion. Consequently, the WS₂ films grown at higher pressures exhibit increased thickness and a dense nanosheet architecture. In contrast, lower pressure (1 torr) reduces the supersaturation level and thereby shifts the growth mode toward lateral expansion where atoms are systematically incorporated into existing nuclei. This leads to the formation of larger WS₂ domains with reduced density. These results demonstrate that precise pressure control is an effective strategy for engineering WS₂ morphology to realize a transition from planar layers to dense, vertical nanosheet structures.

Fig. 1. SEM images of WS₂ synthesized on FTO substrate at various growth pressure: (a) 1 torr, (b) 3 torr, (c) 5 torr, and (d) 10 torr.

Figure 2(a) presents the Raman spectra of the WS₂ nanosheets with various growth pressures. Irrespective of the growth pressure, all WS₂ samples exhibit two distinct peaks at 355.1 and 419.4 cm⁻¹, which are respectively assigned to the E¹ _2g and A_1g phonon modes of 2H-WS₂ ^[14]. The E¹ _2g mode originates from the in-plane vibrational motion of W and S atoms, whereas the A_1g mode arises from the out-of-plane vibration of S atoms^{[10, 15]}, confirming the successful formation of WS₂ under all growth conditions. For detailed chemical identification, the W 4f and S 2p core-level spectra of the WS₂ nanosheets grown at 3 torr (WS₂-3torr) were examined by XPS, as shown in Figures 2(b) and (c). The deconvoluted W 4f spectrum in Figure 2(b) displays characteristic W 4f_7/2, W 4f_5/2, and W 5p_3/2 peaks centered at 33.0, 35.2, and 38.55 eV, respectively, which are consistent with WS₂ ^[10]. A weak feature at 36.6 eV is attributed to WO₃ and indicates minor surface oxidation that is likely caused by exposure to ambient air during sample transfer^[16]. The S 2p spectrum in Figure 2(c) exhibits two distinct peaks at 162.7 and 163.95 eV, corresponding to the S 2p_3/2 and S 2p_1/2 states of WS₂ ^[10]. These results confirm the formation of stoichiometric WS₂ with negligible oxide related impurities.

Fig. 2. (a) Raman spectra of various WS₂ samples (WS₂-1torr, WS₂-3torr, WS₂-5torr, and WS₂-10torr). XPS spectra of (b) W 4f, and (c) S 2p core levels for WS₂-3torr sample.

The PEC activities of various WS₂ photoelectrodes (WS₂-1torr, WS₂-3torr, WS₂-5torr, and WS₂-10torr) were evaluated by linear sweep voltammetry under dark conditions and simulated AM 1.5G illumination. Figure 3(a) presents the photocurrent density−potential (J−V) curves obtained under simulated solar irradiation. All WS₂ samples exhibited clear photoresponses with onset potentials of approximately 0.3 V, irrespective of the growth conditions. In contrast, the dark current gradually increased from around 0.5 V, with no pronounced differences among the samples (Figure 3(b)). This increase in dark current is likely attributable to electrocorrosion of the TMD materials^{[17, 18]}. In contrast to the dark current, which showed similar behavior across all growth conditions, the PEC photocurrent performance was strongly influenced by the morphology of WS₂. Figure 3(c) summarizes the maximum photocurrent densities of each sample. The WS₂-3torr sample exhibited a marked increase in photocurrent compared to WS₂-1torr and delivered the highest PEC performance, reaching 2.94 mA cm^–2 at 0.8 V. However, increasing the growth pressure further to 5 torr and 10 torr, corresponding to WS₂-5torr and WS₂-10torr, did not enhance the PEC performance and instead led to deteriorated performance relative to WS₂-3torr. This behavior can be attributed to the following factors. First, small nanosheets exhibit relatively limited PEC photocurrent because their specific surface area is not sufficiently enhanced. Second, large and excessively packed vertical nanosheets can hinder carrier diffusion and reduce the exposure of active sites due to strong overlap between neighboring sheets, leading to degraded PEC activity^{[11, 19]}. These results indicate that WS₂-3torr, which consists of vertically aligned nanosheets with optimized size and density, delivers the best PEC photocurrent performance. Its optimized morphology provides a balanced structural configuration in which the nanosheet thickness remains comparable to or shorter than the effective carrier diffusion length and thereby suppresses bulk recombination. Simultaneously, the increased edge exposure and interconnected vertical pathways enhance interfacial charge transfer kinetics at the electrolyte interface and contribute to the improved photocurrent response.

