Two-Dimensional Crystal Growth of MoS₂ Thin Films from Sodium Dodecyl Sulfate Micellar Solutions and Wettability Between Solution and Substrate

Abstract

In this study, the growth of 2D MoS₂ thin films on SiO₂/Si substrates was investigated using sodium dodecyl sulfate (SDS) micellar solutions, and the effects of SDS concentration and substrate treatment on crystal growth were evaluated. By increasing the SDS concentration, the wettability was improved, and uniform MoS₂ crystal growth was promoted by micellar formation. When the SDS concentration exceeded 10⁻⁴ mol/L, the static contact angle sharply decreased, indicating uniform 2D material growth. The optimal conditions that enabled a uniform supply of Mo-based precursors were as follows: SDS concentration of 3.5 × 10⁻⁴ mol/L; Na₂MoO₄·2H₂O concentration of 1.7 × 10⁻² mol/L. The results indicate that solution-based processes using SDS are effective for 2D material growth, and they may be a valuable technique in future thin film device fabrication processes.

1. Introduction

Transition metal dichalcogenides (e.g., MoS₂, WS₂, etc.), represented by the formula MX₂ (where M is the transition metal, and X is the chalcogen element), are a class of layered materials that exhibit van der Waals heterostructures [1,2,3]. These heterostructures are formed by weak van der Waals forces between layers, allowing for the creation of stacked materials with unique electronic properties. In particular, bulk MoS₂ transitions from an indirect bandgap (1.29 eV) to a direct bandgap (1.8 eV) when reduced to a monolayer [4,5,6]. MoS₂ is favorable as a channel material for transistors because of its resistance to dangling bond formation [7,8]. Dangling bonds are unsatisfied atomic bonds that occur when atoms at the surface are not fully bonded, leading to defects. The ability of MoS₂ to resist such bond formation minimizes surface defects, making it ideal for transistor applications. In addition, as semiconductor devices scale down below 5 nm, 2D MoS₂ exhibits higher effective mobility than silicon, along with a lower tunneling leakage current [9,10]. Consequently, several researchers are exploring methods for producing high-quality MoS₂ thin films.

However, achieving semiconductor-grade MoS₂ thin films on a wafer scale remains challenging. Two-dimensional MoS₂ can be fabricated using different techniques such as mechanical exfoliation, ion intercalation, physical vapor deposition (PVD), chemical vapor deposition (CVD), and thermal annealing of spin-coated films [11,12,13,14,15,16,17,18,19,20,21,22,23]. Among these methods, CVD is commonly used for both graphene preparation and mass production. However, in the CVD synthesis of graphene using segregation or catalytic functions for carbon atoms, transition metal substrates are generally used [24], but in the CVD synthesis of MoS₂, substrates such as SiO₂/Si, sapphire, and even molten glass are usually used [25,26]. However, from the perspective of the practice of following conventional semiconductor processes, growth on SiO₂/Si is industrially important. Furthermore, SiO₂/Si substrates are also commonly used because of the good visibility of 2D-MoS₂/SiO₂/Si under optical microscopes. Solution-based spin-coating followed by thermal annealing has received considerable attention because of its cost-effectiveness and ease of dopant introduction. This method often results in mixed 2D and 3D growth modes, making it difficult to achieve high-quality crystals comparable to those produced via CVD.

Liu et al. synthesized polycrystalline MoS₂ thin films by coating a substrate with a solvent and performing two-step annealing under Ar/H₂ and Ar/Ar+S atmospheres [27]. They emphasized the importance of an appropriate annealing temperature for achieving high crystallinity in solution-processed MoS₂ films. Vapor–liquid–solid technology has been reported to be able to form larger single-crystal MoS₂ monolayers [28]. Unlike silicon and other thin films where covalent bonds are formed between the thin film and the substrate, in MoS₂ monolayers, weak interactions (van der Waals forces) are formed between the sulfur and the substrate. For this reason, the liquid–solid interface, which has high density and polarization, is thought to be very important for 2D crystal formation. On the other hand, this weak interaction can solve many problems related to heteroepitaxy [29,30]. However, this weak interaction also makes 2D growth difficult to control. In other words, the problem is how to control this interaction. Recent research has focused on controlling crystal growth by adding surfactants or other additives to the solution. For example, Zhu et al. achieved the creation of high-quality, large-area MoS₂ monolayers by adding KOH to the solution [31]. In addition, 2D crystal growth can be enhanced by adding potassium or sodium ion [18,25,32,33,34]. Increasing the hydrophilicity of the substrate through wet chemical or UV treatment has also been reported to facilitate 2D crystal growth [25,35]. Esposito et al. also further focused on the role that density gradient mediums played in the sulfurization process, and demonstrated that these density gradient mediums can also affect the sizes, strains, and point-defects of the resulting MoS₂ crystals [36].

