1 Introduction

WS2 belongs to a general category of material called layered transition metal dichalcogenides (TMDs). They are represented by a general formula MX2, where M is a transition-metal atom, like Mo, W, Ti, etc., and X is a chalcogen atom like S, Se, Te [1]. Recently, different low dimensional semiconductors, i.e., two-dimensional (2D), one-dimensional (1D), and zero-dimensional (0D) TMDs materials have been investigated extensively. They have suitable band gaps of 1-2.5 eV, appreciable stability in ambient, excellent electrical, electronic, and optoelectronic properties [2,3,4,5]. Among various TMDs, the WS2 is reported to be nontoxic, inexpensive and corrosion-resistant. They have very interesting properties, like the appropriate and tunable band gap transition from an indirect band gap of 1.32 eV for bulk WS2 to a direct band of around 2.02 eV for monolayer WS2, high light–matter interaction, high carrier mobility [6], large spin splitting [7], significant light absorption coefficient, very large exciton binding energy [8], strong photoluminescence [9], polarized light emission, etc. These interesting and unique properties can make the WS2 a favorable material for different types of applications, such as in FET [10], memristor [11,12,13,14], light-emitting devices [15], optical modulators [16, 17], water splitting material, photo catalyst and electrochemical catalyst [18], photovoltaic device, Superconducting circuits, Sensors and Bio-Sensors, energy storing materials and many others [19]. Investigation on 0D WS2 quantum dots (QDs) has also gained momentum recently due to their exciting and novel properties, which are pretty different from their 2D, and 1D counterparts (tubes, ribbons). Due to the quantum confinement effect and surface edge effect, the WS2 QDs are reported to exhibit strong photoluminescence at room temperature, even without functionalization. The direct band gap of WS2 QDs (~ 4.0 eV) can be tuned and hence there is ample opportunity for tuned or tailored luminescence at various different wavelength [20,21,22,23] and tailored electronic, optoelectronic properties. The WS2 QDs exhibit higher Photoluminescence (PL) quantum yield (around 15.4%) than 2D sheets. They also show better spin-valley coupling than monolayer WS2 [24, 25]. WS2 QDs, due to their small size of a few nm, low toxicity, solubility in water, biocompatibility and unique features described above, have numerous applications in bio imaging, photo thermal therapy, sensing, catalysis, energy storage and conversion [26]. The electronic and catalytic properties of MoS2, WS2, and NbS2 were also studied by density functional theory (DFT) in context of hydrogen evolution reaction (HER) [27]. Hence, property controlled synthesis of WS2 QDs, by optimizing the synthesis parameters is highly desirable so that the synthesized QDs can be utilized for a targeted application with further improvement in their properties.

In literature, there are reports on the synthesis of WS2 QDs by different conventional methods like hydrothermal route, probe assistant ultra-sonication, and multi-exfoliation based on lithium intercalation. These methods have limitations such as high thermal treatment, long processing time, and hazardous chemical production. Besides those conventional methods, pulsed laser ablation in liquid (PLAL) is emerging as a promising method for synthesizing QDs. It is a single-step, chemical-free simple physical process which operates in the normal ambient condition. In PLAL process various synthesis parameters like pulsed laser duration, laser energy, laser fluence, laser intensity, laser wavelength, liquid level, liquid environment, liquid flow rate, target geometry, and ablation time can be varied, optimized, and controlled to obtain desired nanostructure, in regard of crystallinity, morphology, thickness, area, and stoichiometry [28]. The method has the potential for scalable processing and synthesis [30]. PLAL is showing novelty in preparing low-dimensional colloidal nanostructures, including QDs. Wang et al. investigated the material removal mechanism in laser ablation of WS2 by femtosecond laser with different fluencies. They reported that selective local modulation of material with greater accuracy and quality can be achieved by laser ablation of WS2 [29]. Moreover, Fan et al. showed that TMDC QDs, synthesized by PLAL, can be easily functionalized compared to QDs produced by other conventional methods because of reaction with solvent [30].

So far, there is no systematic study on the effect of laser parameters in controlling the size of WS2 QDs. In the present work, the effect of laser energy on the size of WS2 QDs has been explored extensively. The focus is to study the influence of laser energy on the size of synthesized WS2 QDs so that an easy, feasible, and repeatable design for size modulation of the WS2 QDs can be obtained, which can ultimately lead to property engineering of WS2 QDs.