Fig. 3. (a) Photocurrent density-potential curves, (b) dark current density-potential curves, and (c) highest photocurrent densities of PEC cells with various WS₂ working electrodes (WS₂-1torr, WS₂-3torr, WS₂-5torr, and WS₂-10torr).

To assess the multifunctional applicability of WS₂ with controlled morphology, the samples were further evaluated as NO₂ gas sensors. WS₂ nanosheets were grown on Al₂O₃ substrates patterned with interdigitated Au electrodes under identical pressure conditions. SEM images in Figures 4(a)–(d) reveal morphological trends similar to those observed on FTO substrates and confirm that the morphology of WS₂ can be controlled independently of the substrate. Figure 4(e) shows the Raman spectra of WS₂ synthesized at different pressures on Al₂O₃ substrates. All samples exhibit the characteristic E¹ _2g and A_1g modes of WS₂ ^{[11, 14]}, indicating that the structure of WS₂ is preserved regardless of the substrate. The NO₂ gas sensing performance was evaluated at 30 °C under 100 ppm NO₂, and the corresponding dynamic response curves are presented in Figure 4(f). Among all samples, the WS₂-3torr sensor demonstrates the highest response (R_a/R_g) and the fastest response speed. The WS₂-1torr sensor shows a relatively weak response due to an insufficient density of active adsorption sites. The superior sensing performance of WS₂-3torr is attributed to its high density of accessible edge sites and large effective surface area, which promote efficient NO₂ adsorption and rapid charge transfer at the WS₂–gas interface^[20-22]. As an oxidizing gas, NO₂ acts as an electron acceptor and withdraws electrons from the n-type WS₂ surface upon adsorption, leading to modulation of the depletion layer and a corresponding change in resistance. Therefore, an increased density of exposed edge sites promotes stronger gas adsorption and more pronounced charge transfer and accordingly results in enhanced sensing response. Meanwhile, the WS₂-5torr and WS₂-10torr sensors exhibit degraded sensitivity, which is ascribed to hindered gas diffusion and limited utilization of inner active sites within overly compact nanosheet networks. Notably, both PEC and gas sensing performance exhibit the same dependence on the WS₂ morphology, with the WS₂-3torr sample consistently outperforming the other samples. This suggests that efficient charge transport and an enlarged active surface area are common requirements for both photoelectrochemical and gas sensing processes. Consequently, the optimized vertical WS₂ nanosheet architecture provides a versatile structural platform that simultaneously enhances light-driven electrochemical reactions and gas solid interfacial charge transfer.

Fig. 4. SEM images of WS₂ synthesized on Al₂O₃ substrate at various growth pressure: (a) 1 torr, (b) 3 torr, (c) 5 torr, and (d) 10 torr. (e) Raman spectra and (f) Response curves to 100 ppm NO₂ of various WS₂ sample (WS₂-1torr, WS₂-3torr, WS₂-5torr, and WS₂-10 torr).

4. CONCLUSIONS

In summary, we demonstrated a morphology engineering strategy to realize multifunctional WS₂ for both PEC energy conversion and NO₂ gas sensing applications. By modulating the growth pressure during the MOCVD process, the density and thickness of vertical WS₂ nanosheets were precisely controlled. Among the fabricated samples, WS₂ grown at 3 torr exhibited an optimized nanosheet architecture that resulted in the highest PEC photocurrent density and superior gas sensing performance. The enhanced dual functionality arises from the combined effects of increased surface area, abundant exposed edge sites, and improved charge transport pathways provided by the optimized vertical nanosheet structure. This study suggests morphology control as a universal and effective design principle for broadening the applicability of WS₂ and other TMD materials across diverse technological fields.