Skip to main content
PMC home page
Scientific Reports logoLink to Scientific Reports
. 2015 Dec 14;5:17482. doi: 10.1038/srep17482

Yb- and Er-doped fiber laser Q-switched with an optically uniform, broadband WS2 saturable absorber

M Zhang 1,a,*, Guohua Hu 2,*, Guoqing Hu 1, R C T Howe 2, L Chen 3,b, Z Zheng 1,4, T Hasan 2
PMCID: PMC4677313  PMID: 26657601

Abstract

We demonstrate a ytterbium (Yb) and an erbium (Er)-doped fiber laser Q-switched by a solution processed, optically uniform, few-layer tungsten disulfide saturable absorber (WS2-SA). Nonlinear optical absorption of the WS2-SA in the sub-bandgap region, attributed to the edge-induced states, is characterized by 3.1% and 4.9% modulation depths with 1.38 and 3.83 MW/cm2 saturation intensities at 1030 and 1558 nm, respectively. By integrating the optically uniform WS2-SA in the Yb- and Er-doped laser cavities, we obtain self-starting Q-switched pulses with microsecond duration and kilohertz repetition rates at 1030 and 1558 nm. Our work demonstrates broadband sub-bandgap saturable absorption of a single, solution processed WS2-SA, providing new potential efficacy for WS2 in ultrafast photonic applications.

Two dimensional (2d) nanomaterials such as mono- or few-layer graphene, semiconducting transition metal dichalcogenides (s-TMDs) and black phosphorus exhibit high third-order optical nonlinear susceptibility and ultrafast carrier dynamics, making them attractive for nonlinear photonics and optoelectronics1,2,3. Amongst the 2d materials, s-TMDs are of particular research interest due to their diversity and the distinct yet complementary properties to graphene they offer. TMDs, a family of ~40 different layered materials, have a general formula MX2, where M is a transition metal atom (e.g. Mo, W or Nb) and X is a chalcogen atom (e.g. S, Se or Te). Each TMD layer consists of a single plane of M atoms held between two planes of X atoms by strong covalent bonds. Depending on the coordination and oxidation states of the M atoms, TMDs may behave as metallic, semiconducting or insulating. Similar to other layered materials (LMs), the individual layers in TMD bulk crystals are stacked together by relatively weak van der Waals forces, allowing their exfoliation into single and few layer forms. The optoelectronic properties of s-TMDs are strongly thickness-dependent. For example, the bandgap of s-TMDs typically shifts from indirect for bulk material to direct for monolayer flakes1,4. Collectively, the bandgaps of s-TMDs span the visible and near-infrared spectrum1. Such layer-dependent characteristics make s-TMDs comparable or even superior to the zero-gap graphene for a variety of (opto)electronic and photonic applications1. s-TMDs also offer the possibility of engineering their optical properties for desirable performances5. We note that while black phosphorus has recently attracted a strong interest3,6,7, poor material stability, even for over a few days, remains a significant drawback3.

s-TMDs have been shown to possess remarkable optical and optoelectronic properties, including high optical nonlinear susceptibility8,9, ultrafast carrier dynamics10 and broadband working wavelength range11,12, in addition to robustness and environmental stability. This has led to the demonstration of numerous nonlinear optical phenomena using s-TMDs, including saturable absorption11,12 (i.e. reduced optical absorption with increased intensity of incident light13) and optical parametric processes (i.e. second8,14 and third9 harmonic generation), suggesting that these materials could be a suitable platform for the development of photonic devices. One such potential application exploiting the saturable absorption property is in the generation of short pulses by mode-locking or Q-switching in laser cavities, where saturable absorber (SA) devices act as a passive optical switch to modulate the intra-cavity loss13. Such short-pulsed or ‘ultrafast’ lasers have become an indispensable tool, playing an increasingly important role in a wide range of applications, including biomedical imaging and therapy, materials processing, fundamental research and military15. Unlike graphene, which has a linear dispersion of Dirac electrons enabling broadband saturable absorption, s-TMDs typically have a (bulk and monolayer) bandgap ranging between ~1–2 eV1. Some recent progress has been made in measuring the saturable absorption of s-TMDs in the visible region16,17, suggesting their potential as a SA device to produce short pulses in this spectral range.

Similar nonlinear optical absorption and fast carrier dynamics in s-TMDs have also been reported in the near-infrared region11,18,19,20. In particular, a number of studies have reported the generation of short pulses from fiber lasers operating at ~1.018,19,20, 1.5511,18,20,21,22 and 1.9 μm18,20 using several sulfide and selenide-based s-TMDs, including molybdenum disulfide (MoS2) and molybdenum diselenide (MoSe2). More recently, tungsten disulfide (WS2) has also been reported to achieve mode-locking and Q-switching23,24,25,26,27,28, but the wideband operation of a single composite is yet to be demonstrated. While the operating wavelengths of these lasers correspond to photon energies below either the bulk or monolayer bandgap of these s-TMDs, the processes behind this sub-bandgap absorption are not fully understood. Therefore, it is of significant importance to investigate the governing physical mechanism of saturable absorption of WS2 in the near-infrared region and its applicability as a wideband SA material.