2 Experimental methods

The WS2 QDs were synthesized by PLAL method, as shown in the schematic experimental Fig. 1. In this experiment, a pellet of WS2 was used as a target for laser ablation and the dimensions of the pellet were diameter of 13 mm and thickness of 2 mm. The pellet was placed at the bottom of a 50 ml beaker of diameter 4 cm and height 5.5 cm, as shown in the Fig. 1. The beaker was filled with 8 ml of distilled water and water level above the target surface was 5 mm. In this method, the Q-switched Nd: YAG laser of 532 nm was used. The laser used was 8 nanosecond pulsed laser with a repetition rate of 10 Hz. The laser beam was focused onto the target with the help of a full reflector plane mirror and a convex lens of focus length 25 cm. The laser beam was focused on the immersed target, on a spot size of 2 mm in diameter. The spot cross-sectional area was 3.14 × 10− 3 m2. In this method, the synthesis parameter laser energy was varied, keeping all other parameters, like spot area, level of the liquid, ablation time, dimensions of the target, laser wavelength, and repetition rate unchanged.

Fig. 1
figure 1

Schematic diagram of PLAL of WS2 in water

Full size image

. The ablation was carried out for laser energies 10 mJ, 20 mJ and 40 mJ, keeping the ablation time 5 min, constant for every laser energy. The laser energy density for 10 mJ is 3.185 J/m2, for 20 mJ is 6.369 J/m2 and for 40 mJ is 12.739 J/m2. The top of the beaker was covered with a glass plate during the ablation of the target to avoid splashing of water droplets on the focusing lens. The beaker was placed on a motorized translational stage so that due to continuous translation of the beaker and target, every laser shot can ablate a fresh surface of the target material. Manual stirring of the colloidal solution was also carried out; in between the ablations so that the nanoparicles (NP) formed can be distributed evenly. Moreover, the incident laser beam does not get obstructed by the suspended WS2 nanoflakes, NPs, and QDs. In this way the ablation can be carried out uniformly, during the entire ablation time. The colloidal suspension was centrifuged at 12,000 rpm to segregate WS2 QDs from WS2 nanoflakes.

The Raman spectroscopic characterization of the synthesized WS2 QDs was carried out in backscattering geometry using excitation wavelength of 488 nm, obtained from Argon ion laser and the spectroscope used for recording was Horiba JobiYovon, LabRam HR 800. The TEM imaging of synthesized WS2 QDs was carried out using TEM-JEOL, JEM 2100, operating voltage 200 kV, to estimate the lateral size of QDs.The Photoluminescence (PL) spectra of synthesized WS2 QDs were recorded by Horiba Jobin Yovon Fluoromax4, using 290 nm excitation wavelengths.

2.1 Results & discussion

The Raman spectra of the WS2 samples for excitation wavelength of 488 nm is shown in Fig. 2. It is observed that the prominent Raman peaks, related to nanostructured WS2, available and existing for all three laser energies 10 mJ, 20 mJ and 40 mJ are E12g(Γ) and A1g(Γ) [31]. However, 2LA (M) peak is also observed in 20 mJ sample, along with E12g(Γ) and A1g (Γ). The 2LA peak is a second order Raman peak, contributed by a double resonant Raman process and indicates the uniformity of WS2 monolayer. The shape of the peaks indicates appreciable crystallinity of the synthesized product. The detailed analysis of the spectra indicated, prominent and identified Raman peaks, at 356.78 (E12g) and 420.71 cm− 1 (A1g) for 10 mJ laser energy, at 351.40 (2LA), 357.03 (E12g) and 421.57 cm− 1 (A1g) for 20 mJ laser energy, at 353.21 (E12g) and 419.161 cm− 1 (A1g) for 40 mJ laser energy. The number of prominent peaks in 40 mJ spectra is 2, in 20 mJ spectra is 4 and in 10 mJ spectra is 5. However, the 20 mJ spectra have high similarity with the reported spectra of WS2 and have highest number of identifiable peaks. The E12g peak corresponds to in-plane vibration and the A1g peak corresponds to out of plane vibration. The Peak frequency difference between A1g(Γ) & E12g(Γ) increases from 63.93 to 65.95 cm− 1 for laser energy variation of 10 to 40 mJ. The peak intensity ratio I E12g/I A1g, decreases from 0.81 to 0.74 for laser energy of 10 to 40 mJ. The FWHM for E12g(Γ) increases from 3.98 to 5.2 for laser energies from 10 mJ to 40 mJ and for A1g (Γ), from 3.69 to 4.62, for laser energies from 10 mJ to 40 mJ.

