Growth of MoS₂ films: High-quality monolayered and multilayered material Open Access

Two-dimensional (2D) transition metal dichalcogenides (TMDs) have been widely investigated in diverse contexts due to their novel characteristics as one-layer-thick materials that undergo a quantum confinement effect. The confinement leads to a tunable band structure, abundant in applicable phenomena, such as high electron mobility, superconductivity, photoconductivity, and photothermal effects.^1–5 Such materials have demonstrated superior potential for use in future rapid, flexible, transparent, and effective nanodevices. Furthermore, recent studies have indicated a wide range of bioapplications for 2D materials, such as cancer and coronavirus detection.^6–10 The 2D preparation methodology is modified according to each study’s objective. Most researchers have utilized chemical vapor deposition (CVD) techniques to grow high-quality and large-area samples on different substrates.^11–14 CVD yields excellent results in terms of the size and quality of growth, but achieving a large homogeneous area with large grains remains a challenge.

Extensive efforts have been devoted to developing strategies for synthesizing high-quality large-area uniform films at low cost. The salt-assisted CVD method has recently emerged as an effective method, advancing the progression of TMDs.^15–17 The use of alkali salts, such as NaCl and KCl, has influenced the synthesis of some TMDs that are hard to grow due to their high melting points and low vapor pressure. Introducing molten salts reduces the melting point of the metal precursor, forming intermediate products, such as metal oxychlorides, which increase the mass flux and growth rate, facilitating the generation of a single crystal up to a millimeter in size^15–17 as well as films that are centimeters in area.^18,19 The low cost and high efficiency of salt-assisted growth have encouraged researchers to develop various methods for growing 2D materials to use in the generation of wafer-scale, regular, and clean monolayer films.^15,16,20,21 Despite these achievements, however, the success of homogeneous, scalable growth on diverse substrates remains limited.

In the current study, NaCl-assisted CVD growth was used to grow MoS₂ on different substrates [Si/SiO₂ and fluorine-doped tin oxide coated glass (FTO)], and the growth temperature was adjusted to enhance growth on each substrate. The NaCl to MoO₃ ratio used was calculated based on the equation of the possible reactions between NaCl and MoO₃, which in conjunction with adjustment of other growth parameters yielded good results in terms of film growth.

A CVD quartz-tube furnace was used to grow MoS₂ films, and MoO₃ and sulfur were used as precursors. NaCl was added to MoO₃ to reduce its melting point and enhance growth. An aqueous suspension of NaCl mixed with MoO₃ at a ratio of 40% was prepared in deionized water and sonicated for 45 min in an ultrasonic bath to render it more homogeneous [Fig. 1(a)]. 250 μl of the as-prepared 8:20 mg in 1 ml water was dropped onto a quartz boat and heated to 50 °C to gradually evaporate the water [Fig. 1(b)]. This method ensures that good mixing and distribution of the precursors occurs under the substrate. The boat was placed in the center of a 1.5-in. tube furnace, 21 cm away from 60 mg sulfur. The Si/SiO₂ substrate (300 nm thermal SiO₂ prime grade, purchased from University Wafer, orientation 〈100〉) was positioned face down on a 1-cm-high boat [Fig. 1(c)]. The tube was first purged with a high N₂ flow rate to generate a cleaner environment; then, it was reduced to 5 SCCM to preserve the sublimated vapor until it reached the growth temperature. The furnace temperature was ramped to 780 °C at a rate of 15 °C/min. Sulfur was heated by an external heater outside the furnace that ramped to 250 °C at a rate of 10 °C/s, which was a sufficient temperature to evaporate the sulfur under these experimental conditions and resulted in a reproducible film [Fig. 1(c)]. The sulfur was introduced using an N₂ flow of 20 SCCM, which was sufficient in the short tube and open-side boat. The growth times used ranged from 5 to 10 min, depending on the growth conditions and the required film thickness. After the growth period, the furnace was left to cool down to room temperature.

MoS₂ films were synthesized via the above-described method on FTO-coated glass at 700 °C to prevent deformation of the substrate, which has a low melting point. To adjust for the new growth conditions at low temperature, the growth time was increased to 15 min, and the amount of aqueous suspension was doubled (500 µl of 8:20 mg NaCl:MoO₃ in 1 ml water).