Mono- or few-layer WS2 can be produced through a variety of methods, with mechanical cleavage, chemical vapor deposition (CVD) and solution processing techniques such as ultrasonic assisted liquid phase exfoliation (UALPE) being the most commonly exploited ones. In particular, UALPE allows mass production of chemically pristine mono- and few-layer WS2 flakes under ambient conditions, without the need for the high temperatures and complex transfer procedures associated with CVD. As with the other LMs, WS2 dispersions produced via UALPE can be printed onto optical components such as quartz substrates, mirrors and fiber facets or blended with polymers to form composites for simple integration into a laser cavity, making this strategy very attractive for a wide range photonic and optoelectronic applications12,15,29.

Here, we fabricate a few-layer WS2 polymer composite SA based on UALPE for short pulse generation. The free-standing SA exhibits high spatial uniformity in nonlinear optical properties. The modulation depth of the WS2-SA is 3.1% at 1030 nm and 4.9% at 1558 nm, respectively. Using a single WS2-SA composite, we demonstrate microsecond-duration Q-switched pulses with kilohertz repetition rates in all-fiber Yb- and Er-doped lasers to underscore its applicability as a broadband SA material.

Results

Sample preparation and characterization

We fabricate the SA as a freestanding polymer composite film using WS2 nanoflakes produced by UALPE of bulk WS2 crystals. The UALPE process for WS2 is similar to that used for other LMs such as graphene30,31 and several other TMDs11,12,32,33,34, consisting of two steps. First, bulk crystals of WS2 are mixed with a suitable solvent and exfoliated via mild ultrasonication. This relies on the formation of microbubbles in the solvent resulting from the high-frequency (usually ~20–60 kHz) pressure variations35. Under appropriate frequency, pressure and solvent conditions, these microbubbles become unstable as they grow in size and eventually collapse35. The shockwaves produced from the collapsed microbubbles create strong shear forces that are sufficient to overcome the weak van der Waals forces between the layers in bulk LM crystals, resulting in exfoliation of thinner flakes. The second step involves removal of the unexfoliated, thicker flakes. This is typically achieved via centrifugation or filtration31,36. Thus, UALPE and subsequent processing of WS2 bulk crystals produce exfoliated mono-, bi- and few-layer WS2 flakes in suitable solvents.

The stability of the exfoliated WS2 (and other LMs) dispersions is dependent on minimizing the enthalpy of mixing, ΔH36,37. It has been demonstrated that ‘good’ solvents for exfoliation have dispersive (Inline graphic), polar (Inline graphic) and hydrogen bonding (Inline graphic) Hansen solubility parameters38 that match those empirically derived for WS2 (Inline graphic~18 MPa1/2, Inline graphic~8 MPa1/2, Inline graphic~7.5 MPa1/2)39. However, the best-suited solvents (e.g. N-cyclohexyl-2-pyrrolidone [CHP], cyclohexanone and N-methyl-2-pyrrolidone [NMP]) are challenging for device/composite processing due to their relatively high boiling points (NMP ~202 °C, cyclohexanone ~155 °C, CHP~284 °C). UALPE in pure lower boiling point solvents such as alcohols and water does not typically yield stable dispersions due to relatively large mismatch between Inline graphic, Inline graphic and Inline graphic of these solvents to those of WS2. In this case, surfactants can be used to stabilize the dispersion. Such stable dispersions are necessary for the fabrication of optically homogeneous composite SAs, as the slow drying process would otherwise allow aggregates to form15, leading to scattering losses and unreliable device performance15,40. This approach has therefore been widely used to fabricate 1d and 2d material-based SAs20,29,30,31,41.

It has been experimentally shown with graphene that quasi-2d surfactants with hydrophobic and hydrophilic faces (e.g. sodium deoxycholate [SDC], a bile salt surfactant) are well-suited to exfoliating and stabilizing hydrophobic 2d materials (such as WS2) in water31. For stable dispersions, the surfactant should be present at a concentration in excess of the temperature dependent critical micelle concentration (CMC), defined as the surfactant concentration above which its molecules can spontaneously self-arrange to form micelles in water42. Therefore, if exfoliated hydrophobic flakes (such as WS2) are present in a water-surfactant solution above the corresponding CMC value, the surfactant molecules are expected to encapsulate and stabilize the flakes, preventing their reaggregation and sedimentation31,37,43,44.

We prepare the stable few-layer WS2 dispersion by mixing 100 mg WS2 crystals with ~70 mg SDC surfactant in 10 mL of DI water and sonicating for 12 hours in a bath sonicator at ~15 °C. We select SDC, a di-hydroxy bile salt surfactant over the more commonly used sodium cholate (SC), a tri-hydroxy bile salt due to its higher hydrophobic index, which should allow stronger interaction between the surfactant and WS231. The CMC of SDC in water at room temperature is found to be 4.7 mM (~2 g.L−1)20 by pendant droplet measurement of the surface tension of different concentration SDC solutions (the surface tension of solutions changes rapidly below the CMC, but stabilizes above). Thus the surfactant concentration we use here is ~3.5 times the CMC, sufficient to support the exfoliated WS231,45,46. The sonicated dispersion, containing a mixture of exfoliated and unexfoliated materials, is centrifuged at ~1500 g for one hour. The top 70% of the dispersion, enriched with mono-, bi- and few layer flakes, is then decanted for characterization and SA fabrication.