Moreover, solution-based crystal growth can be regulated by controlling the surface tension of the solution and the surface energy of the substrate. This principle is similar to the use of surfactants in electroplating to control surface tension. Surfactants such as sodium dodecyl sulfate (SDS), cetyltrimethylammonium bromide (CTAB), and polyvinylpyrrolidone (PVP) are commonly used in electroplating to control surface tension and substrate surface energy [37,38,39,40]. SDS is selected over other surfactants because of its anionic nature, which prevents adsorption onto hydrophilic surfaces such as silanol groups on silicon substrates. In addition, SDS forms micelles, which can enhance wettability and promote uniform crystal growth. However, during high-temperature (>700 °C) sulfurization, the initial state of the solution and substrate surface does not affect crystal growth, and studies in this area are limited. By contrast, in ceramic sintering, the dispersion state of the slurry is critical to the final microstructure [41,42]. Therefore, elucidating the role of additives in promoting 2D MoS₂ growth is necessary.

In this study, the role of additives in promoting 2D MoS₂ growth was investigated by spin-coating SDS-containing solutions commonly used in electroplating onto SiO₂/Si substrates, followed by sulfurization to form 2D MoS₂ films. This study aimed to clarify the relationship between growth and wettability. In the absence of effective solutions for increasing crystal size in CVD and PVD, the liquid-phase process in this study has the potential to increase crystal size, and it is important both academically and industrially to work out the details of controlling the liquid-phase process.

2. Materials and Methods

The experimental procedure is shown in Figure 1. Silicon wafers with a 300 nm thick SiO₂ layer (10 mm², Canosis Co., Ltd., Tokyo, Japan) were used as substrates. The substrates were hydrophilized by vacuum ultraviolet (VUV) irradiation using a xenon excimer lamp (USHIO, SH-172C, wavelength 172 nm, Tokyo, Japan). The substrates were placed in a vacuum chamber and irradiated at 100 Pa for 10 min. For the spin-coating solution, sodium molybdate dihydrate (Na₂MoO₄·2H₂O, purity > 99%, Kanto Chemical, Tokyo, Japan) was mixed with a dispersing agent and ultrasonically stirred for 30 min. Sodium dodecyl sulfate (purity >99.0%, Sigma-Aldrich, Louis, MO, USA) was used as the dispersant. Sodium molybdate concentrations were adjusted to five levels: 8.3 × 10⁻³, 1.7 × 10⁻², 2.5 × 10⁻², 4.1 × 10⁻², and 8.3 × 10⁻² (mol/L). In addition, SDS concentrations were adjusted to five levels: 3.5 × 10⁻⁵, 1.7 × 10⁻⁴, 3.5 × 10⁻⁴, 6.9 × 10⁻⁴, and 3.5 × 10⁻³ (mol/L).

The prepared solution was spin-coated onto the substrate using a spin-coater (MIKASA Co., Ltd., MS-B100, Tokyo, Japan). During spin-coating, the rotation speed was increased to 5000 rpm for 10 s; this speed was maintained for another 10 s and then stopped. After spin-coating, the samples were dried in an oven at 90 °C for 30 min. Then, the spin-coated samples were placed in a tube furnace together with 500 mg of sulfur powder (purity > 99.5%, Kanto Chemical, Tokyo, Japan), and sulfurization was performed at 750 °C in accordance with previous studies [43]. After evacuating the furnace to approximately 20 Pa, argon gas was used to return the pressure to atmospheric levels. Evacuation and repressurization were repeated two times, followed by the introduction of 30 sccm of argon as a carrier gas. Sulfur (orthorhombic) has a melting point of 113 °C and a boiling point of 445 °C. The sulfur source was controlled at about 170 °C and was in the liquid phase. Therefore, sulfur gas was supplied at a vapor pressure below the saturated vapor pressure at 170 °C. On the other hand, the melting point of Na₂MoO₄ is 687 °C, and it gradually changes phase with the liquid on the substrate. Therefore, Na₂MoO₄ (l) was distributed on the substrate. In this sulfurization process, it can be predicted that the heterogeneous reaction between sulfur gas and this liquid will reduce Na₂MoO₄ to produce MoS₂ and sodium oxide (Na₂O), and the resulting Na₂O reacts with the silica surface (SiO₂) to form silicate (Na₂SiO₃) [44]. These reactions are as follows:

$$\begin{gathered} \mathrm{S}\left(\mathrm{s}\right)\stackrel{>113 °\mathrm{C}}{\to }\mathrm{S}(\mathrm{l}) \\ \mathrm{S}\left(\mathrm{l}\right)\stackrel{170 °\mathrm{C}}{\iff }{\mathrm{S}}_2(\mathrm{g}) \\ {\mathrm{N}\mathrm{a}}_2\mathrm{M}\mathrm{o}{\mathrm{O}}_4(\mathrm{s})\to {\mathrm{N}\mathrm{a}}_2\mathrm{M}\mathrm{o}{\mathrm{O}}_4(\mathrm{l}) \\ {\mathrm{S}}_2\left(\mathrm{g}\right)+{\mathrm{N}\mathrm{a}}_2\mathrm{M}\mathrm{o}{\mathrm{O}}_4\left(\mathrm{s}\right)\stackrel{>684 °\mathrm{C}}{\to }\mathrm{M}\mathrm{o}{\mathrm{S}}_2\left(\mathrm{s}\right)+{\mathrm{N}\mathrm{a}}_2\mathrm{O}\left(\mathrm{s}\right)+\mathrm{S}{\mathrm{O}}_3(\mathrm{g}) \\ {\mathrm{N}\mathrm{a}}_2\mathrm{O}\left(\mathrm{s}\right)+{\mathrm{S}\mathrm{i}\mathrm{O}}_2\left(\mathrm{s}\right)(\mathrm{s}\mathrm{u}\mathrm{r}\mathrm{f}\mathrm{a}\mathrm{c}\mathrm{e})\to {\mathrm{N}\mathrm{a}}_2\mathrm{S}\mathrm{i}{\mathrm{O}}_3\left(\mathrm{s}\right) \end{gathered}$$

The morphology of the samples was analyzed using a field-emission scanning electron microscope (S-4800, Hitachi, Tokyo, Japan) with an accelerating voltage of 15.0 kV and a working distance of 8.8–8.9 mm. The chemical bonding states of the samples were analyzed by using a Raman microscope (Renishaw inVia, England, UK). Raman analysis was performed using a 532 nm laser (with a maximum power of 150 mW, Samba, Cobolt, Sweden). Light microscopic images were obtained using a light microscope integrated with the Raman system. The static contact angle was measured using a digital optical camera and the level of substrate stage was adjusted and confirmed by a level gauge. The drop volume was 10 μL and the contact angle was calculated using spherical approximation.

3. Results and Discussion

Figure 2 shows the molecular structure of SDS, highlighting the hydrophobic tail and the hydrophilic anionic head group. The anionic head forms the hydrophilic part of the molecule responsible for its interaction with the solution.

Figure 3 shows the static contact angles of aqueous SDS solutions under three conditions: (a) on substrates without VUV treatment, (b) on substrates with VUV treatment, and (c) on VUV-treated substrates with SDS/Na₂MoO₄·2H₂O solution. The original static contact angles photos can be found in Figure S1. In addition, the contact angle obtained here can be ignored because the surface roughness of the silicon wafer used in this experiment is in the nanometer range. The pink band shown in Figure 3 represents the critical micelle concentration (CMC) of SDS in the solution [45,46]. The CMC refers to the concentration at which surfactant molecules begin to aggregate and form micelles in the solution. As shown in Figure 3a, the static contact angle decreased from about 45° to 30° with the increase in SDS concentration, indicating improved wettability caused by the addition of SDS, although the wettability was not significantly high.