The size, shape, quality, and yield of monolayers grown depend on all growth factors, namely, temperature, NaCl:MoO₃ ratio, the amount of oxide and chalcogen precursors, carrier gas flow, substrate position relative to the oxide, boat geometry, and growth time. Diverse NaCl:MoO₃ ratios have been used to grow wide-area films in previous studies. Some studies have used small quantities of salt, just a few micrograms,^22,23 which resulted in a film of several hundred micrometers length or less. In contrast, others have used ratios from 2% to 43%,^15,16,24,25 or very high ratios of NaCl:MoO₃—up to 900%^19,26 (Table I).

It is essential to achieve even mixing and spreading of precursors under the substrate to grow films larger than 1 cm². The schematic diagram in Fig. 1 illustrates solution preparation and the distribution of the solution on the boat. The first parameter that needs to be adjusted is the NaCl:MoO₃ ratio to provide sufficient vapor pressure to yield high nucleation density and promote film growth.^15,16 In the current study, a 40% ratio was calculated based on the following reaction:¹⁵ 

The amount of 40% aqua solution used needs to be tested with respect to the boat’s size and geometry. In the present study, an open-side boat of 4 cm length × 1 cm width × 1 cm height containing a 2-mg NaCl:5-mg MoO₃ solution yielded the growth of a 1 × 2 cm² monolayer film with high crystalline quality (Fig. 2). Increasing the NaCl:MoO₃ ratio to 40% enhances nucleation density, resulting in a film with a wide area after adjusting the rest of the growth parameters. Some studies investigating the use of a salt-assisted growth mechanism whereby increasing salt weight lowered the growth temperature, with evidence of some Na compounds under the 2D crystalline material, are summarized in Table I.^19,26 Due to certain water-soluble Na compounds (Na₂S or Na₂SO₄), the 2D film can be easily transferred to a new substrate without the use of specialized techniques because as Zhang et al.²⁶ demonstrated, deionized water can dissolve the compounds under the 2D film. As a result, the optical and electrical properties of the non-transferred 2D crystal were degraded and affected by the interlayer, which made the transformation a requirement to preserve the quality of the material.^19,26 In the current study, Raman, Photoluminescence (PL), and X-ray Diffraction (XRD) signals indicated the growth of high crystalline monolayer films with no indication of any extra Cl or Na compounds under the 2D film (Figs. 2–4), which is concordant with previous studies that used a small quantity of salt.^{15–17,20,25} However, x-ray photoelectron spectroscopy (XPS) data [Figs. 4(c) and 4(d)] have one weak signal for Na, which is most likely in the form of dopant and not separate compounds (Fig. S3 and Tables S1 and S2).

Figure 2(c) shows the Raman spectra of a monolayer as determined via a 442-nm laser. The two Raman modes E¹_2g and A_1g, with a separation of 18 cm⁻¹, are considered a fingerprint for monolayer growth. The E¹_2g Raman signal has a full width at half maximum (FWHM) of 3.93 cm⁻¹, indicating high crystalline quality.^27–30 The size and orientation of the crystalline grains affect the intensity of the in-plane vibration E¹_2g, but not the intensity of out-of-plane vibration mode A_1g.^27,31 As a result, the ratio between intensities of the two modes E¹_2g/A_1g, larger than one, indicates the high quality of the crystalline structure in terms of grain size and orientation. All the samples grown exhibited an intensity ratio greater than 1.0 (2.5 < E/A < 3.0) and a sharp, intense PL signal [Fig. 2(d)], confirming high crystalline quality, large grain size, and high film orientation.

Increasing the precursor weight, decreasing boat dimensions, or lowering the substrate height with respect to the boat results in non-homogeneous growth, bulk seeds, and/or multilayer films with thick grain boundaries (Fig. S1). Increasing the vapor pressure of precursors on the substrate by the excessive weight of precursors (NaCl–MoO₃ solution or sulfur) could suppress the growth rate, resulting in spots of non-crystalline material [Figs. S1(a) and S1(b)]. Adjusting sulfurization by optimizing the heating temperature leads to a controllable feeding rate and nucleation density, facilitating a more uniform and reproducible film with fewer seeds. Figure S2 shows that decreasing the heating temperature for sulfur leads to a lower percentage of seeding area. Growth on a glass substrate coated by FTO was achieved at 700 °C. Figures 3(a) and 3(b) show the prevalent multilayer material on the sample. Increasing the amount of material leads to multilayer film growth [Fig. 3(c)]. Figure 3(d) shows the Raman signal of MoS₂ grown on FTO. The increase in separations between the two Raman modes from 21.7 to 24.9 cm⁻¹ indicates an increase in the number of layers. XPS data for the growth on FTO indicates the presence of some Mo oxide compounds related to the peak at 234.5 eV in Fig. S3(d).