Fig. 2
figure 2

Raman Spectrum and evolution of prominent and identified peaks with laser energy, using excitation wavelength, λexc = 488 nm of synthesized WS2 QDs, with laser energies 10 mJ, 20 mJ and 40 mJ

Full size image

BerkdemirA. et al. reported that, for 488 nm excitation wavelength, the peak positions for A1g (Γ) are 417.5 cm− 1, 418.9 cm− 1, 419.5 cm− 1 and 420.2 cm− 1, for 1-layer, 2-layer, 3-layer and bulk WS2, respectively. Similarly, the peak positions for E12g (Γ) are 355.9 cm− 1, 355.3 cm− 1, 355.0 cm− 1 and 355.8 cm− 1, for 1-layer, 2-layer, 3-layer and bulk WS2, respectively. Moreover, the intensity ratio of the Raman peaks E12g and A1g decreases from 0.78, 0.62, 0.59 and to 0.53 and the peak frequency difference between A1g (Γ) & E12g (Γ) increases from 61.6, 63.6, 64.5 and to 64.4, for 1-layer, 2-layer, 3-layer and bulk WS2 respectively [31]. Similar trend was observed in our Raman peak data, as discussed above and shown in Figs. 3 and 4. Hence it can be concluded that, with increase in the laser energy, the size or the number of layers for synthesized WS2 QD also increased.

Fig. 3
figure 3

Variation of Intensity ratios of the two Raman modes E12g and A1gwith laser energy

Full size image

From Fig. 3, it is observed that the E12g and A1g intensity ratio decreases with increase of laser energy and it may happen due to increase of number of layers of WS2, inviting Van-der Waals interactions, compared to monolayers, with increase of laser energy which effects the out of plane vibration and hence effects the A1g peak.

Fig. 4
figure 4

Peak frequency difference between A1g(Γ) & E12g(Γ)with laser energy

Full size image
Fig. 5
figure 5

Evolution of FWHM for A1g(Γ) & E12g(Γ) with laser energy

Full size image

Figure 5 shows the variation of FWHM of E12g(Γ) and A1g (Γ) for all the laser energies. It is observed that FWHM increases with laser energy indicating decrease of crystallinity, increase of disorder & strain of the synthesized product. The FWHM for E12g peak increases from 3.97 to 4.53 to 5.22 for laser energies 10 mJ to 20 mJ to 40 mJ, respectively. The FWHM for A1g peak increases from 3.68 to 4.13 to 4.63 for laser energies 10 mJ to 20 mJ to 40 mJ, respectively. The peak at 271.6 cm− 1 for 10 mJ laser energy may correspond to WO3 or 1T WS2 and the peak at 264.6 cm− 1 for 40 mJ laser energy may be for 1T WS2. Presence of 1T WS2 peak can be highly beneficial for using the synthesized product, as electrocatalyst for Hydrogen evolution reaction (HER) applications.

Figure 6 (a), (b) and (c) represent the TEM images of the QDs samples synthesized at three different laser energy. The figures clearly indicate that the size homogeneity of synthesized QDs are better for 10 mJ and 20 mJ, as compared to that of 40 mJ laser energy. The figure also indicates that the synthesized QDs are not spherical in shape but have irregular or elliptical shape.

Fig. 6
figure 6

FETEM images of WS2 QDs synthesized with laser energies of (a) 10 mJ, (b) 20 mJ and (c) 40 mJ

Full size image

The size distribution for QDs synthesized at 10 mJ, 20 mJ and 40 mJ laser energies are presented by Fig. 7. Figure 8 shows the variation of average particles size with the laser energy. It is observed that the average size of the QDs increases with increase in laser energy. Moreover, from Fig. 8, it is evident that the size distribution is narrowest for 20 mJ and broadest for 40 mJ. QDs formed by 10 mJ laser energy have the intermediate size distribution. From the TEM image analysis, it is observed that, with variation of laser energy, synthesized WS2 QDs have average size variation, from about 3.56 nm for 10 mJ to 4.57 nm for 20 mJ and 6.10 nm for 40 mJ laser energy.