Then, the static contact angle was measured on the VUV-treated substrates. The VUV treatment removed hydrocarbon contamination from the substrate surface and made it hydrophilic by generating silanol groups [47]. Consequently, the static contact angle decreased sharply as the SDS concentration exceeded 10⁻⁴ mol/L and approached zero at higher concentrations. This sharp decrease in the static contact angle was due to an increase in surface hydrophilicity, resulting from changes in surface energy after VUV treatment, which made the substrate more water attractive. Therefore, 10⁻⁴ mol/L is the CMC for this system.

At low SDS concentrations, an SDS monolayer is formed at the air–liquid interface, which reduces the static contact angle. At high SDS concentrations, micelles are formed in the solution, thereby reducing the static contact angle and surface tension. Micellar formation at concentrations below the typical CMC is likely due to electrostatic repulsion between the anionic silanol groups on the surface and the anionic SDS head groups [48,49,50,51,52]. Micellar formation in a solution is energetically more favorable than surface adsorption. Next, the effect of adding Na₂MoO₄·2H₂O as the MoS₂ precursor on wettability was investigated. The results indicated that wettability was primarily determined by the SDS concentration, regardless of the presence of Na₂MoO₄·2H₂O. Therefore, SDS micelles facilitate the uniform distribution of MoO₄²⁻ ions throughout the solution, leading to homogeneous growth over the entire surface. The micelles, with their hydrophilic heads, encapsulate and transport the MoO₄²⁻ ions, allowing them to be evenly dispersed in the solution and uniformly delivered to the substrate. If this hypothesis is correct, then the crystal growth of MoS₂ would be more uniform and two dimensional at SDS concentrations above the CMC.

Figure 4 shows the SEM images of MoS₂ crystals at a fixed Na₂MoO₄·2H₂O concentration of 1.7 × 10⁻² mol/L and varying SDS concentrations, as well as at a fixed SDS concentration of 3.5 × 10⁻⁴ mol/L and varying Na₂MoO₄·2H₂O concentrations. As shown in Figure 4, the uniform coverage of crystalline MoS₂ was observed even at the scale of several millimeters under all concentration conditions. In Figure 4(1), the size of the MoS₂ crystals exceeded 200 μm at an SDS concentration of 1.7 × 10⁻⁴ mol/L, and 2D preferential growth was observed. In addition, the MoS₂ coverage rate in the low-magnification SEM images (scale bar = 1 mm) was calculated using ImageJ software, and the results are shown in

Table 1. MoS₂ coverages on SiO₂/Si calculated from Figure 4(1): a fixed Na₂MoO₄·2H₂O concentration of 1.7 × 10⁻² mol/L and varying SDS concentrations.

SDS Concentration (mol/L)	DI Water	3.5 × 10−5	1.7 × 10−4	3.5 × 10−4	6.9 × 10−4	3.5 × 10−3
Coverage	40.5%	49.9%	82.1%	85.1%	92.8%	76.9%

Coverages were obtained by ImageJ software (U. S. National Institutes of Health, Bethesda, MD, USA) (version 1.53t).

and

Table 2. MoS₂ coverages on SiO₂/Si calculated from Figure 4(2): a fixed SDS concentration of 3.5 × 10⁻⁴ mol/L and varying Na₂MoO₄·2H₂O concentrations.

Na2MoO4·2H2O Concentration (mol/L)	DI Water	8.3 × 10−3	1.7 × 10−2	2.5 × 10−2	4.1 × 10−2	8.3 × 10−2
Coverage	40.5%	24.0%	85.1%	98.6%	88.4%	95.3%

Coverages were obtained by ImageJ software.

. As a result, the millimeter scale coverage rate was only 40.5% for the sample prepared with DI water, but the sample with added SDS showed a higher coverage rate than the sample without added SDS. This high coverage rate is extremely important for potential applications in the semiconductor industry. When the SDS concentration exceeded the CMC (3.5 × 10⁻⁴ mol/L), the crystal coverage increased with the increase in Na₂MoO₄·2H₂O concentration, indicating that MoO₄²⁻ ions were uniformly supplied to the entire substrate. In addition, even at the low coverage (24.0%) of the sample in Figure 4(2b), the 2D MoS₂ crystals were uniformly distributed on the substrate. From the above, it is clear that the increase in wettability caused by SDS makes a significant contribution to uniformity.