Dispersion characterization

The optical absorption spectrum of the WS2 dispersion, diluted to 10% v/v to reduce scattering effects47, is shown in Fig. 1(a), with the inset of a photograph of the cuvette containing the dispersion. The spectrum shows the characteristic WS2 excitonic peaks, referred to as A (at ~630 nm) and B (at ~520 nm) according to common nomenclature48. The spectrum can be used to estimate the concentration of WS2 using Beer-Lambert law (Inline graphic), where c is the WS2 concentration (gL−1), l is the distance the light passes through the dispersion (m) and Inline graphic and Inline graphic are the absorption (a.u.) and material dependent optical absorption coefficient (Lg−1m−1) at wavelength λ (nm), respectively. Using a combination of optical absorption spectroscopy and thermogravimetric analysis on a set of dispersions, as we have previously reported for MoS211, we estimate the value of Inline graphic for WS2 ~ 1324 Lg−1m−1. From this, we estimate the concentration of dispersed WS2 to be 0.23 gL−1.

Figure 1.

Figure 1

Open in a new tab

(a) Linear optical absorption of WS2 flakes in a 10% v/v dispersion. Inset: Stable WS2 dispersion (diluted to 10% v/v) in a cuvette: (b) AFM image of a typical WS2 flake; (c) height variations of the flakes along the marked line; (d) AFM distribution of flake thickness; (e) AFM distribution of lateral WS2 flake dimensions.

The distribution of flake thicknesses and lateral dimensions is measured via atomic force microscopy (AFM). Samples are prepared by drop-casting a diluted (5% v/v) WS2 dispersion onto a Si/SiO2 wafer. The wafers are then rinsed with DI water to remove residual surfactant, giving clean and isolated flakes on the substrates. A typical flake is shown in Fig. 1(b). The corresponding height profile, presented in Fig. 1(c), shows a thickness of ~5 nm. The average thickness [Fig. 1(d)], measured across ~400 individual flakes, is (5.0 ± 0.1) nm. We find ~73% of the flakes are Inline graphic5 nm thick, corresponding to <8 layers, assuming ~1 nm for a monolayer flake and 0.6 nm for each subsequent layer49. As shown in Fig. 1(e), the average lateral dimension of the flakes is (62 ± 1) nm.

Composite characterization

To allow integration of the SA into the fiber laser cavity, a free-standing composite film is prepared from the WS2 dispersion using an approach previously demonstrated for other 1d and 2d materials11,12,15,20,29,30,31,32,41,53. The composite is prepared by homogeneously mixing the WS2 dispersion with a 5 wt% aqueous solution of polyvinyl alcohol (PVA) polymer. PVA is used as the host polymer because it does not exhibit strong optical absorption at 1030 and 1558 nm. In addition, it is selected for its solvent compatibility (PVA is water soluble), ease of processability (can be processed at room temperature), robustness and flexibility (dried thin films have high tensile strength and are not brittle). The mixture is poured into a Petri dish and allowed to dry at room temperature in a desiccator, producing a ~30 μm thick, free-standing WS2-polymer composite SA [Fig. 2(a) inset].

Figure 2.

Figure 2

Open in a new tab

(a) Optical micrograpgh of the free-standing WS2-PVA SA composite, confirming absence of aggregates. Inset, photograph of WS2-SA (b). Optical absorbance of a pure PVA and WS2-PVA SA (c). Schematic showing integration of WS2-PVA SA device between two fiber patch cords.

For transmissive type thin film SA devices, optical uniformity without defects (such as cracks, voids, material aggregates etc.) is an important consideration to ensure reliable and repeatable device performance. The uniformity and quality of the WS2-SA is assessed by using optical microscopy, a commonly used technique employed for other nanomaterial based SAs15,20. The optical micrograph, shown in Fig. 2(a), confirms the absence of >1 μm aggregates in the SA composite, which would otherwise have led to undesirable nonsaturable scattering losses47.

Figure 2(b) shows the optical absorption of the composite film. Measurement of a pure PVA film prepared by the same method is also included in Fig. 2(b) as a reference. We note that the WS2-SA composite shows non-zero absorption, even at energies below bandgap of either bulk (~1.3–1.4 eV, 954–886 nm)1,50,51 or monolayer (~2.1 eV, 590 nm 1,51) WS2. Considering contributions from PVA (0.011–0.015 absorbance, 2.5–3.4%) [Fig. 2(b)] the WS2-PVA composite shows a marked increase in absorption (0.046–0.039 absorbance, 10.05%–8.6%), indicating strong optical absorption from the embedded WS2 flakes at both 1030 (1.2 eV) and 1558 nm (0.8 eV). We have recently attributed this sub-bandgap light absorption in other s-TMD materials (MoS211,12,52, MoSe2 20) to edge-induced sub-bandgap states11,12,20,32. We propose that similar edge-induced states arise within the material bandgap in our UALPE WS2 flakes, leading to the observed sub-bandgap light absorption. We note that the contribution from scattering in the sub-bandgap region in the composite is minimal, only ~1% absorption across the operation wavelength (i.e. ~10% of the measured 0.046-0.039 absorbance). This was confirmed by measurements of the WS2-SA composites using an absorption spectrometer with integrating sphere53.

