1 Introduction

Structural uniqueness of layered crystalline materials have enabled researchers to experience the two dimensional materials (2D). 2D nanomaterials consist of 2D layers bounded by vander wal forces. It is interesting here to note that number of layers plays a significant role in determining the properties of 2D materials [1,2,3,4,5]. It has been reported in the literature that few-layer 2D materials exhibit different properties from their bulk counterpart. Due to immense potential of 2D materials, a growing number of researchers have turned their focus to explore this new class of materials [5, 6]. Graphene is found to be most prominent 2D material from the past few decades with high electron mobility but due to zero energy bandgap, it is not suitable for semiconductor device fabrication. Another 2D material Hexagonal Boron nitride (h-BN) shows a large bandgap of 6.5 eV due to which it behaves as an insulator [7,8,9,10]. Further, in this context, a new class of 2D materials known as Transition metal dichalcogenide (TMD), (MX2) (M = Mo, W; X = S, Se, Te) shows a bandgap that lies between graphene and h-BN. Further, material belonging to this category exhibits excellent optical and physical properties, like tuneable bandgap from direct to indirect, high mobility, absence of dangling bonds and spin orbit coupling [10]. Among various TMD materials, MoS2 is the most researched material due to its extraordinary properties such as a catalysis, lubricant industry and electrochemical energy storage. It is noteworthy here that few layers of MoS2 offer attractive properties such as considerable semiconducting band gap of 1.2–1.8 eV, optical properties in visible region electromagnetic spectrum and sufficient surface charge density, making it suitable for fabricating flexible, transparent, low power optical and electronics devices [11,12,13].

According to the relevant literatures, the prevailing research on few-layer MoS2 primarily relies on the exfoliation method [14], thermal evaporation method [15, 16] or chemical vapour deposition method [17, 18]. Aforementioned techniques provide poor area coverage and difficulty in thickness control [14,15,16,17,18]. Consequently, deposition of film on a large area is still a major obstacle that further limits its use in fabrication of optoelectronics devices. In context to this, RF sputtering method is utilized for large area, uniform, high quality crystalline MoS2 thin films that find application in commercial coating, flat panel display and integrated circuits. This method provides a good layer by layer control over thickness of the film which is very difficult to achieve in gas phase CVD technique, also there is no strict thickness requirement of Mo metal as demanded in Mo sulfurization method. Additionally, since sputtering process is a physical vapour deposition method, which is not bounded by process and precursor preparation as in sol–gel technique, thus it is a promising method for depositing 2D TMDs by target. Samartsev et al. showed growth of multilayer film on silicon and glass substrate via magnetron sputtering method. They observed a well-defined grain size of 40 µm diameter using optical microscopy. However, grain size was not promising enough to fall in nano scale domain [19]. Further, Ramos et al. analysed mechanical and structural characteristic of few-layer MoS2 grown by RF sputtering. They concluded that the deposited film was highly porous with outstanding flexibility [20]. Several literatures have reported the successful deposition of MoS2 thin film by RF sputtering process [6,7,8, 13, 21] but the main challenge appeared in controlling the exact metrics of the film. The main obstacle for large-area sputtered high quality MoS2 films are possibly located in the difficulty of disorder control during thin film deposition that can possibly be controlled by optimizing process parameters. The process parameters uphold their distinct advantages that may have positive effects on film growth. Therefore, in the present work we have used distinct sputtering times to analyse their effect on few-layer MoS2 properties. In the present paper, few-layers of MoS2 were deposited on ITO coated glass, n-Si and p-Si substrate for different growth times of 5, 10, 15 and 20 min using RF sputtering method. To the best of our knowledge, for the first time, this study reports a comparative analysis of few-layer MoS2 deposited by using RF sputtering of a MoS2 target. The structural and optical properties have been discussed with reference to sputtering time that is a critical process parameter affecting the film quality. The present analysis prevails the potential of RF sputtering process to be the general method for depositing 2D MoS2 that can effectively be used for fabrication of a wide variety of electronic and optoelectronic devices.

