1 Introduction

Transition metal dichalcogenides have attracted the attention of many due to their exceptional electrical and optical properties. This paper discusses the Raman studies on MoS2, which naturally occurs as molybdenite. MoS2 is similar to graphene as 2D material but has better mechanical and electronic properties than graphene. The stable form of MoS2 is semiconductor in nature, and once it is reduced to a 2D layer, it has salient features which can be used for various optoelectronic applications. The band gap for MoS2 changes from indirect to direct, from bulk to monolayer [1,2,3,4,5]. Single-layer MoS2 with a direct bandgap is widely used in field effect transistors [6,7,8], integrated circuits [9], LEDs(light emitting diodes), and solar cells [10]. Multi-layered MoS2 has various gas sensing applications [11, 12], energy storage, and conversion applications [13, 14].

Raman spectroscopy can be used for the structural characterization of the materials since it is related to the dynamics of the lattice, phonon density of states, etc. It is a powerful tool to determine the number of layers of MoS2 [15,16,17,18,19]. Layer dependent variation in the difference between the fundamental vibration modes of MoS2 is reported earlier [20]. There are several studies on the different interactions in bulk and single-layered MoS2. This paper focuses on the laser-induced electron–phonon interactions in hydrothermally synthesized 2H-MoS2 nanoflakes, the resultant changes happening in the material due to the laser-induced surface temperature changes, and the resonant interactions occurring in the material, including phonon softening, phase change, and Fano resonance.

Fano resonance is a unique electron–phonon interaction in systems that was first reported in 1961. Many systems exhibiting Fano resonance have excellent potential in nano regimes [21,22,23,24,25]. Although various methods exist to study the Fano resonance in a material, being nondestructive and comparatively precise, Raman spectroscopy is a simple and versatile method to check Fano resonance happening in nanomaterials. Fano resonance occurs due to the interference between the continuum and discrete states. Here the electronic continuum is provided by the laser-induced electron plasma, and discrete states are the vibrational phonon modes. Another possible way for Fano resonance is by doping the material, which gives rise to a change in the electronic energy levels [26, 27]. The study of the Fano interactions happening in the material can aid in studying the lattice thermal conductivity, optical properties, and carrier transport in the material [28].

Raman spectroscopy on MoS2 relies on the first-order phonon vibrational modes, namely the in-plane \({E}_{2g}^{1}\) mode due to the intra-layer vibration of S-Mo-S atoms and the out-of-the-plane \({A}_{1g}\) mode due to the interlayer vibration of S–S atoms [29,30,31]. Phonon confinement model which was earlier used to describe the resonance interaction happening in the material has the expression, [32]

$${I}_{L}\left(\omega \right)=\underset{0}{\overset{1}{\int }}\frac{\mathit{exp}\left(\frac{-{k}^{2}{L}^{2}}{4{a}^{2}}\right){d}^{n}\left(k\right)}{{\left[\omega -\omega \left(k\right)\right]}^{2}+{\left(\frac{\gamma }{2}\right)}^{2}}$$
(1)

where a, L, k, \(\upomega \left(\mathrm{k}\right)\), and \(\gamma\) denote the lattice parameter, nanoparticle size, wave vector, phonon dispersion curve, and FWHM (full-width-at-half-maximum) of the vibrational phonon mode under study. For lower laser powers, two significant changes happen in the phonon vibrational modes: asymmetry in the vibrational modes and the peak shift in the vibrational mode.

Here, the asymmetry in the line shape of phonon modes in the Raman spectra is assumed to be from the Fano scattering, and the shift in the phonon peak center is due to the thermal effect of laser-induced heating. A phase change is observed in the material at higher laser powers due to the laser-induced heating. A thorough study of the material response to laser power is done by varying the laser power gradually and taking the corresponding Raman spectra for each laser power. There are earlier reports on the laser thinning of the MoS2 nanolayers and laser heating of the MoS2 layers [33, 34].