The nonlinear optical absorption of the WS2-SA is characterized using an open-aperture Z-scan technique. For this, an ultrashort fiber source operating at 1030 nm is used as the pump light (120 fs pulse duration, 20 MHz pulse repetition rate), split using a 90%:10% fused fiber coupler, the latter enabling monitoring of power used as a reference. The SA composite is swept through the focus of a beam of the remaining port and the transmitted power is recorded as a function of incident intensity on the device. A typical dataset from a single Z-scan measurement, at a fixed transverse position on the WS2-SA, shown in Fig. 3(a), can be well-fitted with the two-level SA model54. From the fit, the following SA parameters are extracted: at 1030 nm, the saturation intensity, Isat,1030 ~ 1.38 MW/cm2, modulation depth, Inline graphic ~ 3.1% and nonsaturable absorption, Inline graphic~6.9%. The same measurement is also carried out at the wavelength of 1558 nm (150 fs pulse duration, 10 MHz pulse repetition rate). The measured parameters are: Isat,1558 ~ 3.83 MW/cm2, Inline graphic ~ 4.9%, and Inline graphic ~ 3.7%, respectively [Fig 3(d)]. Thus, the WS2-SA shows strong saturable absorption in both these wavelengths. The damage threshold of the sample is estimated to be ~1.2 GW/cm2.

Figure 3. Nonlinear optical absorption at (a–c) 1030 nm and (d–f) 1558 nm measured by an open aperture Z-scan technique.

Figure 3

Open in a new tab

(a), (d) Typical datasets from Z-scan experiment. Mapping of the homogeneity of the sample used in the experiment: variation in modulation depth (b), (e), colorbar scale: ~1 –6%; and saturation intensity (c), (f), colorbar scale: 1 ~ –Inline graphic.

The uniformity of the WS2 flakes embedded in the free-standing WS2-PVA composite is next evaluated by raster scanning the sample. This is essential to ensure reliable and repeatable performances of the WS2-SA devices. For this, we measure the nonlinear saturation of the ~30 μm SA with 0.5 mm spatial increments in the XY plane, across a 2 × 2 mm section, at both wavelengths. The extracted data, processed to display the variation in Inline graphic and Inline graphic [Fig. 3(b,e)] and Isat,1030, Isat,1558 [Fig. 3(c,f)] are presented on two-dimensional grids. The spatial increment, i.e. the resolution in the scanning, is limited by the laser spot size during the measurement. Figure 3(b,c,e,f) show clear evidence of optical homogeneity (with standard deviations of 0.13%, 0.17 MW/cm2, 0.13%, 0.11 MW/cm2, respectively), allowing us to obtain repeatable performance across the WS2-SA sample.

Demonstration of Q-switching a fiber laser using few-layer WS2-SA

The demonstrated sub-bandgap saturable absorption of the few-layer WS2-PVA composite at 1030 and 1558 nm indicates that the device could be used to modulate the loss and Q-factor of a fiber laser cavity. This could in turn be exploited to generate a regular train of Q-switched pulses in this spectral region. To explore the potential of using a single WS2-SA for short-pulse generation at different wavelengths, fully fiber-integrated Yb- and Er-doped lasers are constructed. For each laser cavity, a ring configuration is adopted, consisting of entirely isotropic, single-mode fiber. The Yb and Er fiber amplifiers consist of single-mode Yb- and Er-doped active fiber, respectively, co-pumped by a 974 nm pump diode. In addition to the fiber amplifier, each cavity includes a polarization-independent optical isolator to ensure unidirectional propagation, fused-fiber output coupler for both spectral and temporal diagnostics and polarization controller to adjust the net cavity birefringence. The WS2-SA is integrated into the cavities by sandwiching a ~1 mm × 1 mm of the composite between two fiber patch chords, shown in Fig. 2(c). The total cavity length for the Yb- and Er-doped lasers is 66 m and 9 m, respectively.

Q-switched Yb-doped fiber laser characterization

Self-starting Q-switching is obtained from the all-fiber integrated Yb-doped laser [Fig. 4(a)], generating a stable train of pulses, centered at 1030 nm [Fig. 4(e)]. Typical output characteristics of the laser, at 0.5 mW average output power, are shown in Fig. 4. Pulses are generated with 27.2 μs spacing, corresponding to 36.7 kHz repetition rate [Fig. 4(b)] and a full width at half maximum (FWHM) pulse duration of 3.2 μs [Fig. 4(c)]. The radio frequency (RF) spectrum of the output shows a high signal-to-background contrast of 53 dB [Fig. 4(d)], indicating good pulse train stability, comparable to Q-switched fiber lasers based on other 2d layered materials12,55,56.

Figure 4.

Figure 4

Open in a new tab

Q-switched Yb-doped fiber laser: (a) cavity configuration; (b) output pulse train, with a spacing of 27.2 μs; (c) single pulse profile, with 3.2 μs FWHM pulse width; (d) radio frequency spectrum of fundamental frequency on a 5 kHz span, where Inline graphic, with the green trace showing the noise floor of the RF analyzer; (e) measured optical spectrum.