2 Experimental details

MoS2 thin films were deposited on p-type silicon < 100 >, n-type silicon < 100 > and ITO (indium tin oxide) coated glass using RF sputtering technique (equipment model Hind HIVAC). The RF sputtering target was prepared by pure (99.9% Sigma Aldrich) MoS2 powder. At first, a few drops of Poly vinyl alcohol (PVC) and 30 g MoS2 powder was mixed by ball milling machine at 100 rpm for 3 h. Then the target was formed by applying hydraulic pressure of 20 kN on MoS2 powder. Further, target was vacuum annealed at 800 °C for 5 h. After that, n-type and p-type silicon samples were cut into small pieces and cleaned using RCA-1 and RCA-2 cleaning method to remove organic and ionic contaminates from the substrate. In the next step, samples were dipped in diluted HF (10:1) solution for 30 s and rinsed with DI water (6–8 times). ITO samples were cleaned with solution containing 10% NaOH for 3 min followed by acetone and DI water bath. Appropriately cleaned and dried samples were immediately loaded in a vacuum chamber of RF sputtering system. A high vacuum of 10−6 mbar is then established in deposition chamber using turbo pump followed by introduction of Ar gas into the reaction chamber. The flow rate of Ar gas was kept at 30 sccm. The working pressure was fixed at 10−2 mbar and RF power was set at 50W at 13.56 MHz. Further, all the samples were coated at room temperature for different growth times − 5, 10, 15 and 20 min respectively under same sputtering conditions. Post-deposition annealing was done at 300 °C in vacuum for a duration of 2 h to minimize surface and interface contaminations. Surface morphology and crystallinity of as-grown MoS2 film were performed using Field emission scanning electron microscopy (FESEM, Nova NanoSEM 450), atomic force microscopy (MFP-3D Model oxford instruments), X-ray diffraction spectroscopy (Rigaku smart lab in plane grazing angle XRD). Lattice vibrational modes and optical characteristic of as-grown MoS2 film were investigated by Raman spectroscopy (Princeton Instruments Action pro SP-2500) with a laser of 532 nm wavelength, ultraviolet-visible-INR (Shimadzu uv-3600 plus) spectrophotometer respectively.

3 Results and discussions

3.1 Structural analysis & morphological study

3.1.1 X-ray diffraction (XRD) study

X-ray diffraction (XRD) pattern of layered MoS2 films grown by RF sputtering process, have been shown in Fig. 1a–c. The presented XRD results correspond to diffraction pattern of MoS2 film deposited on n-Si, p-Si and ITO coated glass substrate respectively. All the three XRD diffractograms have been obtained for MoS2 film sputter deposited for 5, 10, 15 and 20 min, respectively. Prior to XRD measurement, heat treatment of the samples was performed under vacuum condition. The diffraction graphs exhibit crystalline peaks related to hexagonal MoS2 corresponding to JCPDS card number 00-037-1492. Figure 1a exhibits XRD pattern of MoS2 deposited on ITO coated glass. It can be observed that multiple peaks are present depicting polycrystalline deposition. The average crystallite size of plane has been calculated from the full width at half-maximum (FWHM) intensities of XRD lines using the well-recognized Debye–Scherrer formula,

$$D=\frac{{K\lambda }}{{\beta cos\theta }}$$
Fig. 1
figure 1

XRD pattern of MoS2 deposited at various growth time on a ITO coated glass, b n-silicon, c p-silicon, d bare samples

Full size image

where β is full width half maximum in radian, λ wavelength of X-ray, k is constant, D is the average size of crystallites of plane in the direction perpendicular to the lattice planes for the observed reflection and \(\theta\) is the Bragg angle. For MoS2 film on ITO coated glass substrate, samples sputtered for 5, 10 and 20 min, (006) orientation appeared as preferred orientation with the average crystallite size of plane as 25.6 nm, 24.9 nm and 24.07 nm, respectively. Further for samples sputtered for 15 min, preferred orientation is (008) and crystallite size 56.16 nm. MoS2 film on n-Si and p-Si are (004) and (101) with corresponding crystallite sizes as 57.21 nm and 59.31 for MoS2 film deposited on n-Si substrate for 5 min and 10 min, respectively, and 55.6 nm and 51 nm for films deposited on p-Si substrate for 10 and 20 min, respectively. The phase structures of mono layer MoS2 display sharp diffraction peak along (002) plane, however, normally, when growth of 2D MoS2 takes place, then stacking of (002) plane is considerably subdued and very thin (upto thickness of 7 nm) 2D MoS2 nanostructures are obtained [22]. In the present work, following the Bragg’s equation, d-spacing for the preferred orientation of the plane has been calculated and indexed in the following Table 1. Further, an underestimated approximation of thickness is also done based upon the number of layers obtained from the Raman spectroscopy (to be discussed in section A (iii)), Table 1 is given as follows.