2 Experimental methods

2.1 Materials

Sodium molybdate (Na2MoO4·2H2O) and Thiourea (NH2CSNH2) were purchased from Sigma Aldrich with 99.9% purity and were used throughout the experiment without further purification. Double distilled water is used for washing and as a solvent in the experimental procedures.

2.2 Synthesis of MoS2 nanoflakes

MoS2 nanoflakes are synthesized using the usual hydrothermal method. An aqueous mixture of sodium molybdate (0.005 mol) and thiourea (0.02 mol) in the mol ratio 1:4, which gives the 60 ml solution in double distilled water, was transferred into a 100 ml stainless steel autoclave, was sealed and kept in an oven at a hydrothermal temperature of about 220 ºC for 24 h. Finally, the MoS2 nanoflakes were washed many times with double distilled water and then held for drying in a vacuum oven at 60º C temperature for 12 h. The formed MoS2 was collected and used throughout the experiment.

2.3 Characterizations

The material was analyzed using a Rigaku Ultima III powder X-ray diffractometer in the powder diffraction pattern ranging from 20° to 80° and a 2° min−1 scanning speed. This gives the phase of the nanomaterial formed. The XRD (X-ray diffraction) pattern's peaks show the sample's overall crystalline nature. The sample morphology is studied using a high-resolution transmission electron microscope (HR-TEM) using Joel/JEM 2100 TEM.

Raman spectra for the nanoparticles were taken using the Horiba LabRAM HR Evolution Raman spectrometer in the backscattering configuration in atmospheric conditions. A Silicon wafer is used to calibrate the instrument. Throughout the measurement, the objective lens with a magnification of 100X, a grating with 1800 groves/mm, a laser source with 532 nm excitation wavelength, and 0.5 µm spot size were used. The measurement parameters set were accusation time of 20 s and accumulation 2, during the measurements. The Raman spectra of the same material are studied using a 488 nm laser excitation wavelength for different laser powers.

3 Results and discussions

3.1 Structural and morphological studies

The X-ray diffraction analysis of the prepared MoS2 in Fig. S1 gives 2 \(\theta\) values at 14.05º, 33.48º, 39.57º, 49.43º, and 60.53º. The values correspond to (002), (101), (103), (105), and (110) planes, respectively. These values agree with the JCPDS card number 37–1492 and confirm 2H-MoS2 (Fig. S1). HR-TEM image of the sample gives a nanosheet-like structure, as shown in Fig. 1a. The HR-TEM image with the d-spacing marked is shown in Fig. 1b. The average crystallite size of the MoS2 nanoparticles corresponding to the planes above is calculated using the Scherrer equation as in Eq. (2) and is found to be 6.00 nm. The instrumental broadening for the instrument is 0.04° (using Si standard) is considered for the calculation.

$$D=\frac{k\lambda }{\beta \mathrm{cos}\theta }$$
(2)

where D is the average crystallite size, k is Scherrer's constant with a value of 0.9, \(\lambda\) is the wavelength of the X-rays used (0.154 nm), \(\beta\) is the FWHM in radians, and \(\theta\) is the Bragg angle in Radians.

Fig. 1
figure 1

a HR-TEM image of the sample shows a layer-like structure, b Higher resolution HR-TEM image with the distance between the plane’s ‘d’ marked

Full size image

3.2 Raman analysis

Bulk 2H-MoS2 nanoflakes, belonging to the P63/mmc space group, have a hexagonal structure. Raman spectra are obtained for the bulk 2H-MoS2 nanoflakes at different laser powers for 532 nm and 488 nm laser wavelengths, and the peak position, FWHM, and the asymmetry in the peaks are noted.