The pulse properties in continuous-wave pumped Q-switched lasers rely on nonlinear dynamics in the gain medium and SA. This leads to a dependence of cavity repetition rate and pulse duration on pump power57. A pulse is emitted once the storage energy of the cavity reaches a certain threshold. Therefore, a greater pump power enables higher repetition rates and results in shorter pulses. This is experimentally observed by changing the pump power as the pulse duration is reduced from 6.4 μs to 3.2 μs and the cavity repetition rate is increased from 24.9 kHz to 36.7 kHz [Fig. 5(a)]. The maximum pulse energy is 13.6 nJ, limited by the available pump power. We believe that higher pulse energies could be achieved by further optimizing the laser cavity. The pulse duration could also be further shortened by reducing the length of the laser cavity.

Figure 5.

Figure 5

Open in a new tab

Variation of the pulse duration and repetition rate with average output power for Q-switched (a) Yb-, (b) Er-doped fiber lasers using few-layer WS2-PVA composite SA.

Q-switched Er-doped fiber laser characterization

To demonstrate the sub-bandgap saturable absorption for short pulse generation at a longer wavelength, the same SA is subsequently incorporated into the Er-doped fiber laser described above. Self-starting Q-switching operation is achieved at 1558 nm with an average output power of 6.4 mW. With increasing pump power, the repetition rate is tuned from 79 kHz to 97 kHz, with corresponding pulse duration from 3.4 μs to 1.1 μs as the average output power is increased from 6.4 mW to 16.4 mW [Fig. 5(b)]. The pulse energy is 179.6 nJ at the highest output power. Typical output pulse train properties are shown in Fig. 6(b). The cavity repetition rate is 97.1 kHz, corresponding to a pulse-to-pulse spacing of 10.3 μs. Figure 6(c) shows the single pulse profile, with a FWHM pulse duration of 1.3 μs. The signal-to-noise background contrast of RF spectrum, plotted in Fig. 6(d) is 44 dB, again showing the good pulse train stability of the Er-laser.

Figure 6.

Figure 6

Open in a new tab

Q-switched Er-doped fiber laser: (a) cavity configuration; (b) output pulse train, with a spacing of 10.3 μs; (c) single pulse profile, with 1.3 μs FWHM pulse width; (d) radio frequency spectrum of fundamental frequency on a 20 kHz span, where Inline graphic, with the green trace showing the noise floor of the RF analyzer; (e) measured optical spectrum.

Discussion

Following stable Q-switching results obtained by using the WS2-SA composite at 1030 and 1558 nm, the same experiment is conducted with a ~30 μm thick pure PVA film (fabrication process similar to that used for WS2-SA fabrication, but without the WS2 flakes). No Q-switching is observed at any power level or polarization controller position, confirming that the sub-bandgap saturable absorption arises from the few-layer WS2 flakes.

As discussed before, the WS2-SA exhibits non-zero sub-bandgap absorption at energies below the bulk (~1.3–1.4 eV, 954–886 nm1,50,51) or monolayer (~2.1 eV, 590 nm 1,51) WS2 bandgap. The sub-bandgap absorption and its nonlinear behavior is also verified by Z-scan measurement and demonstration of Q-switched lasers at 1030 (1.2 eV) and 1558 nm (0.8 eV). Similar observation of sub-bandgap saturable absorption in other s-TMDs have also been reported11,18,19,20. For the case of MoS2 and MoSe2, we recently proposed that the sub-bandgap absorption arises from edge-induced sub-bandgap states11,12,20. Our proposal is based on previous experimental observations of increased sub-bandgap absorption in lithographically patterned MoS2 compared to their large crystals due to increased edge to surface area ratio52. From the AFM data and considering approximately square flakes for ease of calculation, we estimate a high 1:6 ratio between the edge (=4 × d × t where d is the lateral dimension and t is the thickness of the flakes) and surface area (=2 Inline graphic) of the WS2 flakes. We thus suggest that the observed sub-bandgap absorption in the WS2-SA is also due to the edge-states, promoted by the high edge to surface area ratio of the UALPE WS2 flakes. This edge-induced sub-bandgap absorption can be saturated at high incident intensities by Pauli blocking, which enables WS2 to act as an SA material in the near-infrared region. Additionally, a distribution of edge-induced states within the bandgap could explain the wideband saturable absorption experimentally observed here, and in recent reports at a number of different laser wavelengths in other s-TMDs11,12,18,19,20.

We note that Wang et al. have also proposed a complementary explanation for this phenomenon based on crystallographic defect states, supported by theoretical bandgap studies by varying the ratio of M and S atoms58. Zhang et al.59 reported nonlinear optical measurements of WS2 and MoS2, grown by high temperature, direct vapor phase sulfurization of pre-deposited metal films. Both materials showed layer dependent saturable and reverse saturable absorption (RSA, due to two-photon absorption) properties. Additionally, the authors demonstrated RSA for 1–3, 18–20, 39–41 layered samples at 1030 nm while SA for 18–20 layers and RSA for 1–3 layer WS2 at 800 nm. We do not observe any such RSA behavior in our samples (<8 layers) during the nonlinear measurements (maximum peak intensity, 420 MW/cm2) and Q-switching operation. Another recent report attributes exciton-exciton interaction in bulk and monolayer MoSe2 (a selenium based s-TMD) for saturable absorption at excitonic resonance60. However, this does not explain the saturable sub-bandgap absorption we observe in our experiments here. A distribution of edge-induced states within the bandgap could better explain the wideband saturable absorption, supported by recent reports on mode-locking or Q-switching at a number of different laser wavelengths in other s-TMDs20,12. We note that grain boundaries and other defects present in s-TMD crystals, such as those grown by CVD may also strongly contribute to such sub-bandgap absorption. Indeed, this may explain ultrafast pulse generation using CVD-grown s-TMDs below the bulk and monolayer bandgap61,62,63,64. Considering the above discussion, we propose that the large edge-to-surface ratio of nanoflakes of WS2, in particular, prepared by UALPE, is likely to be the primary origin of the optical absorption below the fundamental material bandgap. We stress that further theoretical and experimental studies are required to confirm our hypothesis.