Table 1 Thickness approximation using interplane spacing (d)
Full size table

As can be seen from the Table 1, estimated thickness is an underestimated version of the thickness obtained from the other methods (Table 4) since multiplicity of the deposited planes has been taken equal to 1 while calculating ‘d’. Multiplicity would affect the value of d and thereby, the overall thickness of the deposited film. However, in the present work, thickness obtained from other methods indicates that multiplicity is too high therefore, underestimated thickness obtained from the XRD in conjunction with other results indicates that deposited film is 2D in nature. XRD results of bare samples has also been indexed in Fig. 1d.

3.1.2 Atomic force microscopy (AFM)

Surface topography and height profile measurement of thin film MoS2 were carried out using Atomic force microscope. Line scan was used to measure thickness of MoS2 layers deposited on different substrates. Further, the measured average physical thickness corresponding to line shown in Figs. 2a–d, 3a–d, 4a–d lies between (3.6–5.1 nm), (3.65–5.3 nm) and (3.5–5.93 nm) for ITO coated glass, n-Si and p-Si substrates, respectively. Cross sectional profile shows smooth and incessant structure apart from some dot isles formed that may be due to condensation of sulphides atoms. Table 2 shows the estimated thickness of MoS2 film.

Fig. 2
figure 2

AFM thickness profile of MoS2 thin film deposited on ITO coated glass a 5 min, b 10 min, c 15 min, d 20 min

Full size image
Fig. 3
figure 3

AFM thickness profile of thin film MoS2 deposited on n-Si a 5 min, b 10 min, c 15 min, d 20 min

Full size image
Fig. 4
figure 4

AFM thickness profile of thin film MoS2 deposited on p-Si a 5 min, b 10 min, c 15 min, d 20 min

Full size image
Table 2 Estimation thickness from AFM
Full size table

3.1.3 Raman analysis

For confirmation of number of layers, Raman results for samples sputtered for 5, 10, 15 and 20 min was recorded and analysed in detail [23]. Raman spectra for MoS2 films grown on ITO coated glass, n-Si and p-Si substrate have been shown in Fig. 5a–c. The excitation wavelength 532 nm with 2 s exposure time and 15 accumulations was selected. Raman spectroscopy basically provides the chemical fingerprint of MoS2 by two signature peaks i.e. E2g and A1g, where E2g is in-plane vibration of S molecules with respect to Mo, and A1g is the out-of-plane vibration of S molecules opposite to each other as discussed previously. The difference between these two peaks is used to estimate number of layers [24,25,26,27]. The relevant peak (E2g and; A1g) position for MoS2 film deposited on ITO coated glass is shown in Table 3 and it can be depicted that thickness of the films is less than 10 nm. It can be seen from Fig. 5a that slight red shift is observed in E2g peak for 20 min sample due to coulomb interaction between coupled dipole that originates from Mo-S bonds that causes the strain, which increases with number of layers [28]. Further, from Table 3 it can be noted that for n-type and p-type Si substrate, relatively thick film has been deposited as confirmed by AFM height profile that lies between 3.65 to 5.3 nm for n-Si and, 3.5 to 5.9 nm for p-Si ,respectively. Further for n-type and p-type Si substrates the MoS2 deposition has been observed to be ranging between 5 and 6 layers. It can be seen from Fig. 5b that there is red shift in E2g mode due to the in plane tensile strain caused by coulomb interaction [29, 30]. From the Fig. 5a–c, it can be clearly observed that with the increase in sputtering time, the peaks of E2g and A1g appeared to be stronger and sharper depicting substantial improvement in the crystallinity of the film.

