Hydrothermally synthesized 2H-MoS2 under optimized conditions – A structure and morphology analysis

In this study, we obtained the optimized conditions to synthesize pure semiconducting 2H-MoS2 nanomaterial, using a facile and scalable hydrothermal route under the variation of growth parameters such as reaction temperature, reaction time and sulfur precursors. The structural and phase identification of obtained MoS2 powders was analysed using XRD and raman spectroscopy. The reproducible formation of pure 2H-MoS2 phase is reported for the optimized reaction time of 22 h at a temperature of 200 °C using thiourea as sulfur source, with a high yield of 77.4%. FESEM analysis revealed nanoflower-like morphology of average diameter of 300–400 nm with identifiable petals of thickness ∼25 nm for the formed 2H-MoS2 under the optimized conditions. The crystallite size, strain and dislocation density were estimated theoretically using Williamson-Hall plots for the MoS2 formed under the variation of growth temperatures. Tensile strain values were obtained for MoS2 formed using thiourea, which correlated only with phase transitions from mixed 1 T/2H-MoS2 to pure 2H-MoS2. In contrast, only mixed 1 T/2H-MoS2 phase were obtained for MoS2 powders using L-Cysteine, and correspondingly the strain values were extremely small, which may be due to no phase transition observed and presence of nanosheets without curved petal-like features. The results of this study provide optimized condition for the formation of semiconducting 2H-MoS2 nanomaterial by a scalable route. This is useful for low-cost fabrication of flexible nanoelectronic devices such as non-volatile ReRAMs, supercapacitors and sensors based on 2H-MoS2.