In summary, a free-standing few-layer WS2-PVA SA has been fabricated by UALPE of chemically pristine WS2. Using this SA, we have developed self-starting Q-switched Yb- and Er-doped fiber lasers for short pulse generation. We have proposed edge-induced sub-bandgap states in WS2 as the primary reason for broadband saturable absorption in the near IR spectral region. This extends our existing understanding for this phenomenon to a wider class of s-TMDs with regard to their potential for future photonic technologies.

Additional Information

How to cite this article: Zhang, M. et al. Yb- and Er-doped fiber laser Q-switched with an optically uniform, broadband WS2 saturable absorber. Sci. Rep. 5, 17482; doi: 10.1038/srep17482 (2015).

Acknowledgments

The authors thank E.J.R. Kelleher and R.I. Woodward for valuable discussions. M.Z. acknowledges support from Beihang University, China, through a Zhuoyue Bairen Program and T.H. from the Royal Academy of Engineering through a fellowship (Graphlex). This work at Beihang University was supported by 973 Program (2012CB315601), NSFC (61221061/61435002) and the Fundamental Research Funds for the Central Universities.

Footnotes

Author Contributions M.Z., L.C. and T.H. conceived the experiments, Guohua Hu,Guoqing Hu and R.C.T.H. conducted the experiments, M.Z., GuoHua Hu, R.C.T.H. and T.H. analyzed the results, M.Z., L.C., Z.Z. and T.H. wrote the manuscript. All authors reviewed the manuscript.