Fig. 5
figure 5

Raman spectra of MoS2 thin film deposited on a ITO coated glass, b n-Si, c p-Si

Full size image
Table 3 E2g and A1g peak positions for MoS2 film
Full size table

Note that in all samples deposited on ITO coated glass, n-Si and p-Si substrates, as the sputtering time increases, the ratio of intensity increases, this reflects an effective conversion of surface morphology of deposited MoS2 from 2D to 3D nanostructure. In addition to this, Raman spectra of MoS2 deposited on n-Si and p-Si have a broader FWHM (full width half maximum) than those on ITO coated glass, suggesting a better quality films on ITO coated glass.

The most reliable and standard method for confirmation of 2D structure of MoS2 is Raman spectroscopy. As per results obtained from Raman spectroscopy (for deposition on different substrate with different growth time), difference in peaks gives estimation of 5 to 7 layers. Theoretically, for MoS2, thickness of one layer is equal to 0.65 nm that makes us reach to the conclusion that the deposited MoS2 film thicknesses lies within 10 nm for all substrates and, for all growth conditions. 2D nature of MoS2 is associated with the number of layers or in turn thickness of the deposited film therefore in order to confirm the 2D nature, it is essential to determine thickness of the film. The 2D structure of MoS2 is further confirmed (in addition to Raman analysis) by the thickness calculation done with the help of two methods i.e. by using AFM [as discussed in section A (ii)], ellipsometry and by using general method for thickness calculation. The general method involves weight difference method for measuring thickness. The mass of the film is measured before and after deposition of the film. This difference gives the mass of the film (‘m’), now since area of the film (A) and density (d) of the film material are known, therefore thickness of the film (t) can be determined by using relation “t = m/Ad”. However, the approximated thickness is not accurate as density of the material in bulk and thin film is different. The thicknesses obtained for MoS2 samples grown on different substrates under different growth conditions can now be summarized as (as shown in Table 4).

Table 4 Thickness calculations via different methods
Full size table

3.1.4 Field emission scanning electron microscope (FESEM) analysis

Surface Morphologies of Vacuum annealed MoS2 thin film deposited on ITO, n-Si and p-Si have been shown in Figs. 6, 7, 8 respectively. The obtained FESEM results are well aligned with the results of XRD. It is noteworthy here that preferred orientation of the deposition depends upon the nanostructure grown on the substrate i.e., if the nanostructure tends to lie flat on the substrate then the intensity of the Bragg’s reflection increases in comparison to the other nanostructure plane that have 3D geometrical shape [31, 32]. The FESEM results as shown in Fig. 6a, b, d clearly shows that samples with sputtering time of 5, 10 and 20 min, exhibit a cleavage nanostructure that might get aligned parallel to a crystallographic orientation that in turn results in enhanced intensity of specific peaks. Further, sample sputtered for 15 min. shows flat structure, however crystallite size for the same is 56.16 nm (as discussed in XRD section) that indicates lesser number of crystallites (in comparison to samples sputtered for 5, 10 and 20 min) with increased crystallite size due to which lesser Bragg’s reflection will take place and intensity of (008) plane may suppress. Preferred orientation of MoS2 film in n-Si and p-Si are (004) and (101). The corresponding crystallite size (as calculated from the Bragg’s relation) are 57.21 nm and 59.31 for MoS2 film deposited on n-Si substrate for 5 min and 10 min, respectively and 55.6 nm and 51 nm, for films deposited on p-Si substrate for 10 and 20 min, respectively. As can be observed from the FESEM results, nanostructures grown on n-Si and p-Si substrate are three dimensional resulting in deposition of planes repeating in highly different multiplicity. It is to be noted here that Bragg’s rule applies to the lattice plane that are parallel in nature to the substrate surface. Therefore, for every theta, diffractogram appears in accordance with their multiplicity [32, 33]. In case of 3D nanostructure with high multiplicity fraction of different planes increases with respect to the illuminated substrate area, resulting in significant changes of XRD pattern obtained for these structure. Furthermore, in general it can be concluded from the FESEM results that for lesser sputtering time deposition results in distributed flat grains that starts serving as nucleation sites when sputtering time is increased. The prolonged sputtering time therefore results in relatively deposition, as also evident from the Raman results (Table 3).