Both top-down and bottom-up approaches have been employed for the synthesis of MoS 2 viz. liquid exfoliation of bulk materials, mechanical exfoliation as well as chemical vapor deposition (CVD) and hydrothermal/solvothermal methods, respectively. Among the top-down approaches, mechanical exfoliation and liquid exfoliation are prevalently used for obtaining single-layer MoS 2 and multi-layer MoS 2 , respectively.
Chemical Vapor Deposition technique has been used for growth of layer-dependent MoS 2 for biosensing [16,17] and piezoelectric applications [18]. Hydrothermal and solvothermal techniques are inexpensive and facile routes for large scale synthesis of MoS 2 .
Hydrothermal synthesis of MoS 2 has been reported by several groups. Recently, there has been interest in studying low temperature hydrothermal synthesis of defect induced MoS 2 nanosheets for HER catalytic applications [19,20]. Doped MoS 2 viz. p-type, has been synthesized hydrothermally and its photo catalytic hydrogen production has been compared with intrinsic MoS 2 [21]. Study of hydrothermal synthesis of 1 T, 2H and mixed 1 T/2H phases of MoS 2 under variation of different parameters like pH, reaction temperature, reaction time, surfactants and Mo/S ratio has been carried out by different groups [22][23][24]. G H Jetani et al recently, reported formation of polycrystalline MoS 2 with mixed 1 T/2H phase, under variation in pH of precursors, for photosensitivity experiments [25]. Morphology change of MoS 2 from nanosheets to nanoflowers has also been reported under the variation of Mo source from sodium molybdate to molybdenum oxide [26]. Controllable growth of MoS 2 microstructures by the addition of surfactants like CTAB and PEG has been studied and formation of larger size and uniform nanosheet and nanoflower morphologies have been reported with increased concentration of surfactants [27][28][29]. Study of the effect of various alcohols during hydrothermal synthesis of MoS 2 was performed by X Leng et al, who reported MoS 2 with diverse morphologies and properties as a function of the alcohols [30]. The effects of reaction time, temperature and pH of the precursors viz. ammonium molybdate and thioacetamide were studied for the hydrothermal growth of MoS 2 and mechanism proposed for the varying catalytic performance for 1 T, 2H and 1 T/2H mixed phase of MoS 2 [31]. Hence, in above works, variation of the growth parameters was performed and formation of different 1T-, 2H-and mixed 1 T/2H-MoS 2 phases were reported towards different applications.
Few groups reported work towards optimization of 1 T dominant MoS 2 phase. It has been proposed that an intercalation agent must be present, for the formation of 1 T dominant MoS 2 phase, [32]. Hydrazine hydrate intercalation of MoS 2 over a period of 48 h has been carried out for optimized synthesis of stable 1T-MoS 2 for high performance electrocatalytic applications [33]. It has been shown that hydrothermal synthesis under high magnetic field also results in stable synthesis of 1T-MoS 2 [34].
Our aim, in this work, was to synthesize and identify optimized condition for the formation of 2H-MoS 2 phase using hydrothermal synthesis under the influence of variation of reaction temperature, reaction time and sulfur precursors using sodium molybdate dihydrate as Mo source. Structure and phase transition of different polymorphs of MoS 2 samples obtained under varying growth parameters, was investigated systematically using XRD and raman spectroscopy. The morphology of the formed MoS 2 nanostructures was examined using FESEM. Strain developed in the nanostructures was theoretically estimated using Williamson-Hall plots, under the variation of growth temperature and correlated with the phase transitions observed in MoS 2 . 2H-MoS 2 semiconducting phase was reproducibly obtained at reaction time of 22 h and reaction temperature of 200°C with a good yield. The identified optimized condition of synthesis of semiconducting 2H-MoS 2 is useful for lowcost fabrication of MoS 2 nanoelectronic devices. 2. Experimental 2.1. Materials and methods MoS 2 synthesis was performed using hydrothermal route. Sodium molybdate dihydrate, thiourea and L-cysteine were procured from SRL Pvt. Ltd. Chennai and used as such without any purification. First, a dispersion of Mo precursor viz. 2.4195 g of Sodium Molybdate dihydrate (Na 2 MoO 4 .2H 2 O, extrapure AR grade) was prepared in 20 ml of distilled water. To this dispersion, 3.0048 g of thiourea (CH 4 N 2 S, extrapure AR grade) was added and to obtain homogenous dispersion, the mixture was magnetically stirred for 10 min. The Mo to S ratio was kept fixed as 1:4 in all the experiments. Subsequently, the mixture was transferred to 100 ml Teflon coated stainless steel autoclave and kept for heating in a hot air oven at different temperatures (180°C, 200°C, and 220°C) for two different reaction times (20 h and 22 h). The flowchart of the experimental part is depicted in figure 1. Afterwards, the autoclave vessel was cooled down to room temperature and the powders were kept for drying in an oven for 8 h at 80°C. Similar procedure was repeated for the synthesis of MoS 2 with 4.895 g of L-cysteine (C 3 H 7 NO 2 S, extrapure AR grade) as a sulfur precursor.

Characterizations
The crystal structure of as obtained samples was characterized by x-ray Diffraction (Bruker D8 Advance, Germany) using Cu-k α radiation wavelength of 1.5406 Å. A Raman spectrum (solid state lasers, Horiba, 532 nm, 25 mW, France) was employed for the phase identification of the obtained powders. The morphology of the as synthesized powders was determined by field emission scanning electron microscopy (TESCAN-MIRA 3 LMH, Korea).

Results and discussion
MoS 2 synthesis was carried out using hydrothermal route and the phase evolution of 1 T, mixed 1 T/2H and pure 2H-MoS 2 phase was studied under the influence of varying reaction temperature, reaction time and sulfur precursors using XRD and Raman analysis. The morphology of the formed samples was examined using FESEM and the strain values were theoretically estimated using Williamson-Hall plots.