References

  1. Wang Q. H., Kalantar-Zadeh K., Kis A., Coleman J. N. & Strano M. S. Electronics and optoelectronics of two-dimensional transition metal dichalcogenides. Nat. Nanotech. 7, 699–712 (2012). [DOI] [PubMed] [Google Scholar]
  2. Bonaccorso F., Sun Z., Hasan T. & Ferrari A. C. Graphene photonics and optoelectronics. Nat. Photon. 4, 611–622 (2010). [Google Scholar]
  3. Hanlon D., Backes C., Doherty, E., Cucinotta C. S., Berner N. C., Boland C., Lee, K., Harvey A., Lynch P., Gholamvand Z., et al.: Liquid exfoliation of solvent-stabilized few-layer black phosphorus for applications beyond electronics. Nat. Commun. 6, 8563 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Yun W. S., Han S. W., Hong S. C., Kim I. G. & Lee J. D. Thickness and strain effects on electronic structures of transition metal dichalcogenides: 2H-MX2 semiconductors (M = Mo, W; X = S, Se, Te). Phys. Rev. B 85, 033305 (2012). [Google Scholar]
  5. Hong X. et al. Ultrafast charge transfer in atomically thin MoS2/WS2 heterostructures. Nat. Nanotech. 9, 682–686 (2014). [DOI] [PubMed] [Google Scholar]
  6. Li D. et al. Ultrafast pulse generation with black phosphorus. arXiv preprint arXiv:1505.00480 (2015).
  7. Lu S. et al. Broadband nonlinear optical response in multi-layer black phosphorus: an emerging infrared and mid-infrared optical material. Opt. Express 23, 11183–11194 (2015). [DOI] [PubMed] [Google Scholar]
  8. Kumar N. et al. Second harmonic microscopy of monolayer MoS2. Phys. Rev. B 87, 161403 (2013). [Google Scholar]
  9. Wang R. et al. Third-harmonic generation in ultrathin films of MoS2. ACS Appl. Mater. Interfaces 6, 314–318 (2013). [DOI] [PubMed] [Google Scholar]
  10. Wang K. et al. Ultrafast saturable absorption of two-dimensional MoS2 nanosheets. ACS Nano 7, 9260–9267 (2013). [DOI] [PubMed] [Google Scholar]
  11. Zhang M. et al. Solution processed MoS2-PVA composite for sub-bandgap mode-locking of a wideband tunable ultrafast Er: fiber laser. Nano Res. 8, 1522–1534 (2015). [Google Scholar]
  12. Woodward R. et al. Few-layer MoS2 saturable absorbers for short-pulse laser technology: current status and future perspectives [invited]. Photon. Res. 3, A30–A42 (2015). [Google Scholar]
  13. Keller U. Recent developments in compact ultrafast lasers. Nature 424, 831–838 (2003). [DOI] [PubMed] [Google Scholar]
  14. Janisch C. et al. Extraordinary second harmonic generation in tungsten disulfide monolayers. Sci. Rep. 4, 1–5 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hasan T. et al. Nanotube-polymer composites for ultrafast photonics. Adv. Mater. 21, 3874–3899 (2009). [Google Scholar]
  16. Wang K. et al. Broadband ultrafast nonlinear absorption and nonlinear refraction of layered molybdenum dichalcogenide semiconductors. Nanoscale 6, 10530–10535 (2014). [DOI] [PubMed] [Google Scholar]
  17. Zhou K.-G. et al. Optical materials: Size-dependent nonlinear optical properties of atomically thin transition metal dichalcogenide nanosheets. Small 11, 634–634 (2015). [DOI] [PubMed] [Google Scholar]
  18. Luo Z. et al. 1-, 1.5-, and 2-μm fiber lasers Q-switched by a broadband few-layer MoS2 saturable absorber. J. Lightwave Technol. 32, 4077–4084 (2014). [Google Scholar]
  19. Du J. et al. Ytterbium-doped fiber laser passively mode locked by few-layer molybdenum disulfide (MoS2) saturable absorber functioned with evanescent field interaction. Sci. Rep. 4, 1–7 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Woodward R. I., Howe R. C. T., Runcorn T. H. Hu G. Torrisi F., Kelleher, E. J. R., Hasan T.: Wideband saturable absorption in few-layer molybdenum diselenide (MoSe2) for Q-switching Yb-, Er- and Tm-doped fiber lasers. Opt. Express. 23, 20051 (2015). [DOI] [PubMed] [Google Scholar]
  21. Luo Z. et al. Nonlinear optical absorption of few-layer molybdenum diselenide (MoSe2) for passively mode-locked soliton fiber laser [invited]. Photon. Res. 3, A79–A86 (2015). [Google Scholar]
  22. Huang Y. et al. Widely-tunable, passively Q-switched erbium-doped fiber laser with few-layer MoS2 saturable absorber. Opt. Express 22, 25258–25266 (2014). [DOI] [PubMed] [Google Scholar]
  23. Mao D. et al. WS2 mode-locked ultrafast fiber laser. Sci. Rep. 5, 1–7 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kassani S. H. et al. All-fiber Er-doped Q-switched laser based on tungsten disulfide saturable absorber. Opt. Mater. Express 5, 373–379 (2015). [Google Scholar]
  25. Wu K., Zhang X., Wang J., Li X. & Chen J. WS2 as a saturable absorber for ultrafast photonic applications of mode-locked and Q-switched lasers. Opt. Express 23, 11453–11461 (2015). [DOI] [PubMed] [Google Scholar]
  26. Yan P. et al. Microfiber-based WS2-film saturable absorber for ultra-fast photonics. Opt. Mater. Express 5, 479–489 (2015). [Google Scholar]
  27. Khazaeinezhad R., Hosseinzadeh Kassani S., Jeong H., Yeom D.-I. & Oh K. Passively mode-locked fiber laser based on CVD WS2. In CLEO: Applications and Technology, JW2A–74 (Optical Society of America, 2015). [Google Scholar]
  28. Jung M. et al. Mode-locked, 1.94 μm, all-fiberized laser using WS2-based evanescent field interaction. Opt. Express 23, 19996–20006 (2015). [DOI] [PubMed] [Google Scholar]
  29. Sun Z., Hasan T. & Ferrari A. C. Ultrafast lasers mode-locked by nanotubes and graphene. Phys. E 44, 1082–1091 (2012). [Google Scholar]
  30. Sun Z. et al. Graphene mode-locked ultrafast laser. ACS Nano 4, 803–810 (2010). [DOI] [PubMed] [Google Scholar]
  31. Hasan T. et al. Solution-phase exfoliation of graphite for ultrafast photonics. Phys. Stat. Sol. (b) 247, 2953–2957 (2010). [Google Scholar]