Fig. 6
figure 6

SEM image of MoS2 grown on ITO coated glass a 5 min, b 10 min, c 15 min, d 20 min

Full size image
Fig. 7
figure 7

SEM image of MoS2 grown on n-Si a 5 min, b 10 min, c 15 min, d 20 min

Full size image
Fig. 8
figure 8

SEM image of MoS2 deposited on p-Si a ITO 5 min, b 10 min, c 15 min, d 20 min

Full size image

3.2 Optical characterization

3.2.1 UV–vis spectroscopy

It can be observed from the absorption curve (Fig. 9a) that high excitonic features are observed in between 600 and 800 nm, indicating visible region spectrum of light, for diverse sputtering times. These excitonic peaks for (5, 10, 15 and 20 min) samples appear near the band edge, centred around 660–671 nm, that confirms the presence of excitons states (wave vector k = 0). Further, a significant absorption peak in between 434 and 440 nm, corresponds to C transition from deep in valence band to conduction band. So there are two absorption thresholds in the spectrum, due to direct transition from deep in valence band to conduction band [34]. Further, it can be seen that there is suppression of absorption in IR (infrared region), that confirms the formation of 2H–MoS2. It can be seen that absorbance in 10 min sputtered sample is relatively great and broad which can be attributed to distinct random shaped grains structure (Fig. 6b) that exhibit good absorbance. As deposition time increases, surface is more occupied with atoms making surface denser, smoother and thicker, leading to the reduced absorbance as shown in Fig. 9a (for 15 min and 20 min sputtering time). It can be seen that there is variation in the shape of the curve as the relative intensities of the peaks depend on the thickness and quality of the sample. Additionally, interference fringes can be clearly observed in Fig. 9a, that are, basically, due to the ITO coated glass substrate for which interference fringes can be seen in the visible region of electromagnetic spectrum (as shown in Fig. 9d for bare sample). These fringes are results of reflection caused by upper and lower edges of thin film due to which reflected waves interfere with each other causing constructive or destructive interference depending upon thickness of the film. If film is deposited over ITO coated glass substrate and in case the thickness of the film is an odd multiple of quarter wavelength of light incident on it, then it will cancel out both the reflected waves from the surface. As a result, it can be inferred that deposited film will be transparent in this region. On the other hand, if thickness of deposited film is multiple of half of the wavelength of light incident on it, then the two reflected waves superimpose giving rise to constructive interference that clearly indicates reduction in the transmittance whereas reflection is increasing. In present case, MoS2 deposited on ITO coated glass substrate as shown in Fig. 9d, exhibits interference fringes modulated by the basic absorbance characteristics of MoS2, as discussed above. Wavelength and energy corresponding to absorption peaks have been given in Table 5.

Fig. 9
figure 9

Absorbance curve of layered MoS2 on a ITO coated glass, b n-Si, c p-Si, d–f bare ITO coated glass, n-Si and p-Si

Full size image
Table 5 Peak position of absorption spectrum thin film of MoS2 deposited on ITO coated glass
Full size table

Absorption Spectra of vacuum annealed MoS2 film on n-Si and p-Si show two peaks near absorbance edge, which are classified as two direct excitonic transitions (A & B) in brillouin zone [10]. Energy corresponding to these wavelengths and their difference for both (n-Si and p-Si) at various sputtering time (5, 10, 15 and 20 min) has been shown in Tables 6 and 7. It can be observed from Tables 6 and 7 that energy separation of two direct excitonic transitions (A and B) is between the range of 0.06–0.18 eV for both the samples, it is due to the spin orbit coupling that splits the orbits at the top of the valence band in brillouin zone [35]. The energy separation between A and B excitons is in agreement with few-layer deposition of MoS2 film [35]. It can be seen from the Fig. 9b that 10 min sample shows good absorbance which can be attributed to formation of randomly distributed nanoflakes throughout the surface (as shown FESEM image Fig. 7b) that enhances its absorption capability, However in 20 min sample these nanoflakes disappear since the nanoflakes structure are very sensitive to the absorption, this could be the reason of the least absorbance offered by its surface. Likewise, from Fig. 9c, it can be noticed that 15 min sample shows good absorbance, since the grown film exhibit vertically standing flower like morphology that serves as active site for absorption [36].

