Applied Science and Convergence Technology 2018; 27(6): 149-152
Published online November 30, 2018
Copyright © The Korean Vacuum Society.
Bong Ho Kim , Jin Hee Han , Soon Hyeong Kwon , and Young Joon Yoon*
Nano-Convergence Materials Center, Korea Institute of Ceramic Engineering and Technology (KICET), Gyeongsangnam-do 52851, Republic of Korea
Correspondence to:*Corresponding author: E-mail: email@example.com
We demonstrate the preparation of molybdenum disulfide (MoS2) nanopetals by RF magnetron sputtering and electron-beam irradiation (EBI). To investigate the structure and the surface morphology of MoS2 films, atomic force microscopy (AFM), Raman spectroscopy, and scanning electron microscopy (SEM) were carried out. AFM results reveal that the surface roughness of MoS2 increases with increase in deposition. After EBI process, the surface roughness decreases except for MoS2 deposited for 5 min. In addition, EBI process increases the crystallinity of MoS2 films, which is confirmed by Raman spectroscopy. The increase of crystallinity and the change of surface morphology are induced by inelastic scattering, which is the interaction between incident electrons and target material. SEM images display nanopetal structure of EBI-treated MoS2 and we expect EBI-treated MoS2 to be a potential candidate for applications such as gas sensors and hydrogen evolution reactions due to its vertically exposed edge sites.
Keywords: 2D material, Molybdenum disulfide, Nanopetal, Electron beam irradiation, Sputtering
A single-layer graphene presented the advantages of two-dimensional (2D) materials , however, the absence of bandgap in graphene restricts its application in electronic and optoelectronic devices. Recently, other 2D materials with a bandgap, such as transition metal dichalcogenide (TMD), generated interest as an analogous structure of graphene. Molybdenum disulfide (MoS2), which is the most researched TMD, has direct bandgap of 1.8 eV  in single-layer and indirect bandgap of 1.2 eV in bulk . Furthermore, MoS2 exhibits a high carrier mobility of ~200 cm2V−1s−1 in top-gate field-effect transistor (FET) . These unique and outstanding optical and electrical properties make MoS2 a promising material for high performance of electronic and optoelectronic devices.
To utilize the exceptional properties of MoS2 films, mechanical exfoliation, which is the most familiar method previously used for obtaining graphene, was carried out . However, mechanical exfoliation has several disadvantages such as small area, low uniformity, low yield, and difficult thickness control of the film. To overcome this challenge, chemical vapor deposition  and physical vapor deposition  were investigated. Until now, using these methods to single-layer and few-layer (less than 7 atomic layers) of MoS2 films were obtained. However, obtaining multilayer (more than 7 atomic layers) of MoS2 was rarely attempted. Most of the preparation and device application of MoS2 were focused on ultra-thin films. However, a few previous reports of multilayer MoS2 have suggested the potential for its device applications [7–9] since multilayer structure of MoS2 is insensitive to the extrinsic factors  and enhances the current flow of FET .
In this work, we report the preparation of MoS2 nanopetals prepared by RF magnetron sputtering and electron-beam irradiation (EBI) process. EBI process was introduced as an effective method of MoS2 preparation . However, since earlier reports only explored few-layer MoS2 films, various thicknesses of MoS2 films were prepared and the surface morphology change and Raman spectroscopy with EBI process were investigated in this work. For EBI-treated MoS2 films, we observed an increase in crystallinity and decrease in surface roughness when compared to as-deposited MoS2 films. We believe that our results will contribute to broadening the understanding of MoS2 film preparation.
RF magnetron sputtering system (Infovion Inc.) was used for synthesizing MoS2 films of different thicknesses. The purity and diameter of MoS2 target were 99% and 50.8 mm, respectively. The sputtering parameters were as follows: RF power of 20W, Ar flow rate of 10 sccm, working pressure of 5 mTorr, and a working distance of 135 mm. The sputtering time was varied from 5 to 120 min. There was no substrate heating. The SiO2 (100 nm)/Si wafer was used as the substrate and was cleaned by ethanol, acetone, and isopropyl alcohol prior to MoS2 sputtering. After deposition of MoS2, EBI process, the equipment for which was housed in the same chamber as RF magnetron sputtering system, was conducted. The EBI parameters were as follows: RF power of 300 W, DC power of 1000 V, Ar flow rate of 10 sccm, working pressure of 0.8 mTorr, and a working distance of 75 mm. The process time was 1 min. There was no additional heating and the substrate temperature heated of about 100°C was a result of 1 min of EBI.
To determine the thickness of MoS2 films, atomic force microscopy (AFM; WITec, Alpha 300 S) was performed. In addition, the surface roughness of various thicknesses of MoS2 was measured and compared before and after EBI process. Raman spectroscopy (WITec, Alpha 300 S) was carried out to investigate the structural differences between a given film, before and after EBI process, for different thicknesses of MoS2. Laser wavelength with 532 nm and a grating with 1800 grooves × mm−1 were used. Scanning electron microscopy (SEM; Jeol, JSM-6700F) was performed to investigate the change of surface morphology of MoS2.