Influence of variation of reaction temperature
Sodium molybdate dihydrate and thiourea were used as Mo and S precursors, respectively and the hydrothermal reactions were carried out for two different reaction times (20 and 22 h) at three different temperatures viz. 180°C , 200°C, and 220°C, using DI water as a solvent. XRD plots are shown along with the standard JCPDS peaks for ease of comparison, as per practice [35].  figure 2(b). It is known that orthorhombic α -MoO 3 formation takes place around 200°C [36], which is an intermediary product formed. Equation (1), below, depicts the intermediary α -MoO 3 formation in the hydrothermal reaction of sodium molybdate dihydrate and thiourea. Alongside the α -MoO 3 peaks, significantly sharp 1T-MoS 2 peaks corresponding to (002) and (004) planes are slightly shifted to higher 2θ values of 11.39°and 18.60°. This may be on account of altered interlayer stacking in the metastable 1T-MoS 2 , upon increase in temperature, resulting in decrease in c-lattice parameter. Also, small intensity peaks corresponding to (100) and (110) planes of 2H-MoS 2 are seen in figure 2

( )
With increased reaction temperature of 220°C for 20 h, further reaction of the intermediary MoO 3 takes place, leading to the formation of 2H-MoS 2 . This is reflected by the presence of several low intensity 2θ peaks appearing at 13.32°, 32.04°, 33.84°and 56.34°corresponding to (002), (100), (101) and (110) planes of 2H-MoS 2 , as seen in figure 2(c). The Raman spectrum, shown in figure 3, depicts the corresponding evolution of the phases of formed MoS 2 . At lower reaction temperature of 180°C, prominent A 1g out-of-plane and E 1 2g in-plane modes are seen, with A 1g transverse S-S vibrational mode to be stronger than E 1 2g longitudinal Mo-S mode. This clearly points to the development of 2H-MoS 2 phase. With increase in reaction temperature to 200°C, dominant B 2g raman active mode along with J1, J2, J3 and E 1g in-plane vibrational modes appear, confirming the α -MoO 3 phase formation along with 1T-MoS 2 phase. This is in accordance with the XRD shown in figure 2(b). At 220°C, B 2g , J1, J2, J3 modes are significantly reduced and extremely weak A 1g and E 1 2g modes emerge, suggestive of an emerging 2H-MoS 2 phase.
Thus, as the reaction temperature is varied from 180°C to 220°C, we observe a transition in the phase from mixed 1 T/2H-MoS 2 at 180°C, to predominant MoO 3 and 1T-MoS 2 at 200°C and the formation of an emerging 2H-MoS 2 phase at 220°C.  Interestingly, under this condition, no XRD peaks due to 1T-MoS 2 or α -MoO 3 are observed. Thus, at 200°C for 22 h, a pure 2H-MoS 2 semiconducting phase is obtained, as per our objective. Hydrothermal reaction at these conditions was repeated several times, to ascertain the reproducibility of the formation of pure 2H-MoS 2 . Subsequently, at slightly higher growth temperature of 220°C, the prolonged reaction time, once again, shows the formation of metastable 1T-MoS 2 , as reflected by the presence of XRD peaks at 2θ values of 11.33°and 18.34°. This transformation of 2H-MoS 2 phase at 220°C to metastable 1T-MoS 2 may be attributed to possible intercalation by water molecules at higher temperature, resulting in increased d-spacing along c-axis.
The raman spectra, shown in figure 5 confirms the phase transitions obtained for MoS 2 in accordance with structure from the XRD data. Accordingly, weak A 1g , E