Table 6 A and B excitons location in absorption spectrum of thin film MoS2 deposited on n-Si
Full size table
Table 7 A and B excitons location in absorption spectrum of thin film MoS2 deposited on p-Si
Full size table

Transmittance curve of MoS2 thin film deposited for different sputtering time (5, 10, 15 and 20 min) on n-Si, p-Si and ITO coated glass substrate has been shown in Fig. 10.

Fig. 10
figure 10

Transmittance curve of MoS2 deposited on a ITO coated glass, b n-Si, c p-Si, d–f bare substrates

Full size image

It has been observed from the Fig. 10a, b that 20 min sample of ITO and n-Si shows good transmission in range of 500–600 nm and 700–800 nm, respectively, further 10 min sample shows good transmittance (700–800 nm) in case of p-Si. It can be observed that the thick sample transmits more light as observed from the height profile obtained from AFM [37]. Absorbance and transmittance results of bare samples have also been indexed in Figs. 9d–f and 10d–f.

3.2.2 Photoluminescence

Photoluminescence measurement predominantly exhibits recombination of photogenerated electrons with the holes or traps. Recombination of photogenerated electrons with traps or surface states is more prevalent in non-uniform 3D nanostructure [30]. This because of the availability of increased number of surface atoms that gives rise to dangling bonds surface defects [38, 39]. Following the above discussion, MoS2 deposited on ITO coated glass substrate, n-Si and p-Si (Fig. 11a–c) shows significantly good PL characteristics, owing to the MoS2 nanostructure formed as evident from FESEM results. Figure 11a shows photoluminescence spectra of MoS2 deposited on ITO coated glass with relatively strong intensity and peak centred at 1.8 eV, indicating direct band gap in brillouin zone for 10 min sputtering time, which is in agreement of its morphological structure that shows continuous and homogenous distribution of nanoparticles as shown in Fig. 6b. Figure 11b shows PL spectra of layered MoS2 film on n-Si substrate for 15 min sputtered deposited sample, PL peak intensity is highest and it is centred at 1.85 eV, depicting direct band transition from bottom of valence band and top of conduction band in brillouin zone. This may be due to vertically standing nanoflowers structure as observed in FESEM are formed leading to enhanced field emission [40]. Further, no bandgap shift is observed in the remaining n-Si samples, centred at 1.8 eV. It is important to note here that in this work PL is not governed by thickness of the material but this is due to nanostructure and vacancies formed over the surface of few-layer MoS2 film during the variation in growth time followed by vacuum annealing [38]. Moving further towards p-Si shown in Fig. 11c, it can be clearly observed that PL pattern is dominated by film thickness that is directly related to the sputtering time. The highest PL intensity for 5 min sample with peak at 1.79 eV and the lowest for 20 min sample centred at 1.8 eV may be due to the increase in the thickness of films which increases the interlayer coupling in brillouin zone [40, 41].

Fig. 11
figure 11

Emission spectra of a ITO coated glass, b n-Si, c p-Si with excitation wavelength of 460 nm

Full size image

4 Conclusion

In summary, few-layer MoS2 film has been grown on n-Si and p-Si and ITO coated glass substrate under diverse growth time by RF sputtering technique. XRD study reflected that MoS2 deposited on n-Si and p-Si substrate shows better crystallinity in comparison to film grown over ITO coated glass substrate. Raman spectroscopy results confirmed deposition of few-layer MoS2 deposition by its signature peaks of A1g and E2g. Further, FESEM results exhibited thicker deposition with prolonged sputtering time that also got verified from Raman, AFM and XRD results. It has been observed that despite of few layers thick deposition, it is possible to get photoluminescence by optimizing growth time. The nanostructures grown during different deposition times play a key role in nurturing the photoluminescence in thin film uplifting scope of few-layer MoS2 material for application in field emission devices and solar cell. The present study suggests that RF sputtering is a viable route for the deposition of high quality few-layer MoS2 by controlling the growth time.