It is not expected that growth will be analogous on both substrates, Si/SiO₂ and FTO, due to FTO surface roughness compared to prime grade Si/SiO₂. The higher FWHM of the E¹_2g Raman signal (6.7 cm⁻¹) and the lower E¹_2g/A_1g intensity ratio (0.75) are concordant with this expectation. XRD patterns of MoS₂ on the Si/SiO₂ substrate are shown in Fig. 4(a), and XRD patterns of MoS₂ on the FTO substrate are shown in Fig. 4(b). XRD uses a Cu target for x rays, and all XRD patterns were acquired in the range of 2θ from 10° to 100°. For MoS₂ on Si/SiO₂, for 2θ larger than 65°, only Si-related peaks were observed; therefore, the range of our interest is shown in Fig. 4(a) from 10° to 65°. Two peaks related to MoS₂, 002 at 14.64° and 100 at 31.93°, are produced by the Cu x ray. One peak of 400 related to Si is produced by the Cu x ray. Similarly, the XRD pattern of MoS₂ on FTO in Fig. 4(b) shows the 002 peak of MoS₂ at 14.88°, and all the rest of the peaks in the figure are related to the FTO substrate.

Neither sample yielded any peak associated with any impurity, such as Na or related compounds, which could be due to two main reasons: one is that there is no significant concentration of impurity in a crystalline phase that produces a peak in the XRD pattern, and the other is that the impurity and related compounds are in the amorphous phase. To investigate these two possibilities, XPS analysis of the samples was performed. The XPS system is equipped with an Al-x-ray source and operates at an ultra-high vacuum with a base pressure of 3.0 × 10⁻¹⁰ Torr. Advantage software was used to analyze the data and for peak fitting. XPS patterns of MoS₂ on Si/SiO₂ and MoS₂ on FTO substrates are shown in Figs. 4(c) and 4(d), respectively. Two spectra are shown for each sample: one recorded before etching the sample with Ar plasma and another recorded after. The samples were exposed to air, so the Ar etching would remove contaminants from the surface layer. Ar etching has been done with the ion beam energy of 3 keV for the time of exposure of 30 s. The C peak at ∼287 eV in the spectra is due to organic compounds, which disappear in the spectra taken after etching. In Fig. 4(c), peaks related to Mo, S, Si, and O and small peaks related to Na are present. Similarly, in Fig. 4(d), peaks related to Mo, S, Sn, and O and a small convoluted peak at 1070 eV related to Na are present. The Mo and S peaks are produced by the sample, and the Si, Sn, and O peaks are produced by the substrate. As evident from the intensity of the peaks, the trace amount of Na could be dopant in the sample because the Na signal is produced by the 1s⁺¹ state. With the absence of a Cl peak, which would confirm the presence of NaCl, the Na⁺¹ could bond with O or S to form Na₂O or Na₂S. The absence of peaks for Na₂O and Na₂S in Raman spectroscopy indicates that Na exists as a dopant in the MoS₂ sample. High-resolution spectra of Mo, S, and Na-related peaks are provided in Fig. S3 in the supplementary material. The area under the peak is proportional to the atomic concentration in the sample, so the areas of all peaks are provided in Tables S1 and S2 for MoS₂ on Si/SiO₂ and MoS₂ on FTO, respectively.

Using salt to grow monolayered or multilayered MoS₂ enhances the growth size of films on different substrates via an immediate and direct approach. Despite the presence of Na—most likely as a dopant—the film’s premium characteristics are retained. The homogeneity and scale of the material are also optimal. This method could be further tested and improved for additional growth applications and purposes related to size, substrate, quality, or doping.

See supplementary material for extra data, images, and tables.

The authors extend their appreciation to the Deputyship for Research and Innovation, Ministry of Education in Saudi Arabia, for funding this research (Project Number 776).

The authors have no conflicts to disclose.

The data that support the findings of this study are available within the article and its supplementary material.