The normal vibrational modes in MoS2 are shown in Eq. (3) [35]:

$$\Gamma^{{v{\text{i}}b}} = A_{1g} + 2A_{2u} + 2B_{2g} + B_{1u} + 2E_{1u} + 2E_{2g}^{1} + E_{2u} + E_{1g}$$
(3)

The \({A}_{1g}\), \({E}_{1g}\), and \({2E}_{2g}^{1}\) modes are Raman active modes, and \({E}_{1g}\) is absent in the backscattering geometry. So, the remaining first-order modes observed here are the in-plane \({E}_{2g}^{1}\) and out-of-plane \({A}_{1g}\) modes, respectively. Apart from these, a “b” band is observed around 455 cm−1, which disappears at higher laser intensities. Raman spectra for the hydrothermally synthesized 2H-MoS2 nanoflakes are shown in Fig. 2 for various laser powers ranging from 0.05 to 50 mW for 532 nm laser wavelength. For lower laser powers, only fundamental vibrational Raman modes are observed in the Raman spectra, whereas upon increasing laser power, many additional peaks are observed. The appearance of new lines at higher laser powers shows the phase change happening in the spectra due to the increase in the laser-induced surface temperature. A similar experiment is done with changing laser wavelength to 488 nm, and the corresponding Raman spectra for varying laser powers are observed. The Raman spectra for this excitation wavelength for varying laser powers are shown in Fig. S2. For this wavelength, the laser power is varied from 0.045 to 96 mW. Further, the laser power variant studies are systematically carried out for the second excitation wavelength.

Fig. 2
figure 2

Raman Spectra for the hydrothermally synthesized 2H-MoS2 nanoflakes for different laser powers ranging from 0.05 to 50 mW and for 532 nm laser wavelength. The laser power corresponding to each spectrum is marked beside. The inset graph highlights the changes in the fundamental modes of MoS2 (area highlighted with the dotted rectangle) with changing laser powers. The characteristic modes from the 2H-MoS2 and α-MoO3 after phase change are given, respectively. The red dotted lines in the inset graph are the eye guide showing the peak shift for the vibrational modes with changing laser power

Full size image

Incident photons can change the electronic continuum in the material by laser-induced plasma and its interaction with the discrete vibrational phonon modes. From Fig. 2, it can be seen that, after 5 mW laser power, many peaks are visible in the spectrum in addition to the fundamental modes. The peaks at 661 cm−1,818 cm−1, and 991 cm−1 correspond to the Ag mode of α-MoO3. It is the asymmetric stretching vibration modes of the triply coordinated terminal oxygen Mo3=O, symmetric stretching vibrations of doubly coordinated terminal oxygen Mo2=O, and asymmetric stretching vibration of the oxygen Mo=O respectively [36]. As the laser power is further increased to 50 mW, more α-MoO3 peaks are seen because of the temperature-induced phase change of 2H-MoS2 to α-MoO3. This is due to the oxidation of the material in the region of laser irradiation since we conduct the experiment in atmospheric conditions. The material's phase change at higher laser powers is also observed for the laser excitation wavelength 488 nm, which can be confirmed from the appearance of the characteristic α-MoO3 vibrational peaks seen in the Raman spectra at higher laser powers. The laser-induced surface temperature increase in the material is responsible for this phase change in the material. The highest peaks at 661 cm−1,818 cm−1 and 991 cm−1 which corresponds to Ag vibrational mode of α-MoO3 are present in the Raman spectra for higher laser powers for both excitation wavelengths.

The difference between \({E}_{2g}^{1}\) and \({A}_{1g}\) Raman modes is 18.6 cm−1 in monolayer MoS2 [37]. In this study, the difference between the two fundamental modes is above 27 cm−1, which confirms the bulk nature of 2H-MoS2 nanoflakes.