Effect of variation of sulfur sources
Further, the effect of change in sulfur source, as another parameter affecting the formation of MoS 2 was studied. When the sulfur source is changed to L-cysteine, a higher degree of crystalline XRD peaks are observed in the XRD plots of the samples under the reaction time of 20 and 22 h at different reaction temperatures. For the case of 20 h reaction, at lower temperature of 180°C, small intensity XRD peaks at 11.38°, 30.95°and 40.73°related to 1 T/2H-MoS 2 and at 13.08°, 24.96°, 28.41°and 54.78°related to α -MoO 3 are seen in figure 6(a), which may be ascribed to the nucleation of MoS 2 and intermediary MoO 3 . The intermediary MoO 3 byproduct is, once again, formed during the reaction of sodium molybdate with L-cysteine, as given by equation (2)   By extending the reaction time to 22 h, more intermediary α -MoO 3 peaks along with distinct crystalline 1 T/ 2H-MoS2 peaks are obtained, at 180°C and 200°C, pointing to increased formation of intermediary α -MoO 3 along with 1 T/2H-MoS 2 crystallites (see figures S1(a) and S1(b)). Increasing reaction temperature to 220°C, once again leads to the formation of nano 1 T/2H-MoS 2 along with α -MoO 3 .
Raman spectra for corresponding samples obtained with variation in temperature is shown in figure S2. Dominant J3, A1g modes for MoS 2 are observed at all temperatures with small A g , B 1g , B 2g and B 3g modes for MoO 3 . Interestingly, J3 mode shows broadening with increase in temperature, indicating a transition from bulklike to nanophase. This is in corroboration with the broadening of XRD peaks with increase in temperature as seen in figure S1.

Morphology and strain studies
FESEM analysis was performed for MoS 2 powders obtained using thiourea and L-cysteine, as shown in figures 8 and 9, respectively. Figure 8 shows primarily nanoflower morphologies with petal-like features, which developed with increase in reaction temperature from 180°C to 220°C for the reaction time of 22 h. At 180°C, larger nanosphere with poor petal-like features are observed in figure 8(a). The nanoflower morphology with distinct petals are observed for 2H-MoS 2 obtained at optimized condition of 22 h at 200°C using thiourea (see figure 8(b)). A decrease in nanoflower diameter is visible at 220°C in line with the dominant broad 2H-MoS 2 XRD peaks obtained for this sample (see figure 8(c)). The diameter of the agglomerated nanoflower morphologies ranges between 300-400 nm and the average petal thickness is approximately 25 nm.
For the case of L-cysteine, FESEM images for the 20 h reaction sample are analysed. SEM micrograph, in figure 9, shows primarily nanosheets, which has also been previously reported by other groups [37]. MoS 2 sheet-like structures with irregular shapes, are obtained at low reaction temperature of 180°C as seen in figure 9(a). The increase in reaction temperature to 200°C, allows crystallization to result in well-formed intermediary MoO 3 nanorods along with few 1 T/2H-MoS 2 nanosheets, as revealed clearly in SEM micrograph (see figure 9(b)). This is in corroboration with XRD and raman spectra for MoS 2 obtained using L-cysteine (see figures 6 and 7), where high intensity MoO 3 peaks are observed along with 1 T/2H-MoS 2 peaks. Further, at elevated temperature of 220°C, the intermediary MoO 3 grows to form nanosheet-like structure corresponding to 1 T/2H-MoS 2 phase (see figure 9(c)). The average lateral size of the nanosheets lies between 200-400 nm. The thickness of the nanosheets is approximately 40-50 nm (see figure 9). Hence, in the case of L-cysteine as sulfur precursor, predominantly nanosheets formation takes place instead of nanoflowers, as compared with the case of thiourea.
We analyzed the crystallite size, strain and dislocation density only for the well-formed MoS 2 samples using thiourea at 22 h and L-cysteine at 20 h, as a function of variation in growth temperature. This choice was based where D is the crystalline size, β is the full width at half maximum of the peak intensity; θ is the diffraction angle; λ is the wavelength (0.15406 nm); and ε is the crystalline strain. The calculated crystallite size was compared with the experimental values obtained from SEM micrographs and the theoretical strain values were correlated with the phase transitions observed from Raman spectra. Table 2 shows the calculated crystallite size, strain and dislocation density for the indicated synthesis conditions, in the case of thiourea as the precursor, considering 2θ values of dominant XRD peaks. Figure 10 shows the plots of β cos θ versus 4sinθ for the dominant planes, obtained in the case of thiourea at reaction temperatures of 180°C, 200°C and 220°C, respectively for 22 h reaction. Linear fit of the data of these plots provides the values of R 2 , the strain (ε) from slope and crystallite size (D) from the y-intercept. The strain values obtained for the sample synthesized at reaction temperatures of 180°C, 200°C and 220°C were 0.02004, 0.02174 and 0.01405. It is observed that with increase in growth temperature, the strain does not follow a monotonic trend. At 180°C, the expanded interlayer spacing of 0.784 nm, along the (002) c-axis, observed in XRD figure 4(a), corresponds to an estimated strain value of 0.02004, which may be attributed to tensile strain along c-axis. The presence of J3, E 1 2g and A 1g raman modes corresponding to mixed 1 T/2H-MoS 2 phase (see figure 5), at 180°C shows a transition to dominant E 1 2g and A 1g modes for pure 2H-MoS 2 at 200°C and also (002) plane of 2H-MoS 2 appears at 2θ values of 13.73°instead of 14.96°, suggesting interlayer expansion along c-axis. This is reflected by an increased strain value of 0.02174 (see table 2). It is known that small changes in the lattice structure in 2D materials may alter the lattice strains. Hence, strain-driven 2H to 1 T transition have been observed in TMDCs [38,39]. This strain value is higher than that of the bulk MoS 2 strain ∼0.019, which may be on account of the formed nanophase-MoS 2 [25]. At 220°C, the decreased strain value of 0.01405, once again alludes to phase transition from pure 2H-MoS 2 to mixed 1 T/2H-MoS 2 phase. The decreased in the strain value, in this case, may be on the account of altered interlayer spacing due to intercalation of water molecules. Crystallite sizes of 14.   For the case of L-cysteine as sulfur source, nanosheets and nanorod-like morphologies corresponding to MoS 2 and intermediary MoO 3 are seen in SEM micrograph in figure 9. Crystallite size, strain and dislocation density values for MoS 2 using L-cysteine, were also estimated from W-H plots, (see table S1). Extremely small strain values of −0.0016, −0.0037 and 0.002 were obtained as a function of increase in growth temperatures for 20 h. Under this condition, a mixed 1 T/2H-MoS 2 phase was obtained at different temperatures, with morphologies changing from irregular crystallites to nanorods and nanosheets. The decreased strain values may result from absence of any curved petal-like features of the formed MoS 2 nanostructures.