Fig. 7
figure 7

Particle size distribution of WS2 QDs synthesized with laser energies of 10 mJ, 20 mJ and 40 mJ

Full size image
Fig. 8
figure 8

Evolution of synthesized WS2 QDs particle size and its distribution with laser energy

Full size image

Figure 9 shows the PL spectra of the synthesized WS2 QDs at the laser energies of 10 mJ, 20 mJ and 40 mJ. It is observed that the luminescence of QDs, formed by all the laser energies is quiet strong, as is also expected due to quantum confinement and increase in band gap due to size reduction. All the PL spectra are broaden in the range 300 nm to 500 nm because of the size distribution of the QDs samples.

Fig. 9
figure 9

Evolution of PL spectrum with laser energy

Full size image

Moreover, after performing peak deconvolution of PL spectrum for laser energies, 10 mJ, 20 mJ and 40 mJ, it is observed that they have multiple peaks, indicating size distribution or variation of WS2 QDs size, synthesized at particular laser energy. The deconvoluted PL spectra of the QDs samples for 10 mJ, 20 mJ and 40 mJ laser energies are presented in Fig. 10.

Fig. 10
figure 10

Deconvoluted PL spectrum for WS2 QDs synthesized by (a) 10 mJ, (b) 20 mJ and (c) 40 mJ laser energy

Full size image

Among the multiple peaks in the deconvoluted PL spectra, the PL peak position of having maximum area intensity of the 10 mJ, 20 mJ and 40 mJ QDs samples are 369.0 nm (3.36 eV), 396.8 nm (3.12 eV) and 400.4 nm (3.09 eV), respectively. Figure 11 shows the variation of those PL peaks position with laser energy, indicating the red shift of PL peaks with increasing laser energy. It is observed that the PL peaks get shifted towards longer wavelength with increasing laser energy, indicating increase in size of WS2 QD. This red shift is also expected because with the increase in the size of WS2 QDs, the band gap gets narrower and hence low frequency or high wavelength fluorescence is expected. However, it is observed that the red shift for 10 mJ QDs to 20 mJ QDs are more intense in comparison to red shift observed from 20 mJ QDs to 40 mJ QDs.

Fig. 11
figure 11

Red shift of PL peak position with laser energy

Full size image

In order to better understand the influence of laser energy on ablation of WS2 and formation of QDs and its size, the understanding of laser- matter interaction is required and how the reflectivity and absorption coefficient of the surface to be ablated changes with laser fluence is to be understood for better control on property controlled synthesis. A detailed discussion on this issue, in regard of WS2 QDs is made available by Wang et al. [29] and in general, for PLAL method by Fazio et al. [27]. Investigation on size modulation of MoS2 QDs by varying laser energy in PLAL was carried out by Sunitha et al. and they have reported similar results [32], as is obtained in the present study. The WS2 QD can be used for electrochemical detection [33], fluorescent probes [34], White light-emitting diodes [35], and sensitive fluorescent sensor [36].

3 Conclusion

The present study indicates that WS2 QDs can be efficiently fabricated by the PLAL technique. The prominent peaks of 2LA (M), E12g(Γ), and A1g (Γ) vibrational mode were present in the Raman spectra. The E12g(Γ) and A1g (Γ) peak difference increases and the peak intensity ratio decreases with laser energy, indicating increase in size of QDs. The FWHMs for Raman peaks increase with laser energy, indicating an increase in disorder and strain with laser energy. The FETEM images give an idea about the size and shape of synthesized QDs. The size distribution of QDs, as obtained from FETEM image, clearly indicates that the mean size of synthesized QDs increases with laser energy. The multiple peaks in PL spectra indicates the size distribution of QDs for particular laser energy. The red shift of PL peaks indicates the narrowing of band gap and increase in size of QDs with laser energy. The present study indicates that the size modulation of WS2 QDs is feasible by variation of laser energy or fluence in PLAL method. As the sizes of the QDs are directly related to their band gap, band gap engineering and hence tailoring of electronic and optoelectronic properties and responses of WS2 QDs can be easily achieved by PLAL method. Moreover, the crystallinity, size homogeneity, reduction of defect, purity and other desired properties of WS2 QDs may be enhanced if other synthesis parameters of PLAL method can be controlled, varied and optimized. The major takeaways from this paper is the findings, that variation of synthesis parameters of PLAL can influence the size of the WS2 QDs. Moreover, from this finding, it is clear that many more synthesis parameters of PLAL, not explored here, can be varied, controlled, and optimized to influence and enhance the desired morphology and properties of WS2 QDs. The synthesized QDs can further be investigated for different targeted applications. Further investigation in this regard may make the PLAL method more precise, controllable, and scalable.