Figure 1(a) shows the height profile for MoS2 films, obtained by AFM. The thicknesses of MoS2 films deposited for 5, 30, 60, and 120 min were 4.5, 7.5, 22, and 34 nm, respectively. Figure 1(b) shows a correlation of the thickness to the deposition time. The trend line for deposition rate is linear and has a coefficient of determination (R2) as 0.9679, which indicates a high reliability of the trend.
Raman spectroscopy is an effective method for determining the crystallinity of MoS2 because crystalline MoS2 exhibits two Raman peaks of E12g (~380 cm−1) and A1g (~408 cm−1) while amorphous MoS2 exhibits no peaks. Figure 2 shows the Raman spectra of MoS2 films of different thicknesses before and after the EBI process. DC power and process time of EBI process were 1000 V and 1 min, respectively. After EBI process, the intensities of the two typical Raman peaks increased regardless of the thickness, which indicates the increase of crystallinity. However, the appearance of Raman peaks was observed for the as-deposited MoS2 films even before the EBI process. With increase in deposition time, the initially deposited MoS2 became the seed layer for MoS2 deposited thereafter. As a result, MoS2 films with longer deposition time (60 min and 120 min) could grow as crystalline structures while MoS2 films with shorter deposition time (5 min and 30 min) grew as amorphous structure.
Figures 3(a)–(d) shows AFM images for the surface morphology and roughness value of the as-deposited MoS2 films for different thicknesses. Those of EBI-treated MoS2 films are depicted in Figs. 3(e)–(h). The measured area was 5 × 5 μm2. The surface roughness of the as-deposited MoS2 films deposited for 5 min was 0.565 nm, which indicates that the surface morphology is very flat. It was similar to the surface roughness of the SiO2/Si substrate of 0.524 nm. However, the surface roughness of MoS2 films increased as the thickness of MoS2 films increased, and as-deposited MoS2 deposited for 120 min showed the surface roughness of 1.506 nm. Interestingly, the surface roughness of MoS2 increased after EBI process only in case of MoS2 deposited for 5 min. Other samples showed a decrease in the surface roughness after the EBI process. The surface roughness change is due to the interaction between incident electrons of EBI and the target material of MoS2. In this experiment, owing to the large mass difference between atomic nuclei and electron and energy of electrons as low as 1000 V, inelastic scattering is more dominant than elastic scattering . Inelastic scattering induces local excitation, bond breaking, and atomic rearrangement. A sequence of processes transforms amorphous as-deposited MoS2 into crystalline MoS2 and changes the surface roughness.
To verify the surface morphology change of EBI-treated MoS2 films, SEM was performed. Figure 4 shows the SEM images of as-deposited and EBI-treated MoS2 films, which were deposited for 120 min. The scale bar is 100 nm. Both as-deposited and EBI-treated MoS2 displayed nanopetal shape similar to other sputtered MoS2 films [14, 15]. The nanopetal structure of as-deposited MoS2 was densely compacted, however, after EBI process, the density of nanopetal structure decreased because of the sharper morphology of nanopetals. As illustrated by Figs. 4(c) and 4(d), EBI-treated MoS2 grew uniformly with exposed edge sites. Since these vertically exposed edge sites of MoS2 play an important role in improving catalytic activity and sensing property , we expected EBI-treated MoS2 to be a potential candidate for gas sensor and hydrogen evolution reaction (HER). Furthermore, because morphology of MoS2 is significantly affected by fabrication parameters [14,15], optimized sputtering and EBI parameters for more dense and uniform morphology would be required to application of EBI-treated MoS2 in gas sensor and HER.
The MoS2 films were synthesized by RF magnetron sputtering and EBI process. Deposition rate of MoS2 was derived and the effect of EBI on MoS2 films was investigated by Raman spectroscopy, AFM, and SEM. After undergoing EBI process, MoS2 films exhibited higher intensities of Raman peaks, thus indicating higher crystallinity. For MoS2 for longer deposition time (60 and 120 min), even the as-deposited MoS2 exhibited two typical Raman peaks while MoS2 with shorter deposition time (5 and 30 min) exhibited none. Surface roughness of MoS2 films increased with increasing deposition time. EBI-treated MoS2 films showed decreased surface roughness except for MoS2 film deposited for 5 min. SEM images displayed sharp nanopetal structure of EBI-treated MoS2 films, which is expected to be a potential candidate for sensing and catalytic applications.
This research was supported by the Basic Science Research Program of the National Research Foundation (NRF) of Korea funded by the Ministry of Science and ICT (Grant number; NRF-2017R1D1A1B03032923).