Conclusion
In summary, a facile hydrothermal synthesis route was adopted to obtain pure, well-formed 2H-MoS 2 nanostructures by optimizing synthesis parameters such as reaction temperature, time and sulfur precursors. The optimized condition for forming pure semiconducting 2H-MoS 2 phase was obtained at 200°C for 22 h reaction time using thiourea as sulfur precursor, as verified by the XRD and raman analysis. A number of trials were repeated at this condition to ascertain the reproducibility of the formed 2H-MoS 2 phase. The yield of the formed product was 77.4%, being much higher than at other reaction temperatures and time. In the case of L-cysteine as sulfur precursor, reaction at 20 and 22 h yielded mixed 1 T/2H-MoS 2 phases. MoS 2 nanoflower morphology with petal-like features were obtained with thiourea, while as for L-cysteine, nanosheets and intermediary nanorods were observed. The strain in the formed MoS 2 powders using thiourea, was estimated using W-H plots and the values of 0.02004, 0.02174 and 0.01405 for 22 h were obtained, with increase in reaction temperature. The changes in the tensile strain are related only to phase transitions from 1 T/2H-MoS 2 to 2H-MoS 2 on account of lattice structure changes. The strain values were extremely small and showed very little variation with reaction temperature for MoS 2 powders obtained using L-Cysteine, which may be due to nonchanging 1 T/2H-MoS 2 phase and nanosheets without curved petal-like features. Hence, the results of this study can be used for low cost fabrication of various rigid and flexible nanoelectronic devices such as non-volatile ReRAM devices, strain sensors and bio-sensors using pure semiconducting 2H-MoS 2 obtained in a scalable manner.