  32. Woodward R. et al. Tunable Q-switched fiber laser based on saturable edge-state absorption in few-layer molybdenum disulfide (MoS2). Opt. Express 22, 31113–31122 (2014). [DOI] [PubMed] [Google Scholar]
  33. Luo Z.-C. et al. Microfiber-based few-layer black phosphorus saturable absorber for ultra-fast fiber laser. arXiv preprint arXiv:1505.03035 (2015). [DOI] [PubMed]
  34. Hernandez Y. et al. High-yield production of graphene by liquid-phase exfoliation of graphite. Nat. Nanotech 3, 563–568 (2008). [DOI] [PubMed] [Google Scholar]
  35. Mason T. J. & Lorimer J. P. Applied sonochemistry. The uses of power ultrasound in chemistry and processing 1–48 (2002). [Google Scholar]
  36. Coleman J. N. et al. Two-dimensional nanosheets produced by liquid exfoliation of layered materials. Science 331, 568–571 (2011). [DOI] [PubMed] [Google Scholar]
  37. Coleman J. N. Liquid-phase exfoliation of nanotubes and graphene. Adv. Funct. Mater. 19, 3680–3695 (2009). [Google Scholar]
  38. Hansen C. M. Hansen solubility parameters: a user’s handbook (CRC press, 2007). [Google Scholar]
  39. Cunningham G. et al. Solvent exfoliation of transition metal dichalcogenides: dispersibility of exfoliated nanosheets varies only weakly between compounds. ACS Nano 6, 3468–3480 (2012). [DOI] [PubMed] [Google Scholar]
  40. Wong W., Pun E. & Chan K. Er3+-Yb3+ codoped polymeric optical waveguide amplifiers. Appl. Phys. Lett. 84, 176 (2004). [Google Scholar]
  41. Martinez A. & Sun Z. Nanotube and graphene saturable absorbers for fibre lasers. Nat. Photon. 7, 842–845 (2013). [Google Scholar]
  42. Butt H.-J., Graf K. & Kappl M. Physics and Chemistry of Interfaces (John Wiley & Sons, 2006). [Google Scholar]
  43. Lotya M. et al. Liquid phase production of graphene by exfoliation of graphite in surfactant/water solutions. J. Am. Chem. Soc. 131, 3611–3620 (2009). [DOI] [PubMed] [Google Scholar]
  44. Smith R. J. et al. Large-scale exfoliation of inorganic layered compounds in aqueous surfactant solutions. Adv. Mater. 23, 3944–3948 (2011). [DOI] [PubMed] [Google Scholar]
  45. Sugihara G. et al. Thermodynamic study on the Langmuir adsorption of various bile salts including taurine and glycine conjugates onto graphite in water. Langmuir 16, 1825–1833 (2000). [Google Scholar]
  46. Miyajima K., Machida K., Taga T., Komatsu H. & Nakagaki M. Correlation between the hydrophobic nature of monosaccharides and cholates, and their hydrophobic indices. J. Chem. Soc. 84, 2537–2544 (1988). [Google Scholar]
  47. Bohren C. F. & Huffman D. R. Absorption and scattering of light by small particles (John Wiley & Sons, 2008). [Google Scholar]
  48. Beal A., Knights J. & Liang W. Transmission spectra of some transition metal dichalcogenides. II. Group VIA: trigonal prismatic coordination. J. Phys. C: Sol. State Phys. 5, 3540 (1972). [Google Scholar]
  49. Gutiérrez H. R. et al. Extraordinary room-temperature photoluminescence in triangular WS2 monolayers. Nano Lett. 13, 3447–3454 (2012). [DOI] [PubMed] [Google Scholar]
  50. Kam K. & Parkinson B. Detailed photocurrent spectroscopy of the semiconducting group VIB transition metal dichalcogenides. J. Phys. Chem. 86, 463–467 (1982). [Google Scholar]
  51. Kuc A., Zibouche N. & Heine T. Influence of quantum confinement on the electronic structure of the transition metal sulfide TS2. Phys. Rev. B 83, 245213 (2011). [Google Scholar]
  52. Roxlo C., Chianelli R., Deckman H., Ruppert A. & Wong P. Bulk and surface optical absorption in molybdenum disulfide. J. Vac. Sci. Technol. A 5, 555–557 (1987). [Google Scholar]
  53. Howe R. C. T., Hu G., Yang Z., & Hasan T. Functional inks of graphene, metal dichalcogenides and black phosphorous for photonics and (opto)electronics. Proc. SPIE Nanoscience+Engineering 9553–26 (2015). [Google Scholar]
  54. Haus H. A. Theory of mode locking with a fast saturable absorber. J. Appl. Phys. 46, 3049–3058 (1975). [Google Scholar]
  55. Liu J., Wu S., Yang Q.-H. & Wang P. Stable nanosecond pulse generation from a graphene-based passively Q-switched Yb-doped fiber laser. Opt. Lett. 36, 4008–4010 (2011). [DOI] [PubMed] [Google Scholar]
  56. Popa D. et al. Graphene Q-switched, tunable fiber laser. Appl. Phys. Lett. 98, 073106 (2011). [Google Scholar]
  57. Degnan J. J. Optimization of passively Q-switched lasers. IEEE J. Quantum. Electron. 31, 1890–1901 (1995). [Google Scholar]
  58. Wang S. et al. Broadband few-layer MoS2 saturable absorbers. Adv. Mater. 26, 3538–3544 (2014). [DOI] [PubMed] [Google Scholar]
  59. Zhang S. et al. Direct observation of degenerate two-photon absorption and its saturation of WS2 and MoS2 monolayer and few-layer films. ACS Nano 9, 7142–7150 (2015). [DOI] [PubMed] [Google Scholar]
  60. Kumar N. et al. Exciton-exciton annihilation in MoSe2 monolayers. Phys. Rev. B 89, 125427 (2014). [Google Scholar]
  61. O’Brien M. et al. Transition metal dichalcogenide growth via close proximity precursor supply. Sci. Rep. 4, 1–7 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Zhang X.-Q., Lin C.-H., Tseng Y.-W., Huang K.-H. & Lee Y.-H. Synthesis of lateral heterostructures of semiconducting atomic layers. Nano Lett. 15, 410–415 (2014). [DOI] [PubMed] [Google Scholar]
  63. Lv R. et al. Transition metal dichalcogenides and beyond: Synthesis, properties, and applications of single-and few-layer nanosheets. Acc. Chem. Res. 48, 56–64 (2014). [DOI] [PubMed] [Google Scholar]
  64. Liu K.-K. et al. Growth of large-area and highly crystalline MoS2 thin layers on insulating substrates. Nano Lett. 12, 1538–1544 (2012). [DOI] [PubMed] [Google Scholar]

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group

RESOURCES