3.2.1 Laser-induced phonon softening.

The irradiation of the bulk 2H-MoS2 nanoflakes with different laser powers causes the softening of the fundamental vibrational modes in the Raman spectra. Both the fundamental vibrational modes soften upon increasing the incident laser power, and the shift is clearly shown in Fig. 3a. The laser dependent shifts in the centre of the peaks for laser excitation wavelengths 488 nm and 532 nm are given in Fig. 3b, c respectively. The Raman spectra for the fundamental vibrational phonon modes, which show laser power dependant softening behaviour, for 488 nm excitation are shown in Fig. S3. For \({E}_{2g}^{1}\) mode, there is a slight hardening in the smaller laser powers and then shows softening behaviour for an increase in laser power for 532 nm laser excitation wavelength. From Fig. 3c, the intra-layer mode \({E}_{2g}^{1}\) softens up to 6 cm−1 upon an increase in laser power (0.05–50 mW), and the interlayer vibrational mode \({A}_{1g}\) softens only 3 cm−1 which shows that the S-Mo-S vibrational bond is more dependent on variation in laser power. Considering the power-dependent study of bulk 2H-MoS2 nanoflakes upon irradiation with 488 nm laser wavelength, the change in the phonon modes \({E}_{2g}^{1}\) and \({A}_{1g}\) are 7 cm−1 and 4 cm−1(Fig. 3b). In both cases, the phonon softening occurs with increasing laser power. The softening of the bonds and the reduction in the force constant upon higher laser excitation power is the reason for this behaviour.

Fig. 3
figure 3

a Laser-induced Raman shift in MoS2 nanoflakes. The changes in the first order, \({E}_{2g}^{1}\), and \({A}_{1g}\), for 532 nm laser excitation are shown in the graph. (The dotted lines are the original spectra, and the solid lines are the Lorentzian fit for the original spectra. The blue-dotted vertical line is the eye guide to the phonon softening in the fundamental bands as the laser power varies). The logarithmic laser power dependant changes in the centre of the peak for \({E}_{2g}^{1}\) and \({A}_{1g}\) band for b 488 nm laser wavelength, c 532 nm laser wavelength

Full size image

3.2.2 Fano resonance

Asymmetric Raman modes may happen for many reasons, including the quantum confinement effect, electron Raman scattering, nonlinear Fano scattering, etc. [21] Due to the bulk-like nature of the 2H-MoS2 nanoflakes, we rule out the asymmetry caused by the quantum confinement effect. In semiconductors, resonant Fano scattering happens due to the interference between the vibrational phonons and the electronic continuum [32, 38, 39]. The electronic continuum can occur due to doping, the intrinsic effect due to phonon confinement, and inhomogeneous heating [24, 32, 40]. In this study, pure and bulk like 2H-MoS2 nanoflakes are used to eliminate the first two possibilities. Both resonance and antiresonance Fano asymmetry is observed in the fundamental vibrational modes, which is the characteristic property of Fano interaction [39, 40], thus, the asymmetry due to Fano resonance can be considered here.

The nature of the interference gives the position of asymmetry in the given vibrational mode. If the interference is constructive, the asymmetry will be seen on the higher wavenumber side. If the interference is destructive, the asymmetry will be seen in the lower wavenumber side of the vibrational phonon mode. In this study, asymmetry is seen in both the fundamental vibrational modes of bulk 2H-MoS2 nanoflakes, and this asymmetry is fitted using the Fano line shape function, which has the expression:

$$I\left(\omega \right)=\frac{{\left(q+\varepsilon \right)}^{2}}{\left(1+{\varepsilon }^{2}\right)}+A$$
(4)

where \(\varepsilon =\frac{\omega -{\omega }_{0}}{\Gamma }\), q denotes the asymmetry strength, A- background, \({\omega }_{0}\)– center of frequency, \(\Gamma\)-width parameter, and I – the intensity. The value 1/|q| denotes the coupling strength of the material and is related to the electron–phonon coupling in the material [21].

The Fano line shape function was fitted for different laser powers for the Raman spectra for fundamental vibrational phonon modes for both the excitation wavelengths and could be perfectly fitted with the experimental values. The value of the Fano asymmetry factor changes with changing laser power in both excitation wavelengths. This is due to the variation in the interaction between the vibrational level and the electronic continuum in the material. Fano function in Eq. (4), fitted for different phonon modes, is plotted in Fig. 4a, b. The variation of Fano asymmetry strength with logarithmic laser power is given in Fig. 4c, d. The Fano-fitted spectra for 488 nm laser excitation are shown in Fig. S4. The change in the material phase affects the asymmetry seen since the electron distribution changes with phase change.