Figure 5 shows the Raman spectroscopy results for different SDS (1) and Na₂MoO₄·2H₂O (2) concentrations. The Raman spectra were measured at the center of each optical microscopic image. The typical Raman modes of MoS₂, namely A_1g and E¹_2g, were measured, and the difference in Raman shift between the two modes was used to estimate the number of layers.

In the optical microscopic images, the bright regions (B1) were identified as bulk MoS₂ based on the Raman spectra. The relatively dark areas (D1) corresponded to MoS₂ with 2–4 layers, whereas the darker areas (D2) were identified as monolayer MoS₂ [53,54]. The brightest areas (D3) indicated the bare SiO₂ surface without MoS₂ layers. For clarity, the SiO₂ surface was marked with a purple mask in the optical images. The purple mask was applied to the area (D1) to make it easier to distinguish.

When the SDS concentration was zero (Figure 5(1)), large areas of bulk MoS₂ were observed. At an SDS concentration of 3.5 × 10⁻⁵ mol/L, the formation of bulk MoS₂ was suppressed, although the crystal size did not increase. When the SDS concentration was further increased to 3.5 × 10⁻⁴ mol/L, the entire substrate was covered with MoS₂, although the number of layers also increased, as indicated by the growth in regions (D1) and (B1).

When the Na₂MoO₄·2H₂O concentration was varied (Figure 5(2)), MoS₂ crystals covered the entire substrate at 1.7 × 10⁻² mol/L. At lower concentrations, the areas of exposed SiO₂ were prominent, whereas, at higher Na₂MoO₄·2H₂O concentrations, the size of regions (B1) and (D1) increased. This finding indicates that the balance between SDS and Na₂MoO₄·2H₂O concentrations determines not only the uniformity of MoO₄²⁻ ion supply, but also the uniformity of MoS₂ crystal growth.

Based on these results, the optimal conditions for the uniform supply of Mo-based precursors throughout the substrate were as follows: an SDS concentration of 3.5 × 10⁻⁴ mol/L and a Na₂MoO₄·2H₂O concentration of 1.7 × 10⁻² mol/L. This phenomenon may be due to the surface charges of silanol groups, the charge of SDS, and the formation of micelles (Figure 6). However, the range of MoO₄²⁻ ion concentrations suitable for uniform growth is limited. Future work should focus on optimizing the surfactant by introducing branched structures to the hydrophobic tail of SDS, thereby increasing the solvent exclusion volume of the micelle. The potential application of micelle-based solvents in controlling MoS₂ crystal growth also warrants further investigation.

4. Conclusions

In this study, the growth of 2D MoS₂ thin films was investigated using SDS micellar solutions, and the effects of SDS concentration and substrate surface treatment on crystal growth were evaluated. We confirmed that SDS improved wettability and promoted MoS₂ crystal growth through micellar formation in a solution. When the SDS concentration exceeded 10⁻⁴ mol/L, the static contact angle decreased sharply, enabling the uniform 2D growth of MoS₂ on the substrate. In addition, by optimizing the concentration of Na₂MoO₄·2H₂O, MoO₄²⁻ ions could be supplied uniformly over the surface, allowing for controlled crystal size.

The optimal growth conditions were as follows: SDS concentration of 3.5 × 10⁻⁴ mol/L and Na₂MoO₄·2H₂O concentration of 1.7 × 10⁻² mol/L. These results indicate that using SDS is effective for low-cost and efficient growth of 2D materials, making it a promising approach for future thin film fabrication. The results of this study show that general additives such as KOH can preferentially induce a 2D growth mode due to their anion-wetting-promoting effect at high temperatures, but that SDS can control both the wettability of the solution and substrate at room temperature and the anion-wetting-promoting effect at high temperatures. In other words, the simultaneous control of the substrate coating process and the crystallization process in the solution method will be the basis for producing even higher quality MoS₂ crystals in the future. Moreover, further optimization of material synthesis based on micellar formation could promote advanced control of crystal growth.