Fig. 4
figure 4

Fano fitted graphs for 532 nm Laser wavelength for powers ranging from 0.5 to 50 mW for a mode \({E}_{2g}^{1}\), b mode \({A}_{1g}\) (dotted curve represent the original spectra, and the continuous curve represents the spectra fitted with Fano function in Eq. (4)). c The logarithmic laser power dependence of the asymmetry strength 1/|q| for Raman modes \({E}_{2g}^{1}\) and \({A}_{1g}\) for 532 nm laser excitation wavelength. d The logarithmic laser power dependence of the asymmetry strength 1/|q| for Raman modes \({E}_{2g}^{1}\) and \({A}_{1g}\) for 488 nm laser excitation wavelength

Full size image

The value of the asymmetry factor is found to be negative for the fundamental mode \({E}_{2g}^{1}\), showing an asymmetry in the lower wavenumber side, and positive for the fundamental mode \({A}_{1g}\), showing an asymmetry in the higher wavenumber side. Not only does the asymmetry change with the laser power, but it changes with laser wavelength also. The value 1/|q| is the asymmetric strength and is directly related to electron–phonon coupling in the material upon laser irradiation. The asymmetry factor less than 1 shows antiresonance, and the asymmetry factor greater than 1 shows resonance. The laser-induced change in the asymmetry strength of the in-plane S–Mo–S vibration upon laser excitation wavelength 532 nm, as in Fig. 4c, shows a variation from 0.15 to 0.26 on increasing the laser power from 0.05 to 25 mW. The slight dip in the asymmetry strength upon increasing laser power after 2.5 mW can be attributed to the laser-induced changes in the phase and the interplay between the electron plasma of the MoS2 and α-MoO3 phases. The asymmetry strength increases up to 25 mW and then faces a more significant dip at 50 mW. The asymmetry strength of the interplanar S–S vibration mode shows relatively less change upon an increase in laser power apart from an abrupt increase in the asymmetry strength from 0.04 to 0.3 on irradiation with 50 mW laser power. Compared to this, the \({E}_{2g}^{1}\) and \({A}_{1g}\) modes show slightly different behavior upon irradiation of the material with a 488 nm laser, as in Fig. 4d. This can be attributed to the change in electron–phonon coupling in the material. The dependence of out-of-the plane \({A}_{1g}\) mode is relatively less than \({E}_{2g}^{1}\) mode in both the excitation wavelengths.

3.2.3 Laser induced surface temperature calculation and phase change.

Depending on the applications, the surface temperature calculation of the materials is essential in material fabrication. For the manufacturing of temperature-stable materials, one should know the temperature-withstanding capability of the material. Raman intensity measurements are a beneficial technique among temperature calculation techniques. The Stokes and anti-Stokes Raman lines are used for the temperature calculation of the material surface on laser irradiation. Unlike the conventional contact processes with a low areal resolution, Raman spectroscopy is a non-contact method that can go up to an aerial resolution of 1 µm [42,43,44,45,46]. The temperature is found using Raman spectroscopy either by peak shifting method based on the change in bond length which in turn is dependent on the temperature of the material, or the intensity ratio method which is proportional to the occupancy probability [21]. The intensity ratio of the Stokes and the anti-Stokes lines equals the ratio of the Boltzmann probability of occupancy of the ground and excited states. Out of the two spectra, the anti-Stokes spectra are more dependent on the temperature of the surface since the higher levels are populated at higher temperatures. According to the Boltzmann distribution, the intensity ratio between the levels is:

$$\frac{{I}_{As}}{{I}_{s}}={\left(\frac{{v}_{0}+{v}_{m}}{{v}_{0}-{v}_{m}}\right)}^{4}{e}^{\left(\frac{-h{v}_{m}}{{k}_{B}T}\right)}$$
(5)

where \({I}_{s}\) and \({I}_{As}\) are the intensities of Stokes and anti-Stokes lines, \({v}_{0}\) and \({v}_{m}\) are the frequency of the incident laser and the centre frequency of the vibrational mode, respectively, h is Planck's constant, \({k}_{B}\) is the Boltzmann constant and T is the temperature of the surface at the point of irradiation of the laser, which can be calculated from the above parameters as:

$$T=\frac{h{v}_{m}}{{k}_{B}\left(\mathit{ln}\left(\frac{{I}_{s}}{{I}_{As}}\right)+4\mathit{ln}\left(\frac{{v}_{0}+{v}_{m}}{{v}_{0}-{v}_{m}}\right)\right)}$$
(6)

The upper-level population is high at higher temperatures, the anti-Stokes line intensifies, and the probability ratio reaches unity at higher temperatures. The Stokes and anti-Stokes Raman spectra are taken for the temperature calculation, and the spectra are plotted in Fig. 5a. The fundamental A1g mode is used for calculating the surface temperature. At higher temperatures, the bonds break, and MoO3 is formed, which gives the peaks of MoS2 and\({\propto -MoO}_{3}\). The laser dependent temperature change in the material is shown in Fig. 5b, The temperature calculated using Eq. (6), for 0.5 mW, 1.25 mW, 2.5 mW, 5 mW, and 12.5 mW are respectively 61 ºC, 149 ºC, 327 ºC, 247 ºC, and 2150 ºC. The MoS2 converts to MoOx at temperatures around 400 ºC and about 3 m W [36, 47]. This may be the reason for the slight dip in the surface temperature after 2.5 mW laser power. After 5 mW, the temperature increases with an increase in laser power.

Fig. 5
figure 5

a The Stokes and anti-Stokes lines for various laser powers are shown in the graph. (Spectra on the left (-ve x-axis) correspond to the Raman anti-stokes line, and the spectra on the Right correspond to Raman stokes line (positive x-axis).) The dotted curve stands for the experimentally obtained spectra, and the solid curves are the peak-fitted spectra. b The laser power vs. temperature graph for the fundamental \({A}_{1g}\) Raman vibrational phonon mode for 532 nm laser excitation wavelength 532 nm

Full size image

The presence of MoO3 at higher laser powers can be confirmed using HR-TEM image, and XRD data analysis. Analysing the HR-TEM image of the material after irradiation with 532 nm laser light, as in Fig. 6a, it can be concluded that, apart from the presence of \({\mathrm{MoS}}_{2}\), something else is present in the material. The presence of \({\mathrm{\alpha }-\mathrm{MoO}}_{3}\) is confirmed by comparing the XRD pattern with JCPDS card number 75–0912. The XRD pattern is shown in Fig. S5. Thus, irradiation of the material with a 532 nm laser for higher powers causes an increase in temperature and thus induces a phase change in the material. The d spacing for two phases is calculated from HR-TEM image and marked in Fig. 6b.

Fig. 6
figure 6

a HR-TEM image of the laser irradiated 2H-MoS2 nanoflakes, b higher resolution HR-TEM image of the sample with the interplanar distance marked

Full size image

4 Conclusion

We have conducted a detailed study on laser-induced phonon softening, Fano resonance, surface temperature calculation, and phase change in the Raman spectra of hydrothermally synthesized bulk 2H-MoS2 nanoflakes. The surface temperature calculated for the various laser power increased with increasing laser power up to 2.5 mW. Phase change starts around 3 mW, and the temperature is found to increases again up to maximum laser power (50 mW). The asymmetry in Raman spectra indicates the Fano resonance happening due to the electron–phonon interaction between the vibrational phonons and the electron continuum in the material. Fano asymmetry dependence on the changing laser power is studied in detail, in this work. The asymmetry increases with the increase in laser power for two different laser excitation wavelengths. Thus, the electron–phonon interactions in the materials strengthen with the laser excitation power for both the excitation wavelengths.